From 65ab8385b67854792e89267907f9fcb27e779f95 Mon Sep 17 00:00:00 2001
From: Divy Le Ray <divy@chelsio.com>
Date: Wed, 4 Feb 2009 16:31:39 -0800
Subject: cxgb3: Fix lro switch

The LRO switch is always set to 1 in the rx processing loop.
It breaks the accelerated iSCSI receive traffic.
Fix its computation.

Signed-off-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/cxgb3/sge.c b/drivers/net/cxgb3/sge.c
index 379a132..d31791f 100644
--- a/drivers/net/cxgb3/sge.c
+++ b/drivers/net/cxgb3/sge.c
@@ -2276,8 +2276,7 @@ no_mem:
 		} else if ((len = ntohl(r->len_cq)) != 0) {
 			struct sge_fl *fl;
 
-			if (eth)
-				lro = qs->lro_enabled && is_eth_tcp(rss_hi);
+			lro &= eth && is_eth_tcp(rss_hi);
 
 			fl = (len & F_RSPD_FLQ) ? &qs->fl[1] : &qs->fl[0];
 			if (fl->use_pages) {
-- 
cgit v0.10.2


From 1fbe49328f7442090439addddf441fb5b3186e71 Mon Sep 17 00:00:00 2001
From: Andy Fleming <afleming@freescale.com>
Date: Wed, 4 Feb 2009 16:37:40 -0800
Subject: gianfar: Fix BD_LENGTH_MASK definition

BD_LENGTH_MASK is supposed to catch the low 16-bits of the status field, not
the low byte.  The old way, we would never be able to clean up tx packets with
sizes divisible by 256.

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h
index b1a8334..eaa8689 100644
--- a/drivers/net/gianfar.h
+++ b/drivers/net/gianfar.h
@@ -312,7 +312,7 @@ extern const char gfar_driver_version[];
 #define ATTRELI_EI(x) (x)
 
 #define BD_LFLAG(flags) ((flags) << 16)
-#define BD_LENGTH_MASK		0x00ff
+#define BD_LENGTH_MASK		0x0000ffff
 
 /* TxBD status field bits */
 #define TXBD_READY		0x8000
-- 
cgit v0.10.2


From b98ac702f49042ab0c382b839465b95a2bd0cd65 Mon Sep 17 00:00:00 2001
From: Andy Fleming <afleming@freescale.com>
Date: Wed, 4 Feb 2009 16:38:05 -0800
Subject: gianfar: Fix potential soft reset race

SOFT_RESET must be asserted for at least 3 TX clocks in order for it to work
properly.  The syncs in the gfar_write() commands have been hiding this, but
we need to guarantee it.

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 3f7eab4..acae2d8 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -351,6 +351,9 @@ static int gfar_probe(struct of_device *ofdev,
 	/* Reset MAC layer */
 	gfar_write(&priv->regs->maccfg1, MACCFG1_SOFT_RESET);
 
+	/* We need to delay at least 3 TX clocks */
+	udelay(2);
+
 	tempval = (MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
 	gfar_write(&priv->regs->maccfg1, tempval);
 
-- 
cgit v0.10.2