linux-kernel - Re: [PATCH v2 1/1] workqueue: fix data race with the pwq->stats[] increment

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <ZO-er9rbuXiqwfU6@slm.duckdns.org>
Date:   Wed, 30 Aug 2023 09:55:27 -1000
From:   Tejun Heo <tj@...nel.org>
To:     Mirsad Todorovac <mirsad.todorovac@....unizg.hr>
Cc:     linux-kernel@...r.kernel.org,
        Lai Jiangshan <jiangshanlai@...il.com>
Subject: Re: [PATCH v2 1/1] workqueue: fix data race with the pwq->stats[]
 increment

Hello,

On Wed, Aug 30, 2023 at 09:48:53PM +0200, Mirsad Todorovac wrote:
> These might just not have been triggered yet, but logically they seem suspicious for a data race.
> I still have to verify if they are called from a locked context ...
> 
> Of course, as the original developer, you will have a clearer picture. I have been led by KCSAN
> to this particular detail.

These are just stats and pwq's are already per-cpu (most of the time). If
they trip kcsan and there's no easy way to work around like here, we should
add a wrapper to inc the stat and annotate it so that kcsan ignores it.

Thanks.

-- 
tejun