All of lore.kernel.org
 help / color / mirror / Atom feed
From: John Garry <john.garry@huawei.com>
To: Christoph Hellwig <hch@lst.de>
Cc: <linux-block@vger.kernel.org>,
	Bart Van Assche <bvanassche@acm.org>,
	Hannes Reinecke <hare@suse.com>,
	Thomas Gleixner <tglx@linutronix.de>
Subject: Re: blk-mq: improvement CPU hotplug (simplified version) v4
Date: Thu, 28 May 2020 09:29:40 +0100	[thread overview]
Message-ID: <b98f055f-6f38-a47c-965d-b6bcf4f5563f@huawei.com> (raw)
In-Reply-To: <20200527180644.514302-1-hch@lst.de>

On 27/05/2020 19:06, Christoph Hellwig wrote:
> Hi all,
> 
> this series ensures I/O is quiesced before a cpu and thus the managed
> interrupt handler is shut down.
> 
> This patchset tries to address the issue by the following approach:
> 
>   - before the last cpu in hctx->cpumask is going to offline, mark this
>     hctx as inactive
> 
>   - disable preempt during allocating tag for request, and after tag is
>     allocated, check if this hctx is inactive. If yes, give up the
>     allocation and try remote allocation from online CPUs
> 
>   - before hctx becomes inactive, drain all allocated requests on this
>     hctx
> 
> The guts of the changes are from Ming Lei, I just did a bunch of prep
> cleanups so that they can fit in more nicely.  The series also depends
> on my "avoid a few q_usage_counter roundtrips v3" series.
> 
> Thanks John Garry for running lots of tests on arm64 with this previous
> version patches and co-working on investigating all kinds of issues.
> 
> A git tree is available here:
> 
>      git://git.infradead.org/users/hch/block.git blk-mq-hotplug.3
> 
> Gitweb:
> 
>      http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/blk-mq-hotplug.3
> 
> Changes since v3:
>    - add two new patches to clean up the magic -1 tag values
>    - improve a few commit messages and comments
>    - cleanup the blk_mq_all_tag_iter implementation
>    - add a msleep to the cpu hot unplug case in __blk_mq_alloc_request
> 
> Changes since v2:
>    - don't disable preemption and use smp calls
> 

I tested this again, so:
Tested-by: John Garry <john.garry@huawei.com>

As an aside, I'm not familiar with blktests, but it may be possible to 
add something to test this. I'll look.

Cheers

  parent reply	other threads:[~2020-05-28  8:30 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-27 18:06 blk-mq: improvement CPU hotplug (simplified version) v4 Christoph Hellwig
2020-05-27 18:06 ` [PATCH 1/8] blk-mq: remove the bio argument to ->prepare_request Christoph Hellwig
2020-05-27 18:16   ` Johannes Thumshirn
2020-05-27 18:06 ` [PATCH 2/8] blk-mq: simplify the blk_mq_get_request calling convention Christoph Hellwig
2020-05-27 18:17   ` Johannes Thumshirn
2020-05-27 18:06 ` [PATCH 3/8] blk-mq: move more request initialization to blk_mq_rq_ctx_init Christoph Hellwig
2020-05-27 18:16   ` Hannes Reinecke
2020-05-28  9:50   ` Johannes Thumshirn
2020-05-27 18:06 ` [PATCH 4/8] blk-mq: rename BLK_MQ_TAG_FAIL to BLK_MQ_NO_TAG Christoph Hellwig
2020-05-27 18:14   ` Johannes Thumshirn
2020-05-27 18:17   ` Hannes Reinecke
2020-05-27 22:38   ` Bart Van Assche
2020-05-27 18:06 ` [PATCH 5/8] blk-mq: use BLK_MQ_NO_TAG in more places Christoph Hellwig
2020-05-27 18:15   ` Johannes Thumshirn
2020-05-27 18:18   ` Hannes Reinecke
2020-05-27 22:38   ` Bart Van Assche
2020-05-27 18:06 ` [PATCH 6/8] blk-mq: open code __blk_mq_alloc_request in blk_mq_alloc_request_hctx Christoph Hellwig
2020-05-27 18:06 ` [PATCH 7/8] blk-mq: add blk_mq_all_tag_iter Christoph Hellwig
2020-05-27 18:21   ` Hannes Reinecke
2020-05-27 22:52   ` Bart Van Assche
2020-05-27 18:06 ` [PATCH 8/8] blk-mq: drain I/O when all CPUs in a hctx are offline Christoph Hellwig
2020-05-27 18:26   ` Hannes Reinecke
2020-05-27 23:09   ` Bart Van Assche
2020-05-28  1:46     ` Ming Lei
2020-05-28  3:33       ` Bart Van Assche
2020-05-28  5:19         ` Ming Lei
2020-05-28 13:37           ` Bart Van Assche
2020-05-28 17:21             ` Paul E. McKenney
2020-05-29  1:53               ` Ming Lei
2020-05-29  3:07                 ` Paul E. McKenney
2020-05-29  3:53                   ` Ming Lei
2020-05-29 18:13                     ` Paul E. McKenney
2020-05-29 19:55                       ` Bart Van Assche
2020-05-29 21:12                         ` Paul E. McKenney
2020-05-29  1:13             ` Ming Lei
2020-05-27 20:07 ` blk-mq: improvement CPU hotplug (simplified version) v4 Bart Van Assche
2020-05-27 20:31   ` John Garry
2020-05-29 13:26     ` Christoph Hellwig
2020-05-28  8:29 ` John Garry [this message]
2020-05-29 13:53 Christoph Hellwig
2020-05-29 16:23 ` Jens Axboe

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=b98f055f-6f38-a47c-965d-b6bcf4f5563f@huawei.com \
    --to=john.garry@huawei.com \
    --cc=bvanassche@acm.org \
    --cc=hare@suse.com \
    --cc=hch@lst.de \
    --cc=linux-block@vger.kernel.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
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.