projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d20a1b8) | patch
[AMDGPU] Remove CC exception for Promote Alloca Limits
authorpvanhout <pierre.vanhoutryve@amd.com>
Wed, 12 Apr 2023 11:04:55 +0000 (13:04 +0200)
committerpvanhout <pierre.vanhoutryve@amd.com>
Thu, 13 Apr 2023 06:48:34 +0000 (08:48 +0200)
commitfd1d60873fdce6e908c9865ddf925f2616fccd55
tree03ba0a64e7ec5a078ab62adaaba1b49996f77833tree | snapshot
parentd20a1b87f689a716e16477b8562fd820a72a6a8bcommit | diff
[AMDGPU] Remove CC exception for Promote Alloca Limits

Apparently it was used to work around some issue that has been fixed.
Removing it helps with high scratch usage observed in some cases due to failed alloca promotion.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D145586
llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/vector-alloca-limits.ll diff | blob | history
Domain: System / Toolchain;