[X86][SSE] Enable target shuffle combining to combine multiple shuffle inputs.

[RKSimon]

Summary

We currently only support combining target shuffles that consist of a single source input (plus elements known to be undef/zero).

This patch generalizes the recursive combining of the target shuffle to collect all the inputs, merging any duplicates along the way, into a full set of src ops and its shuffle mask.

We uncover a number of cases where we have failed to combine a unary shuffle because the input has been duplicated and separated during lowering.

This will allow us to combine to 2-input shuffles in a future patch.

[RKSimon]

ping?

[andreadb]

Hi Simon,

Overall the patch lgtm.
My major concern was related to the complexity of your new algorithm. 'combineX86ShufflesRecursively' is now called for every "used" shuffle operand. Have you noticed any particular cases where this function is now expensive?
However, given that we already limit the maximum recursion depth to 8, I don't expect this to be much of an issue.

-Andrea

[RKSimon]

In D22859#504618, @andreadb wrote:

Hi Simon,

Overall the patch lgtm.
My major concern was related to the complexity of your new algorithm. 'combineX86ShufflesRecursively' is now called for every "used" shuffle operand. Have you noticed any particular cases where this function is now expensive?
However, given that we already limit the maximum recursion depth to 8, I don't expect this to be much of an issue.

-Andrea

Thanks Andrea.

I haven't found anything to indicate that the search has become notably more costly to builds - between the limits on depth and hasOneUse(), many shuffles already early out. If anybody can create a repro then I can investigate ways to manipulate the max search depth but this hasn't proved to be necessary yet.

llvm/trunk/lib/Target/X86/X86ISelLowering.cpp

 /// 1) Collapse generic shuffles to specialized single instructions when
 ///    equivalent. In most cases, this is just an encoding size win, but
 ///    sometimes we will collapse multiple generic shuffles into a single
 ///    special-purpose shuffle.
 /// 2) Look for sequences of shuffle instructions with 3 or more total
 ///    instructions, and replace them with the slightly more expensive SSSE3
 ///    PSHUFB instruction if available. We do this as the last combining step
 ///    to ensure we avoid using PSHUFB if we can implement the shuffle with
-///    a suitable short sequence of other instructions. The PHUFB will either
+///    a suitable short sequence of other instructions. The PSHUFB will either
 ///    use a register or have to read from memory and so is slightly (but only
 ///    slightly) more expensive than the other shuffle instructions.
 ///
 /// Because this is inherently a quadratic operation (for each shuffle in
 /// a chain, we recurse up the chain), the depth is limited to 8 instructions.
 /// This should never be an issue in practice as the shuffle lowering doesn't
 /// produce sequences of more than 8 instructions.
 ///
 /// FIXME: We will currently miss some cases where the redundant shuffling
 /// would simplify under the threshold for PSHUFB formation because of
 /// combine-ordering. To fix this, we should do the redundant instruction
 /// combining in this recursive walk.
-static bool combineX86ShufflesRecursively(SDValue Op, SDValue Root,
+static bool combineX86ShufflesRecursively(ArrayRef<SDValue> SrcOps,
+                                          int SrcOpIndex, SDValue Root,
                                           ArrayRef<int> RootMask,
                                           int Depth, bool HasVariableMask,
                                           SelectionDAG &DAG,
                                           TargetLowering::DAGCombinerInfo &DCI,
                                           const X86Subtarget &Subtarget) {
   // Bound the depth of our recursive combine because this is ultimately
   // quadratic in nature.
   if (Depth > 8)
     return false;
 
   // Directly rip through bitcasts to find the underlying operand.
+  SDValue Op = SrcOps[SrcOpIndex];
   Op = peekThroughOneUseBitcasts(Op);
 
   MVT VT = Op.getSimpleValueType();
   if (!VT.isVector())
     return false; // Bail if we hit a non-vector.
 
