Merge remote-tracking branch 'linus/master' into merge

Conflicts:
	Documentation/hwmon/ina2xx
	arch/powerpc/Kconfig
	arch/powerpc/boot/dts/b4860emu.dts
	arch/powerpc/boot/dts/b4qds.dtsi
	arch/powerpc/boot/dts/fsl/b4si-post.dtsi
	arch/powerpc/boot/dts/fsl/qoriq-sec6.0-0.dtsi
	arch/powerpc/boot/dts/p1023rdb.dts
	arch/powerpc/boot/dts/t4240emu.dts
	arch/powerpc/boot/dts/t4240qds.dts
	arch/powerpc/configs/85xx/p1023_defconfig
	arch/powerpc/configs/corenet32_smp_defconfig
	arch/powerpc/configs/corenet64_smp_defconfig
	arch/powerpc/configs/mpc85xx_smp_defconfig
	arch/powerpc/include/asm/cputable.h
	arch/powerpc/include/asm/device.h
	arch/powerpc/include/asm/epapr_hcalls.h
	arch/powerpc/include/asm/kvm_host.h
	arch/powerpc/include/asm/mpic.h
	arch/powerpc/include/asm/pci.h
	arch/powerpc/include/asm/ppc-opcode.h
	arch/powerpc/include/asm/ppc_asm.h
	arch/powerpc/include/asm/reg_booke.h
	arch/powerpc/kernel/epapr_paravirt.c
	arch/powerpc/kernel/process.c
	arch/powerpc/kernel/prom.c
	arch/powerpc/kernel/setup-common.c
	arch/powerpc/kernel/setup_32.c
	arch/powerpc/kernel/setup_64.c
	arch/powerpc/kernel/smp.c
	arch/powerpc/kernel/swsusp_asm64.S
	arch/powerpc/kernel/swsusp_booke.S
	arch/powerpc/kvm/book3s_pr.c
	arch/powerpc/kvm/booke.c
	arch/powerpc/kvm/booke.h
	arch/powerpc/kvm/e500.c
	arch/powerpc/kvm/e500.h
	arch/powerpc/kvm/e500_emulate.c
	arch/powerpc/kvm/e500mc.c
	arch/powerpc/kvm/powerpc.c
	arch/powerpc/perf/e6500-pmu.c
	arch/powerpc/platforms/85xx/Kconfig
	arch/powerpc/platforms/85xx/Makefile
	arch/powerpc/platforms/85xx/b4_qds.c
	arch/powerpc/platforms/85xx/c293pcie.c
	arch/powerpc/platforms/85xx/corenet_ds.c
	arch/powerpc/platforms/85xx/corenet_ds.h
	arch/powerpc/platforms/85xx/p1023_rds.c
	arch/powerpc/platforms/85xx/p2041_rdb.c
	arch/powerpc/platforms/85xx/p3041_ds.c
	arch/powerpc/platforms/85xx/p4080_ds.c
	arch/powerpc/platforms/85xx/p5020_ds.c
	arch/powerpc/platforms/85xx/p5040_ds.c
	arch/powerpc/platforms/85xx/smp.c
	arch/powerpc/platforms/85xx/t4240_qds.c
	arch/powerpc/platforms/Kconfig
	arch/powerpc/sysdev/Makefile
	arch/powerpc/sysdev/fsl_mpic_timer_wakeup.c
	arch/powerpc/sysdev/fsl_msi.c
	arch/powerpc/sysdev/fsl_pci.c
	arch/powerpc/sysdev/fsl_pci.h
	arch/powerpc/sysdev/fsl_soc.h
	arch/powerpc/sysdev/mpic.c
	arch/powerpc/sysdev/mpic_timer.c
	drivers/Kconfig
	drivers/clk/Kconfig
	drivers/clk/clk-ppc-corenet.c
	drivers/cpufreq/Kconfig.powerpc
	drivers/cpufreq/Makefile
	drivers/cpufreq/ppc-corenet-cpufreq.c
	drivers/crypto/caam/Kconfig
	drivers/crypto/caam/Makefile
	drivers/crypto/caam/ctrl.c
	drivers/crypto/caam/desc_constr.h
	drivers/crypto/caam/intern.h
	drivers/crypto/caam/jr.c
	drivers/crypto/caam/regs.h
	drivers/dma/fsldma.c
	drivers/hwmon/ina2xx.c
	drivers/iommu/Kconfig
	drivers/iommu/fsl_pamu.c
	drivers/iommu/fsl_pamu.h
	drivers/iommu/fsl_pamu_domain.c
	drivers/iommu/fsl_pamu_domain.h
	drivers/misc/Makefile
	drivers/mmc/card/block.c
	drivers/mmc/core/core.c
	drivers/mmc/host/sdhci-esdhc.h
	drivers/mmc/host/sdhci-pltfm.c
	drivers/mtd/nand/fsl_ifc_nand.c
	drivers/net/ethernet/freescale/gianfar.c
	drivers/net/ethernet/freescale/gianfar.h
	drivers/net/ethernet/freescale/gianfar_ethtool.c
	drivers/net/phy/at803x.c
	drivers/net/phy/phy_device.c
	drivers/net/phy/vitesse.c
	drivers/pci/msi.c
	drivers/staging/Kconfig
	drivers/staging/Makefile
	drivers/uio/Kconfig
	drivers/uio/Makefile
	drivers/uio/uio.c
	drivers/usb/host/ehci-fsl.c
	drivers/vfio/Kconfig
	drivers/vfio/Makefile
	include/crypto/algapi.h
	include/linux/iommu.h
	include/linux/mmc/sdhci.h
	include/linux/msi.h
	include/linux/netdev_features.h
	include/linux/phy.h
	include/linux/skbuff.h
	include/net/ip.h
	include/uapi/linux/vfio.h
	net/core/ethtool.c
	net/ipv4/route.c
	net/ipv6/route.c

