From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 2/7] nvme: switch AEN processing to use blk_execute_rq_nowait
Date: Thu, 24 Sep 2015 15:38:50 +0200 [thread overview]
Message-ID: <20150924133850.GA6350@lst.de> (raw)
In-Reply-To: <alpine.LNX.2.00.1509232128470.23840@localhost.lm.intel.com>
On Wed, Sep 23, 2015@11:16:06PM +0000, Keith Busch wrote:
> AEN needs to release the request immediately. That's why it uses
> a "reserved" tag because it is tracked by the driver instead of
> the block layer: there is no guarantee the controller will ever
> return it, which deadlocks blk-mq's hot cpu notification. See
> 1efccc9ddb98fd533169669160201b027562af7e for details.
>
> Other than that, love the rest of the series.
Unfortunately we really need to keep the struct request around to
avoid special casing AENs in the completion path.
Yigal, is there a nice easy way to trigger the CPU hotplug migration
through sysfs or another userspace interface? I think I can solve
this without freeing the AEN early with a little more thought.
next prev parent reply other threads:[~2015-09-24 13:38 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-21 18:40 nvme completion path optimizations and fixes Christoph Hellwig
2015-09-21 18:40 ` [PATCH 1/7] blk-mq: fix racy updates of rq->errors Christoph Hellwig
2015-09-27 7:35 ` Sagi Grimberg
2015-09-21 18:40 ` [PATCH 2/7] nvme: switch AEN processing to use blk_execute_rq_nowait Christoph Hellwig
2015-09-23 23:16 ` Keith Busch
2015-09-24 13:38 ` Christoph Hellwig [this message]
2015-09-24 13:51 ` Keith Busch
2015-09-25 22:11 ` Christoph Hellwig
2015-09-21 18:40 ` [PATCH 3/7] nvme: switch delete SQ/CQ to blk_execute_rq_nowait Christoph Hellwig
2015-09-21 18:40 ` [PATCH 4/7] nvme: switch abort " Christoph Hellwig
2015-09-21 18:40 ` [PATCH 5/7] nvme: simplify completion handling Christoph Hellwig
2015-09-21 18:40 ` [PATCH 6/7] nvme: properly free resources for cancelled command Christoph Hellwig
2015-09-21 18:40 ` [PATCH 7/7] nvme: micro optimize nvme_submit_priv Christoph Hellwig
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=20150924133850.GA6350@lst.de \
--to=hch@lst.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 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).