   assert(Root.getSimpleValueType().isVector() &&
          "Shuffles operate on vector types!");
   assert(VT.getSizeInBits() == Root.getSimpleValueType().getSizeInBits() &&
          "Can only combine shuffles of the same vector register size.");
 
   // Extract target shuffle mask and resolve sentinels and inputs.
   SDValue Input0, Input1;
   SmallVector<int, 16> OpMask;
   if (!resolveTargetShuffleInputs(Op, Input0, Input1, OpMask))
     return false;
 
+  // Add the inputs to the Ops list, avoiding duplicates.
+  SmallVector<SDValue, 8> Ops(SrcOps.begin(), SrcOps.end());
+
+  int InputIdx0 = -1, InputIdx1 = -1;
+  for (int i = 0, e = Ops.size(); i < e; ++i) {
+    SDValue BC = peekThroughBitcasts(Ops[i]);
+    if (Input0 && BC == peekThroughBitcasts(Input0))
+      InputIdx0 = i;
+    if (Input1 && BC == peekThroughBitcasts(Input1))
+      InputIdx1 = i;
+  }
+
+  if (Input0 && InputIdx0 < 0) {
+    InputIdx0 = SrcOpIndex;
+    Ops[SrcOpIndex] = Input0;
+  }
+  if (Input1 && InputIdx1 < 0) {
+    InputIdx1 = Ops.size();
+    Ops.push_back(Input1);
+  }
+
   assert(VT.getVectorNumElements() == OpMask.size() &&
          "Different mask size from vector size!");
   assert(((RootMask.size() > OpMask.size() &&
            RootMask.size() % OpMask.size() == 0) ||
           (OpMask.size() > RootMask.size() &&
            OpMask.size() % RootMask.size() == 0) ||
           OpMask.size() == RootMask.size()) &&
          "The smaller number of elements must divide the larger.");
for (int i = 0; i < MaskWidth; ++i) {
     int RootIdx = i / RootRatio;
     if (RootMask[RootIdx] < 0) {
       // This is a zero or undef lane, we're done.
       Mask.push_back(RootMask[RootIdx]);
       continue;
     }
 
     int RootMaskedIdx = RootMask[RootIdx] * RootRatio + i % RootRatio;
+
+    // Just insert the scaled root mask value if it references an input other
+    // than the SrcOp we're currently inserting.
+    if ((RootMaskedIdx < (SrcOpIndex * MaskWidth)) ||
+        (((SrcOpIndex + 1) * MaskWidth) <= RootMaskedIdx)) {
+      Mask.push_back(RootMaskedIdx);
+      continue;
+    }
+
+    RootMaskedIdx %= MaskWidth;
+
     int OpIdx = RootMaskedIdx / OpRatio;
     if (OpMask[OpIdx] < 0) {
       // The incoming lanes are zero or undef, it doesn't matter which ones we
       // are using.
       Mask.push_back(OpMask[OpIdx]);
       continue;
     }
 
-    // Ok, we have non-zero lanes, map them through.
-    Mask.push_back(OpMask[OpIdx] * OpRatio +
-                   RootMaskedIdx % OpRatio);
+    // Ok, we have non-zero lanes, map them through to one of the Op's inputs.
+    int OpMaskedIdx = OpMask[OpIdx] * OpRatio + RootMaskedIdx % OpRatio;
+    OpMaskedIdx %= MaskWidth;
+
+    if (OpMask[OpIdx] < (int)OpMask.size()) {
+      assert(0 <= InputIdx0 && "Unknown target shuffle input");
+      OpMaskedIdx += InputIdx0 * MaskWidth;
+    } else {
+      assert(0 <= InputIdx1 && "Unknown target shuffle input");
+      OpMaskedIdx += InputIdx1 * MaskWidth;
+    }
+
+    Mask.push_back(OpMaskedIdx);
   }
 
