projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1445b77) | patch
AMDGPU/GlobalISel: Insert waterfall loop for vector indexing
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 29 Mar 2019 03:54:56 +0000 (03:54 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 29 Mar 2019 03:54:56 +0000 (03:54 +0000)
commit5fddf091879e195d1f3ac83e8384ec1ff2882c23
tree9d2b770b44634e07da8c531c5a5517d36d7a3ca2tree | snapshot
parent1445b77e8c6da8e43ba12f86b010852b97aea4fbcommit | diff
AMDGPU/GlobalISel: Insert waterfall loop for vector indexing

The register index can only really be an SGPR. Lie that a VGPR index
is legal, and then rewrite the instruction in a waterfall loop to
handle the index.

llvm-svn: 357235
llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.h diff | blob | history
llvm/test/CodeGen/AMDGPU/GlobalISel/regbankselect-extract-vector-elt.mir diff | blob | history
Domain: System / Toolchain;