]> asedeno.scripts.mit.edu Git - linux.git/commit
ipv4: factorize sk_wmem_alloc updates done by __ip_append_data()
authorEric Dumazet <edumazet@google.com>
Sat, 31 Mar 2018 20:16:25 +0000 (13:16 -0700)
committerDavid S. Miller <davem@davemloft.net>
Sun, 1 Apr 2018 18:08:21 +0000 (14:08 -0400)
commit694aba690de062cf27b28a5e56e7a5a7185b0a1c
tree1fbdc27ecff23dc1babcf50d1fd9884d84d800ee
parentc07255020551cadae8bf903f2c5e1fcbad731bac
ipv4: factorize sk_wmem_alloc updates done by __ip_append_data()

While testing my inet defrag changes, I found that the senders
could spend ~20% of cpu cycles in skb_set_owner_w() updating
sk->sk_wmem_alloc for every fragment they cook.

The solution to this problem is to use alloc_skb() instead
of sock_wmalloc() and manually perform a single sk_wmem_alloc change.

Similar change for IPv6 is provided in following patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv4/ip_output.c