Don't combine a shuffle of two BUILD_VECTORs in general.

[efriedma]

Summary

This combine causes problems in many situations. (Not sure exactly when it's safe; suggestions welcome.)

Fixes https://llvm.org/bugs/show_bug.cgi?id=31364 . Partially fixes https://llvm.org/bugs/show_bug.cgi?id=31301.

It's possible the change to combineShuffleOfScalars is a little too restrictive... but I'm not sure what a good heuristic would be.

[mkuper]

lib/CodeGen/SelectionDAG/DAGCombiner.cpp

14099 ↗

(On Diff #81499)

Just to make sure I understand - the reason you're removing the hasOneUse constraint is that for the case where the BUILD_VECTOR ends up a constant pool load, it's better to just load the combined vector instead of shuffling even if other users exist?

14361 ↗

(On Diff #81499)

Is there a reason this has to land in the same patch as the other change?
Also, could you explain why you consider splats free? On x86, at least, I don't see why that would be the case.

test/CodeGen/ARM/vzip.ll

270 ↗

(On Diff #81499)

Regenerate this with the update script?

test/CodeGen/X86/mmx-bitcast.ll

83 ↗

(On Diff #81499)

This looks like a common enough pattern that I'm not excited about regressing it at the expense of the "zip" example.

test/CodeGen/X86/vector-shuffle-128-v16.ll

670 ↗

(On Diff #81499)

Same as above.

[efriedma]

lib/CodeGen/SelectionDAG/DAGCombiner.cpp

14099 ↗

(On Diff #81499)

We might end up with something slightly more expensive in some cases, but it seems like a worthwhile tradeoff in most situations. The hasOneUse() check won't reliably preserve constants anyway; other parts of the code like to fold away bitcasts and insert undefs into constants.

14361 ↗

(On Diff #81499)

I can land it separately, sure.

Not all broadcasts are free, but some are; for example, on x86 with AVX2, a splat can be folded into a load to form vpbroadcastb. I'll add an x86 testcase which covers this.

test/CodeGen/ARM/vzip.ll

270 ↗

(On Diff #81499)

There is no update script for ARM, as far as I know; it's possible I should spend some time writing one.

test/CodeGen/X86/mmx-bitcast.ll

83 ↗

(On Diff #81499)

Mmm... maybe. I would appreciate suggestions.

[efriedma]

Split into multiple patches.

[RKSimon]

Not sure about this, would we be better off attempting a SHUFFLE(BUILD_VECTOR(), BUILD_VECTOR()) -> SHUFFLE(BUILD_VECTOR(), UNDEF) combine replacing repeated insertions with shuffles?

[mkuper]

test/CodeGen/ARM/vzip.ll

270 ↗

(On Diff #81499)

Right, forgot, I'm too used to x86, sorry. :-)

test/CodeGen/X86/mmx-bitcast.ll

83 ↗

(On Diff #81499)

I don't really have good ones.

My knee-jerk reaction is "do the (simplifying!) combine, and improve build_vector lowering". I guess, in general, that would require looking at the BUILD_VECTOR and trying to find the best way to split it into the best sequence of BUILD_VECTORs and target shuffles. After all, there's nothing guaranteeing the pre-combine BUILD_VECTORs were optimal in any way. What if you're shuffling two BUILD_VECTORs that are splats of the same value?

But this all sounds a bit scary.

[efriedma]

More "conservative" patch. This just tries to specifically detect splat-ish cases.

Not sure this actually covers every interesting case, but I don't have any other ideas.

[efriedma]

Not sure about this, would we be better off attempting a SHUFFLE(BUILD_VECTOR(), BUILD_VECTOR()) -> SHUFFLE(BUILD_VECTOR(), UNDEF) combine replacing repeated insertions with shuffles?

This doesn't reliably produce better code; for example, see https://llvm.org/bugs/show_bug.cgi?id=31301. The particular shuffle written by the user is cheaper than anything we're realistically going to produce from SHUFFLE(BUILD_VECTOR(), UNDEF) unless we add a special case specifically for ZIP(splat, splat), and I don't have the time to write code to special-case every possible permutation of SHUFFLE(splat, splat)/SHUFFLE(SHUFFLE(splat, splat), SHUFFLE(splat, splat))/etc. for every target supported by LLVM.

test/CodeGen/X86/mmx-bitcast.ll

83 ↗

(On Diff #81499)

(Phabricator won't let me delete this...)

