[all-commits] [llvm/llvm-project] 352fef: [InstCombine] Negator - sink sinkable negations

[Roman Lebedev via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 352fef3f11f5ccb2ddc8e16cecb7302a54721e9f
      https://github.com/llvm/llvm-project/commit/352fef3f11f5ccb2ddc8e16cecb7302a54721e9f
  Author: Roman Lebedev <lebedev.ri at gmail.com>
  Date:   2020-04-21 (Tue, 21 Apr 2020)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/CMakeLists.txt
    M llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
    M llvm/lib/Transforms/InstCombine/InstCombineInternal.h
    A llvm/lib/Transforms/InstCombine/InstCombineNegator.cpp
    M llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
    M llvm/test/Transforms/InstCombine/and-or-icmps.ll
    M llvm/test/Transforms/InstCombine/fold-sub-of-not-to-inc-of-add.ll
    M llvm/test/Transforms/InstCombine/high-bit-signmask-with-trunc.ll
    M llvm/test/Transforms/InstCombine/high-bit-signmask.ll
    M llvm/test/Transforms/InstCombine/sub-of-negatible.ll
    M llvm/test/Transforms/InstCombine/sub.ll
    M llvm/test/Transforms/InstCombine/zext-bool-add-sub.ll

  Log Message:
  -----------
  [InstCombine] Negator - sink sinkable negations

Summary:
As we have discussed previously (e.g. in D63992 / D64090 / [[ https://bugs.llvm.org/show_bug.cgi?id=42457 | PR42457 ]]), `sub` instruction
can almost be considered non-canonical. While we do convert `sub %x, C` -> `add %x, -C`,
we sparsely do that for non-constants. But we should.

Here, i propose to interpret `sub %x, %y` as `add (sub 0, %y), %x` IFF the negation can be sinked into the `%y`

This has some potential to cause endless combine loops (either around PHI's, or if there are some opposite transforms).
For former there's `-instcombine-negator-max-depth` option to mitigate it, should this expose any such issues
For latter, if there are still any such opposing folds, we'd need to remove the colliding fold.
In any case, reproducers welcomed!

Reviewers: spatel, nikic, efriedma, xbolva00

Reviewed By: spatel

Subscribers: xbolva00, mgorny, hiraditya, reames, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D68408