[all-commits] [llvm/llvm-project] c23cbe: [VectorUtils] add IR-level analysis for widening o...

[RotateRight via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: c23cbefd9d73e8551b0c8b6cc3f14bb6b067ca92
      https://github.com/llvm/llvm-project/commit/c23cbefd9d73e8551b0c8b6cc3f14bb6b067ca92
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-04-12 (Sun, 12 Apr 2020)

  Changed paths:
    M llvm/include/llvm/Analysis/VectorUtils.h
    M llvm/lib/Analysis/VectorUtils.cpp
    M llvm/unittests/Analysis/VectorUtilsTest.cpp

  Log Message:
  -----------
  [VectorUtils] add IR-level analysis for widening of shuffle mask

This is similar to the recent move/addition of "scaleShuffleMask" (D76508),
but there are a couple of differences:

1. The existing x86 helper (canWidenShuffleElements) always tries to
   divide-by-2, so it gets called iteratively and wouldn't handle the
   general case of non-pow-2 length.
2. The existing x86 code handles "SM_SentinelZero" - we don't have
   that in IR, but this code should be safe to use with that or other
   special (negative) values.

The motivation is to enable shuffle folds in instcombine/vector-combine
that are similar to D76844 and D76727, but in the reverse-bitcast direction.
Those patterns are visible in the tests for D40633.

Differential Revision: https://reviews.llvm.org/D77881