[PATCH] D57243: GlobalISel: Allow bitcount ops to have different result type

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: aemerson, aditya_nandakumar, volkan, paquette, dsanders.
Herald added subscribers: tpr, javed.absar, kristof.beyls, rovka, nhaehnle, wdng, jvesely.

For AMDGPU the result is always 32-bit for 64-bit inputs.


https://reviews.llvm.org/D57243

Files:
  include/llvm/CodeGen/GlobalISel/LegalizerHelper.h
  include/llvm/Target/GenericOpcodes.td
  lib/CodeGen/GlobalISel/LegalizerHelper.cpp
  lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
  lib/Target/ARM/ARMLegalizerInfo.cpp
  test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
  test/CodeGen/AMDGPU/GlobalISel/legalize-ctlz-zero-undef.mir
  test/CodeGen/AMDGPU/GlobalISel/legalize-ctlz.mir
  test/CodeGen/AMDGPU/GlobalISel/legalize-ctpop.mir
  test/CodeGen/AMDGPU/GlobalISel/legalize-cttz-zero-undef.mir
  test/CodeGen/AMDGPU/GlobalISel/legalize-cttz.mir
  unittests/CodeGen/GlobalISel/LegalizerHelperTest.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D57243.183554.patch
Type: text/x-patch
Size: 30078 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190125/3300978c/attachment.bin>