linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: keith.busch@intel.com (Keith Busch)
Subject: [PATCH 5/5] NVMe: IO queue deletion re-write
Date: Sat, 2 Jan 2016 21:30:09 +0000	[thread overview]
Message-ID: <20160102213008.GA10969@localhost.localdomain> (raw)
In-Reply-To: <20160102170730.GA30184@infradead.org>

On Sat, Jan 02, 2016@09:07:30AM -0800, Christoph Hellwig wrote:
> Just curious:  why do we even both with the async queue shutdown.  At
> least the controllers I have access to complete queue deletions very
> quickly.

The async deletion was written for a bug reporting "hang" on a device
removal. The "hang" was the controller taking on the order of 100's msec
to delete a queue (sometimes >1sec if lots of commands queued). This
controller had 2k queues, and took ~15 minutes to remove serially. Async
deletion brought it down to ~20 seconds, so looked like a good idea.

It wasn't a controller I make, so I personally don't care about
parallelizing queue deletion. The driver's been this way for so long
though, I don't have a good way to know how beneficial this feature
is anymore.

  reply	other threads:[~2016-01-02 21:30 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-30 17:27 [PATCH 0/5] NVMe fixes and updates for 4.5 Keith Busch
2015-12-30 17:27 ` [PATCH 1/5] NVMe: Fix admin queue ring wrap Keith Busch
2015-12-30 17:49   ` Christoph Hellwig
2015-12-30 17:56     ` Keith Busch
2015-12-30 17:53   ` Jens Axboe
2015-12-30 18:09     ` Christoph Hellwig
2015-12-30 17:27 ` [PATCH 2/5] NVMe: Use a retryable error code on reset Keith Busch
2015-12-30 17:52   ` Christoph Hellwig
2015-12-30 18:09     ` Keith Busch
2015-12-30 18:18       ` Christoph Hellwig
2015-12-30 17:27 ` [PATCH 3/5] NVMe: Remove queue freezing on resets Keith Busch
2015-12-30 17:53   ` Christoph Hellwig
2015-12-30 20:44     ` Sagi Grimberg
2015-12-30 20:42   ` Sagi Grimberg
2015-12-31 17:19     ` Sagi Grimberg
2015-12-30 17:27 ` [PATCH 4/5] NVMe: Shutdown controller only for power-off Keith Busch
2015-12-30 17:58   ` Christoph Hellwig
2015-12-30 18:02     ` Keith Busch
2015-12-30 18:14       ` Christoph Hellwig
2015-12-30 17:27 ` [PATCH 5/5] NVMe: IO queue deletion re-write Keith Busch
2015-12-30 18:04   ` Christoph Hellwig
2015-12-30 19:07     ` Keith Busch
2016-01-02 17:07       ` Christoph Hellwig
2016-01-02 21:30         ` Keith Busch [this message]
2016-01-03 11:40           ` Christoph Hellwig
2016-01-03 15:43             ` Keith Busch
2016-01-03 16:17               ` Christoph Hellwig
2016-01-03 16:26                 ` Keith Busch
2016-01-03 18:04                   ` Keith Busch
2016-01-03 17:05             ` Sagi Grimberg

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=20160102213008.GA10969@localhost.localdomain \
    --to=keith.busch@intel.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).