From 87fd308cfc6b2e880bf717a740bd5c58d2aed10c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 13 Jul 2010 05:24:20 +0000 Subject: net: skb_tx_hash() fix relative to skb_orphan_try() commit fc6055a5ba31e2 (net: Introduce skb_orphan_try()) added early orphaning of skbs. This unfortunately added a performance regression in skb_tx_hash() in case of stacked devices (bonding, vlans, ...) Since skb->sk is now NULL, we cannot access sk->sk_hash anymore to spread tx packets to multiple NIC queues on multiqueue devices. skb_tx_hash() in this case only uses skb->protocol, same value for all flows. skb_orphan_try() can copy sk->sk_hash into skb->rxhash and skb_tx_hash() can use this saved sk_hash value to compute its internal hash value. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 723a34710ad..4b05fdf762a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1911,8 +1911,16 @@ static int dev_gso_segment(struct sk_buff *skb) */ static inline void skb_orphan_try(struct sk_buff *skb) { - if (!skb_tx(skb)->flags) + struct sock *sk = skb->sk; + + if (sk && !skb_tx(skb)->flags) { + /* skb_tx_hash() wont be able to get sk. + * We copy sk_hash into skb->rxhash + */ + if (!skb->rxhash) + skb->rxhash = sk->sk_hash; skb_orphan(skb); + } } int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, @@ -1998,8 +2006,7 @@ u16 skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb) if (skb->sk && skb->sk->sk_hash) hash = skb->sk->sk_hash; else - hash = (__force u16) skb->protocol; - + hash = (__force u16) skb->protocol ^ skb->rxhash; hash = jhash_1word(hash, hashrnd); return (u16) (((u64) hash * dev->real_num_tx_queues) >> 32); -- cgit v1.2.3 From 91a72a70594e5212c97705ca6a694bd307f7a26b Mon Sep 17 00:00:00 2001 From: Doug Kehn Date: Wed, 14 Jul 2010 18:02:16 -0700 Subject: net/core: neighbour update Oops When configuring DMVPN (GRE + openNHRP) and a GRE remote address is configured a kernel Oops is observed. The obserseved Oops is caused by a NULL header_ops pointer (neigh->dev->header_ops) in neigh_update_hhs() when void (*update)(struct hh_cache*, const struct net_device*, const unsigned char *) = neigh->dev->header_ops->cache_update; is executed. The dev associated with the NULL header_ops is the GRE interface. This patch guards against the possibility that header_ops is NULL. This Oops was first observed in kernel version 2.6.26.8. Signed-off-by: Doug Kehn Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/neighbour.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net/core') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 6ba1c0eece0..a4e0a7482c2 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -949,7 +949,10 @@ static void neigh_update_hhs(struct neighbour *neigh) { struct hh_cache *hh; void (*update)(struct hh_cache*, const struct net_device*, const unsigned char *) - = neigh->dev->header_ops->cache_update; + = NULL; + + if (neigh->dev->header_ops) + update = neigh->dev->header_ops->cache_update; if (update) { for (hh = neigh->hh; hh; hh = hh->hh_next) { -- cgit v1.2.3 From b0f77d0eae0c58a5a9691a067ada112ceeae2d00 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Wed, 14 Jul 2010 20:50:29 -0700 Subject: net: fix problem in reading sock TX queue Fix problem in reading the tx_queue recorded in a socket. In dev_pick_tx, the TX queue is read by doing a check with sk_tx_queue_recorded on the socket, followed by a sk_tx_queue_get. The problem is that there is not mutual exclusion across these calls in the socket so it it is possible that the queue in the sock can be invalidated after sk_tx_queue_recorded is called so that sk_tx_queue get returns -1, which sets 65535 in queue_index and thus dev_pick_tx returns 65536 which is a bogus queue and can cause crash in dev_queue_xmit. We fix this by only calling sk_tx_queue_get which does the proper checks. The interface is that sk_tx_queue_get returns the TX queue if the sock argument is non-NULL and TX queue is recorded, else it returns -1. sk_tx_queue_recorded is no longer used so it can be completely removed. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- net/core/dev.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'net/core') diff --git a/net/core/dev.c b/net/core/dev.c index 4b05fdf762a..0ea10f849be 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2029,12 +2029,11 @@ static inline u16 dev_cap_txqueue(struct net_device *dev, u16 queue_index) static struct netdev_queue *dev_pick_tx(struct net_device *dev, struct sk_buff *skb) { - u16 queue_index; + int queue_index; struct sock *sk = skb->sk; - if (sk_tx_queue_recorded(sk)) { - queue_index = sk_tx_queue_get(sk); - } else { + queue_index = sk_tx_queue_get(sk); + if (queue_index < 0) { const struct net_device_ops *ops = dev->netdev_ops; if (ops->ndo_select_queue) { -- cgit v1.2.3