Sergey Sharybin
83ce02879f
Cycles: Fix possible race condition when generating Beckmann table
...
Two issues here:
- Checking table size to be non-zero is not a proper way to go here. This is
because we first resize the table and then fill it in. So it was possible that
non-initialized table was used.
Trickery with using temporary memory and then doing table.swap() might work,
but we can not guarantee that table size will be set after the data pointer.
- Mutex guard was useless, because every thread was using own mutex. Need to
make mutex guard static so all threads are using same mutex.
2017-10-06 21:06:15 +05:00
..
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-10-05 17:57:58 +02:00
2017-03-29 13:41:11 +02:00
2017-10-04 21:11:14 +02:00
2017-10-04 21:11:14 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-07-03 12:12:27 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-05-07 14:40:58 +02:00
2017-05-07 14:40:58 +02:00
2017-08-23 12:35:23 +02:00
2017-08-20 16:51:53 +02:00
2017-09-04 13:14:54 +02:00
2017-08-09 04:24:03 -04:00
2017-08-07 14:01:24 +02:00
2017-08-07 14:01:24 +02:00
2017-10-05 17:57:58 +02:00
2017-03-29 13:41:11 +02:00
2017-10-05 17:57:58 +02:00
2017-03-29 13:41:11 +02:00
2017-08-08 07:12:04 -04:00
2017-03-29 13:41:11 +02:00
2017-08-20 17:37:22 +02:00
2017-08-20 17:37:22 +02:00
2017-08-17 12:44:32 +02:00
2017-08-10 09:21:02 +02:00
2017-08-20 16:51:53 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-08-08 07:12:04 -04:00
2017-08-08 07:12:04 -04:00
2017-10-04 21:11:14 +02:00
2017-08-15 01:22:40 +02:00
2017-10-06 21:06:15 +05:00
2017-10-06 21:06:15 +05:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-08-20 16:51:53 +02:00
2017-08-20 14:27:51 +02:00
2017-03-29 13:41:11 +02:00
2017-03-29 13:41:11 +02:00
2017-08-17 02:00:44 +02:00
2017-08-15 01:22:40 +02:00