diff options
| author | David S. Miller <davem@davemloft.net> | 2008-11-05 03:36:01 -0800 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2008-11-05 03:36:01 -0800 |
| commit | 518a09ef11f8454f4676125d47c3e775b300c6a5 (patch) | |
| tree | a22f9f9de58e91d3287bcd21277e9378de49e18f /net/ipv4/tcp.c | |
| parent | efb9a8c28ca0edd9e2572117105ebad9bbc0c368 (diff) | |
| download | lwn-518a09ef11f8454f4676125d47c3e775b300c6a5.tar.gz lwn-518a09ef11f8454f4676125d47c3e775b300c6a5.zip | |
tcp: Fix recvmsg MSG_PEEK influence of blocking behavior.
Vito Caputo noticed that tcp_recvmsg() returns immediately from
partial reads when MSG_PEEK is used. In particular, this means that
SO_RCVLOWAT is not respected.
Simply remove the test. And this matches the behavior of several
other systems, including BSD.
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp.c')
| -rw-r--r-- | net/ipv4/tcp.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index eccb7165a80c..c5aca0bb116a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1374,8 +1374,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, sk->sk_state == TCP_CLOSE || (sk->sk_shutdown & RCV_SHUTDOWN) || !timeo || - signal_pending(current) || - (flags & MSG_PEEK)) + signal_pending(current)) break; } else { if (sock_flag(sk, SOCK_DONE)) |