[all-commits] [llvm/llvm-project] ce97ce: [VectorCombine] try to form a better extractelement

[RotateRight via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: ce97ce3a5d72163fa217f367b1f8c8cab84c5963
      https://github.com/llvm/llvm-project/commit/ce97ce3a5d72163fa217f367b1f8c8cab84c5963
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-04-03 (Fri, 03 Apr 2020)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/VectorCombine.cpp
    M llvm/test/Transforms/VectorCombine/X86/extract-binop.ll
    M llvm/test/Transforms/VectorCombine/X86/extract-cmp.ll

  Log Message:
  -----------
  [VectorCombine] try to form a better extractelement

Extracting to the same index that we are going to insert back into
allows forming select ("blend") shuffles and enables further transforms.

Admittedly, this is a quick-fix for a more general problem that I'm
hoping to solve by adding transforms for patterns that start with an
insertelement.

But this might resolve some regressions known to be caused by the
extract-extract transform (although I have not gotten more details on
those yet).

In the motivating case from PR34724:
https://bugs.llvm.org/show_bug.cgi?id=34724

The combination of subsequent instcombine and codegen transforms gets us this improvement:

  vmovshdup	%xmm0, %xmm2    ## xmm2 = xmm0[1,1,3,3]
  vhaddps	%xmm1, %xmm1, %xmm4
  vmovshdup	%xmm1, %xmm3    ## xmm3 = xmm1[1,1,3,3]
  vaddps	%xmm0, %xmm2, %xmm0
  vaddps	%xmm1, %xmm3, %xmm1
  vshufps	$200, %xmm4, %xmm0, %xmm0 ## xmm0 = xmm0[0,2],xmm4[0,3]
  vinsertps	$177, %xmm1, %xmm0, %xmm0 ## xmm0 = zero,xmm0[1,2],xmm1[2]

  -->

  vmovshdup	%xmm0, %xmm2    ## xmm2 = xmm0[1,1,3,3]
  vhaddps	%xmm1, %xmm1, %xmm1
  vaddps	%xmm0, %xmm2, %xmm0
  vshufps	$200, %xmm1, %xmm0, %xmm0 ## xmm0 = xmm0[0,2],xmm1[0,3]

Differential Revision: https://reviews.llvm.org/D76623