[all-commits] [llvm/llvm-project] ca78dd: Considering VF when getting vector call costs.

Paschalis Mpeis via All-commits all-commits at lists.llvm.org
Mon Jan 15 06:01:12 PST 2024


  Branch: refs/heads/users/paschalis-mpeis/loop-vectorizer-frem-scalable
  Home:   https://github.com/llvm/llvm-project
  Commit: ca78dd0180568a08f528b089f8a7af95115ae6c8
      https://github.com/llvm/llvm-project/commit/ca78dd0180568a08f528b089f8a7af95115ae6c8
  Author: Paschalis Mpeis <Paschalis.Mpeis at arm.com>
  Date:   2024-01-15 (Mon, 15 Jan 2024)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    A llvm/test/Analysis/CostModel/AArch64/arith-fp-frem.ll
    R llvm/test/Analysis/CostModel/AArch64/arith-fp-sve-frem.ll

  Log Message:
  -----------
  Considering VF when getting vector call costs.

Added neon tests as well.




More information about the All-commits mailing list