diff options
| author | Eric Dumazet <edumazet@google.com> | 2015-05-30 09:16:53 -0700 |
|---|---|---|
| committer | Ziyan <jaraidaniel@gmail.com> | 2016-01-05 18:20:38 +0100 |
| commit | 6c6baea5a8e9de8e542b98027d32e93abe744f36 (patch) | |
| tree | 1a92cb93946c3949c25f2bd2fb2987e595996f21 /net/rds/info.h | |
| parent | 2cfb35e5936ef93b8bfc09a72042ccbd4df34743 (diff) | |
| download | kernel_samsung_tuna-6c6baea5a8e9de8e542b98027d32e93abe744f36.zip kernel_samsung_tuna-6c6baea5a8e9de8e542b98027d32e93abe744f36.tar.gz kernel_samsung_tuna-6c6baea5a8e9de8e542b98027d32e93abe744f36.tar.bz2 | |
udp: fix behavior of wrong checksums
We have two problems in UDP stack related to bogus checksums :
1) We return -EAGAIN to application even if receive queue is not empty.
This breaks applications using edge trigger epoll()
2) Under UDP flood, we can loop forever without yielding to other
processes, potentially hanging the host, especially on non SMP.
This patch is an attempt to make things better.
We might in the future add extra support for rt applications
wanting to better control time spent doing a recv() in a hostile
environment. For example we could validate checksums before queuing
packets in socket receive queue.
Change-Id: I9355321ac7ee564d56c342fa7738b918052bf308
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/rds/info.h')
0 files changed, 0 insertions, 0 deletions