diff options
| author | Paolo Abeni <pabeni@redhat.com> | 2018-04-04 14:30:01 +0200 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2018-04-04 11:53:08 -0400 |
| commit | 9e8445a56c253f01f6716ac14526a7dae6bc0d46 (patch) | |
| tree | 8ebe5055cb50b610f373a711c9198ac2ffe392aa /net/ipv4/ip_output.c | |
| parent | 2a37ce25d9f2071ab6cce7f70fd5bd4bf4a50ee7 (diff) | |
| download | linux-9e8445a56c253f01f6716ac14526a7dae6bc0d46.tar.bz2 | |
net: avoid unneeded atomic operation in ip*_append_data()
After commit 694aba690de0 ("ipv4: factorize sk_wmem_alloc updates
done by __ip_append_data()") and commit 1f4c6eb24029 ("ipv6:
factorize sk_wmem_alloc updates done by __ip6_append_data()"),
when transmitting sub MTU datagram, an addtional, unneeded atomic
operation is performed in ip*_append_data() to update wmem_alloc:
in the above condition the delta is 0.
The above cause small but measurable performance regression in UDP
xmit tput test with packet size below MTU.
This change avoids such overhead updating wmem_alloc only if
wmem_alloc_delta is non zero.
The error path is left intentionally unmodified: it's a slow path
and simplicity is preferred to performances.
Fixes: 694aba690de0 ("ipv4: factorize sk_wmem_alloc updates done by __ip_append_data()")
Fixes: 1f4c6eb24029 ("ipv6: factorize sk_wmem_alloc updates done by __ip6_append_data()")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/ip_output.c')
| -rw-r--r-- | net/ipv4/ip_output.c | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 94cacae76aca..4c11b810a447 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1090,7 +1090,8 @@ alloc_new_skb: length -= copy; } - refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc); + if (wmem_alloc_delta) + refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc); return 0; error_efault: |