[PATCH] D137811: InstCombine: Port amdgcn.class intrinsic combines to is.fpclass

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: sepavloff, spatel, kpn, andrew.w.kaylor, efriedma, cameron.mcinally, jcranmer, jyknight.
Herald added subscribers: nlopes, kosarev, foad, kerbowa, hiraditya, jvesely.
Herald added a project: All.
arsenm requested review of this revision.
Herald added a subscriber: wdng.
Herald added a project: LLVM.

The intrinsics are identical, except llvm.amdgcn.class doesn't require
immarg for the test mask.


https://reviews.llvm.org/D137811

Files:
  llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp
  llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
  llvm/lib/Transforms/InstCombine/InstCombineInternal.h
  llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
  llvm/test/Transforms/InstCombine/is_fpclass.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D137811.474648.patch
Type: text/x-patch
Size: 25077 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20221111/9ca41381/attachment.bin>