[all-commits] [llvm/llvm-project] 864977: [VectorCombine] Scalarize vector load/extract.

[Florian Hahn via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 86497785d540e59eaca24bed4219ddec183cbc9b
      https://github.com/llvm/llvm-project/commit/86497785d540e59eaca24bed4219ddec183cbc9b
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2021-05-24 (Mon, 24 May 2021)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/VectorCombine.cpp
    M llvm/test/Transforms/VectorCombine/AArch64/load-extractelement-scalarization.ll
    M llvm/test/Transforms/VectorCombine/X86/load-inseltpoison.ll
    M llvm/test/Transforms/VectorCombine/X86/load.ll

  Log Message:
  -----------
  [VectorCombine] Scalarize vector load/extract.

This patch adds a new combine that tries to scalarize chains of
`extractelement (load %ptr), %idx` to `load (gep %ptr, %idx)`. This is
profitable when extracting only a few elements out of a large vector.

At the moment, `store (extractelement (load %ptr), %idx), %ptr`
operations on large vectors result in huge code in the backend.

This can easily be triggered by using the matrix extension, e.g.
https://clang.godbolt.org/z/qsccPdPf4

This should complement D98240.

Reviewed By: spatel

Differential Revision: https://reviews.llvm.org/D100273