Eric Dumazet f4a775d144 net: introduce __skb_header_release()
While profiling TCP stack, I noticed one useless atomic operation
in tcp_sendmsg(), caused by skb_header_release().

It turns out all current skb_header_release() users have a fresh skb,
that no other user can see, so we can avoid one atomic operation.

Introduce __skb_header_release() to clearly document this.

This gave me a 1.5 % improvement on TCP_RR workload.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-26 15:40:06 -04:00
..
2014-09-10 12:40:10 -07:00
2014-07-01 23:46:21 -07:00
2014-08-22 12:23:11 -07:00
2014-09-10 12:40:10 -07:00
2013-09-23 16:52:45 -04:00
2014-09-07 16:07:25 -07:00
2014-09-26 00:26:48 -04:00
2014-08-13 18:13:19 -06:00
2014-09-10 14:00:58 -07:00
2013-11-21 13:09:42 -05:00