linux-kernel - Re: [PATCH 2/7] futex: factor out the futex wake handling

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

Message-ID: <3e2ba7d7-0431-7558-2ece-89ce266f7792@kernel.dk>
Date:   Wed, 12 Jul 2023 09:02:52 -0600
From:   Jens Axboe <axboe@...nel.dk>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     io-uring@...r.kernel.org, linux-kernel@...r.kernel.org,
        tglx@...utronix.de, mingo@...hat.com
Subject: Re: [PATCH 2/7] futex: factor out the futex wake handling

On 7/12/23 2:58?AM, Peter Zijlstra wrote:
> On Tue, Jul 11, 2023 at 06:47:00PM -0600, Jens Axboe wrote:
>> In preparation for having another waker that isn't futex_wake_mark(),
>> add a wake handler in futex_q. No extra data is associated with the
>> handler outside of struct futex_q itself. futex_wake_mark() is defined as
>> the standard wakeup helper, now set through futex_q_init like other
>> defaults.
> 
> Urgh... so if I understand things right, you're going to replace this
> with a custom wake function that does *NOT* put the task on the wake_q.
> 
> The wake_q will thus be empty and the task does not get woken up. I'm
> presuming someone gets a notification instead somewhere somehow.

Right. The custom wake function is responsible for waking the task, if
that is what needs to happen. On the io_uring side, the callback would
queue work for the original task to post a completion event, which would
then also wake it up obviously.

> I might've been nice to mention some of this somewhere ...

I'll expand the commit message a bit, it is a bit light.

-- 
Jens Axboe