Commit 21b29e74 authored by Eric Dumazet's avatar Eric Dumazet Committed by Jakub Kicinski
Browse files

tcp: take care of zero tp->window_clamp in tcp_set_rcvlowat() 



Some applications (like selftests/net/tcp_mmap.c) call SO_RCVLOWAT
on their listener, before accept().

This has an unfortunate effect on wscale selection in
tcp_select_initial_window() during 3WHS.

For instance, tcp_mmap was negotiating wscale 4, regardless
of tcp_rmem[2] and sysctl_rmem_max.

Do not change tp->window_clamp if it is zero
or bigger than our computed value.

Zero value is special, it allows tcp_select_initial_window()
to enable autotuning.

Note that SO_RCVLOWAT use on listener is probably not wise,
because tp->scaling_ratio has a default value, possibly wrong.

Fixes: d1361840 ("tcp: fix SO_RCVLOWAT and RCVBUF autotuning")
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Reviewed-by: default avatarKuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: default avatarNeal Cardwell <ncardwell@google.com>
Link: https://patch.msgid.link/20251003184119.2526655-1-edumazet@google.com


Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
parent 95920c2e

net/ipv4/tcp.c

+4 −1
Original line number Diff line number Diff line
@@ -1791,6 +1791,7 @@ EXPORT_IPV6_MOD(tcp_peek_len); 
/* Make sure sk_rcvbuf is big enough to satisfy SO_RCVLOWAT hint */

int tcp_set_rcvlowat(struct sock *sk, int val)

{

	struct tcp_sock *tp = tcp_sk(sk);

	int space, cap;



	if (sk->sk_userlocks & SOCK_RCVBUF_LOCK)
@@ -1809,7 +1810,9 @@ int tcp_set_rcvlowat(struct sock *sk, int val) 
	space = tcp_space_from_win(sk, val);

	if (space > sk->sk_rcvbuf) {

		WRITE_ONCE(sk->sk_rcvbuf, space);

		WRITE_ONCE(tcp_sk(sk)->window_clamp, val);



		if (tp->window_clamp && tp->window_clamp < val)

			WRITE_ONCE(tp->window_clamp, val);

	}

	return 0;

}