linux-kernel - Re: [PATCH V2 1/4] posix-timers: Make next_posix_timer_id an atomic

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: <87wmdkh9mo.ffs@tglx>
Date: Thu, 20 Feb 2025 17:19:11 +0100
From: Thomas Gleixner <tglx@...utronix.de>
To: Eric Dumazet <edumazet@...gle.com>
Cc: Anna-Maria Behnsen <anna-maria@...utronix.de>, Frederic Weisbecker
 <frederic@...nel.org>, linux-kernel <linux-kernel@...r.kernel.org>,
 Benjamin Segall <bsegall@...gle.com>, Eric Dumazet
 <eric.dumazet@...il.com>
Subject: Re: [PATCH V2 1/4] posix-timers: Make next_posix_timer_id an atomic_t

On Thu, Feb 20 2025 at 16:55, Eric Dumazet wrote:
> On Thu, Feb 20, 2025 at 3:32 PM Thomas Gleixner <tglx@...utronix.de> wrote:
>>
>> On Thu, Feb 20 2025 at 15:04, Thomas Gleixner wrote:
>> > On Thu, Feb 20 2025 at 09:49, Eric Dumazet wrote:
>> >> On Thu, Feb 20, 2025 at 9:09 AM Thomas Gleixner <tglx@...utronix.de> wrote:
>> >>> > This allows the following patch to use RCU.
>> >>>
>> >>> Your patch ordering is slightly off by two :)
>> >>>
>> >>> And it fails to explain for what RCU can be used....
>> >>
>> >> This is explained in the following patches.
>> >
>> > The changelog of a patch has to be self contained. The 'following patch'
>> > has no meaning when the patch is merged.
>>
>> That said, please just fold this into the patch which actually does this RCU
>> lookup upfront. The change is trivial enough that it does not really
>> require to be seperate. If the lockless increment would cause issues,
>> then the subsequent RCU lookup is the least of the worries :)
>
> I can squash all patches into a single one if you prefer.

Please don't. The wraparound race prevention has nothing to do with the
lookup optimization.

Thanks,

        tglx