[all-commits] [llvm/llvm-project] 0cf0be: [InstCombine] fix operands of shouldChangeType() f...

[RotateRight via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 0cf0be993c382f15c277de6d0db8be6a236c7c50
      https://github.com/llvm/llvm-project/commit/0cf0be993c382f15c277de6d0db8be6a236c7c50
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-02-04 (Tue, 04 Feb 2020)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
    M llvm/test/CodeGen/AMDGPU/diverge-switch-default.ll
    M llvm/test/CodeGen/AMDGPU/reqd-work-group-size.ll
    M llvm/test/Transforms/InstCombine/cast_phi.ll
    M llvm/test/Transforms/InstCombine/icmp-div-constant.ll
    M llvm/test/Transforms/InstCombine/icmp-mul-zext.ll

  Log Message:
  -----------
  [InstCombine] fix operands of shouldChangeType() for casted phi transform

This is a bug noted in the recent D72733 and seen
in the similar transform just above the changed source code.

I added tests with illegal types and zexts to show the bug -
we could transform legal phi ops to illegal, etc. I did not add
tests with trunc because we won't see any diffs on those patterns.
That is because InstCombiner::SliceUpIllegalIntegerPHI() appears to
do those transforms independently of datalayout. It can also create
more casts than are present in existing code.

There are some existing regression tests that do not include a
datalayout that would be altered by this fix. I assumed that the
lack of a datalayout in those regression files is an oversight, so
I added the minimal layout (make i32 legal) necessary to preserve
behavior on those tests.

Differential Revision: https://reviews.llvm.org/D73907