[PATCH] D82990: [AMDGPU] Limit promote alloca to vector with VGPR budget
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 1 12:59:23 PDT 2020
arsenm added inline comments.
================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp:439
+ // Use up to 1/4 of available register budget for vectorization.
+ if (DL.getTypeSizeInBits(AllocaTy) * 4 > MaxVGPRs * 32) {
----------------
This seems like a huge default. I thought we previously limited this to 16 VGPRs. There should also probably be a cl::opt for this too
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D82990/new/
https://reviews.llvm.org/D82990
More information about the llvm-commits
mailing list