[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Ysxes81GjdNK3yE+@bombadil.infradead.org>
Date: Mon, 11 Jul 2022 10:32:35 -0700
From: Luis Chamberlain <mcgrof@...nel.org>
To: Aaron Tomlin <atomlin@...hat.com>
Cc: rostedt@...dmis.org, cl@...ux.com, pmladek@...e.com,
mbenes@...e.cz, christophe.leroy@...roup.eu,
akpm@...ux-foundation.org, linux-kernel@...r.kernel.org,
linux-modules@...r.kernel.org, atomlin@...mlin.com,
ghalat@...hat.com, oleksandr@...alenko.name, neelx@...hat.com,
daniel.thompson@...aro.org, hch@...radead.org, tglx@...utronix.de,
adrian.hunter@...el.com, linux-rt-users@...r.kernel.org
Subject: Re: [PATCH v3 -next 1/1] module: kallsyms: Ensure preemption in
add_kallsyms() with PREEMPT_RT
On Mon, Jul 11, 2022 at 06:17:19PM +0100, Aaron Tomlin wrote:
> The commit 08126db5ff73 ("module: kallsyms: Fix suspicious rcu usage")
> under PREEMPT_RT=y, disabling preemption introduced an unbounded
> latency since the loop is not fixed. This change caused a regression
> since previously preemption was not disabled and we would dereference
> RCU-protected pointers explicitly. That being said, these pointers
> cannot change.
>
> Before kallsyms-specific data is prepared/or set-up, we ensure that
> the unformed module is known to be unique i.e. does not already exist
> (see load_module()). Therefore, we can fix this by using the common and
> more appropriate RCU flavour as this section of code can be safely
> preempted.
>
> Reported-by: Steven Rostedt <rostedt@...dmis.org>
> Fixes: 08126db5ff73 ("module: kallsyms: Fix suspicious rcu usage")
> Signed-off-by: Aaron Tomlin <atomlin@...hat.com>
Thanks! Queued up!
Luis
Powered by blists - more mailing lists