[all-commits] [llvm/llvm-project] 584e9b: [LV] Calculate max feasible scalable VF.

[sdesmalen-arm via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 584e9b6e4b4987b882719923e640eed854613d91
      https://github.com/llvm/llvm-project/commit/584e9b6e4b4987b882719923e640eed854613d91
  Author: Sander de Smalen <sander.desmalen at arm.com>
  Date:   2021-04-28 (Wed, 28 Apr 2021)

  Changed paths:
    M llvm/include/llvm/Transforms/Vectorize/LoopVectorize.h
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/test/Transforms/LoopVectorize/AArch64/scalable-reductions.ll
    A llvm/test/Transforms/LoopVectorize/AArch64/scalable-vf-analysis.ll
    M llvm/test/Transforms/LoopVectorize/AArch64/scalable-vf-hint.ll
    M llvm/test/Transforms/LoopVectorize/scalable-vf-hint.ll

  Log Message:
  -----------
  [LV] Calculate max feasible scalable VF.

This patch also refactors the way the feasible max VF is calculated,
although this is NFC for fixed-width vectors.

After this change scalable VF hints are no longer truncated/clamped
to a shorter scalable VF, nor does it drop the 'scalable flag' from
the suggested VF to vectorize with a similar VF that is fixed.

Instead, the hint is ignored which means the vectorizer is free
to find a more suitable VF, using the CostModel to determine the
best possible VF.

Reviewed By: c-rhodes, fhahn

Differential Revision: https://reviews.llvm.org/D98509