[PATCH] D47009: AMDGPU: Add pass to optimize reqd_work_group_size

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: rampitec, kzhuravl, cfang, b-sumner, yaxunl.
Herald added subscribers: t-tye, tpr, dstuttard, mgorny, nhaehnle, wdng.

Eliminate loads from the dispatch packet when they will have
a known value.

      

Also pattern match the code used by the library to handle partial
workgroup dispatches, which isn't necessary if reqd_work_group_size
is used.


https://reviews.llvm.org/D47009

Files:
  lib/Target/AMDGPU/AMDGPU.h
  lib/Target/AMDGPU/AMDGPULowerKernelAttributes.cpp
  lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
  lib/Target/AMDGPU/CMakeLists.txt
  test/CodeGen/AMDGPU/reqd-work-group-size.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D47009.147284.patch
Type: text/x-patch
Size: 33108 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180517/1457bd73/attachment.bin>