Linux-Block Archive on lore.kernel.org
 help / color / Atom feed
From: Ming Lei <ming.lei@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Jens Axboe <axboe@kernel.dk>, Christoph Hellwig <hch@lst.de>,
	linux-kernel@vger.kernel.org, linux-block@vger.kernel.org,
	John Garry <john.garry@huawei.com>,
	Bart Van Assche <bvanassche@acm.org>,
	Hannes Reinecke <hare@suse.com>,
	io-uring@vger.kernel.org, Peter Zijlstra <peterz@infradead.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
Date: Fri, 22 May 2020 09:57:19 +0800
Message-ID: <20200522015719.GB755458@T590> (raw)
In-Reply-To: <87pnaxt9nv.fsf@nanos.tec.linutronix.de>

On Thu, May 21, 2020 at 08:39:16PM +0200, Thomas Gleixner wrote:
> Ming,
> 
> Ming Lei <ming.lei@redhat.com> writes:
> > On Thu, May 21, 2020 at 10:13:59AM +0200, Thomas Gleixner wrote:
> >> Ming Lei <ming.lei@redhat.com> writes:
> >> > On Thu, May 21, 2020 at 12:14:18AM +0200, Thomas Gleixner wrote:
> >> > - otherwise, the kthread just retries and retries to allocate & release,
> >> > and sooner or later, its time slice is consumed, and migrated out, and the
> >> > cpu hotplug handler will get chance to run and move on, then the cpu is
> >> > shutdown.
> >> 
> >> 1) This is based on the assumption that the kthread is in the SCHED_OTHER
> >>    scheduling class. Is that really a valid assumption?
> >
> > Given it is unlikely path, we can add msleep() before retrying when INACTIVE bit
> > is observed by current thread, and this way can avoid spinning and should work
> > for other schedulers.
> 
> That should work, but pretty is something else
> 
> >> 
> >> 2) What happens in the following scenario:
> >> 
> >>    unplug
> >> 
> >>      mq_offline
> >>        set_ctx_inactive()
> >>        drain_io()
> >>        
> >>    io_kthread()
> >>        try_queue()
> >>        wait_on_ctx()
> >> 
> >>    Can this happen and if so what will wake up that thread?
> >
> > drain_io() releases all tag of this hctx, then wait_on_ctx() will be waken up
> > after any tag is released.
> 
> drain_io() is already done ...
> 
> So looking at that thread function:
> 
> static int io_sq_thread(void *data)
> {
> 	struct io_ring_ctx *ctx = data;
> 
>         while (...) {
>               ....
> 	      to_submit = io_sqring_entries(ctx);
> 
> --> preemption
> 
> hotplug runs
>    mq_offline()
>       set_ctx_inactive();
>       drain_io();
>       finished();
> 
> --> thread runs again
> 
>       mutex_lock(&ctx->uring_lock);
>       ret = io_submit_sqes(ctx, to_submit, NULL, -1, true);
>       mutex_unlock(&ctx->uring_lock);
> 
>       ....
> 
>       if (!to_submit || ret == -EBUSY)
>           ...
>       	  wait_on_ctx();
> 
> Can this happen or did drain_io() already take care of the 'to_submit'
> items and the call to io_submit_sqes() turns into a zero action ?
> 
> If the above happens then nothing will wake it up because the context
> draining is done and finished.

As Jens replied, you mixed the ctx from io uring and blk-mq, both are in
two worlds.

Any wait in this percpu kthread should just wait for generic resource,
not directly related with blk-mq's inactive hctx. Once this thread is
migrated to other online cpu, it will move on.


Thanks,
Ming


  parent reply index

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-18  6:39 blk-mq: improvement CPU hotplug (simplified version) v2 Christoph Hellwig
2020-05-18  6:39 ` [PATCH 1/9] blk-mq: split out a __blk_mq_get_driver_tag helper Christoph Hellwig
2020-05-18  6:39 ` [PATCH 2/9] blk-mq: remove the bio argument to ->prepare_request Christoph Hellwig
2020-05-18  6:39 ` [PATCH 3/9] blk-mq: simplify the blk_mq_get_request calling convention Christoph Hellwig
2020-05-18  6:39 ` [PATCH 4/9] blk-mq: merge blk_mq_rq_ctx_init into __blk_mq_alloc_request Christoph Hellwig
2020-05-18  6:39 ` [PATCH 5/9] blk-mq: don't set data->ctx and data->hctx in blk_mq_alloc_request_hctx Christoph Hellwig
2020-05-18  8:32   ` Thomas Gleixner
2020-05-18  9:31     ` Ming Lei
2020-05-18 10:42       ` Thomas Gleixner
2020-05-18 11:54         ` Ming Lei
2020-05-18 13:16           ` Christoph Hellwig
2020-05-18 14:11             ` Ming Lei
2020-05-18 16:56               ` Christoph Hellwig
2020-05-18 18:38                 ` Thomas Gleixner
2020-05-18 18:45                   ` Christoph Hellwig
2020-05-18 18:59                     ` Thomas Gleixner
2020-05-19  1:54                 ` Ming Lei
2020-05-19 15:30                   ` Christoph Hellwig
2020-05-20  1:18                     ` Ming Lei
2020-05-20  3:04                       ` Ming Lei
2020-05-20  8:03                         ` io_uring vs CPU hotplug, was " Christoph Hellwig
2020-05-20 14:45                           ` Jens Axboe
2020-05-20 15:20                             ` Jens Axboe
2020-05-20 15:31                               ` Christoph Hellwig
2020-05-20 19:41                               ` Thomas Gleixner
2020-05-20 20:18                                 ` Jens Axboe
2020-05-20 22:14                                   ` Thomas Gleixner
2020-05-20 22:40                                     ` Jens Axboe
2020-05-21  2:27                                     ` Ming Lei
2020-05-21  8:13                                       ` Thomas Gleixner
2020-05-21  9:23                                         ` Ming Lei
2020-05-21 18:39                                           ` Thomas Gleixner
2020-05-21 18:45                                             ` Jens Axboe
2020-05-21 20:00                                               ` Thomas Gleixner
2020-05-22  1:57                                             ` Ming Lei [this message]
2020-05-18 18:47             ` Thomas Gleixner
2020-05-18 13:18           ` Thomas Gleixner
2020-05-18  6:39 ` [PATCH 6/9] blk-mq: don't set data->ctx and data->hctx in __blk_mq_alloc_request Christoph Hellwig
2020-05-18  6:39 ` [PATCH 7/9] blk-mq: disable preemption during allocating request tag Christoph Hellwig
2020-05-18  6:39 ` [PATCH 8/9] blk-mq: add blk_mq_all_tag_iter Christoph Hellwig
2020-05-18  6:39 ` [PATCH 9/9] blk-mq: drain I/O when all CPUs in a hctx are offline Christoph Hellwig
2020-05-18  8:42   ` John Garry
2020-05-18  9:21     ` Ming Lei
2020-05-18 11:49 ` blk-mq: improvement CPU hotplug (simplified version) v2 John Garry
2020-05-19 15:30   ` Christoph Hellwig
2020-05-19 17:17     ` John Garry
2020-05-20 14:35     ` John Garry

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200522015719.GB755458@T590 \
    --to=ming.lei@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=hare@suse.com \
    --cc=hch@lst.de \
    --cc=io-uring@vger.kernel.org \
    --cc=john.garry@huawei.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-Block Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-block/0 linux-block/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-block linux-block/ https://lore.kernel.org/linux-block \
		linux-block@vger.kernel.org
	public-inbox-index linux-block

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-block


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git