[all-commits] [llvm/llvm-project] 2955cc: [ARM] Improve costs for FMin/Max reductions

[David Green via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2955cc15ff2242b53cce5e0318ed6867f1b06714
      https://github.com/llvm/llvm-project/commit/2955cc15ff2242b53cce5e0318ed6867f1b06714
  Author: David Green <david.green at arm.com>
  Date:   2023-09-04 (Mon, 04 Sep 2023)

  Changed paths:
    M llvm/lib/Target/ARM/ARMTargetTransformInfo.cpp
    M llvm/lib/Target/ARM/ARMTargetTransformInfo.h
    M llvm/test/Analysis/CostModel/ARM/intrinsic-cost-kinds.ll
    M llvm/test/Analysis/CostModel/ARM/reduce-fminmax.ll

  Log Message:
  -----------
  [ARM] Improve costs for FMin/Max reductions

Similar to the other reductions, this changes the cost of fmin/fmax reductions
under MVE/NEON to perform vector operations until the types need to be
scalarized. The fp16 vectors can perform a VREV+FMIN/FMAX to skip a step of the
reduction, and otherwise need lanewise extract fro the top lanes.