Jeff Garzik
7fb6ec287a
[libata] fix PIO completion race ...
Make sure we that completion is the final action we take; prior to this
change, another CPU may have changed ap->pio_task_state before we tested
it a final time.
Spotted by, and original patch by Albert Lee @ IBM.
Also includes a minor optimization: eliminate a ton of unnecessary
queue_work() calls, simply by jumping to the beginning of the FSM
function ata_pio_task().
2005-09-16 06:01:48 -04:00
2005-09-06 17:52:54 -05:00
2005-09-06 17:52:54 -05:00
2005-08-23 18:43:42 -07:00
2005-09-07 17:31:27 -07:00
2005-09-10 11:11:48 -05:00
2005-09-12 20:02:18 -07:00
2005-09-09 18:11:16 -05:00
2005-08-28 11:34:11 -05:00
2005-09-10 10:06:30 -07:00
2005-09-08 15:07:08 -07:00
2005-09-08 15:07:08 -07:00
2005-09-10 10:16:27 -07:00
2005-08-28 11:33:52 -05:00
2005-09-10 10:06:30 -07:00
2005-08-14 15:43:39 -07:00
2005-09-06 17:26:06 -05:00
2005-09-10 10:06:30 -07:00
2005-09-09 16:43:37 -05:00
2005-09-16 06:01:48 -04:00
2005-08-30 21:52:18 -04:00
2005-08-30 03:37:42 -04:00
2005-09-09 16:43:37 -05:00
2005-09-05 00:06:16 -07:00
2005-09-07 17:17:34 -07:00
2005-09-07 17:17:33 -07:00
2005-09-10 10:06:30 -07:00
2005-09-09 14:03:48 -07:00
2005-08-30 22:48:51 -05:00
2005-09-08 23:07:29 -04:00
2005-08-30 05:42:52 -04:00
2005-08-30 05:42:52 -04:00
2005-08-28 20:18:39 -04:00
2005-09-07 16:57:46 -07:00
2005-08-30 05:42:52 -04:00
2005-09-14 08:22:44 -04:00
2005-08-30 05:42:52 -04:00
2005-08-30 05:42:52 -04:00
2005-09-08 15:07:08 -07:00
2005-08-29 15:59:42 -04:00
2005-08-30 05:42:52 -04:00
2005-09-06 17:23:43 -05:00
2005-09-06 17:52:54 -05:00
2005-09-06 17:52:54 -05:00
2005-09-10 21:02:25 -05:00
2005-09-09 10:28:17 -05:00
2005-09-10 14:43:25 -05:00
2005-09-09 10:28:17 -05:00
2005-08-28 11:18:35 -05:00
2005-09-09 18:43:56 -05:00
2005-08-28 11:34:10 -05:00
2005-08-12 12:27:44 -05:00
2005-09-06 17:52:54 -05:00
2005-09-09 16:37:05 -05:00
2005-08-28 11:34:07 -05:00
2005-08-28 11:40:00 -05:00
2005-08-28 11:14:12 -05:00
2005-09-06 17:52:54 -05:00
7fb6ec287a