projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 7d4a799) | patch
[AMDGPU] Extend WorkGroupID* codegen for compute shaders
authorChristudasan Devadasan <Christudasan.Devadasan@amd.com>
Wed, 1 Mar 2023 07:51:57 +0000 (13:21 +0530)
committerChristudasan Devadasan <Christudasan.Devadasan@amd.com>
Wed, 8 Mar 2023 02:06:19 +0000 (07:36 +0530)
commit2171f04c121ed27fe2705d3f116df052daeb40a5
tree6f61e29a784505cc168dc8c51d0f467bc5da6487tree | snapshot
parent7d4a799f4a0b27b0544dbbdd5bf57e938458e687commit | diff
[AMDGPU] Extend WorkGroupID* codegen for compute shaders

Currently, the codegen support for llvm.amdgcn.workgroup.id*
intrinsics are enabled only for compute kernels. In addition,
this patch enables their selection for compute shaders on
subtargets that have architected SGPRs.

Differential Revision: https://reviews.llvm.org/D145045
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/lower-work-group-id-intrinsics.ll [new file with mode: 0644] blob
Domain: System / Toolchain;