projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3f00e10) | patch
[AMDGPU] Preserve deadness of vcc when shrinking instructions
authorJay Foad <jay.foad@amd.com>
Fri, 22 Oct 2021 10:18:11 +0000 (11:18 +0100)
committerJay Foad <jay.foad@amd.com>
Fri, 22 Oct 2021 13:22:24 +0000 (14:22 +0100)
commit74cd4dee20e85cb2b071283a5caa8f924539e1b1
treeb8216bce94f6fd39c5373d9deec802ade5a82772tree | snapshot
parent3f00e10bddd7bb2bdd14b5cbecd000e6288588bfcommit | diff
[AMDGPU] Preserve deadness of vcc when shrinking instructions

This doesn't have any effect on codegen now, but it might do in the
future if we shrink instructions before post-RA scheduling, which is
sensitive to live vs dead defs.

Differential Revision: https://reviews.llvm.org/D112305
llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/shrink-carry.mir diff | blob | history
Domain: System / Toolchain;