diff options
| author | Eric Dumazet <edumazet@google.com> | 2014-04-03 09:28:10 -0700 |
|---|---|---|
| committer | Ben Hutchings <ben@decadent.org.uk> | 2014-06-09 13:29:00 +0100 |
| commit | fa79aa5e1b89bb5ec90fbd757513e907ca970ba5 (patch) | |
| tree | 677f8222b6daec95262de7eca9550359f3cdf1dd /net | |
| parent | 93a1554e0ca3cbabfb90d8d6edeb1680597283a3 (diff) | |
| download | lwn-fa79aa5e1b89bb5ec90fbd757513e907ca970ba5.tar.gz lwn-fa79aa5e1b89bb5ec90fbd757513e907ca970ba5.zip | |
net-gro: reset skb->truesize in napi_reuse_skb()
[ Upstream commit e33d0ba8047b049c9262fdb1fcafb93cb52ceceb ]
Recycling skb always had been very tough...
This time it appears GRO layer can accumulate skb->truesize
adjustments made by drivers when they attach a fragment to skb.
skb_gro_receive() can only subtract from skb->truesize the used part
of a fragment.
I spotted this problem seeing TcpExtPruneCalled and
TcpExtTCPRcvCollapsed that were unexpected with a recent kernel, where
TCP receive window should be sized properly to accept traffic coming
from a driver not overshooting skb->truesize.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Diffstat (limited to 'net')
| -rw-r--r-- | net/core/dev.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/net/core/dev.c b/net/core/dev.c index 7bcf37df0ce9..854da1571997 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3648,6 +3648,7 @@ static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb) skb->vlan_tci = 0; skb->dev = napi->dev; skb->skb_iif = 0; + skb->truesize = SKB_TRUESIZE(skb_end_offset(skb)); napi->skb = skb; } |