[PATCH] D31482: [AMDGPU][GFX9] Set +fp32-denormals for >=gfx900 unless -cl-denorms-are-zero is set
Matt Arsenault via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Wed Mar 29 17:55:58 PDT 2017
arsenm accepted this revision.
arsenm added a comment.
This revision is now accepted and ready to land.
LGTM with f32 clarification
================
Comment at: lib/Basic/Targets.cpp:2114
+ static bool hasFullSpeedFMA(StringRef GPUName) {
+ return parseAMDGCNName(GPUName) >= GK_GFX9;
----------------
FMAF32?
================
Comment at: lib/Basic/Targets.cpp:2115
+ static bool hasFullSpeedFMA(StringRef GPUName) {
+ return parseAMDGCNName(GPUName) >= GK_GFX9;
+ }
----------------
We should probably add a new subtarget feature for this, but that's a separate patch
https://reviews.llvm.org/D31482
More information about the cfe-commits
mailing list