projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3f17dc8) | patch
[AMDGPU] Fix liveness for loops in si-optimize-exec-masking-pre-ra
authorCarl Ritson <carl.ritson@amd.com>
Thu, 30 Jun 2022 03:26:47 +0000 (12:26 +0900)
committerCarl Ritson <carl.ritson@amd.com>
Thu, 30 Jun 2022 06:26:50 +0000 (15:26 +0900)
commitd0f6641615755e3dd27bc71390e6ae866dfef9ec
treec428b89c9a68f47154539c81a912cc1e6cb6f900tree | snapshot
parent3f17dc89d3f24e6c9e41f25dd88488beeaf5f3f5commit | diff
[AMDGPU] Fix liveness for loops in si-optimize-exec-masking-pre-ra

Follow up to D127894, new liveness update code needs to handle
the case where S_ANDN2 input must be extended through loops when
V_CNDMASK_B32 has been hoisted.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D128800
llvm/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/opt-exec-masking-pre-ra-update-liveness.mir diff | blob | history
Domain: System / Toolchain;