[all-commits] [llvm/llvm-project] 9162b7: DADCombiner: Don't simplify the token factor if th...

[Changpeng Fang via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 9162b70e510475a044ee98332c9cdabf094b0e9b
      https://github.com/llvm/llvm-project/commit/9162b70e510475a044ee98332c9cdabf094b0e9b
  Author: Changpeng Fang <changpeng.fang at gmail.com>
  Date:   2020-07-25 (Sat, 25 Jul 2020)

  Changed paths:
    M llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
    A llvm/test/CodeGen/AMDGPU/token-factor-inline-limit-test.ll

  Log Message:
  -----------
  DADCombiner: Don't simplify the token factor if the node's number of operands already exceeds TokenFactorInlineLimit

Summary:
  In parallelizeChainedStores, a TokenFactor was created with the size greater than 3000.
We found that DAGCombiner::visitTokenFactor will consume a huge amount of time on
such nodes. Since the number of operands already exceeds TokenFactorInlineLimit, we propose
to give up simplification with the consideration of compile time.

Reviewers:
  @spatel, @arsenm

Differential Revision:
  https://reviews.llvm.org/D84204