From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dan Williams Subject: [PATCH 03/18] scsi: fix eh wakeup (scsi_schedule_eh vs scsi_restart_operations) Date: Sun, 06 May 2012 11:18:12 -0700 Message-ID: <20120506181812.17912.19252.stgit@dwillia2-linux.jf.intel.com> References: <20120506181611.17912.3798.stgit@dwillia2-linux.jf.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20120506181611.17912.3798.stgit@dwillia2-linux.jf.intel.com> Sender: linux-ide-owner@vger.kernel.org To: linux-scsi@vger.kernel.org Cc: Tejun Heo , linux-ide@vger.kernel.org, Tom Jackson List-Id: linux-scsi@vger.kernel.org Rapid ata hotplug on a libsas controller results in cases where libsas is waiting indefinitely on eh to perform an ata probe. A race exists between scsi_schedule_eh() and scsi_restart_operations() in the case when scsi_restart_operations() issues i/o to other devices in the sas domain. When this happens the host state transitions from SHOST_RECOVERY (set by scsi_schedule_eh) back to SHOST_RUNNING and ->host_busy is non-zero so we put the eh thread to sleep even though ->host_eh_scheduled is active. Before putting the error handler to sleep we need to check if the host_state needs to return to SHOST_RECOVERY for another trip through eh. Cc: Tejun Heo Reported-by: Tom Jackson Tested-by: Tom Jackson Signed-off-by: Dan Williams --- drivers/scsi/scsi_error.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 386f0c5..cc8dc8c 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1687,6 +1687,20 @@ static void scsi_restart_operations(struct Scsi_Host *shost) * requests are started. */ scsi_run_host_queues(shost); + + /* + * if eh is active and host_eh_scheduled is pending we need to re-run + * recovery. we do this check after scsi_run_host_queues() to allow + * everything pent up since the last eh run a chance to make forward + * progress before we sync again. Either we'll immediately re-run + * recovery or scsi_device_unbusy() will wake us again when these + * pending commands complete. + */ + spin_lock_irqsave(shost->host_lock, flags); + if (shost->host_eh_scheduled) + if (scsi_host_set_state(shost, SHOST_RECOVERY)) + WARN_ON(scsi_host_set_state(shost, SHOST_CANCEL_RECOVERY)); + spin_unlock_irqrestore(shost->host_lock, flags); } /**