[PATCH] D73666: AMDGPU/GlobalISel: Adjust image load register type based on dmask

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: nhaehnle, kerbowa.
Herald added subscribers: Petar.Avramovic, jfb, hiraditya, t-tye, tpr, dstuttard, rovka, yaxunl, wdng, jvesely, kzhuravl.
Herald added a project: LLVM.
arsenm added parent revisions: D73446: AMDGPU/GlobalISel: Legalize a16 images, D73447: AMDGPU/GlobalISel: Legalize non-a16 non-NSA images, D73444: AMDGPU/GlobalISel: Legalize TFE image result loads, D72796: AMDGPU/GlobalISel: Legalize unpacked d16 image operations.

Trim elements that won't be written. The equivalent still needs to be
done for writes. Also start widening 3 elements to 4
elements. Selection will get the count from the dmask.


https://reviews.llvm.org/D73666

Files:
  llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.dim.a16.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.load.2d.d16.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.load.2d.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D73666.241294.patch
Type: text/x-patch
Size: 110830 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200129/824dc690/attachment.bin>