Re: [PATCH v8 09/13] libsas: enforce eh strategy handlers only in eh context

On Thu, Mar 1, 2012 at 6:29 AM, James Bottomley
[off-list ref] wrote:

quoted

Hmm, what about something like:

  if (current != host->ehandler) {
    schedule_reset_to_run_in_eh_context():
    wait_for_eh();
  } else
    do_reset();

I think that would work for me ... as long as the wait doesn't cause
entangled deadlocks (I can't think of any at the moment, but I'll think
a bit more deeply about it).

So there is a deadlock due to:

commit d7a1bb0a04ca835bffc0a91e64ab827dfba7d8f5
Author: James Smart [off-list ref]
Date:   Wed Mar 8 14:50:12 2006 -0500

    [SCSI] Block I/O while SG reset operation in progress - the midlayer patch

    The scsi midlayer portion of the patch

    Signed-off-by: James Smart [off-list ref]
    Signed-off-by: James Bottomley [off-list ref]

...this adds shost->tmf_in_progress to scsi_host_in_recovery(), so I
can't "wait for eh" because the exit condition for that wait is
"!scsi_host_in_recovery()".  But since sg_reset is opened O_NONBLOCK
by default and that is likely the only non-eh reason for calling the
eh_{bus|device}_reset_handler routines I'll just make this routine
asynchronously queue a reset and return.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html