   // Handle the all undef/zero cases early.
   if (llvm::all_of(Mask, [](int Idx) { return Idx == SM_SentinelUndef; })) {
     DCI.CombineTo(Root.getNode(), DAG.getUNDEF(Root.getValueType()));
     return true;
   }
   if (llvm::all_of(Mask, [](int Idx) { return Idx < 0; })) {
     // TODO - should we handle the mixed zero/undef case as well? Just returning
     // a zero mask will lose information on undef elements possibly reducing
     // future combine possibilities.
     DCI.CombineTo(Root.getNode(), getZeroVector(Root.getSimpleValueType(),
                                                 Subtarget, DAG, SDLoc(Root)));
     return true;
   }
 
-  int MaskSize = Mask.size();
-  bool UseInput0 = std::any_of(Mask.begin(), Mask.end(),
-                  [MaskSize](int Idx) { return 0 <= Idx && Idx < MaskSize; });
-  bool UseInput1 = std::any_of(Mask.begin(), Mask.end(),
-                  [MaskSize](int Idx) { return MaskSize <= Idx; });
-
-  // At the moment we can only combine unary shuffle mask cases.
-  if (UseInput0 && UseInput1)
-    return false;
-  else if (UseInput1) {
-    std::swap(Input0, Input1);
-    ShuffleVectorSDNode::commuteMask(Mask);
+  // Remove unused shuffle source ops.
+  SmallVector<SDValue, 8> UsedOps;
+  for (int i = 0, e = Ops.size(); i < e; ++i) {
+    int lo = UsedOps.size() * MaskWidth;
+    int hi = lo + MaskWidth;
+    if (std::any_of(Mask.begin(), Mask.end(),
+                     [lo, hi](int i) { return (lo <= i) && (i < hi); })) {
+      UsedOps.push_back(Ops[i]);
+      continue;
     }
-
-  assert(Input0 && "Shuffle with no inputs detected");
+    for (int &M : Mask)
+      if (lo <= M)
+        M -= MaskWidth;
+  }
+  assert(!UsedOps.empty() && "Shuffle with no inputs detected");
+  Ops = UsedOps;
 
   HasVariableMask |= isTargetShuffleVariableMask(Op.getOpcode());
 
-  // See if we can recurse into Input0 (if it's a target shuffle).
-  if (Op->isOnlyUserOf(Input0.getNode()) &&
-      combineX86ShufflesRecursively(Input0, Root, Mask, Depth + 1,
+  // See if we can recurse into each shuffle source op (if it's a target shuffle).
+  for (int i = 0, e = Ops.size(); i < e; ++i)
+    if (Ops[i].getNode()->hasOneUse() || Op->isOnlyUserOf(Ops[i].getNode()))
+      if (combineX86ShufflesRecursively(Ops, i, Root, Mask, Depth + 1,
                                         HasVariableMask, DAG, DCI, Subtarget))
         return true;
 
+  // At the moment we can only combine unary shuffle mask cases.
+  if (Ops.size() != 1)
+    return false;
+
   // Minor canonicalization of the accumulated shuffle mask to make it easier
   // to match below. All this does is detect masks with sequential pairs of
   // elements, and shrink them to the half-width mask. It does this in a loop
   // so it will reduce the size of the mask to the minimal width mask which
   // performs an equivalent shuffle.
   SmallVector<int, 16> WidenedMask;
   while (Mask.size() > 1 && canWidenShuffleElements(Mask, WidenedMask)) {
     Mask = std::move(WidenedMask);
   }
 
-  return combineX86ShuffleChain(Input0, Root, Mask, Depth, HasVariableMask, DAG,
+  return combineX86ShuffleChain(Ops[0], Root, Mask, Depth, HasVariableMask, DAG,
                                 DCI, Subtarget);
 }
 
