[all-commits] [llvm/llvm-project] ed9df5: [Passes] Run sinking/hoisting in SimplifyCFG earlier.

[Florian Hahn via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ed9df5bd2f50b2199204cc6e50910ba52dd5e93e
      https://github.com/llvm/llvm-project/commit/ed9df5bd2f50b2199204cc6e50910ba52dd5e93e
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2021-04-30 (Fri, 30 Apr 2021)

  Changed paths:
    M llvm/lib/Passes/PassBuilder.cpp
    M llvm/lib/Transforms/IPO/PassManagerBuilder.cpp
    M llvm/test/Transforms/PGOProfile/Inputs/thinlto_cspgo_bar_use.ll
    M llvm/test/Transforms/PGOProfile/cspgo_profile_summary.ll
    M llvm/test/Transforms/PGOProfile/thinlto_cspgo_use.ll
    M llvm/test/Transforms/PhaseOrdering/AArch64/hoisting-sinking-required-for-vectorization.ll

  Log Message:
  -----------
  [Passes] Run sinking/hoisting in SimplifyCFG earlier.

Hoisting and sinking instructions out of conditional blocks enables
additional vectorization by:

1. Executing memory accesses unconditionally.
2. Reducing the number of instructions that need predication.

After disabling early hoisting / sinking, we miss out on a few
vectorization opportunities. One of those is causing a ~10% performance
regression in one of the Geekbench benchmarks on AArch64.

This patch tires to recover the regression by running hoisting/sinking
as part of a SimplifyCFG run after LoopRotate and before LoopVectorize.

Note that in the legacy pass-manager, we run LoopRotate just before
vectorization again and there's no SimplifyCFG run in between, so the
sinking/hoisting may impact the later run on LoopRotate. But the impact
should be limited and the benefit of hosting/sinking at this stage
should outweigh the risk of not rotating.

Compile-time impact looks slightly positive for most cases.
http://llvm-compile-time-tracker.com/compare.php?from=2ea7fb7b1c045a7d60fcccf3df3ebb26aa3699e5&to=e58b4a763c691da651f25996aad619cb3d946faf&stat=instructions

NewPM-O3: geomean -0.19%
NewPM-ReleaseThinLTO: geoman -0.54%
NewPM-ReleaseLTO-g: geomean -0.03%

With a few benchmarks seeing a notable increase, but also some
improvements.

Alternative to D101290.

Reviewed By: lebedev.ri

Differential Revision: https://reviews.llvm.org/D101468