[all-commits] [llvm/llvm-project] 70e3c9: [BasicAA] Always strip single-argument phi nodes

[Nikita Popov via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 70e3c9a8b6684c49576aee38b5c90b18fdf00d2c
      https://github.com/llvm/llvm-project/commit/70e3c9a8b6684c49576aee38b5c90b18fdf00d2c
  Author: Nikita Popov <nikita.ppv at gmail.com>
  Date:   2021-02-18 (Thu, 18 Feb 2021)

  Changed paths:
    M llvm/include/llvm/IR/Value.h
    M llvm/lib/Analysis/BasicAliasAnalysis.cpp
    M llvm/lib/Analysis/GlobalsModRef.cpp
    M llvm/lib/IR/Value.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUAliasAnalysis.cpp
    M llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
    M llvm/test/Analysis/BasicAA/phi-aa.ll

  Log Message:
  -----------
  [BasicAA] Always strip single-argument phi nodes

We can always look through single-argument (LCSSA) phi nodes when
performing alias analysis. getUnderlyingObject() already does this,
but stripPointerCastsAndInvariantGroups() does not. We still look
through these phi nodes with the usual aliasPhi() logic, but
sometimes get sub-optimal results due to the restrictions on value
equivalence when looking through arbitrary phi nodes. I think it's
generally beneficial to keep the underlying object logic and the
pointer cast stripping logic in sync, insofar as it is possible.

With this patch we get marginally better results:

  aa.NumMayAlias | 5010069 | 5009861
  aa.NumMustAlias | 347518 | 347674
  aa.NumNoAlias | 27201336 | 27201528
  ...
  licm.NumPromoted | 1293 | 1296

I've renamed the relevant strip method to stripPointerCastsForAliasAnalysis(),
as we're past the point where we can explicitly spell out everything
that's getting stripped.

Differential Revision: https://reviews.llvm.org/D96668