projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 29ac3a5) | patch
AMDGPU: Don't rematerialize mov with implicit operands
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Mon, 4 Feb 2019 22:26:21 +0000 (22:26 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Mon, 4 Feb 2019 22:26:21 +0000 (22:26 +0000)
commitcba0c6d0c9d2c610c566e4b46b8f0315a2cfebbc
treec7c50c770ff0471d43d0088c24ddbe2f0a744f49tree | snapshot
parent29ac3a5b822ba8c097a3ae78d983cdb94da43dd4commit | diff
AMDGPU: Don't rematerialize mov with implicit operands

This was pulling the mov used for register indexing on gfx9 out of the
loop.

llvm-svn: 353101
llvm/lib/Target/AMDGPU/SIInstrInfo.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/no-remat-indirect-mov.mir [new file with mode: 0644] blob
Domain: System / Toolchain;