[all-commits] [llvm/llvm-project] 90ea81: [LV] Widen freeze instead of scalarizing it

[lizhijin1024 via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 90ea81fcb28e48c080e1a9016dd5262eeb33d0ce
      https://github.com/llvm/llvm-project/commit/90ea81fcb28e48c080e1a9016dd5262eeb33d0ce
  Author: lizhijin <lizhijin3 at huawei.com>
  Date:   2022-05-19 (Thu, 19 May 2022)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    A llvm/test/Transforms/LoopVectorize/vector-freeze.ll

  Log Message:
  -----------
  [LV] Widen freeze instead of scalarizing it

This patch changes the strategy for vectorizing freeze instrucion, from
replicating multiple times to widening according to selected VF.

Fixes #54992

Reviewed By: fhahn

Differential Revision: https://reviews.llvm.org/D125016