linux-kernel - Re: [PATCH] debugobjects: Fix potential data race in debug_objects

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: <871q5rdnae.ffs@tglx>
Date: Fri, 24 May 2024 21:26:17 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: Breno Leitao <leitao@...ian.org>, Andrew Morton <akpm@...ux-foundation.org>
Cc: paulmck@...nel.org, "open list:DEBUGOBJECTS:"
 <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] debugobjects: Fix potential data race in
 debug_objects_maxchain

On Thu, May 09 2024 at 04:06, Breno Leitao wrote:
> -	if (cnt > debug_objects_maxchain)
> -		debug_objects_maxchain = cnt;
> +	if (cnt > READ_ONCE(debug_objects_maxchain))
> +		WRITE_ONCE(debug_objects_maxchain, cnt);

The data race is actually harmless as this is just used for debugfs
statistics.

I'm pretty sure that the rest of these debug variables are racy as
well. Can we please annotate all of them without waiting for KCSAN to
detect them one by one?

Thanks,

        tglx