projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f9e2a62) | patch
[AMDGPU] Fix -amdgpu-inline-arg-alloca-cost
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Fri, 12 Mar 2021 17:19:46 +0000 (09:19 -0800)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Fri, 12 Mar 2021 18:19:50 +0000 (10:19 -0800)
commitb7b99b0799fa73ebf57155e5105adc03803e2ed6
treea12bac54517fd761e19995f0e1908af480feb97btree | snapshot
parentf9e2a62cc594c96194908a3ac4804caa07f86ba6commit | diff
[AMDGPU] Fix -amdgpu-inline-arg-alloca-cost

Before D94153 this threshold was in a pre-scaled units.
After D94153 inlining threshold multiplier is not applied
to this portion of the threshold anymore. Restore the
threshold by applying the multiplier.

Differential Revision: https://reviews.llvm.org/D98362
llvm/lib/Analysis/InlineCost.cpp diff | blob | history
llvm/test/Transforms/Inline/AMDGPU/amdgpu-inline-alloca-argument-cost.ll [new file with mode: 0644] blob
Domain: System / Toolchain;