All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mike Snitzer <snitzer@redhat.com>
To: Mikulas Patocka <mpatocka@redhat.com>
Cc: dm-devel@redhat.com, "Alasdair G. Kergon" <agk@redhat.com>
Subject: Re: dm-mpath: do not change SCSI device handler
Date: Wed, 3 Apr 2013 09:32:34 -0400	[thread overview]
Message-ID: <20130403133234.GA2197@redhat.com> (raw)
In-Reply-To: <Pine.LNX.4.64.1304022000390.27554@file.rdu.redhat.com>

On Tue, Apr 02 2013 at  8:04pm -0400,
Mikulas Patocka <mpatocka@redhat.com> wrote:

> Hi
> 
> This fixes BZ 912245 and 902595.
> 
> Mikulas
> 
> ---
> 
> dm-mpath: do not change SCSI device handler
> 
> This patch prevents the multipath target from changing the device handler.
> This fixes a kernel crash that can happen when changing the device
> handler.
> 
> When we reload a multipath device, there are two instances of the
> multipath target - the first instance that is active and the second
> instance that is being constructed with "ctr" method.
> 
> If the multipath constructor finds out that the device is using a
> different device handler, it detaches the existing handler and attaches a
> new handler. However, the first instance of the multipath target still
> exists and processes requests. If the first instance sends some
> path-management request with scsi_dh_activate and the second instance
> detaches the device handler while the path-management request is in
> flight, a crash happens. The reason for the crash is that the endio
> routine for the path-management request is working with structures that
> were freed when the handler was detached.
> 
> There is no practical need to change device handlers on an active device,
> so this patch disables it.
> 
> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
> 
> ---
>  drivers/md/dm-mpath.c |   13 +++++++------
>  1 file changed, 7 insertions(+), 6 deletions(-)
> 
> Index: linux-3.9-rc5-fast/drivers/md/dm-mpath.c
> ===================================================================
> --- linux-3.9-rc5-fast.orig/drivers/md/dm-mpath.c	2013-04-02 21:54:25.000000000 +0200
> +++ linux-3.9-rc5-fast/drivers/md/dm-mpath.c	2013-04-03 01:15:04.000000000 +0200
> @@ -618,12 +618,13 @@ static struct pgpath *parse_path(struct 
>  		 */
>  		r = scsi_dh_attach(q, m->hw_handler_name);
>  		if (r == -EBUSY) {
> -			/*
> -			 * Already attached to different hw_handler:
> -			 * try to reattach with correct one.
> -			 */
> -			scsi_dh_detach(q);
> -			r = scsi_dh_attach(q, m->hw_handler_name);
> +			ti->error = "a different device handler is already "
> +				"attached";

Don't really need to wrap this to multiple lines (helps searching).

> +			DMERR("A different device handler for device %s is "
> +				"attached. You need to deactivate "
> +				"the device to change device handler.",
> +				p->path.dev->name);
> +			goto bad;
>  		}
>  
>  		if (r < 0) {

We should include the currently attached scsi_dh in the DMERR message,
e.g.:
	attached_handler_name = scsi_dh_attached_handler_name(q, GFP_KERNEL);
	DMERR("The %s device handler for devices %s is already attached.  ...",
	      attached_handler_name, p->path.dev->name);
	kfree(attached_handler_name);
	goto bad;

Otherwise, looks good.

Acked-by: Mike Snitzer <snitzer@redhat.com>

  reply	other threads:[~2013-04-03 13:32 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-03  0:04 [PATCH] dm-mpath: do not change SCSI device handler Mikulas Patocka
2013-04-03 13:32 ` Mike Snitzer [this message]
2013-04-03 20:54   ` Mikulas Patocka
2013-04-04  6:47 ` [PATCH] " Hannes Reinecke
2013-04-04 12:24   ` Mike Snitzer
2013-04-04 12:55     ` Mikulas Patocka
2013-04-04 13:16       ` Mike Snitzer
2013-04-04 13:36         ` Mikulas Patocka
2013-04-04 14:20           ` Mike Snitzer
2013-04-04 15:13             ` Mikulas Patocka
2013-04-04 15:38               ` Mikulas Patocka
2013-04-08 21:50         ` [PATCH 1/2] [SCSI] scsi_dh: add scsi_dh_alloc_data Mike Snitzer
2013-04-08 21:50           ` [PATCH 2/2] dm mpath: attach scsi_dh during table resume Mike Snitzer
2013-04-22 22:33             ` Mike Snitzer
2013-04-25 13:48               ` Mikulas Patocka
2013-04-25 14:17                 ` Mike Snitzer
2013-04-25 14:50                   ` Mikulas Patocka
2013-04-25 15:27                     ` Bryn M. Reeves
2013-04-25 15:37                       ` Mike Snitzer
2013-04-25 15:44                         ` Bryn M. Reeves
2013-04-25 15:31                     ` Mike Snitzer
2013-04-26  6:05                       ` Hannes Reinecke
2013-04-26 13:29                         ` Mike Snitzer

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=20130403133234.GA2197@redhat.com \
    --to=snitzer@redhat.com \
    --cc=agk@redhat.com \
    --cc=dm-devel@redhat.com \
    --cc=mpatocka@redhat.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.