[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220124094623.7ca137e0@kicinski-fedora-PC1C0HJN.hsd1.ca.comcast.net>
Date: Mon, 24 Jan 2022 09:46:23 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Eric Dumazet <edumazet@...gle.com>
Cc: Brian Vazquez <brianvv@...gle.com>,
Jeffrey Ji <jeffreyjilinux@...il.com>,
"David S . Miller" <davem@...emloft.net>,
LKML <linux-kernel@...r.kernel.org>,
netdev <netdev@...r.kernel.org>, jeffreyji <jeffreyji@...gle.com>
Subject: Re: [PATCH net-next] net-core: add InMacErrors counter
On Mon, 24 Jan 2022 09:39:22 -0800 Eric Dumazet wrote:
> On Mon, Jan 24, 2022 at 9:29 AM Jakub Kicinski <kuba@...nel.org> wrote:
>
> > That much it's understood, but it's a trade off. This drop point
> > existed for 20 years, why do we need to consume extra memory now?
>
> Being able to tell after the facts, that such a drop reason had
> occured can help investigations.
>
> nstat -a | grep InMac
>
> Jeffrey, what about _also_ adding the kfree_skb_reason(), since this
> seems to also help ?
Yes, and please add a proper justification to the commit message,
with real life examples of cases where this drop point has proven
to be the culprit. Right now the commit message says gives the example
of trafgen :(
Powered by blists - more mailing lists