Matt Arsenault
fc84925208
AMDGPU: Fix target builtins for gfx10
...
This wasn't setting some of the features from older generations.
llvm-svn: 364123
2019-06-22 01:30:00 +00:00
Stanislav Mekhanoshin
cafccd7a53
[AMDGPU] gfx1011/gfx1012 clang support
...
Differential Revision: https://reviews.llvm.org/D63308
llvm-svn: 363345
2019-06-14 00:33:59 +00:00
Stanislav Mekhanoshin
91792f1b93
[AMDGPU] gfx1010 clang target
...
Differential Revision: https://reviews.llvm.org/D61875
llvm-svn: 360634
2019-05-13 23:15:59 +00:00
Stanislav Mekhanoshin
1d9f286ecb
[AMDGPU] rename vi-insts into gfx8-insts
...
Differential Revision: https://reviews.llvm.org/D60293
llvm-svn: 357792
2019-04-05 18:25:00 +00:00
Stanislav Mekhanoshin
1607a37308
[AMDGPU] Split dot-insts feature
...
Differential Revision: https://reviews.llvm.org/D57972
llvm-svn: 353588
2019-02-09 00:34:41 +00:00
Stanislav Mekhanoshin
6332f4d0d4
[AMDGPU] Separate feature dot-insts
...
Differential Revision: https://reviews.llvm.org/D56525
llvm-svn: 350794
2019-01-10 03:25:47 +00:00
Matt Arsenault
45bc148093
AMDGPU: Fix enabling denormals by default on pre-VI targets
...
Fast FMAF is not a sufficient condition to enable denormals.
Before VI, enabling denormals caused F32 instructions to
run at F64 speeds.
llvm-svn: 339278
2018-08-08 17:48:37 +00:00
Matt Arsenault
31c895ecdf
AMDGPU: Add builtin for s_dcache_wb
...
llvm-svn: 339110
2018-08-07 07:49:13 +00:00
Matt Arsenault
24f3924709
AMDGPU: Add builtin for s_dcache_inv_vol
...
llvm-svn: 339109
2018-08-07 07:49:04 +00:00
Matt Arsenault
d2da3c20d7
AMDGPU: Add Vega12 and Vega20
...
Changes by
Matt Arsenault
Konstantin Zhuravlyov
llvm-svn: 331216
2018-04-30 19:08:27 +00:00
