[all-commits] [llvm/llvm-project] e6345b: [SLP]Improve lookup of the buildvector top inserte...

[Andrew V. Tischenko via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e6345bf6442259447c451d2a60e74a89589e3c7a
      https://github.com/llvm/llvm-project/commit/e6345bf6442259447c451d2a60e74a89589e3c7a
  Author: Alexey Bataev <a.bataev at outlook.com>
  Date:   2022-08-29 (Mon, 29 Aug 2022)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
    M llvm/test/Transforms/SLPVectorizer/X86/vec_list_bias_external_insert_shuffled.ll

  Log Message:
  -----------
  [SLP]Improve lookup of the buildvector top insertelement instruction.

When estimating the cost of the in-tree vectorized scalars in
buildvector sequences, need to take into account the vectorized
insertelement instruction. The top of the buildvector seuences is the
topmost vectorized insertelement instruction, because it will have
> than 1 use after the vectorization.

For the affected test case improves througput from 21 to 16 (per
llvm-mca).

Differential Revision: https://reviews.llvm.org/D132740