public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
From: Bart Van Assche <Bart.VanAssche@sandisk.com>
To: "James.Bottomley@HansenPartnership.com"
	<James.Bottomley@HansenPartnership.com>,
	"bblock@linux.vnet.ibm.com" <bblock@linux.vnet.ibm.com>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
	"maxg@mellanox.com" <maxg@mellanox.com>,
	"israelr@mellanox.com" <israelr@mellanox.com>,
	"hare@suse.de" <hare@suse.de>,
	"martin.petersen@oracle.com" <martin.petersen@oracle.com>
Subject: Re: [PATCH v3 3/4] sd: Make synchronize cache upon shutdown asynchronous
Date: Thu, 20 Apr 2017 22:52:22 +0000	[thread overview]
Message-ID: <1492728740.2642.14.camel@sandisk.com> (raw)
In-Reply-To: <1492726397.21601.16.camel@HansenPartnership.com>

On Thu, 2017-04-20 at 15:13 -0700, James Bottomley wrote:
> How is that possible?  Once the device goes into the CANCEL state, it
> no longer can be found by starget_for_each_device() because
> scsi_device_get() returns NULL ...

scsi_target_block() is not serialized against __scsi_remove_device(). I think
the following sequence of events can cause a queue to be stopped for a device
in the CANCEL state:
(a) scsi_target_block() triggers a call to scsi_get_device().
(b) __scsi_remove_device() is called from the context of another thread.
(c) __scsi_remove_device() changes the device state into SDEV_CANCEL.
(d) scsi_internal_device_block() calls blk_mq_stop_hw_queue().

Bart.

  parent reply	other threads:[~2017-04-20 22:52 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-17 17:34 [PATCH v3 0/4] Avoid that __scsi_remove_device() hangs Bart Van Assche
2017-04-17 17:34 ` [PATCH v3 1/4] Introduce scsi_start_queue() Bart Van Assche
2017-04-17 17:34 ` [PATCH v3 2/4] Introduce scsi_execute_async() Bart Van Assche
2017-04-17 17:34 ` [PATCH v3 3/4] sd: Make synchronize cache upon shutdown asynchronous Bart Van Assche
2017-04-18 14:44   ` Benjamin Block
2017-04-18 15:34     ` Bart Van Assche
2017-04-18 15:56     ` James Bottomley
2017-04-18 16:06       ` Bart Van Assche
2017-04-18 23:47       ` Bart Van Assche
2017-04-18 23:56         ` James Bottomley
2017-04-19  0:02           ` Bart Van Assche
2017-04-19  0:05             ` James Bottomley
2017-04-19 18:42           ` Bart Van Assche
2017-04-20 21:59           ` Bart Van Assche
2017-04-20 22:13             ` James Bottomley
2017-04-20 22:27               ` Bart Van Assche
2017-04-20 22:52               ` Bart Van Assche [this message]
2017-04-23 17:28                 ` James Bottomley
2017-04-24 21:46                   ` Bart Van Assche
2017-04-26 18:53       ` Bart Van Assche
2017-04-28 18:45         ` James Bottomley
2017-04-24  7:14   ` [lkp-robot] [sd] ab1218235c: INFO:possible_recursive_locking_detected kernel test robot
2017-04-17 17:34 ` [PATCH v3 4/4] Avoid that __scsi_remove_device() hangs Bart Van Assche
2017-04-18 11:58 ` [PATCH v3 0/4] " Israel Rukshin
2017-04-18 15:40   ` Bart Van Assche

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=1492728740.2642.14.camel@sandisk.com \
    --to=bart.vanassche@sandisk.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=bblock@linux.vnet.ibm.com \
    --cc=hare@suse.de \
    --cc=israelr@mellanox.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=maxg@mellanox.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