1 files changed, 131 insertions, 67 deletions

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 961a718..1417d01 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -112,7 +112,8 @@
 #define RT_FL_TOS(oldflp4) \
 	((oldflp4)->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK))
 
-#define IP_MAX_MTU	0xFFF0
+/* IPv4 datagram length is stored into 16bit field (tot_len) */
+#define IP_MAX_MTU	0xFFFF
 
 #define RT_GC_TIMEOUT (300*HZ)
 
@@ -298,7 +299,7 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v)
 	seq_printf(seq,"%08x  %08x %08x %08x %08x %08x %08x %08x "
 		   " %08x %08x %08x %08x %08x %08x %08x %08x %08x \n",
 		   dst_entries_get_slow(&ipv4_dst_ops),
-		   st->in_hit,
+		   0, /* st->in_hit */
 		   st->in_slow_tot,
 		   st->in_slow_mc,
 		   st->in_no_route,
@@ -306,16 +307,16 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v)
 		   st->in_martian_dst,
 		   st->in_martian_src,
 
-		   st->out_hit,
+		   0, /* st->out_hit */
 		   st->out_slow_tot,
 		   st->out_slow_mc,
 
-		   st->gc_total,
-		   st->gc_ignored,
-		   st->gc_goal_miss,
-		   st->gc_dst_overflow,
-		   st->in_hlist_search,
-		   st->out_hlist_search
+		   0, /* st->gc_total */
+		   0, /* st->gc_ignored */
+		   0, /* st->gc_goal_miss */
+		   0, /* st->gc_dst_overflow */
+		   0, /* st->in_hlist_search */
+		   0  /* st->out_hlist_search */
 		);
 	return 0;
 }
@@ -439,12 +440,12 @@ static inline int ip_rt_proc_init(void)
 
 static inline bool rt_is_expired(const struct rtable *rth)
 {
-	return rth->rt_genid != rt_genid(dev_net(rth->dst.dev));
+	return rth->rt_genid != rt_genid_ipv4(dev_net(rth->dst.dev));
 }
 
 void rt_cache_flush(struct net *net)
 {
-	rt_genid_bump(net);
+	rt_genid_bump_ipv4(net);
 #ifdef CONFIG_AS_FASTPATH
 	if (route_flush_fn)
 		route_flush_fn();
@@ -573,10 +574,25 @@ static inline void rt_free(struct rtable *rt)
 
 static DEFINE_SPINLOCK(fnhe_lock);
 
