All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bart Van Assche <bvanassche@acm.org>
To: Jianchao Wang <jianchao.w.wang@oracle.com>,
	axboe@kernel.dk, ming.lei@redhat.com, bart.vanassche@wdc.com,
	sagi@grimberg.me, keith.busch@intel.com, jthumshirn@suse.de,
	jsmart2021@gmail.com
Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-nvme@lists.infradead.org
Subject: Re: [PATCH 2/3] blk-core: introduce queue close feature
Date: Wed, 05 Sep 2018 08:57:17 -0700	[thread overview]
Message-ID: <1536163037.11534.6.camel@acm.org> (raw)
In-Reply-To: <1536120586-3378-3-git-send-email-jianchao.w.wang@oracle.com>

On Wed, 2018-09-05 at 12:09 +0800, Jianchao Wang wrote:
> blk queue freeze is often used to prevent new IO from entering
> request queue. However, becuase we kill the percpu-ref
> q_usage_counter when freeze queue, we have to drain the request
> queue when unfreeze. This is unnecessary for just preventing new
> IO. In addition, If there is IO timeout or other issue when unfreeze
> the queue, the scenario could be very tricky.
> 
> So we introduce BLK_QUEUE_GATE_CLOSED to implement a light-weight
> queue close feature base on the queue_gate to prevent new IO from
> comming in queue which will not need to drain the queue any more.

Does the "queue gate close" feature cause blk_get_request() /
blk_mq_get_request() to block until blk_clear_queue_closed() is called? If
so, I think we need a better name for this feature. How about calling these
two operations suspend and resume?

Thanks,

Bart.

WARNING: multiple messages have this Message-ID (diff)
From: bvanassche@acm.org (Bart Van Assche)
Subject: [PATCH 2/3] blk-core: introduce queue close feature
Date: Wed, 05 Sep 2018 08:57:17 -0700	[thread overview]
Message-ID: <1536163037.11534.6.camel@acm.org> (raw)
In-Reply-To: <1536120586-3378-3-git-send-email-jianchao.w.wang@oracle.com>

On Wed, 2018-09-05@12:09 +0800, Jianchao Wang wrote:
> blk queue freeze is often used to prevent new IO from entering
> request queue. However, becuase we kill the percpu-ref
> q_usage_counter when freeze queue, we have to drain the request
> queue when unfreeze. This is unnecessary for just preventing new
> IO. In addition, If there is IO timeout or other issue when unfreeze
> the queue, the scenario could be very tricky.
> 
> So we introduce BLK_QUEUE_GATE_CLOSED to implement a light-weight
> queue close feature base on the queue_gate to prevent new IO from
> comming in queue which will not need to drain the queue any more.

Does the "queue gate close" feature cause blk_get_request() /
blk_mq_get_request() to block until blk_clear_queue_closed() is called? If
so, I think we need a better name for this feature. How about calling these
two operations suspend and resume?

Thanks,

Bart.

  reply	other threads:[~2018-09-05 15:57 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-05  4:09 [PATCH 0/3] Introduce a light-weight queue close feature Jianchao Wang
2018-09-05  4:09 ` Jianchao Wang
2018-09-05  4:09 ` [PATCH 1/3] blk-core: migrate preempt-only mode to queue_gate Jianchao Wang
2018-09-05  4:09   ` Jianchao Wang
2018-09-05  4:09 ` [PATCH 2/3] blk-core: introduce queue close feature Jianchao Wang
2018-09-05  4:09   ` Jianchao Wang
2018-09-05 15:57   ` Bart Van Assche [this message]
2018-09-05 15:57     ` Bart Van Assche
2018-09-06  1:31     ` jianchao.wang
2018-09-06  1:31       ` jianchao.wang
2018-09-05  4:09 ` [PATCH 3/3] nvme-pci: use queue close instead of queue freeze Jianchao Wang
2018-09-05  4:09   ` Jianchao Wang
2018-09-05 22:09   ` Ming Lei
2018-09-05 22:09     ` Ming Lei
2018-09-06  1:28     ` jianchao.wang
2018-09-06  1:28       ` jianchao.wang
2018-09-06 13:07       ` Ming Lei
2018-09-06 13:07         ` Ming Lei
2018-09-06 14:19         ` jianchao.wang
2018-09-06 14:19           ` jianchao.wang
2018-09-07 16:23           ` Ming Lei
2018-09-07 16:23             ` Ming Lei
2018-09-10  1:49             ` jianchao.wang
2018-09-10  1:49               ` jianchao.wang
2018-09-05 15:48 ` [PATCH 0/3] Introduce a light-weight queue close feature Bart Van Assche
2018-09-05 15:48   ` Bart Van Assche
2018-09-06  1:35   ` jianchao.wang
2018-09-06  1:35     ` jianchao.wang
2018-09-05 21:27 ` Ming Lei
2018-09-05 21:27   ` Ming Lei
2018-09-06  1:51   ` jianchao.wang
2018-09-06  1:51     ` jianchao.wang
2018-09-06 12:57     ` Ming Lei
2018-09-06 12:57       ` Ming Lei
2018-09-06 13:55       ` jianchao.wang
2018-09-06 13:55         ` jianchao.wang
2018-09-07 16:21         ` Ming Lei
2018-09-07 16:21           ` Ming Lei

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=1536163037.11534.6.camel@acm.org \
    --to=bvanassche@acm.org \
    --cc=axboe@kernel.dk \
    --cc=bart.vanassche@wdc.com \
    --cc=jianchao.w.wang@oracle.com \
    --cc=jsmart2021@gmail.com \
    --cc=jthumshirn@suse.de \
    --cc=keith.busch@intel.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=ming.lei@redhat.com \
    --cc=sagi@grimberg.me \
    /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.