5660bb6bc9
AMDGPU: Remove denormal subtarget features
...
Switch to using the denormal-fp-math/denormal-fp-math-f32 attributes.
2020-04-02 17:17:12 -04:00
75cf30918f
AMDGPU: Assume f32 denormals are enabled by default
...
This will likely introduce catastrophic performance regressions on
older subtargets, but should be correct. A follow up change will
remove the old fp32-denormals subtarget features, and switch to using
the new denormal-fp-math/denormal-fp-math-f32 attributes. Frontends
should be making sure to add the denormal-fp-math-f32 attribute when
appropriate to avoid performance regressions.
2020-04-02 17:17:12 -04:00
c43e67bfff
[AMDGPU] gfx1011/gfx1012 targets
...
Differential Revision: https://reviews.llvm.org/D63307
llvm-svn: 363344
2019-06-14 00:33:31 +00:00
700b1ef54d
AMDGPU: Fix V_FMA_F16 selection on GFX9
...
GFX9 should select opsel version.
Differential Revision: https://reviews.llvm.org/D54545
llvm-svn: 347265
2018-11-19 21:10:16 +00:00
7d1532d333
AMDGPU: Fix check lines in fdot2 test:
...
GCN900 -> GFX900
llvm-svn: 346925
2018-11-15 02:42:04 +00:00
c370d7b33d
[AMDGPU] [AMDGPU] Support a fdot2 pattern.
...
Summary: Optimize fma((float)S0.x, (float)S1.x fma((float)S0.y, (float)S1.y, z))
-> fdot2((v2f16)S0, (v2f16)S1, (float)z)
Author: FarhanaAleen
Reviewed By: rampitec, b-sumner
Subscribers: AMDGPU
Differential Revision: https://reviews.llvm.org/D49146
llvm-svn: 337198
2018-07-16 18:19:59 +00:00
Matt Arsenault