[X86] Prefer VPERMQ over VPERM2F128 for any unary shuffle, not just the ones that can be done with a insertf128

The early out for AVX2 in lowerV2X128VectorShuffle is positioned in a weird spot below some shuffle mask equivalency checks.

But I think we want to allow VPERMQ for any unary shuffle.

Differential Revision: https://reviews.llvm.org/D37893

llvm-svn: 313373

1 files changed, 4 insertions, 3 deletions

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index ff00d9e5e00..4df7621bee2 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -12151,6 +12151,10 @@ static SDValue lowerV2X128VectorShuffle(const SDLoc &DL, MVT VT, SDValue V1,
                                         const APInt &Zeroable,
                                         const X86Subtarget &Subtarget,
                                         SelectionDAG &DAG) {
+  // With AVX2, use VPERMQ/VPERMPD for unary shuffles to allow memory folding.
+  if (Subtarget.hasAVX2() && V2.isUndef())
+    return SDValue();
+
   SmallVector<int, 4> WidenedMask;
   if (!canWidenShuffleElements(Mask, WidenedMask))
     return SDValue();
@@ -12174,9 +12178,6 @@ static SDValue lowerV2X128VectorShuffle(const SDLoc &DL, MVT VT, SDValue V1,
     // subvector.
     bool OnlyUsesV1 = isShuffleEquivalent(V1, V2, Mask, {0, 1, 0, 1});
     if (OnlyUsesV1 || isShuffleEquivalent(V1, V2, Mask, {0, 1, 4, 5})) {
-      // With AVX2, use VPERMQ/VPERMPD to allow memory folding.
-      if (Subtarget.hasAVX2() && V2.isUndef())
-        return SDValue();
 
       // With AVX1, use vperm2f128 (below) to allow load folding. Otherwise,
       // this will likely become vinsertf128 which can't fold a 256-bit memop.