+static void fnhe_flush_routes(struct fib_nh_exception *fnhe)
+{
+	struct rtable *rt;
+
+	rt = rcu_dereference(fnhe->fnhe_rth_input);
+	if (rt) {
+		RCU_INIT_POINTER(fnhe->fnhe_rth_input, NULL);
+		rt_free(rt);
+	}
+	rt = rcu_dereference(fnhe->fnhe_rth_output);
+	if (rt) {
+		RCU_INIT_POINTER(fnhe->fnhe_rth_output, NULL);
+		rt_free(rt);
+	}
+}
+
 static struct fib_nh_exception *fnhe_oldest(struct fnhe_hash_bucket *hash)
 {
 	struct fib_nh_exception *fnhe, *oldest;
-	struct rtable *orig;
 
 	oldest = rcu_dereference(hash->chain);
 	for (fnhe = rcu_dereference(oldest->fnhe_next); fnhe;
@@ -584,11 +600,7 @@ static struct fib_nh_exception *fnhe_oldest(struct fnhe_hash_bucket *hash)
 		if (time_before(fnhe->fnhe_stamp, oldest->fnhe_stamp))
 			oldest = fnhe;
 	}
-	orig = rcu_dereference(oldest->fnhe_rth);
-	if (orig) {
-		RCU_INIT_POINTER(oldest->fnhe_rth, NULL);
-		rt_free(orig);
-	}
+	fnhe_flush_routes(oldest);
 	return oldest;
 }
 
@@ -602,11 +614,25 @@ static inline u32 fnhe_hashfun(__be32 daddr)
 	return hval & (FNHE_HASH_SIZE - 1);
 }
 
+static void fill_route_from_fnhe(struct rtable *rt, struct fib_nh_exception *fnhe)
+{
+	rt->rt_pmtu = fnhe->fnhe_pmtu;
+	rt->dst.expires = fnhe->fnhe_expires;
+
+	if (fnhe->fnhe_gw) {
+		rt->rt_flags |= RTCF_REDIRECTED;
+		rt->rt_gateway = fnhe->fnhe_gw;
+		rt->rt_uses_gateway = 1;
+	}
+}
+
 static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
 				  u32 pmtu, unsigned long expires)
 {
 	struct fnhe_hash_bucket *hash;
 	struct fib_nh_exception *fnhe;
+	struct rtable *rt;
+	unsigned int i;
 	int depth;
 	u32 hval = fnhe_hashfun(daddr);
 
@@ -635,8 +661,15 @@ static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
 			fnhe->fnhe_gw = gw;
 		if (pmtu) {
 			fnhe->fnhe_pmtu = pmtu;
-			fnhe->fnhe_expires = expires;
+			fnhe->fnhe_expires = max(1UL, expires);
 		}
+		/* Update all cached dsts too */
+		rt = rcu_dereference(fnhe->fnhe_rth_input);
+		if (rt)
+			fill_route_from_fnhe(rt, fnhe);
+		rt = rcu_dereference(fnhe->fnhe_rth_output);
+		if (rt)
+			fill_route_from_fnhe(rt, fnhe);
 	} else {
 		if (depth > FNHE_RECLAIM_DEPTH)
 			fnhe = fnhe_oldest(hash);
@@ -648,10 +681,27 @@ static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
 			fnhe->fnhe_next = hash->chain;
 			rcu_assign_pointer(hash->chain, fnhe);
 		}
+		fnhe->fnhe_genid = fnhe_genid(dev_net(nh->nh_dev));
 		fnhe->fnhe_daddr = daddr;
 		fnhe->fnhe_gw = gw;
 		fnhe->fnhe_pmtu = pmtu;
 		fnhe->fnhe_expires = expires;
+
+		/* Exception created; mark the cached routes for the nexthop
+		 * stale, so anyone caching it rechecks if this exception
+		 * applies to them.
+		 */
+		rt = rcu_dereference(nh->nh_rth_input);
+		if (rt)
+			rt->dst.obsolete = DST_OBSOLETE_KILL;
+
+		for_each_possible_cpu(i) {
+			struct rtable __rcu **prt;
+			prt = per_cpu_ptr(nh->nh_pcpu_rth_output, i);
+			rt = rcu_dereference(*prt);
+			if (rt)
+				rt->dst.obsolete = DST_OBSOLETE_KILL;
+		}
 	}
 
