The locking in 9e79e12554d "[SCSI] pm8001: Fix for sata io circular
lock dependency." is wrong. We should revert this patch.
diff --git a/drivers/scsi/pm8001/pm8001_hwi.c b/drivers/scsi/pm8001/pm8001_hwi.c
index a3de306..68695b7 100644
--- a/drivers/scsi/pm8001/pm8001_hwi.c
+++ b/drivers/scsi/pm8001/pm8001_hwi.c
@@ -1901,7 +1901,7 @@ mpi_sata_completion(struct pm8001_hba_info *pm8001_ha, void *piomb)
{
struct sas_task *t;
struct pm8001_ccb_info *ccb;
- unsigned long flags;
+ unsigned long flags = 0;
u32 param;
u32 status;
u32 tag;
@@ -2040,7 +2040,9 @@ mpi_sata_completion(struct pm8001_hba_info *pm8001_ha, void *piomb)
ts->stat = SAS_QUEUE_FULL;
pm8001_ccb_task_free(pm8001_ha, t, ccb, tag);
mb();/*in order to force CPU ordering*/
+ spin_unlock_irqrestore(&pm8001_ha->lock, flags);
t->task_done(t);
+ spin_lock_irqsave(&pm8001_ha->lock, flags);
return;
}
break;
We can't just call spin_unlock_irqrestore(&pm8001_ha->lock, 0);
It makes no sense to call spin_lock_irqsave() and then return
immediately. We could call spin_lock_irq() if we don't plan to
restore the state of flags.
regards,
dan carpenter
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
