[all-commits] [llvm/llvm-project] ae9a34: [ARM] Fix DAG combine loop in reduction distribution

[David Green via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ae9a346ef8620851b54682b4b6738788af709578
      https://github.com/llvm/llvm-project/commit/ae9a346ef8620851b54682b4b6738788af709578
  Author: David Green <david.green at arm.com>
  Date:   2021-08-12 (Thu, 12 Aug 2021)

  Changed paths:
    M llvm/lib/Target/ARM/ARMISelLowering.cpp
    M llvm/test/CodeGen/Thumb2/mve-vecreduce-slp.ll

  Log Message:
  -----------
  [ARM] Fix DAG combine loop in reduction distribution

Given a constant operand, the MVE and DAGCombine combines could fight,
each redistributing in the opposite order. Add a guard to the MVE
vecreduce distribution to prevent that.