Craig Topper 678b07fe3c [AVX-512] Remove masking from 512-bit vpermil builtins. The backend now has versions without masking so wrap it with select.
This will allow the backend to constant fold these to generic shuffle vectors like 128-bit and 256-bit without having to working about handling masking.

llvm-svn: 289351
2016-12-11 01:26:52 +00:00
..
2016-07-26 05:52:37 +00:00
2016-07-26 05:52:37 +00:00
2016-07-26 05:52:37 +00:00
2016-07-26 05:52:37 +00:00
2016-07-26 05:52:37 +00:00
2016-04-04 23:14:14 +00:00
2016-07-26 15:21:18 +00:00
2016-07-26 05:52:37 +00:00
2016-07-26 05:52:37 +00:00
2016-02-19 06:54:47 +00:00
2015-11-02 19:32:23 +00:00
2014-10-02 20:53:50 +00:00
2016-11-09 17:02:07 +00:00
2014-10-02 20:53:50 +00:00
2016-06-14 21:02:05 +00:00
2015-11-07 00:48:18 +00:00
2016-04-28 14:13:28 +00:00
2016-03-07 14:22:46 +00:00
2015-09-14 21:27:36 +00:00
2015-01-24 17:39:36 +00:00
2016-02-20 09:23:41 +00:00
2016-02-20 09:23:41 +00:00
2016-04-11 07:48:59 +00:00
2015-08-11 00:19:54 +00:00
2016-05-04 23:37:40 +00:00
2016-09-01 11:29:21 +00:00
2016-06-14 21:02:05 +00:00
2016-09-16 22:56:12 +00:00
2016-05-19 17:47:47 +00:00
2016-06-14 21:02:05 +00:00
2016-01-28 22:56:41 +00:00
2016-09-13 19:43:33 +00:00
2016-06-24 13:11:15 +00:00
2016-07-06 07:13:49 +00:00
2015-01-24 17:39:36 +00:00
2015-01-24 17:39:36 +00:00
2014-08-14 15:14:51 +00:00
2016-06-14 21:02:05 +00:00