 	fnhe->fnhe_stamp = jiffies;
@@ -930,12 +980,9 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
 	if (mtu < ip_rt_min_pmtu)
 		mtu = ip_rt_min_pmtu;
 
-	if (!rt->rt_pmtu) {
-		dst->obsolete = DST_OBSOLETE_KILL;
-	} else {
-		rt->rt_pmtu = mtu;
-		dst->expires = max(1UL, jiffies + ip_rt_mtu_expires);
-	}
+	if (rt->rt_pmtu == mtu &&
+	    time_before(jiffies, dst->expires - ip_rt_mtu_expires / 2))
+		return;
 
 	rcu_read_lock();
 	if (fib_lookup(dev_net(dst->dev), fl4, &res) == 0) {
@@ -997,6 +1044,10 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
 	bool new = false;
 
 	bh_lock_sock(sk);
+
+	if (!ip_sk_accept_pmtu(sk))
+		goto out;
+
 	rt = (struct rtable *) __sk_dst_get(sk);
 
 	if (sock_owned_by_user(sk) || !rt) {
@@ -1076,11 +1127,11 @@ static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie)
 	 * DST_OBSOLETE_FORCE_CHK which forces validation calls down
 	 * into this function always.
 	 *
-	 * When a PMTU/redirect information update invalidates a
-	 * route, this is indicated by setting obsolete to
-	 * DST_OBSOLETE_KILL.
+	 * When a PMTU/redirect information update invalidates a route,
+	 * this is indicated by setting obsolete to DST_OBSOLETE_KILL or
+	 * DST_OBSOLETE_DEAD by dst_free().
 	 */
-	if (dst->obsolete == DST_OBSOLETE_KILL || rt_is_expired(rt))
+	if (dst->obsolete != DST_OBSOLETE_FORCE_CHK || rt_is_expired(rt))
 		return NULL;
 	return dst;
 }
@@ -1189,10 +1240,7 @@ static unsigned int ipv4_mtu(const struct dst_entry *dst)
 			mtu = 576;
 	}
 
-	if (mtu > IP_MAX_MTU)
-		mtu = IP_MAX_MTU;
-
-	return mtu;
+	return min_t(unsigned int, mtu, IP_MAX_MTU);
 }
 
 static struct fib_nh_exception *find_exception(struct fib_nh *nh, __be32 daddr)
@@ -1222,34 +1270,36 @@ static bool rt_bind_exception(struct rtable *rt, struct fib_nh_exception *fnhe,
 	spin_lock_bh(&fnhe_lock);
 
