]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
tcp: consider recv buf for the initial window scale
authorSoheil Hassas Yeganeh <soheil@google.com>
Fri, 29 Jul 2016 13:34:02 +0000 (09:34 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 16 Aug 2016 07:29:03 +0000 (09:29 +0200)
commitd92f45a046e909bfb8b292b2f7d56ccbedf48d55
treed2d224a3eda74b83a80329b3f125cec0e95f86cf
parent8e22cf2218d26d168f0d7ccf550512dfe8c4c6a5
tcp: consider recv buf for the initial window scale

[ Upstream commit f626300a3e776ccc9671b0dd94698fb3aa315966 ]

tcp_select_initial_window() intends to advertise a window
scaling for the maximum possible window size. To do so,
it considers the maximum of net.ipv4.tcp_rmem[2] and
net.core.rmem_max as the only possible upper-bounds.
However, users with CAP_NET_ADMIN can use SO_RCVBUFFORCE
to set the socket's receive buffer size to values
larger than net.ipv4.tcp_rmem[2] and net.core.rmem_max.
Thus, SO_RCVBUFFORCE is effectively ignored by
tcp_select_initial_window().

To fix this, consider the maximum of net.ipv4.tcp_rmem[2],
net.core.rmem_max and socket's initial buffer space.

Fixes: b0573dea1fb3 ("[NET]: Introduce SO_{SND,RCV}BUFFORCE socket options")
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Suggested-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv4/tcp_output.c