[PATCH] D137843: AMDGPU: Fold llvm.amdgcn.sqrt(undef)

[Jay Foad via Phabricator via llvm-commits]

foad added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp:355
   switch (IID) {
+  case Intrinsic::amdgcn_sqrt: {
+    Value *Src = II.getArgOperand(0);
----------------
Just merge with the amdgcn_rsq case?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D137843/new/

https://reviews.llvm.org/D137843