 /// \brief Get the PSHUF-style mask from PSHUF node.
 ///
 /// This is a very minor wrapper around getTargetShuffleMask to easy forming v4
 /// PSHUF-style masks that can be reused with such instructions.
 static SmallVector<int, 4> getPSHUFShuffleMask(SDValue N) {
static SDValue combineShuffle(SDNode *N, SelectionDAG &DAG,
   SmallVector<SDValue, 16> Elts;
   for (unsigned i = 0, e = VT.getVectorNumElements(); i != e; ++i)
     Elts.push_back(getShuffleScalarElt(N, i, DAG, 0));
 
   if (SDValue LD = EltsFromConsecutiveLoads(VT, Elts, dl, DAG, true))
     return LD;
 
   if (isTargetShuffle(N->getOpcode())) {
-    if (SDValue Shuffle =
-            combineTargetShuffle(SDValue(N, 0), DAG, DCI, Subtarget))
+    SDValue Op(N, 0);
+    if (SDValue Shuffle = combineTargetShuffle(Op, DAG, DCI, Subtarget))
       return Shuffle;
 
     // Try recursively combining arbitrary sequences of x86 shuffle
     // instructions into higher-order shuffles. We do this after combining
     // specific PSHUF instruction sequences into their minimal form so that we
     // can evaluate how many specialized shuffle instructions are involved in
     // a particular chain.
     SmallVector<int, 1> NonceMask; // Just a placeholder.
     NonceMask.push_back(0);
-    if (combineX86ShufflesRecursively(SDValue(N, 0), SDValue(N, 0), NonceMask,
+    if (combineX86ShufflesRecursively({Op}, 0, Op, NonceMask,
                                       /*Depth*/ 1, /*HasPSHUFB*/ false, DAG,
                                       DCI, Subtarget))
       return SDValue(); // This routine will use CombineTo to replace N.
   }
 
   return SDValue();
 }
 

llvm/trunk/test/CodeGen/X86/avx512-build-vector.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -march=x86-64 -mtriple=x86_64-apple-darwin -mcpu=knl | FileCheck %s
 
 define <16 x i32> @test2(<16 x i32> %x) {
 ; CHECK-LABEL: test2:
 ; CHECK:       ## BB#0:
 ; CHECK-NEXT:    vpternlogd $255, %zmm1, %zmm1, %zmm1
 ; CHECK-NEXT:    vpaddd %zmm1, %zmm0, %zmm0
 ; CHECK-NEXT:    retq
    %res = add <16 x i32><i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1>, %x
    ret <16 x i32>%res
 }
 
 define <16 x float> @test3(<4 x float> %a) {
 ; CHECK-LABEL: test3:
 ; CHECK:       ## BB#0:
-; CHECK-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; CHECK-NEXT:    vxorps %xmm2, %xmm2, %xmm2
-; CHECK-NEXT:    vmovss {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3]
-; CHECK-NEXT:    vmovss {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3]
-; CHECK-NEXT:    vshufps {{.*#+}} xmm0 = xmm1[1,0],xmm0[0,1]
-; CHECK-NEXT:    vinsertf128 $1, %xmm0, %ymm2, %ymm0
-; CHECK-NEXT:    vxorps %ymm1, %ymm1, %ymm1
+; CHECK-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; CHECK-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[8,9,10,11,0,1,2,3],zero,zero,zero,zero
+; CHECK-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
+; CHECK-NEXT:    vpxor %ymm1, %ymm1, %ymm1
 ; CHECK-NEXT:    vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
 ; CHECK-NEXT:    retq
   %b = extractelement <4 x float> %a, i32 2
   %c = insertelement <16 x float> <float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float undef, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00>, float %b, i32 5
   %b1 = extractelement <4 x float> %a, i32 0
   %c1 = insertelement <16 x float> %c, float %b1, i32 6
   ret <16 x float>%c1
 }

llvm/trunk/test/CodeGen/X86/clear_upper_vector_element_bits.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=SSE --check-prefix=SSE2
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
 
