* Please queue commit e33d0ba8047b049 for stable
@ 2014-05-22 21:09 Cong Wang
2014-05-22 21:14 ` David Miller
0 siblings, 1 reply; 4+ messages in thread
From: Cong Wang @ 2014-05-22 21:09 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Eric Dumazet
Hi, David
Please queue the following commit for stable:
commit e33d0ba8047b049c9262fdb1fcafb93cb52ceceb
Author: Eric Dumazet <edumazet@google.com>
Date: Thu Apr 3 09:28:10 2014 -0700
net-gro: reset skb->truesize in napi_reuse_skb()
Recycling skb always had been very tough...
This time it appears GRO layer can accumulate skb->truesize
adjustments made by drivers when they attach a fragment to skb.
skb_gro_receive() can only subtract from skb->truesize the used part
of a fragment.
I spotted this problem seeing TcpExtPruneCalled and
TcpExtTCPRcvCollapsed that were unexpected with a recent kernel, where
TCP receive window should be sized properly to accept traffic coming
from a driver not overshooting skb->truesize.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
unless Eric disagrees.
If I understand it correctly, it fixes a real bug for all drivers calling
napi_gro_frags(). But I am not sure which version of kernels it should
apply to, it seems to be a long-standing bug.
Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Please queue commit e33d0ba8047b049 for stable
2014-05-22 21:09 Please queue commit e33d0ba8047b049 for stable Cong Wang
@ 2014-05-22 21:14 ` David Miller
2014-05-22 21:27 ` Eric Dumazet
0 siblings, 1 reply; 4+ messages in thread
From: David Miller @ 2014-05-22 21:14 UTC (permalink / raw)
To: cwang; +Cc: netdev, edumazet
From: Cong Wang <cwang@twopensource.com>
Date: Thu, 22 May 2014 14:09:07 -0700
> Please queue the following commit for stable:
>
> commit e33d0ba8047b049c9262fdb1fcafb93cb52ceceb
> Author: Eric Dumazet <edumazet@google.com>
> Date: Thu Apr 3 09:28:10 2014 -0700
>
> net-gro: reset skb->truesize in napi_reuse_skb()
...
> unless Eric disagrees.
>
> If I understand it correctly, it fixes a real bug for all drivers calling
> napi_gro_frags(). But I am not sure which version of kernels it should
> apply to, it seems to be a long-standing bug.
Queued up, Eric what versions does it apply to?
Thanks.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Please queue commit e33d0ba8047b049 for stable
2014-05-22 21:14 ` David Miller
@ 2014-05-22 21:27 ` Eric Dumazet
2014-05-22 21:50 ` David Miller
0 siblings, 1 reply; 4+ messages in thread
From: Eric Dumazet @ 2014-05-22 21:27 UTC (permalink / raw)
To: David Miller; +Cc: cwang, netdev, edumazet
On Thu, 2014-05-22 at 17:14 -0400, David Miller wrote:
> Queued up, Eric what versions does it apply to?
All stable versions have the bug.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Please queue commit e33d0ba8047b049 for stable
2014-05-22 21:27 ` Eric Dumazet
@ 2014-05-22 21:50 ` David Miller
0 siblings, 0 replies; 4+ messages in thread
From: David Miller @ 2014-05-22 21:50 UTC (permalink / raw)
To: eric.dumazet; +Cc: cwang, netdev, edumazet
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 22 May 2014 14:27:06 -0700
> On Thu, 2014-05-22 at 17:14 -0400, David Miller wrote:
>
>> Queued up, Eric what versions does it apply to?
>
> All stable versions have the bug.
Ok, thanks.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2014-05-22 21:50 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-05-22 21:09 Please queue commit e33d0ba8047b049 for stable Cong Wang
2014-05-22 21:14 ` David Miller
2014-05-22 21:27 ` Eric Dumazet
2014-05-22 21:50 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).