All of lore.kernel.org
 help / color / mirror / Atom feed
From: keith.busch@intel.com (Keith Busch)
Subject: [PATCH] NVMe: Avoid interrupt disable during queue init.
Date: Fri, 22 May 2015 14:48:13 +0000 (UTC)	[thread overview]
Message-ID: <alpine.LNX.2.00.1505221446410.15930@localhost.lm.intel.com> (raw)
In-Reply-To: <CAGgvQNQHaZBRzedkLmYxCBtDyxq7j2wr=e0vAYfw5r+KKeVbxg@mail.gmail.com>

On Thu, 21 May 2015, Parav Pandit wrote:
> On Fri, May 22, 2015@1:04 AM, Keith Busch <keith.busch@intel.com> wrote:
>> The q_lock is held to protect polling from reading inconsistent data.
>
> ah, yes. I can see the nvme_kthread can poll the CQ while its getting
> created through the nvme_resume().
> I think this opens up other issue.
>
> nvme_kthread() should,
>
> Instead of,
> struct nvme_queue *nvmeq = dev->queues[i];
>
> it should do,
> struct nvme_queue *nvmeq = rcu_dereference(dev->queues[i]);

The rcu protection on nvme queues was removed with the blk-mq conversion
as we rely on that layer for h/w access.

WARNING: multiple messages have this Message-ID (diff)
From: Keith Busch <keith.busch@intel.com>
To: Parav Pandit <parav.pandit@avagotech.com>
Cc: Keith Busch <keith.busch@intel.com>,
	linux-nvme@lists.infradead.org,
	Matthew Wilcox <willy@linux.intel.com>,
	Jens Axboe <axboe@kernel.dk>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] NVMe: Avoid interrupt disable during queue init.
Date: Fri, 22 May 2015 14:48:13 +0000 (UTC)	[thread overview]
Message-ID: <alpine.LNX.2.00.1505221446410.15930@localhost.lm.intel.com> (raw)
In-Reply-To: <CAGgvQNQHaZBRzedkLmYxCBtDyxq7j2wr=e0vAYfw5r+KKeVbxg@mail.gmail.com>

On Thu, 21 May 2015, Parav Pandit wrote:
> On Fri, May 22, 2015 at 1:04 AM, Keith Busch <keith.busch@intel.com> wrote:
>> The q_lock is held to protect polling from reading inconsistent data.
>
> ah, yes. I can see the nvme_kthread can poll the CQ while its getting
> created through the nvme_resume().
> I think this opens up other issue.
>
> nvme_kthread() should,
>
> Instead of,
> struct nvme_queue *nvmeq = dev->queues[i];
>
> it should do,
> struct nvme_queue *nvmeq = rcu_dereference(dev->queues[i]);

The rcu protection on nvme queues was removed with the blk-mq conversion
as we rely on that layer for h/w access.

  reply	other threads:[~2015-05-22 14:48 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-22  0:12 [PATCH] NVMe: Avoid interrupt disable during queue init Parav Pandit
2015-05-22  0:12 ` Parav Pandit
2015-05-21 18:39 ` Jens Axboe
2015-05-21 18:39   ` Jens Axboe
2015-05-21 19:14   ` Parav Pandit
2015-05-21 19:14     ` Parav Pandit
2015-05-21 19:34 ` Keith Busch
2015-05-21 19:34   ` Keith Busch
2015-05-22  4:15   ` Parav Pandit
2015-05-22  4:15     ` Parav Pandit
2015-05-22 14:48     ` Keith Busch [this message]
2015-05-22 14:48       ` Keith Busch
2015-05-22 14:51       ` Parav Pandit
2015-05-22 14:51         ` Parav Pandit
2015-05-22 15:11         ` Keith Busch
2015-05-22 15:11           ` Keith Busch
2015-05-22 16:03           ` Parav Pandit
2015-05-22 16:03             ` Parav Pandit
2015-05-22 16:23             ` Keith Busch
2015-05-22 16:23               ` Keith Busch
2015-05-22 16:48               ` Parav Pandit
2015-05-22 16:48                 ` Parav Pandit
2015-05-22 17:07                 ` Keith Busch
2015-05-22 17:07                   ` Keith Busch
2015-05-22 17:33                   ` Parav Pandit
2015-05-22 17:33                     ` Parav Pandit
2015-05-22 17:47                     ` Keith Busch
2015-05-22 17:47                       ` Keith Busch
2015-05-22 18:18                       ` Parav Pandit
2015-05-22 18:18                         ` Parav Pandit

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=alpine.LNX.2.00.1505221446410.15930@localhost.lm.intel.com \
    --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 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.