[PATCH] tcp: cubic: fix bug in bictcp_acked()

[PATCH] tcp: cubic: fix bug in bictcp_acked()

[Eric Dumazet]

From: Eric Dumazet <edumazet@google.com>

While investigating about strange increase of retransmit rates
on hosts ~24 days after boot, Van found hystart was disabled
if ca->epoch_start was 0, as following condition is true
when tcp_time_stamp high order bit is set.

(s32)(tcp_time_stamp - ca->epoch_start) < HZ

Quoting Van :

 At initialization & after every loss ca->epoch_start is set to zero so
 I believe that the above line will turn off hystart as soon as the 2^31
 bit is set in tcp_time_stamp & hystart will stay off for 24 days.
 I think we've observed that cubic's restart is too aggressive without
 hystart so this might account for the higher drop rate we observe.

Diagnosed-by: Van Jacobson <vanj@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
---
 net/ipv4/tcp_cubic.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index a9077f4..2b8671d 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -414,7 +414,7 @@ static void bictcp_acked(struct sock *sk, u32 cnt, s32 rtt_us)
 		return;
 
 	/* Discard delay samples right after fast recovery */
-	if ((s32)(tcp_time_stamp - ca->epoch_start) < HZ)
+	if (ca->epoch_start && (s32)(tcp_time_stamp - ca->epoch_start) < HZ)
 		return;
 
 	delay = (rtt_us << 3) / USEC_PER_MSEC;

Re: [PATCH] tcp: cubic: fix bug in bictcp_acked()

[Neal Cardwell]

On Mon, Aug 5, 2013 at 11:05 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> While investigating about strange increase of retransmit rates
> on hosts ~24 days after boot, Van found hystart was disabled
> if ca->epoch_start was 0, as following condition is true
> when tcp_time_stamp high order bit is set.
>
> (s32)(tcp_time_stamp - ca->epoch_start) < HZ
>
> Quoting Van :
>
>  At initialization & after every loss ca->epoch_start is set to zero so
>  I believe that the above line will turn off hystart as soon as the 2^31
>  bit is set in tcp_time_stamp & hystart will stay off for 24 days.
>  I think we've observed that cubic's restart is too aggressive without
>  hystart so this might account for the higher drop rate we observe.
>
> Diagnosed-by: Van Jacobson <vanj@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Neal Cardwell <ncardwell@google.com>
> Cc: Yuchung Cheng <ycheng@google.com>

Acked-by: Neal Cardwell <ncardwell@google.com>

Nice catch!

neal

Re: [PATCH] tcp: cubic: fix bug in bictcp_acked()

[David Miller]

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 05 Aug 2013 20:05:12 -0700

> From: Eric Dumazet <edumazet@google.com>
> 
> While investigating about strange increase of retransmit rates
> on hosts ~24 days after boot, Van found hystart was disabled
> if ca->epoch_start was 0, as following condition is true
> when tcp_time_stamp high order bit is set.
> 
> (s32)(tcp_time_stamp - ca->epoch_start) < HZ
> 
> Quoting Van :
> 
>  At initialization & after every loss ca->epoch_start is set to zero so
>  I believe that the above line will turn off hystart as soon as the 2^31
>  bit is set in tcp_time_stamp & hystart will stay off for 24 days.
>  I think we've observed that cubic's restart is too aggressive without
>  hystart so this might account for the higher drop rate we observe.
> 
> Diagnosed-by: Van Jacobson <vanj@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Applied and queued up for -stable, thanks.