 ;
 ; PR6455 'Clear Upper Bits' Patterns
 ;
 
 define <2 x i64> @_clearupper2xi64a(<2 x i64>) nounwind {
 ; SSE-LABEL: _clearupper2xi64a:
 ; SSE:       # BB#0:
-; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; SSE-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; SSE-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE-NEXT:    andps {{.*}}(%rip), %xmm0
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: _clearupper2xi64a:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
+; AVX-NEXT:    vandps {{.*}}(%rip), %xmm0, %xmm0
 ; AVX-NEXT:    retq
   %x0 = extractelement <2 x i64> %0, i32 0
   %x1 = extractelement <2 x i64> %0, i32 1
   %trunc0 = trunc i64 %x0 to i32
   %trunc1 = trunc i64 %x1 to i32
   %ext0 = zext i32 %trunc0 to i64
   %ext1 = zext i32 %trunc1 to i64
   %v0 = insertelement <2 x i64> undef, i64 %ext0, i32 0
   %v1 = insertelement <2 x i64> %v0,   i64 %ext1, i32 1
   ret <2 x i64> %v1
 }
 
 define <4 x i32> @_clearupper4xi32a(<4 x i32>) nounwind {
 ; SSE-LABEL: _clearupper4xi32a:
 ; SSE:       # BB#0:
-; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; SSE-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
-; SSE-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[3,1,2,3]
-; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
-; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
-; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
-; SSE-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE-NEXT:    andps {{.*}}(%rip), %xmm0
 ; SSE-NEXT:    retq
 ;
 ; AVX1-LABEL: _clearupper4xi32a:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vandps {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: _clearupper4xi32a:

llvm/trunk/test/CodeGen/X86/vector-shuffle-256-v16.ll

 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[0,1,0,1,0,1,8,9,0,1,0,1,0,1,0,1]
 ; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
 ; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v16i16_00_00_00_12_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,0,1,8,9,0,1,0,1,0,1,0,1,16,17,16,17,16,17,16,17,16,17,16,17,16,17,16,17]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 0, i32 0, i32 12, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <16 x i16> %shuffle
 }
 
 define <16 x i16> @shuffle_v16i16_00_00_13_00_00_00_00_00_00_00_00_00_00_00_00_00(<16 x i16> %a, <16 x i16> %b) {
 ; AVX1-LABEL: shuffle_v16i16_00_00_13_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
 ; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3],xmm1[4,5,6,7]
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[0,1,0,1,10,11,0,1,0,1,0,1,0,1,0,1]
 ; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
 ; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v16i16_00_00_13_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,10,11,0,1,0,1,0,1,0,1,0,1,16,17,16,17,16,17,16,17,16,17,16,17,16,17,16,17]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 0, i32 13, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <16 x i16> %shuffle
 }
 
 define <16 x i16> @shuffle_v16i16_00_14_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<16 x i16> %a, <16 x i16> %b) {
 ; AVX1-LABEL: shuffle_v16i16_00_14_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
 ; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3],xmm1[4,5,6,7]
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[0,1,12,13,0,1,0,1,0,1,0,1,0,1,0,1]
 ; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
 ; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v16i16_00_14_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,12,13,0,1,0,1,0,1,0,1,0,1,0,1,16,17,16,17,16,17,16,17,16,17,16,17,16,17,16,17]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 0, i32 14, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <16 x i16> %shuffle
 }
 
 define <16 x i16> @shuffle_v16i16_15_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<16 x i16> %a, <16 x i16> %b) {
 ; AVX1-LABEL: shuffle_v16i16_15_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
 ; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3],xmm1[4,5,6,7]
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[14,15,0,1,0,1,0,1,0,1,0,1,0,1,0,1]
 ; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
 ; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v16i16_15_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[14,15,0,1,0,1,0,1,0,1,0,1,0,1,0,1,16,17,16,17,16,17,16,17,16,17,16,17,16,17,16,17]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <16 x i16> %a, <16 x i16> %b, <16 x i32> <i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <16 x i16> %shuffle
 }
 
