[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220927172618.58f238d6@kernel.org>
Date: Tue, 27 Sep 2022 17:26:18 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Duoming Zhou <duoming@....edu.cn>
Cc: linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
isdn@...ux-pingi.de
Subject: Re: [PATCH V3] mISDN: fix use-after-free bugs in l1oip timer
handlers
On Sat, 24 Sep 2022 10:18:42 +0800 Duoming Zhou wrote:
> + del_timer_sync(&hc->keep_tl);
> + del_timer_sync(&hc->timeout_tl);
> + cancel_work_sync(&hc->workq);
> + del_timer_sync(&hc->keep_tl);
> cancel_work_sync(&hc->workq);
Why not add a bit which will indicate that the device is shutting
down and check it in places which schedule the timer?
I think that's much easier to reason about and we won't need to do
this rep cancel procedure.
Powered by blists - more mailing lists