[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240212070213.504d6930@kernel.org>
Date: Mon, 12 Feb 2024 07:02:13 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Jamal Hadi Salim <jhs@...atatu.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, edumazet@...gle.com,
pabeni@...hat.com, Marcelo Ricardo Leitner <marcelo.leitner@...il.com>,
Davide Caratti <dcaratti@...hat.com>, xiyou.wangcong@...il.com,
jiri@...nulli.us, shmulik.ladkani@...il.com
Subject: Re: [PATCH net] net/sched: act_mirred: use the backlog for mirred
ingress
On Mon, 12 Feb 2024 09:51:20 -0500 Jamal Hadi Salim wrote:
> > The test Davide added in commit ca22da2fbd69 ("act_mirred: use the backlog
> > for nested calls to mirred ingress") hangs our testing VMs every 10 or so
> > runs, with the familiar tcp_v4_rcv -> tcp_v4_rcv deadlock reported by
> > lockdep.
> >
> > In the past there was a concern that the backlog indirection will
> > lead to loss of error reporting / less accurate stats. But the current
> > workaround does not seem to address the issue.
>
> Let us run some basic tests on this first - it's a hairy spot.
Thanks!
> Also, IIRC Cong had some reservations about this in the past. Can't
> remember what they were.
He was worried that the overlimits stats are no longer counted when we
decouple (and Davide's selftest actually checks for that so I had to
modify it).
I'm not married to the solution in any way but bugs much less serious
get three letter acronyms, we can't pretend this one doesn't exist.
Powered by blists - more mailing lists