Commits · 1d9f286ecb81dd72a4f640d29e292fbc03400ea2 · Lorenzo Albano / LLVM bpEVL

Apr 05, 2019
- [AMDGPU] rename vi-insts into gfx8-insts · 1d9f286e
  Stanislav Mekhanoshin authored Apr 05, 2019
```
Differential Revision: https://reviews.llvm.org/D60293

llvm-svn: 357792
```
  1d9f286e
Feb 09, 2019
- [AMDGPU] Split dot-insts feature · 1607a373
  Stanislav Mekhanoshin authored Feb 09, 2019
```
Differential Revision: https://reviews.llvm.org/D57972

llvm-svn: 353588
```
  1607a373
Jan 10, 2019
- [AMDGPU] Separate feature dot-insts · 6332f4d0
  Stanislav Mekhanoshin authored Jan 10, 2019
```
Differential Revision: https://reviews.llvm.org/D56525

llvm-svn: 350794
```
  6332f4d0
Aug 08, 2018

AMDGPU: Fix enabling denormals by default on pre-VI targets · 45bc1480

Matt Arsenault authored Aug 08, 2018

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

45bc1480

Aug 07, 2018
- AMDGPU: Add builtin for s_dcache_wb · 31c895ec
  Matt Arsenault authored Aug 07, 2018
```
llvm-svn: 339110
```
  31c895ec
- AMDGPU: Add builtin for s_dcache_inv_vol · 24f39247
  Matt Arsenault authored Aug 07, 2018
```
llvm-svn: 339109
```
  24f39247
Apr 30, 2018
- AMDGPU: Add Vega12 and Vega20 · d2da3c20
  Matt Arsenault authored Apr 30, 2018
```
Changes by
  Matt Arsenault
  Konstantin Zhuravlyov

llvm-svn: 331216
```
  d2da3c20