From: Bart Van Assche <bvanassche@acm.org>
To: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: linux-scsi <linux-scsi@vger.kernel.org>,
Joe Lawrence <jdl1291@gmail.com>, Tejun Heo <tj@kernel.org>,
Chanho Min <chanho.min@lge.com>,
David Milburn <dmilburn@redhat.com>,
Hannes Reinecke <hare@suse.de>,
Mike Christie <michaelc@cs.wisc.edu>
Subject: Re: [PATCH v11 6/9] Make scsi_remove_host() wait until error handling finished
Date: Tue, 25 Jun 2013 19:47:04 +0200 [thread overview]
Message-ID: <51C9D798.6000206@acm.org> (raw)
In-Reply-To: <1372182057.2806.43.camel@dabdike>
On 06/25/13 19:40, James Bottomley wrote:
> If I look at what we actually do: all the HBAs treat scsi_remove_host as
> a waited for transition. The reason this works is the loop over
> __scsi_remove_device() in scsi_forget_host(). By the time that loop
> returns, every scsi_device is gone (and so is every target). Because
> blk_cleanup_queue() induces a synchronous wait for the queue to die in
> __scsi_remove_device(), there can be no outstanding I/O and no eh
> activity for the device when it returns (and no possibility of starting
> any). Thus at the end of scsi_forget_host, we have no devices to start
> I/O and no eh activity, so the final put will be the last.
With the patch at the start of this thread everything works like you
described above. However, without that patch EH activity can continue
after scsi_remove_device() has returned. I have seen that occurring
several times while testing SCSI LLD patches.
Bart.
next prev parent reply other threads:[~2013-06-25 17:47 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-12 12:48 [PATCH v11 0/9] More device removal fixes Bart Van Assche
2013-06-12 12:49 ` [PATCH v11 1/9] Fix race between starved list and device removal Bart Van Assche
2013-06-24 15:38 ` James Bottomley
2013-06-24 16:16 ` Bart Van Assche
2013-06-24 16:23 ` James Bottomley
2013-06-24 17:24 ` Mike Christie
2013-06-24 17:49 ` James Bottomley
2013-06-12 12:51 ` [PATCH v11 2/9] Remove get_device() / put_device() pair from scsi_request_fn() Bart Van Assche
2013-06-24 1:29 ` Mike Christie
2013-06-24 2:36 ` James Bottomley
2013-06-24 7:13 ` Bart Van Assche
2013-06-24 13:34 ` James Bottomley
2013-06-24 15:43 ` Bart Van Assche
2013-06-12 12:52 ` [PATCH v11 3/9] Avoid calling __scsi_remove_device() twice Bart Van Assche
2013-06-23 21:35 ` Mike Christie
2013-06-24 6:29 ` Bart Van Assche
2013-06-24 17:38 ` James Bottomley
2013-06-25 8:37 ` Bart Van Assche
2013-06-25 13:44 ` James Bottomley
2013-06-25 15:23 ` Bart Van Assche
2013-06-12 12:53 ` [PATCH v11 4/9] Disallow changing the device state via sysfs into "deleted" Bart Van Assche
2013-06-24 1:05 ` Mike Christie
2013-06-24 6:35 ` Bart Van Assche
2013-06-24 17:59 ` James Bottomley
2013-06-25 8:41 ` Bart Van Assche
2013-06-25 13:42 ` James Bottomley
2013-06-12 12:54 ` [PATCH v11 5/9] Avoid saving/restoring interrupt state inside scsi_remove_host() Bart Van Assche
2013-06-24 1:06 ` Mike Christie
2013-06-12 12:55 ` [PATCH v11 6/9] Make scsi_remove_host() wait until error handling finished Bart Van Assche
2013-06-24 1:15 ` Mike Christie
2013-06-24 6:49 ` Bart Van Assche
2013-06-24 19:19 ` James Bottomley
2013-06-24 20:04 ` Mike Christie
2013-06-24 22:27 ` James Bottomley
2013-06-25 2:26 ` Mike Christie
2013-06-25 2:56 ` Michael Christie
2013-06-25 9:01 ` Bart Van Assche
2013-06-25 13:45 ` James Bottomley
2013-06-25 15:31 ` Bart Van Assche
2013-06-25 16:13 ` Michael Christie
2013-06-25 17:40 ` James Bottomley
2013-06-25 17:47 ` Bart Van Assche [this message]
2014-01-30 19:46 ` Bart Van Assche
2014-01-31 5:58 ` James Bottomley
2014-01-31 7:52 ` Bart Van Assche
2013-06-25 11:13 ` Bart Van Assche
2013-06-12 12:56 ` PATCH v11 7/9] Avoid that scsi_device_set_state() triggers a race Bart Van Assche
2013-06-12 12:57 ` [PATCH v11 8/9] Save and restore host_scribble during error handling Bart Van Assche
2013-06-24 1:21 ` Mike Christie
2013-06-24 2:08 ` James Bottomley
2013-06-12 12:58 ` [PATCH v11 9/9] Avoid reenabling I/O after the transport became offline Bart Van Assche
-- strict thread matches above, loose matches on Subject: below --
2013-06-24 10:17 RE:[PATCH v11 6/9] Make scsi_remove_host() wait until error handling finished Jack Wang
2013-06-24 10:53 ` [PATCH " 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=51C9D798.6000206@acm.org \
--to=bvanassche@acm.org \
--cc=James.Bottomley@HansenPartnership.com \
--cc=chanho.min@lge.com \
--cc=dmilburn@redhat.com \
--cc=hare@suse.de \
--cc=jdl1291@gmail.com \
--cc=linux-scsi@vger.kernel.org \
--cc=michaelc@cs.wisc.edu \
--cc=tj@kernel.org \
/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).