diff options
Diffstat (limited to 'llvm/lib/Target/AMDGPU/AMDGPUFeatures.td')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/AMDGPUFeatures.td | 23 |
1 files changed, 7 insertions, 16 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUFeatures.td b/llvm/lib/Target/AMDGPU/AMDGPUFeatures.td index ea3952c316e4d..db00f8f711a33 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUFeatures.td +++ b/llvm/lib/Target/AMDGPU/AMDGPUFeatures.td @@ -18,15 +18,6 @@ def FeatureFMA : SubtargetFeature<"fmaf", "Enable single precision FMA (not as fast as mul+add, but fused)" >; -// Some instructions do not support denormals despite this flag. Using -// fp32 denormals also causes instructions to run at the double -// precision rate for the device. -def FeatureFP32Denormals : SubtargetFeature<"fp32-denormals", - "FP32Denormals", - "true", - "Enable single precision denormal handling" ->; - class SubtargetFeatureLocalMemorySize <int Value> : SubtargetFeature< "localmemorysize"#Value, "LocalMemorySize", @@ -38,16 +29,16 @@ def FeatureLocalMemorySize0 : SubtargetFeatureLocalMemorySize<0>; def FeatureLocalMemorySize32768 : SubtargetFeatureLocalMemorySize<32768>; def FeatureLocalMemorySize65536 : SubtargetFeatureLocalMemorySize<65536>; -class SubtargetFeatureWavefrontSize <int Value> : SubtargetFeature< - "wavefrontsize"#Value, - "WavefrontSize", - !cast<string>(Value), +class SubtargetFeatureWavefrontSize <int ValueLog2> : SubtargetFeature< + "wavefrontsize"#!shl(1, ValueLog2), + "WavefrontSizeLog2", + !cast<string>(ValueLog2), "The number of threads per wavefront" >; -def FeatureWavefrontSize16 : SubtargetFeatureWavefrontSize<16>; -def FeatureWavefrontSize32 : SubtargetFeatureWavefrontSize<32>; -def FeatureWavefrontSize64 : SubtargetFeatureWavefrontSize<64>; +def FeatureWavefrontSize16 : SubtargetFeatureWavefrontSize<4>; +def FeatureWavefrontSize32 : SubtargetFeatureWavefrontSize<5>; +def FeatureWavefrontSize64 : SubtargetFeatureWavefrontSize<6>; class SubtargetFeatureGeneration <string Value, string FeatureName, string Subtarget, |