 define <16 x i16> @shuffle_v16i16_00_00_00_00_00_00_00_00_08_08_08_08_08_08_08_08(<16 x i16> %a, <16 x i16> %b) {
 ; AVX1-LABEL: shuffle_v16i16_00_00_00_00_00_00_00_00_08_08_08_08_08_08_08_08:

llvm/trunk/test/CodeGen/X86/vector-shuffle-256-v32.ll

 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,zero,zero,zero,zero,zero,xmm2[8],zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0,0,0,0,0,0],zero,xmm0[0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_00_00_00_00_00_24_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,0,0,0,0,0,8,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 24, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_00_00_00_00_00_25_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_00_00_00_00_00_25_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,zero,zero,zero,zero,xmm2[9],zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0,0,0,0,0],zero,xmm0[0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_00_00_00_00_25_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,0,0,0,0,9,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 25, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_00_00_00_00_26_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_00_00_00_00_26_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,zero,zero,zero,xmm2[10],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0,0,0,0],zero,xmm0[0,0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_00_00_00_26_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,0,0,0,10,0,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 26, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_00_00_00_27_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_00_00_00_27_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,zero,zero,xmm2[11],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0,0,0],zero,xmm0[0,0,0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_00_00_27_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,0,0,11,0,0,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 0, i32 27, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_00_00_28_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_00_00_28_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,zero,xmm2[12],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0,0],zero,xmm0[0,0,0,0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_00_28_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,0,12,0,0,0,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 0, i32 28, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_00_29_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_00_29_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,zero,xmm2[13],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,0],zero,xmm0[0,0,0,0,0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_00_29_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,0,13,0,0,0,0,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 0, i32 29, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_00_30_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_00_30_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm2 = zero,xmm2[14],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0],zero,xmm0[0,0,0,0,0,0,0,0,0,0,0,0,0,0]
 ; AVX1-NEXT:    vpor %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_00_30_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,14,0,0,0,0,0,0,0,0,0,0,0,0,0,0,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 0, i32 30, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 
 define <32 x i8> @shuffle_v32i8_31_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00(<32 x i8> %a, <32 x i8> %b) {
 ; AVX1-LABEL: shuffle_v32i8_31_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm1
 ; AVX1-NEXT:    movl $128, %eax
 ; AVX1-NEXT:    vmovd %eax, %xmm2
 ; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm2
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
 ; AVX1-NEXT:    vpor %xmm0, %xmm2, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: shuffle_v32i8_31_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00_00:
 ; AVX2:       # BB#0:
-; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm0[2,3,0,1]
-; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,3,0,1]
 ; AVX2-NEXT:    movl $15, %eax
 ; AVX2-NEXT:    vmovd %eax, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> %b, <32 x i32> <i32 31, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
   ret <32 x i8> %shuffle
 }
 

llvm/trunk/test/CodeGen/X86/vector-shuffle-combining.ll

 define void @combine_test23(<8 x float> %v, <2 x float>* %ptr) {
 ; SSE-LABEL: combine_test23:
 ; SSE:       # BB#0:
 ; SSE-NEXT:    movups %xmm0, (%rdi)
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: combine_test23:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT:    vinsertps {{.*#+}} xmm1 = xmm0[0,1],xmm1[0],xmm0[3]
-; AVX-NEXT:    vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3]
 ; AVX-NEXT:    vmovups %xmm0, (%rdi)
 ; AVX-NEXT:    vzeroupper
 ; AVX-NEXT:    retq
   %idx2 = getelementptr inbounds <2 x float>, <2 x float>* %ptr, i64 1
   %shuffle0 = shufflevector <8 x float> %v, <8 x float> undef, <2 x i32> <i32 0, i32 1>
   %shuffle1 = shufflevector <8 x float> %v, <8 x float> undef, <2 x i32> <i32 2, i32 3>
   store <2 x float> %shuffle0, <2 x float>* %ptr, align 8
   store <2 x float> %shuffle1, <2 x float>* %idx2, align 8