projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ca4bf58) | patch
[AMDGPU][GlobalISel] GlobalISel for flat scratch
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Mon, 21 Dec 2020 20:31:06 +0000 (12:31 -0800)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Wed, 23 Dec 2020 00:33:06 +0000 (16:33 -0800)
commitd15119a02d92274cd7f779f4bb8485b1020110e0
tree9b95e3e1108b092d5c364964a93437bec258e2f3tree | snapshot
parentca4bf58e4ee5951473a861716193063c5ef83e9acommit | diff
[AMDGPU][GlobalISel] GlobalISel for flat scratch

It does not seem to fold offsets but this is not specific
to the flat scratch as getPtrBaseWithConstantOffset() does
not return the split for these tests unlike its SDag
counterpart.

Differential Revision: https://reviews.llvm.org/D93670
llvm/lib/Target/AMDGPU/AMDGPUGISel.td diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.h diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/GlobalISel/flat-scratch.ll [new file with mode: 0644] blob
Domain: System / Toolchain;