[mkuper]

This LGTM with a comment nit.
I'm sure there still are cases it'll pessimize, but I think it's an improvement over what we had. Thanks, Eli!

lib/CodeGen/SelectionDAG/DAGCombiner.cpp

14132 ↗

(On Diff #81622)

"a non-constant BUILD_VECTOR operand".

llvm/trunk/lib/CodeGen/SelectionDAG/DAGCombiner.cpp

for (unsigned I = 0; I != NumConcats; ++I) {
     }
   }
 
   return DAG.getNode(ISD::CONCAT_VECTORS, SDLoc(N), VT, Ops);
 }
 
 // Attempt to combine a shuffle of 2 inputs of 'scalar sources' -
 // BUILD_VECTOR or SCALAR_TO_VECTOR into a single BUILD_VECTOR.
-// This combine is done in the following cases:
-// 1. Both N0,N1 are BUILD_VECTOR's composed of constants or undefs.
-// 2. Only one of N0,N1 is a BUILD_VECTOR composed of constants or undefs -
-//    Combine iff that node is ALL_ZEROS. We prefer not to combine a
-//    BUILD_VECTOR of all constants to allow efficient materialization of
-//    constant vectors, but the ALL_ZEROS is an exception because
-//    zero-extension matching seems to rely on having BUILD_VECTOR nodes with
-//    zero padding between elements. FIXME: Eliminate this exception for
-//    ALL_ZEROS constant vectors.
-// 3. Neither N0,N1 are composed of only constants.
+//
+// SHUFFLE(BUILD_VECTOR(), BUILD_VECTOR()) -> BUILD_VECTOR() is always
+// a simplification in some sense, but it isn't appropriate in general: some
+// BUILD_VECTORs are substantially cheaper than others. The general case
+// of a BUILD_VECTOR requires inserting each element individually (or
+// performing the equivalent in a temporary stack variable). A BUILD_VECTOR of
+// all constants is a single constant pool load.  A BUILD_VECTOR where each
+// element is identical is a splat.  A BUILD_VECTOR where most of the operands
+// are undef lowers to a small number of element insertions.
+//
+// To deal with this, we currently use a bunch of mostly arbitrary heuristics.
+// We don't fold shuffles where one side is a non-zero constant, and we don't
+// fold shuffles if the resulting BUILD_VECTOR would have duplicate
+// non-constant operands. This seems to work out reasonably well in practice.
 static SDValue combineShuffleOfScalars(ShuffleVectorSDNode *SVN,
                                        SelectionDAG &DAG,
                                        const TargetLowering &TLI) {
   EVT VT = SVN->getValueType(0);
   unsigned NumElts = VT.getVectorNumElements();
   SDValue N0 = SVN->getOperand(0);
   SDValue N1 = SVN->getOperand(1);
 
   if (!N0->hasOneUse() || !N1->hasOneUse())
     return SDValue();
   // If only one of N1,N2 is constant, bail out if it is not ALL_ZEROS as
   // discussed above.
   if (!N1.isUndef()) {
     bool N0AnyConst = isAnyConstantBuildVector(N0.getNode());
     bool N1AnyConst = isAnyConstantBuildVector(N1.getNode());
     if (N0AnyConst && !N1AnyConst && !ISD::isBuildVectorAllZeros(N0.getNode()))
       return SDValue();
     if (!N0AnyConst && N1AnyConst && !ISD::isBuildVectorAllZeros(N1.getNode()))
       return SDValue();
   }
 
   SmallVector<SDValue, 8> Ops;
+  SmallSet<SDValue, 16> DuplicateOps;
   for (int M : SVN->getMask()) {
     SDValue Op = DAG.getUNDEF(VT.getScalarType());
     if (M >= 0) {
       int Idx = M < (int)NumElts ? M : M - NumElts;
       SDValue &S = (M < (int)NumElts ? N0 : N1);
       if (S.getOpcode() == ISD::BUILD_VECTOR) {
         Op = S.getOperand(Idx);
       } else if (S.getOpcode() == ISD::SCALAR_TO_VECTOR) {
         if (Idx == 0)
           Op = S.getOperand(0);
       } else {
         // Operand can't be combined - bail out.
         return SDValue();
       }
     }
+
+    // Don't duplicate a non-constant BUILD_VECTOR operand; semantically, this is
+    // fine, but it's likely to generate low-quality code if the target can't
+    // reconstruct an appropriate shuffle.
+    if (!Op.isUndef() && !isa<ConstantSDNode>(Op) && !isa<ConstantFPSDNode>(Op))
+      if (!DuplicateOps.insert(Op).second)
+        return SDValue();
+
     Ops.push_back(Op);
   }
   // BUILD_VECTOR requires all inputs to be of the same type, find the
   // maximum type and extend them all.
   EVT SVT = VT.getScalarType();
   if (SVT.isInteger())
     for (SDValue &Op : Ops)
       SVT = (SVT.bitsLT(Op.getValueType()) ? Op.getValueType() : SVT);

llvm/trunk/test/CodeGen/ARM/vtrn.ll

 
 ; Here we get a build_vector node, where all the incoming extract_element
 ; values do modify the type. However, we get different input types, as some of
 ; them get truncated from i32 to i8 (from comparing cmp0 with cmp1) and some of
 ; them get truncated from i16 to i8 (from comparing cmp2 with cmp3).
 define <8 x i8> @vtrn_mismatched_builvector0(<8 x i8> %tr0, <8 x i8> %tr1,
                                              <4 x i32> %cmp0, <4 x i32> %cmp1,
                                              <4 x i16> %cmp2, <4 x i16> %cmp3) {
-  ; CHECK-LABEL: vtrn_mismatched_builvector0
+  ; CHECK-LABEL: vtrn_mismatched_builvector0:
   ; CHECK: vmovn.i32
-  ; CHECK: vtrn
   ; CHECK: vbsl
   %c0 = icmp ult <4 x i32> %cmp0, %cmp1
   %c1 = icmp ult <4 x i16> %cmp2, %cmp3
   %c = shufflevector <4 x i1> %c0, <4 x i1> %c1, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7>
   %rv = select <8 x i1> %c, <8 x i8> %tr0, <8 x i8> %tr1
   ret <8 x i8> %rv
 }
 
