[InstCombine] Teach InstCombine how to handle an obfuscated splat.

An obfuscated splat is where the frontend poorly generates code for a splat
using several different shuffles to create the splat, i.e.,

  %A = load <4 x float>* %in_ptr, align 16
  %B = shufflevector <4 x float> %A, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 undef, i32 undef>
  %C = shufflevector <4 x float> %B, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 4, i32 undef>
  %D = shufflevector <4 x float> %C, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 2, i32 4>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@166061 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Michael Gottesman 2012-10-16 21:29:38 +00:00
parent c5ac87d067
commit 4932bbe20c
2 changed files with 16 additions and 2 deletions

View File

@ -636,8 +636,11 @@ Instruction *InstCombiner::visitShuffleVectorInst(ShuffleVectorInst &SVI) {
// If LHS's width is changed, shift the mask value accordingly.
// If newRHS == NULL, i.e. LHSOp0 == RHSOp0, we want to remap any
// references to RHSOp0 to LHSOp0, so we don't need to shift the mask.
if (eltMask >= 0 && newRHS != NULL)
// references from RHSOp0 to LHSOp0, so we don't need to shift the mask.
// If newRHS == newLHS, we want to remap any references from newRHS to
// newLHS so that we can properly identify splats that may occur due to
// obfuscation accross the two vectors.
if (eltMask >= 0 && newRHS != NULL && newLHS != newRHS)
eltMask += newLHSWidth;
}

View File

@ -0,0 +1,11 @@
; RUN: opt -instcombine -S %s | FileCheck %s
define void @test(<4 x float> *%in_ptr, <4 x float> *%out_ptr) {
%A = load <4 x float>* %in_ptr, align 16
%B = shufflevector <4 x float> %A, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 undef, i32 undef>
%C = shufflevector <4 x float> %B, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 4, i32 undef>
%D = shufflevector <4 x float> %C, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
; CHECK: %D = shufflevector <4 x float> %A, <4 x float> undef, <4 x i32> zeroinitializer
store <4 x float> %D, <4 x float> *%out_ptr
ret void
}