projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d0af732) | patch
[AMDGPU] Move perfhint analysis
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Thu, 8 Jul 2021 18:00:57 +0000 (11:00 -0700)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Wed, 21 Jul 2021 20:06:49 +0000 (13:06 -0700)
commitd01b34ed31467073bd1a0f0040da4d74db20475e
tree8857ed6413c3c400cf0a881a061327a72ef08c32tree | snapshot
parentd0af732bd00c0390fbcc6f900e8a59eb3954106ccommit | diff
[AMDGPU] Move perfhint analysis

This is SCC pass, moving it to the end of SCC PM saves one
Function PM. This needs the analysis to take into account
memory access width since it is now places after the
load/store optimizer (D105651).

Differential Revision: https://reviews.llvm.org/D105652
llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/llc-pipeline.ll diff | blob | history
Domain: System / Toolchain;