 	if (daddr == fnhe->fnhe_daddr) {
-		struct rtable *orig = rcu_dereference(fnhe->fnhe_rth);
-		if (orig && rt_is_expired(orig)) {
+		struct rtable __rcu **porig;
+		struct rtable *orig;
+		int genid = fnhe_genid(dev_net(rt->dst.dev));
+
+		if (rt_is_input_route(rt))
+			porig = &fnhe->fnhe_rth_input;
+		else
+			porig = &fnhe->fnhe_rth_output;
+		orig = rcu_dereference(*porig);
+
+		if (fnhe->fnhe_genid != genid) {
+			fnhe->fnhe_genid = genid;
 			fnhe->fnhe_gw = 0;
 			fnhe->fnhe_pmtu = 0;
 			fnhe->fnhe_expires = 0;
+			fnhe_flush_routes(fnhe);
+			orig = NULL;
 		}
-		if (fnhe->fnhe_pmtu) {
-			unsigned long expires = fnhe->fnhe_expires;
-			unsigned long diff = expires - jiffies;
-
-			if (time_before(jiffies, expires)) {
-				rt->rt_pmtu = fnhe->fnhe_pmtu;
-				dst_set_expires(&rt->dst, diff);
-			}
-		}
-		if (fnhe->fnhe_gw) {
-			rt->rt_flags |= RTCF_REDIRECTED;
-			rt->rt_gateway = fnhe->fnhe_gw;
-			rt->rt_uses_gateway = 1;
-		} else if (!rt->rt_gateway)
+		fill_route_from_fnhe(rt, fnhe);
+		if (!rt->rt_gateway)
 			rt->rt_gateway = daddr;
 
-		rcu_assign_pointer(fnhe->fnhe_rth, rt);
-		if (orig)
-			rt_free(orig);
+		if (!(rt->dst.flags & DST_NOCACHE)) {
+			rcu_assign_pointer(*porig, rt);
+			if (orig)
+				rt_free(orig);
+			ret = true;
+		}
 
 		fnhe->fnhe_stamp = jiffies;
-		ret = true;
 	}
 	spin_unlock_bh(&fnhe_lock);
 
@@ -1418,7 +1468,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 #endif
 	rth->dst.output = ip_rt_bug;
 
-	rth->rt_genid	= rt_genid(dev_net(dev));
+	rth->rt_genid	= rt_genid_ipv4(dev_net(dev));
 	rth->rt_flags	= RTCF_MULTICAST;
 	rth->rt_type	= RTN_MULTICAST;
 	rth->rt_is_input= 1;
@@ -1481,6 +1531,7 @@ static int __mkroute_input(struct sk_buff *skb,
 			   struct in_device *in_dev,
 			   __be32 daddr, __be32 saddr, u32 tos)
 {
+	struct fib_nh_exception *fnhe;
 	struct rtable *rth;
 	int err;
 	struct in_device *out_dev;
@@ -1527,8 +1578,13 @@ static int __mkroute_input(struct sk_buff *skb,
 		}
 	}
 
+	fnhe = find_exception(&FIB_RES_NH(*res), daddr);
 	if (do_cache) {
-		rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input);
+		if (fnhe != NULL)
+			rth = rcu_dereference(fnhe->fnhe_rth_input);
+		else
+			rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input);
+
 		if (rt_cache_valid(rth)) {
 			skb_dst_set_noref(skb, &rth->dst);
 			goto out;
@@ -1543,7 +1599,7 @@ static int __mkroute_input(struct sk_buff *skb,
 		goto cleanup;
 	}
 
-	rth->rt_genid = rt_genid(dev_net(rth->dst.dev));
+	rth->rt_genid = rt_genid_ipv4(dev_net(rth->dst.dev));
 	rth->rt_flags = flags;
 	rth->rt_type = res->type;
 	rth->rt_is_input = 1;
@@ -1556,7 +1612,7 @@ static int __mkroute_input(struct sk_buff *skb,
 	rth->dst.input = ip_forward;
 	rth->dst.output = ip_output;
 
-	rt_set_nexthop(rth, daddr, res, NULL, res->fi, res->type, itag);
+	rt_set_nexthop(rth, daddr, res, fnhe, res->fi, res->type, itag);
 	skb_dst_set(skb, &rth->dst);
 out:
 	err = 0;
@@ -1714,7 +1770,7 @@ local_input:
 	rth->dst.tclassid = itag;
 #endif
 
