diff options
Diffstat (limited to 'llvm/lib/Target/AMDGPU/SIISelLowering.cpp')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp index 925fdce757a..ebc686ca83d 100644 --- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp +++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp @@ -3272,12 +3272,17 @@ bool SITargetLowering::isFMAFasterThanFMulAndFAdd(EVT VT) const { VT = VT.getScalarType(); switch (VT.getSimpleVT().SimpleTy) { - case MVT::f32: + case MVT::f32: { // This is as fast on some subtargets. However, we always have full rate f32 // mad available which returns the same result as the separate operations // which we should prefer over fma. We can't use this if we want to support // denormals, so only report this in these cases. - return Subtarget->hasFP32Denormals() && Subtarget->hasFastFMAF32(); + if (Subtarget->hasFP32Denormals()) + return Subtarget->hasFastFMAF32() || Subtarget->hasDLInsts(); + + // If the subtarget has v_fmac_f32, that's just as good as v_mac_f32. + return Subtarget->hasFastFMAF32() && Subtarget->hasDLInsts(); + } case MVT::f64: return true; case MVT::f16: |
