[PATCH] D156235: [MachineBlockPlacement] Remove the pad limit for no-fallthrough loops
Eli Friedman via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 26 11:48:55 PDT 2023
efriedma added inline comments.
================
Comment at: llvm/lib/CodeGen/MachineBlockPlacement.cpp:2954
+ if (MaxBytesForAlignmentOverride.getNumOccurrences() == 0
+ && (ChainBB == LoopHeader || MLI->getLoopFor(LayoutPred) != L))
+ ChainBB->setAlignment(Align, 0);
----------------
The `!LayoutPred->isSuccessor(ChainBB)` check already ensures the padding will never be executed. Given that, I guess the remaining checks here are to try to maximize icache hits. In that context, why are loop headers special? Do we care if LayoutPred is part of a subloop of L?
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D156235/new/
https://reviews.llvm.org/D156235
More information about the llvm-commits
mailing list