projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 2f0750d) | patch
[AMDGPU] Add merging into S_BUFFER_LOAD_DWORDX8_IMM
authorPiotr Sobczak <Piotr.Sobczak@amd.com>
Mon, 30 Aug 2021 08:41:19 +0000 (10:41 +0200)
committerPiotr Sobczak <Piotr.Sobczak@amd.com>
Thu, 2 Sep 2021 14:26:25 +0000 (16:26 +0200)
commit30d6c39bca6cf5b2a9a166fb04954cfcd7b34dec
treed939e4ed7f0617f4aafc03b85491d3b534887662tree | snapshot
parent2f0750dd2ec09b39ee3f05b4be73b14fce8828b3commit | diff
[AMDGPU] Add merging into S_BUFFER_LOAD_DWORDX8_IMM

Extend SILoadStoreOptimizer to merge into DWORDX8 variant of S_BUFFER_LOAD.

Merging into DWORDX2 and DWORDX4 variants is handled already.

Differential Revision: https://reviews.llvm.org/D108909
llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/merge-sbuffer-load.mir [new file with mode: 0644] blob
Domain: System / Toolchain;