From 57fc2e335fd3c2f898ee73570dc81426c28dc7b4 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 21 Jun 2012 23:25:32 -0700 Subject: [SCSI] fix eh wakeup (scsi_schedule_eh vs scsi_restart_operations) 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. Since i/o that is released by scsi_restart_operations has been blocked for at least one eh cycle, this implementation allows those i/o's to run before another eh cycle starts to discourage hung task timeouts. Cc: Reported-by: Tom Jackson Tested-by: Tom Jackson Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index d0f71e5d065f..804f63270e37 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); } /** -- cgit v1.2.3-55-g7522 From b9d5c6b7ef570bea0d22746944d7b58fa7f17b13 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 21 Jun 2012 23:25:42 -0700 Subject: [SCSI] cleanup setting task state in scsi_error_handler() A quick reading of scsi_error_handler() one could come away with the impression that it does its wakeup event check while the task state is TASK_RUNNING. In fact it sets TASK_INTERRUPTIBLE at the bottom of the loop, but that is ~50 lines down. Just set TASK_INTERRUPTIBLE at the top of loop and be done. Signed-off-by: Dan Williams Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 804f63270e37..4a6381c87253 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1818,15 +1818,14 @@ int scsi_error_handler(void *data) * We never actually get interrupted because kthread_run * disables signal delivery for the created thread. */ - set_current_state(TASK_INTERRUPTIBLE); while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); if ((shost->host_failed == 0 && shost->host_eh_scheduled == 0) || shost->host_failed != shost->host_busy) { SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler scsi_eh_%d sleeping\n", shost->host_no)); schedule(); - set_current_state(TASK_INTERRUPTIBLE); continue; } @@ -1863,7 +1862,6 @@ int scsi_error_handler(void *data) scsi_restart_operations(shost); if (!shost->eh_noresume) scsi_autopm_put_host(shost); - set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); -- cgit v1.2.3-55-g7522