linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 09/12] nvme: properly free resources for cancelled command
Date: Tue, 10 Nov 2015 09:13:57 +0100	[thread overview]
Message-ID: <20151110081357.GA21708@lst.de> (raw)
In-Reply-To: <20151109201232.GC5386@localhost.localdomain>

On Mon, Nov 09, 2015@08:12:33PM +0000, Keith Busch wrote:
> > Haha, so we at least can root cause this now.  Can you point me
> > to the caller that cares?  I'd really like to get rid of the special
> > case of passing a negative errno here, so I'd like to figure out how
> > we could pass this information on instead.
> 
> The "set_queue_count()" was the function that cared, but looks like
> patch 8/12 makes this think a timeout is an aborted status.

I'm still a bit confused on the semantics you want/need.  If the 
Set Features for set_queue_count times out we'll call the reset handler,
which because we are inside the probe handler will remove the device.
How do we care about the return value in that case?

Can you write down a few sentences on why/how we care?  I'll volunteer
to put them into the driver in comment form once we have all this sorted
out so that anyone touching the driver in the future won't be as confused.

  reply	other threads:[~2015-11-10  8:13 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-07  8:44 nvme completion path optimizations and fixes V3 Christoph Hellwig
2015-11-07  8:44 ` [PATCH 01/12] block: fix blk_abort_request for blk-mq drivers Christoph Hellwig
2015-11-07  8:44 ` [PATCH 02/12] block: defer timeouts to a workqueue Christoph Hellwig
2015-11-07  8:44 ` [PATCH 03/12] nvme: factor out a nvme_unmap_data helper Christoph Hellwig
2015-11-09 18:47   ` Keith Busch
2015-11-09 18:56     ` Christoph Hellwig
2015-11-07  8:44 ` [PATCH 04/12] nvme: factor out a few helpers from req_completion Christoph Hellwig
2015-11-07  8:44 ` [PATCH 05/12] nvme: switch delete SQ/CQ to blk_execute_rq_nowait Christoph Hellwig
2015-11-07  8:45 ` [PATCH 06/12] nvme: switch abort " Christoph Hellwig
2015-11-09 21:33   ` Keith Busch
2015-11-09 21:46     ` Jens Axboe
2015-11-10  5:56       ` Christoph Hellwig
2015-11-07  8:45 ` [PATCH 07/12] nvme: special case AEN requests Christoph Hellwig
2015-11-09 21:43   ` Keith Busch
2015-11-09 21:48     ` Jens Axboe
2015-11-10  5:57       ` Christoph Hellwig
2015-11-07  8:45 ` [PATCH 08/12] nvme: simplify completion handling Christoph Hellwig
2015-11-07  8:45 ` [PATCH 09/12] nvme: properly free resources for cancelled command Christoph Hellwig
2015-11-09 18:57   ` Keith Busch
2015-11-09 19:25     ` Christoph Hellwig
2015-11-09 20:12       ` Keith Busch
2015-11-10  8:13         ` Christoph Hellwig [this message]
2015-11-10 16:03           ` Keith Busch
2015-11-10 20:28             ` Keith Busch
2015-11-16 10:05               ` Christoph Hellwig
2015-11-07  8:45 ` [PATCH 10/12] nvme: meta_sg doesn't have to be an array Christoph Hellwig
2015-11-07  8:45 ` [PATCH 11/12] nvme: merge iod and cmd_info Christoph Hellwig
2015-11-07  8:45 ` [PATCH 12/12] block: remove REQ_NO_TIMEOUT flag 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=20151110081357.GA21708@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).