projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 7de383c) | patch
[AMDGPU] use scalar shift for SALU users in frame index elimination
authoralex-t <alexander.timofeev@amd.com>
Sat, 12 Mar 2022 14:38:11 +0000 (17:38 +0300)
committeralex-t <alexander.timofeev@amd.com>
Tue, 22 Mar 2022 10:43:23 +0000 (11:43 +0100)
commit0a488cba2c1b207cec49c8873c7d6b2af75f2d19
tree3df6899db3a7e617eef915ffed8e5b0bde70fa1btree | snapshot
parent7de383c892135ac7ecb67460c677aed72efdb76acommit | diff
[AMDGPU] use scalar shift for SALU users in frame index elimination

In the frame index lowering we have to insert shift and add
instructions to adjust stack object access.  We need to take care of the stack
object user kind and use scalar shift/add for scalar users.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D121524
llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/frame-index.mir [new file with mode: 0644] blob
Domain: System / Toolchain;