[PATCH] tcp: do not overshoot window_clamp in tcp_rcv_space_adjust()

[PATCH] tcp: do not overshoot window_clamp in tcp_rcv_space_adjust()

[Guenter Roeck]

From: Eric Dumazet <edumazet@google.com>

commit 02db55718d53f9d426cee504c27fb768e9ed4ffe upstream.

While rcvbuf is properly clamped by tcp_rmem[2], rcvwin
is left to a potentially too big value.

It has no serious effect, since :
1) tcp_grow_window() has very strict checks.
2) window_clamp can be mangled by user space to any value anyway.

tcp_init_buffer_space() and companions use tcp_full_space(),
we use tcp_win_from_space() to avoid reloading sk->sk_rcvbuf

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Wei Wang <weiwan@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
 net/ipv4/tcp_input.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 125b49c166a4..f0caff3139ed 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -647,7 +647,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
 			sk->sk_rcvbuf = rcvbuf;
 
 			/* Make the window clamp follow along.  */
-			tp->window_clamp = rcvwin;
+			tp->window_clamp = tcp_win_from_space(rcvbuf);
 		}
 	}
 	tp->rcvq_space.space = copied;
-- 
2.7.4

Re: [PATCH] tcp: do not overshoot window_clamp in tcp_rcv_space_adjust()

[Guenter Roeck]

This is supposed to be a backport for v4.14.y and older. Sorry for the confusion.

Guenter

On 06/17/2018 10:37 PM, Guenter Roeck wrote:
> From: Eric Dumazet <edumazet@google.com>
> 
> commit 02db55718d53f9d426cee504c27fb768e9ed4ffe upstream.
> 
> While rcvbuf is properly clamped by tcp_rmem[2], rcvwin
> is left to a potentially too big value.
> 
> It has no serious effect, since :
> 1) tcp_grow_window() has very strict checks.
> 2) window_clamp can be mangled by user space to any value anyway.
> 
> tcp_init_buffer_space() and companions use tcp_full_space(),
> we use tcp_win_from_space() to avoid reloading sk->sk_rcvbuf
> 
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
> Acked-by: Wei Wang <weiwan@google.com>
> Acked-by: Neal Cardwell <ncardwell@google.com>
> Signed-off-by: David S. Miller <davem@davemloft.net>
> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
> ---
>   net/ipv4/tcp_input.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
> index 125b49c166a4..f0caff3139ed 100644
> --- a/net/ipv4/tcp_input.c
> +++ b/net/ipv4/tcp_input.c
> @@ -647,7 +647,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
>   			sk->sk_rcvbuf = rcvbuf;
>   
>   			/* Make the window clamp follow along.  */
> -			tp->window_clamp = rcvwin;
> +			tp->window_clamp = tcp_win_from_space(rcvbuf);
>   		}
>   	}
>   	tp->rcvq_space.space = copied;
> 

Re: [PATCH] tcp: do not overshoot window_clamp in tcp_rcv_space_adjust()

[Greg KH]

On Sun, Jun 17, 2018 at 10:38:54PM -0700, Guenter Roeck wrote:
> This is supposed to be a backport for v4.14.y and older. Sorry for the confusion.

Not a problem, thanks for the patch, now queued up.

greg k-h