Sergey Sharybin
ae41a08288
Cycles: Attempt to work around compilation of sm_20 and sm_21
...
Disabled forceinline for those architectures, which seems to be compiling
successfully more often.
There might be ~3% slowdown based on quick tests, but better be rendering
something rather than failing to compile kernels again and again.
Those architectures will be doomed for abandon once we'll switch to toolkit 9.
2017-09-08 18:37:54 +02:00
..
2017-09-08 15:12:37 +02:00
2017-07-23 23:25:12 +02:00
2017-08-25 17:46:15 +02:00
2017-09-06 18:12:45 +02:00
2017-08-09 14:08:58 +02:00
2017-08-24 03:43:02 +02:00
2017-07-22 23:58:51 +02:00
2017-09-05 15:48:34 +02:00
2017-09-06 13:51:45 +02:00
2017-08-13 01:19:07 +02:00
2017-08-13 01:19:07 +02:00
2017-08-24 03:43:02 +02:00
2017-03-08 00:52:41 -05:00
2017-05-24 17:23:54 +02:00
2017-09-08 18:37:54 +02:00
2017-08-08 15:27:25 +02:00
2015-05-09 19:52:40 +05:00
2017-08-24 03:43:02 +02:00
2015-04-28 00:38:32 +05:00
2017-08-09 14:08:58 +02:00
2017-08-08 07:12:04 -04:00
2017-05-09 17:07:17 +02:00
2017-09-06 13:51:48 +02:00
2017-03-29 13:41:11 +02:00
2014-03-29 13:03:49 +01:00
2017-08-20 16:51:53 +02:00
2017-09-05 15:48:34 +02:00
2017-08-19 18:14:16 +02:00
2017-08-19 18:14:16 +02:00
2017-08-20 23:32:20 +02:00
2017-09-05 15:48:34 +02:00
2017-09-05 15:48:34 +02:00
2017-09-05 15:48:34 +02:00
2017-05-07 14:40:58 +02:00
2017-06-10 03:51:18 -04:00
2017-08-20 23:32:20 +02:00
2017-08-24 03:43:02 +02:00
2017-09-05 16:52:45 +02:00
2017-08-24 03:43:02 +02:00
2017-09-06 18:12:45 +02:00
2017-08-24 03:43:02 +02:00
2017-08-24 03:43:02 +02:00
2017-03-08 00:52:41 -05:00
2017-05-07 14:40:58 +02:00