Here's a fix for hotplug events. The useage of queue_delayed_work seems
to broke the fifo for processing of firmware events. After several iterations
of adding and removing cabling connected to jbods, the devices are not
getting added becuase kernel thread is activited out of order.
Signed-off-by: Eric Moore <eric.moore@xxxxxxx>
diff -uaprN a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
--- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c 2009-04-13 15:58:57.000000000 -0600
+++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c 2009-04-21 15:09:21.000000000 -0600
@@ -119,7 +119,7 @@ struct sense_info {
*/
struct fw_event_work {
struct list_head list;
- struct delayed_work work;
+ struct work_struct work;
struct MPT2SAS_ADAPTER *ioc;
u8 VF_ID;
u8 host_reset_handling;
@@ -2007,8 +2007,8 @@ _scsih_fw_event_add(struct MPT2SAS_ADAPT
spin_lock_irqsave(&ioc->fw_event_lock, flags);
list_add_tail(&fw_event->list, &ioc->fw_event_list);
- INIT_DELAYED_WORK(&fw_event->work, _firmware_event_work);
- queue_delayed_work(ioc->firmware_event_thread, &fw_event->work, 1);
+ INIT_WORK(&fw_event->work, _firmware_event_work);
+ queue_work(ioc->firmware_event_thread, &fw_event->work);
spin_unlock_irqrestore(&ioc->fw_event_lock, flags);
}
@@ -2052,7 +2052,7 @@ _scsih_fw_event_requeue(struct MPT2SAS_A
return;
spin_lock_irqsave(&ioc->fw_event_lock, flags);
- queue_delayed_work(ioc->firmware_event_thread, &fw_event->work, delay);
+ queue_work(ioc->firmware_event_thread, &fw_event->work);
spin_unlock_irqrestore(&ioc->fw_event_lock, flags);
}
@@ -4961,7 +4961,7 @@ static void
_firmware_event_work(struct work_struct *work)
{
struct fw_event_work *fw_event = container_of(work,
- struct fw_event_work, work.work);
+ struct fw_event_work, work);
unsigned long flags;
struct MPT2SAS_ADAPTER *ioc = fw_event->ioc;
--
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
