[all-commits] [llvm/llvm-project] e49043: [CostModel][X86] Update BITREVERSE costs for GFNI ...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e49043512dbdc68319093da46e95a1e331ef837e
      https://github.com/llvm/llvm-project/commit/e49043512dbdc68319093da46e95a1e331ef837e
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-04-17 (Wed, 17 Apr 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86TargetTransformInfo.cpp
    M llvm/test/Analysis/CostModel/X86/bitreverse-codesize.ll
    M llvm/test/Analysis/CostModel/X86/bitreverse-latency.ll
    M llvm/test/Analysis/CostModel/X86/bitreverse-sizelatency.ll
    M llvm/test/Analysis/CostModel/X86/bitreverse.ll

  Log Message:
  -----------
  [CostModel][X86] Update BITREVERSE costs for GFNI targets

Inspired by the recent patches by @shamithoke - we have real scheduler model numbers for GFNI instructions now, allowing us to calculate an upper bounds costs table instead of performing it analytically.


  Commit: 4a5ab13bf5a94ec7f0eabaf24dfe1a5ee720b860
      https://github.com/llvm/llvm-project/commit/4a5ab13bf5a94ec7f0eabaf24dfe1a5ee720b860
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-04-17 (Wed, 17 Apr 2024)

  Changed paths:
    M llvm/test/Transforms/VectorCombine/AArch64/select-shuffle.ll
    M llvm/test/Transforms/VectorCombine/AArch64/shuffletoidentity.ll
    M llvm/test/Transforms/VectorCombine/AArch64/vecreduce-shuffle.ll

  Log Message:
  -----------
  [VectorCombine] Remove single quotes from "-passes=vector-combine"

These confuse the update_test_checks.py script when run by DOS cmd.exe


Compare: https://github.com/llvm/llvm-project/compare/dbafcd7e6d5b...4a5ab13bf5a9

To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications