linux-kernel - Re: io_uring vs CPU hotplug, was Re: [PATCH 5/9] blk-mq: don't set data->ctx and data->hctx in blk_mq_alloc_request

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: <87k115t5x3.fsf@nanos.tec.linutronix.de>
Date:   Thu, 21 May 2020 22:00:08 +0200
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Jens Axboe <axboe@...nel.dk>, Ming Lei <ming.lei@...hat.com>
Cc:     Christoph Hellwig <hch@....de>, linux-kernel@...r.kernel.org,
        linux-block@...r.kernel.org, John Garry <john.garry@...wei.com>,
        Bart Van Assche <bvanassche@....org>,
        Hannes Reinecke <hare@...e.com>, io-uring@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>
Subject: Re: io_uring vs CPU hotplug, was Re: [PATCH 5/9] blk-mq: don't set data->ctx and data->hctx in blk_mq_alloc_request_hctx

Jens Axboe <axboe@...nel.dk> writes:
> Again, this is mixing up io_uring and blk-mq. Maybe it's the fact that
> both use 'ctx' that makes this confusing. On the blk-mq side, the 'ctx'
> is the per-cpu queue context, for io_uring it's the io_uring instance.

Yes, that got me horribly confused. :)

> io_sq_thread() doesn't care about any sort of percpu mappings, it's
> happy as long as it'll keep running regardless of whether or not the
> optional pinned CPU is selected and then offlined.

Fair enough.

So aside of the potential spin forever if the uring thread is lifted to
an RT scheduling class, this looks all good.

Though I assume that if that thread is pinned and an admin pushs it into
RT scheduling the spinning live lock can happen independent of cpu
hotplug.

Thanks,

        tglx