-	rth->rt_genid = rt_genid(net);
+	rth->rt_genid = rt_genid_ipv4(net);
 	rth->rt_flags 	= flags|RTCF_LOCAL;
 	rth->rt_type	= res.type;
 	rth->rt_is_input = 1;
@@ -1728,8 +1784,12 @@ local_input:
 		rth->dst.error= -err;
 		rth->rt_flags 	&= ~RTCF_LOCAL;
 	}
-	if (do_cache)
-		rt_cache_route(&FIB_RES_NH(res), rth);
+	if (do_cache) {
+		if (unlikely(!rt_cache_route(&FIB_RES_NH(res), rth))) {
+			rth->dst.flags |= DST_NOCACHE;
+			rt_add_uncached_list(rth);
+		}
+	}
 	skb_dst_set(skb, &rth->dst);
 	err = 0;
 	goto out;
@@ -1871,7 +1931,7 @@ static struct rtable *__mkroute_output(const struct fib_result *res,
 
 		fnhe = find_exception(nh, fl4->daddr);
 		if (fnhe)
-			prth = &fnhe->fnhe_rth;
+			prth = &fnhe->fnhe_rth_output;
 		else {
 			if (unlikely(fl4->flowi4_flags &
 				     FLOWI_FLAG_KNOWN_NH &&
@@ -1899,7 +1959,7 @@ add:
 
 	rth->dst.output = ip_output;
 
-	rth->rt_genid = rt_genid(dev_net(dev_out));
+	rth->rt_genid = rt_genid_ipv4(dev_net(dev_out));
 	rth->rt_flags	= flags;
 	rth->rt_type	= type;
 	rth->rt_is_input = 0;
@@ -2028,7 +2088,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4)
 							      RT_SCOPE_LINK);
 			goto make_route;
 		}
-		if (fl4->saddr) {
+		if (!fl4->saddr) {
 			if (ipv4_is_multicast(fl4->daddr))
 				fl4->saddr = inet_select_addr(dev_out, 0,
 							      fl4->flowi4_scope);
@@ -2189,7 +2249,7 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or
 		rt->rt_iif = ort->rt_iif;
 		rt->rt_pmtu = ort->rt_pmtu;
 
-		rt->rt_genid = rt_genid(net);
+		rt->rt_genid = rt_genid_ipv4(net);
 		rt->rt_flags = ort->rt_flags;
 		rt->rt_type = ort->rt_type;
 		rt->rt_gateway = ort->rt_gateway;
@@ -2445,19 +2505,22 @@ static int ip_rt_gc_interval __read_mostly  = 60 * HZ;
 static int ip_rt_gc_min_interval __read_mostly	= HZ / 2;
 static int ip_rt_gc_elasticity __read_mostly	= 8;
 
-static int ipv4_sysctl_rtcache_flush(ctl_table *__ctl, int write,
+static int ipv4_sysctl_rtcache_flush(struct ctl_table *__ctl, int write,
 					void __user *buffer,
 					size_t *lenp, loff_t *ppos)
 {
+	struct net *net = (struct net *)__ctl->extra1;
+
 	if (write) {
-		rt_cache_flush((struct net *)__ctl->extra1);
+		rt_cache_flush(net);
+		fnhe_genid_bump(net);
 		return 0;
 	}
 
 	return -EINVAL;
 }
 
-static ctl_table ipv4_route_table[] = {
+static struct ctl_table ipv4_route_table[] = {
 	{
 		.procname	= "gc_thresh",
 		.data		= &ipv4_dst_ops.gc_thresh,
@@ -2624,7 +2687,8 @@ static __net_initdata struct pernet_operations sysctl_route_ops = {
 
 static __net_init int rt_genid_init(struct net *net)
 {
-	atomic_set(&net->rt_genid, 0);
+	atomic_set(&net->ipv4.rt_genid, 0);
+	atomic_set(&net->fnhe_genid, 0);
 	get_random_bytes(&net->ipv4.dev_addr_genid,
 			 sizeof(net->ipv4.dev_addr_genid));
 	return 0;