[PATCH] D76836: [AMDGPU] Add G16 support to image instructions

[Sebastian Neubauer via Phabricator via llvm-commits]

Flakebi updated this revision to Diff 252894.
Flakebi added a comment.

Thank you for the fast review!

I added G16 to GlobalISel legalization. Is there a part where the image instructions get selected for GlobalISel? I didn’t find it and I guess it doesn’t pick up G16 instructions automatically.

gfx9 does not have G16 (addresses are 32 bit, derivatives 16 bit) only A16 (addresses and derivatives are 16 bit), so it’s tested in `unsupported-image-g16.ll`.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76836/new/

https://reviews.llvm.org/D76836

Files:
  llvm/lib/Target/AMDGPU/AMDGPU.td
  llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
  llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
  llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
  llvm/lib/Target/AMDGPU/MIMGInstructions.td
  llvm/lib/Target/AMDGPU/SIISelLowering.cpp
  llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
  llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
  llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.dim.a16.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.sample.a16.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.sample.g16.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.g16.encode.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.g16.ll
  llvm/test/CodeGen/AMDGPU/unsupported-image-g16.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D76836.252894.patch
Type: text/x-patch
Size: 428610 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200326/66154093/attachment-0001.bin>