linux-scsi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: James Bottomley <jbottomley@parallels.com>
To: "bvanassche@acm.org" <bvanassche@acm.org>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
	"hch@infradead.org" <hch@infradead.org>,
	"hare@suse.de" <hare@suse.de>,
	"pbonzini@redhat.com" <pbonzini@redhat.com>,
	"axboe@kernel.dk" <axboe@kernel.dk>,
	"jdl1291@gmail.com" <jdl1291@gmail.com>
Subject: Re: [PATCH 1/3] Remove two cancel_delayed_work() calls from the error handler
Date: Tue, 27 May 2014 08:09:49 +0000	[thread overview]
Message-ID: <1401178172.14454.12.camel@dabdike> (raw)
In-Reply-To: <5384476F.6010705@acm.org>

On Tue, 2014-05-27 at 10:06 +0200, Bart Van Assche wrote:
> On 05/26/14 17:23, Paolo Bonzini wrote:
> > Il 26/05/2014 17:14, Bart Van Assche ha scritto:
> >> diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
> >> index 88d46fe..c972eab 100644
> >> --- a/drivers/scsi/scsi.c
> >> +++ b/drivers/scsi/scsi.c
> >> @@ -334,7 +334,7 @@ void scsi_put_command(struct scsi_cmnd *cmd)
> >>      list_del_init(&cmd->list);
> >>      spin_unlock_irqrestore(&cmd->device->list_lock, flags);
> >>
> >> -    cancel_delayed_work(&cmd->abort_work);
> >> +    WARN_ON_ONCE(delayed_work_pending(&cmd->abort_work));
> >>
> >>      __scsi_put_command(cmd->device->host, cmd);
> >>  }
> >> diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
> >> index f17aa7a..5232583 100644
> >> --- a/drivers/scsi/scsi_error.c
> >> +++ b/drivers/scsi/scsi_error.c
> >> @@ -193,7 +193,7 @@ scsi_abort_command(struct scsi_cmnd *scmd)
> >>          SCSI_LOG_ERROR_RECOVERY(3,
> >>              scmd_printk(KERN_INFO, scmd,
> >>                      "scmd %p previous abort failed\n", scmd));
> >> -        cancel_delayed_work(&scmd->abort_work);
> >> +        WARN_ON_ONCE(delayed_work_pending(&scmd->abort_work));
> >>          return FAILED;
> >>      }
> >>
> >>
> > 
> > I still prefer a BUG_ON in scsi_put_command, but anyway: series
> > 
> > Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
> 
> Hello Paolo,
> 
> As you probably know scsi_put_command() can get called from softirq
> context. A BUG_ON() in that context might make it unnecessary hard for a
> user to collect call traces.

Why?  The messages dumped are the same, the trace just starts from the
IRQ context ... I don't see what the problem is.

The question isn't ease of gathering the data, it's correctness.  The
point is that if the assert fails we have a free of an in-use command
leading to a nasty use after free ... the machine state is hosed at that
point.

James


  reply	other threads:[~2014-05-27  8:09 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-26 15:12 Make SCSI error handler code easier to understand Bart Van Assche
2014-05-26 15:14 ` [PATCH 1/3] Remove two cancel_delayed_work() calls from the error handler Bart Van Assche
2014-05-26 15:15   ` [PATCH 2/3] block: Introduce blk_rq_completed() Bart Van Assche
2014-05-26 15:27     ` James Bottomley
2014-05-27  7:49       ` Bart Van Assche
2014-05-27  7:52         ` hch
2014-05-27  8:00           ` James Bottomley
2014-05-27  8:23         ` James Bottomley
2014-05-27  9:00           ` Bart Van Assche
2014-05-27 10:21             ` James Bottomley
2014-05-27 10:47               ` Paolo Bonzini
2014-05-27 10:59                 ` James Bottomley
2014-05-27 11:13                   ` Paolo Bonzini
2014-05-27 11:26                     ` James Bottomley
2014-05-27 11:52                       ` Paolo Bonzini
2014-05-27 11:57                         ` James Bottomley
2014-05-27  5:40     ` Hannes Reinecke
2014-05-26 15:23   ` [PATCH 1/3] Remove two cancel_delayed_work() calls from the error handler Paolo Bonzini
2014-05-26 15:25     ` James Bottomley
2014-05-27  8:06     ` Bart Van Assche
2014-05-27  8:09       ` James Bottomley [this message]
2014-05-27  8:36         ` Bart Van Assche
2014-05-27  8:56           ` James Bottomley
2014-05-27  9:06             ` Paolo Bonzini
2014-05-27  5:40   ` Hannes Reinecke
2014-05-27  6:08     ` Bart Van Assche
2014-05-27  6:22       ` Hannes Reinecke
2014-05-26 15:15 ` [PATCH 3/3] Make SCSI error handler code easier to understand Bart Van Assche
2014-05-27  5:42   ` Hannes Reinecke
2014-05-28 20:15 ` Joe Lawrence
2014-05-29 11:33   ` James Bottomley

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=1401178172.14454.12.camel@dabdike \
    --to=jbottomley@parallels.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=hare@suse.de \
    --cc=hch@infradead.org \
    --cc=jdl1291@gmail.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=pbonzini@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 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).