[PATCH] D80889: AMDGPU: Fix not using scalar loads for global reads in shaders

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: nhaehnle, foad, tpr, mareko, dstuttard.
Herald added subscribers: kerbowa, hiraditya, t-tye, yaxunl, wdng, jvesely, kzhuravl.
Herald added a project: LLVM.

The pass which infers when it's legal to load a global address space
as SMRD was only considering amdgpu_kernel, and ignoring the shader
entry type calling conventions.


https://reviews.llvm.org/D80889

Files:
  llvm/lib/Target/AMDGPU/AMDGPUAnnotateUniformValues.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/mubuf-global.ll
  llvm/test/CodeGen/AMDGPU/global-saddr.ll
  llvm/test/CodeGen/AMDGPU/infer-uniform-load-shader.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D80889.267499.patch
Type: text/x-patch
Size: 8920 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200531/8e516187/attachment.bin>