[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAAf2ycmZHti95WaBR3s+L5Epm1q7sXmvZ-EqCK=-oZj=45tOwQ@mail.gmail.com>
Date: Mon, 18 Nov 2024 22:46:17 +0000
From: Donald Hunter <donald.hunter@...hat.com>
To: Kuniyuki Iwashima <kuniyu@...zon.com>
Cc: donald.hunter@...il.com, davem@...emloft.net, edumazet@...gle.com,
horms@...nel.org, kuba@...nel.org, netdev@...r.kernel.org, pabeni@...hat.com
Subject: Re: [PATCH net-next v1] net: af_unix: clean up spurious drop reasons
On Mon, 18 Nov 2024 at 18:03, Kuniyuki Iwashima <kuniyu@...zon.com> wrote:
>
> From: Donald Hunter <donald.hunter@...il.com>
> Date: Sat, 16 Nov 2024 09:42:36 +0000
> > Use consume_skb() instead of kfree_skb() in the happy paths to clean up
>
> Both are the unhappy paths and kfree_skb() should be used.
>
> I have some local patches for drop reasons for AF_UNIX and
> can post them after net-next is open if needed.
It would be really helpful if you could share the drop reasons you
have assigned to these. I am seeing ~100/s coming from containerd
which I will need to investigate if it is any kind of error condition.
Thanks,
Donald.
Powered by blists - more mailing lists