[PATCH] D136918: [AMDGPU] Scheduler: fix RP calculation for a MBB with one successor
Valery Pykhtin via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Sat Oct 29 09:27:45 PDT 2022
vpykhtin added a comment.
> ! In D136918#3894272 <https://reviews.llvm.org/D136918#3894272>, @arsenm wrote:
> The custom pressure tracking's treatment of sub registers always seemed broken to me. The register tuples do have a stronger constraint than the same number of individual registers. Pressure is always going to be a fuzzier metric than exact liveness
When I look into this now I tend to agree. We should always account for the full superreg even if only one subreg is live because RenameIndependentSubregs should split it earlier into individual registers. That would simplify RP tracking a lot.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D136918/new/
https://reviews.llvm.org/D136918
More information about the llvm-commits
mailing list