[PATCH net] tcp: fastopen: avoid negative sk_forward_alloc

Subsystems: networking [general], networking [tcp], the rest

STALE3694d

3 messages, 3 authors, 2016-09-08 · open the first message on its own page

[PATCH net] tcp: fastopen: avoid negative sk_forward_alloc

From: Eric Dumazet <hidden>
Date: 2016-09-07 15:34:12

From: Eric Dumazet <redacted>

When DATA and/or FIN are carried in a SYN/ACK message or SYN message,
we append an skb in socket receive queue, but we forget to call
sk_forced_mem_schedule().

Effect is that the socket has a negative sk->sk_forward_alloc as long as
the message is not read by the application.

Josh Hunt fixed a similar issue in commit d22e15371811 ("tcp: fix tcp
fin memory accounting")
 
Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: Eric Dumazet <redacted>
---
 net/ipv4/tcp_fastopen.c |    1 +
 1 file changed, 1 insertion(+)
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 62a5751d4fe1..4e777a3243f9 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -150,6 +150,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
 	tp->segs_in = 0;
 	tcp_segs_in(tp, skb);
 	__skb_pull(skb, tcp_hdrlen(skb));
+	sk_forced_mem_schedule(sk, skb->truesize);
 	skb_set_owner_r(skb, sk);
 
 	TCP_SKB_CB(skb)->seq++;

Re: [PATCH net] tcp: fastopen: avoid negative sk_forward_alloc

From: Josh Hunt <hidden>
Date: 2016-09-07 19:55:02

On 09/07/2016 10:34 AM, Eric Dumazet wrote:
quoted hunk
From: Eric Dumazet <redacted>

When DATA and/or FIN are carried in a SYN/ACK message or SYN message,
we append an skb in socket receive queue, but we forget to call
sk_forced_mem_schedule().

Effect is that the socket has a negative sk->sk_forward_alloc as long as
the message is not read by the application.

Josh Hunt fixed a similar issue in commit d22e15371811 ("tcp: fix tcp
fin memory accounting")

Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: Eric Dumazet <redacted>
---
  net/ipv4/tcp_fastopen.c |    1 +
  1 file changed, 1 insertion(+)
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 62a5751d4fe1..4e777a3243f9 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -150,6 +150,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
  	tp->segs_in = 0;
  	tcp_segs_in(tp, skb);
  	__skb_pull(skb, tcp_hdrlen(skb));
+	sk_forced_mem_schedule(sk, skb->truesize);
  	skb_set_owner_r(skb, sk);

  	TCP_SKB_CB(skb)->seq++;
The change makes sense to me. Thanks Eric!

Reviewed-by: Josh Hunt <redacted>

Re: [PATCH net] tcp: fastopen: avoid negative sk_forward_alloc

From: David Miller <davem@davemloft.net>
Date: 2016-09-08 23:08:32

From: Eric Dumazet <redacted>
Date: Wed, 07 Sep 2016 08:34:11 -0700
From: Eric Dumazet <redacted>

When DATA and/or FIN are carried in a SYN/ACK message or SYN message,
we append an skb in socket receive queue, but we forget to call
sk_forced_mem_schedule().

Effect is that the socket has a negative sk->sk_forward_alloc as long as
the message is not read by the application.

Josh Hunt fixed a similar issue in commit d22e15371811 ("tcp: fix tcp
fin memory accounting")
 
Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: Eric Dumazet <redacted>
Applied and queued up for -stable.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help