Justin Hibbits
ce7b8e55e3
Stop idle threads on power9 in the idle task until an interrupt.
...
This reduces the CPU cycle wastage on power9, which is SMT4. Any idle
thread that's spinning is simply starving working threads on the same core
of valuable resources.
This can be reduced further by taking more advantage of the PSSCR supported
states, as well as permitting state loss, as is currently done for power8.
The currently implemented stop state is the lowest latency, which may still
consume resources.
2018-05-27 20:24:24 +00:00
..
2017-11-18 14:26:50 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2018-02-23 04:04:25 +00:00
2017-11-27 15:09:59 +00:00
2017-11-20 19:43:44 +00:00
2017-12-19 09:59:20 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2018-05-20 03:23:17 +00:00
2017-11-27 15:09:59 +00:00
2018-05-10 03:59:48 +00:00
2017-11-27 15:09:59 +00:00
2016-10-13 06:56:23 +00:00
2017-11-27 15:09:59 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2018-03-06 14:28:37 +00:00
2018-05-26 04:33:19 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-20 19:43:44 +00:00
2017-11-27 15:09:59 +00:00
2018-05-05 15:48:39 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-18 14:26:50 +00:00
2018-04-20 03:19:44 +00:00
2018-02-17 20:59:12 +00:00
2017-11-20 19:43:44 +00:00
2018-02-24 01:46:56 +00:00
2017-11-27 15:09:59 +00:00
2018-01-29 04:33:41 +00:00
2017-11-20 19:43:44 +00:00
2017-11-30 15:48:35 +00:00
2018-02-01 05:31:24 +00:00
2018-05-05 16:00:02 +00:00
2017-11-20 19:43:44 +00:00
2018-04-06 17:35:35 +00:00
2017-11-20 19:43:44 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2018-05-27 20:24:24 +00:00
2018-03-07 17:08:07 +00:00
2018-02-12 14:48:05 +00:00
2017-11-20 19:43:44 +00:00
2017-11-20 19:43:44 +00:00
2018-05-19 03:23:46 +00:00
2017-11-20 19:43:44 +00:00
2017-11-27 15:09:59 +00:00
2017-11-27 15:09:59 +00:00
2018-04-03 00:45:38 +00:00