tcp: refine memory limit test in tcp_fragment()
authorEric Dumazet <edumazet@google.com>
Fri, 21 Jun 2019 13:09:55 +0000 (06:09 -0700)
committerJason Liu <jason.hui.liu@nxp.com>
Fri, 6 Sep 2019 03:11:35 +0000 (11:11 +0800)
commit0e2dd42ae521801c007ea97fbce93b179405eca3
treed2b8a653fba17d97e2209239d630e62790f827f4
parent5c4cd7b751df3971046fe6e0b2445b18d8fd0415
tcp: refine memory limit test in tcp_fragment()

commit b6653b3629e5b88202be3c9abc44713973f5c4b4 upstream.

tcp_fragment() might be called for skbs in the write queue.

Memory limits might have been exceeded because tcp_sendmsg() only
checks limits at full skb (64KB) boundaries.

Therefore, we need to make sure tcp_fragment() wont punish applications
that might have setup very low SO_SNDBUF values.

Fixes: f070ef2ac667 ("tcp: tcp_fragment() should apply sane memory limits")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Christoph Paasch <cpaasch@apple.com>
Tested-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit dad3a9314ac95dedc007bc7dacacb396ea10e376)
net/ipv4/tcp_output.c