 ; Here we get a build_vector node, where half the incoming extract_element
 ; values do not modify the type (the values form cmp2), but half of them do
 ; (from the icmp operation).
 define <8 x i8> @vtrn_mismatched_builvector1(<8 x i8> %tr0, <8 x i8> %tr1,
                            <4 x i32> %cmp0, <4 x i32> %cmp1, <4 x i8> *%cmp2_ptr) {
-  ; CHECK-LABEL: vtrn_mismatched_builvector1
+  ; CHECK-LABEL: vtrn_mismatched_builvector1:
   ; We need to extend the 4 x i8 to 4 x i16 in order to perform the vtrn
   ; CHECK: vmovl
-  ; CHECK: vtrn.8
   ; CHECK: vbsl
   %cmp2_load = load <4 x i8>, <4 x i8> * %cmp2_ptr, align 4
   %cmp2 = trunc <4 x i8> %cmp2_load to <4 x i1>
   %c0 = icmp ult <4 x i32> %cmp0, %cmp1
   %c = shufflevector <4 x i1> %c0, <4 x i1> %cmp2, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7>
   %rv = select <8 x i1> %c, <8 x i8> %tr0, <8 x i8> %tr1
   ret <8 x i8> %rv
 }
 
 ; The shuffle mask is half a vtrn; we duplicate the half to produce the
 ; full result.
 define void @lower_twice_no_vtrn(<4 x i16>* %A, <4 x i16>* %B, <8 x i16>* %C) {
 entry:
-  ; CHECK-LABEL: lower_twice_no_vtrn
+  ; CHECK-LABEL: lower_twice_no_vtrn:
   ; CHECK: @ BB#0:
   ; CHECK-NEXT: vldr d16, [r1]
   ; CHECK-NEXT: vldr d18, [r0]
   ; CHECK-NEXT: vtrn.16 d18, d16
   ; CHECK-NEXT: vorr d17, d16, d16
   ; CHECK-NEXT: vst1.64 {d16, d17}, [r2]
   ; CHECK-NEXT: mov pc, lr
   %tmp1 = load <4 x i16>, <4 x i16>* %A
   %tmp2 = load <4 x i16>, <4 x i16>* %B
   %0 = shufflevector <4 x i16> %tmp1, <4 x i16> %tmp2, <8 x i32> <i32 undef, i32 5, i32 3, i32 7, i32 1, i32 5, i32 3, i32 7>
   store <8 x i16> %0, <8 x i16>* %C
   ret void
 }
 
 ; The shuffle mask is half a vtrn; we duplicate the half to produce the
 ; full result.
 define void @upper_twice_no_vtrn(<4 x i16>* %A, <4 x i16>* %B, <8 x i16>* %C) {
 entry:
-  ; CHECK-LABEL: upper_twice_no_vtrn
+  ; CHECK-LABEL: upper_twice_no_vtrn:
   ; CHECK: @ BB#0:
   ; CHECK-NEXT: vldr d16, [r1]
   ; CHECK-NEXT: vldr d18, [r0]
   ; CHECK-NEXT: vtrn.16 d18, d16
   ; CHECK-NEXT: vorr d19, d18, d18
   ; CHECK-NEXT: vst1.64 {d18, d19}, [r2]
   ; CHECK-NEXT: mov pc, lr
   %tmp1 = load <4 x i16>, <4 x i16>* %A
   %tmp2 = load <4 x i16>, <4 x i16>* %B
   %0 = shufflevector <4 x i16> %tmp1, <4 x i16> %tmp2, <8 x i32> <i32 0, i32 undef, i32 2, i32 6, i32 0, i32 4, i32 2, i32 6>
   store <8 x i16> %0, <8 x i16>* %C
   ret void
 }

llvm/trunk/test/CodeGen/ARM/vzip.ll

entry:
   ; CHECK: vext.16 d16, d16, d17, #3
   ; CHECK: vzip
   %tmp1 = load <8 x i16>, <8 x i16>* %A
   %0 = shufflevector <8 x i16> %tmp1, <8 x i16> undef, <4 x i32> <i32 4, i32 4, i32 5, i32 3>
   store <4 x i16> %0, <4 x i16>* %B
   ret void
 }
 
-; FIXME: This should generate a vzip
 define <8 x i8> @vdup_zip(i8* nocapture readonly %x, i8* nocapture readonly %y)  {
 entry:
   ; CHECK-LABEL: vdup_zip:
-  ; CHECK:      ldrb    r0, [r0]
-  ; CHECK-NEXT: ldrb    r1, [r1]
-  ; CHECK-NEXT: vmov.8  d16[0], r0
-  ; CHECK-NEXT: vmov.8  d16[1], r1
-  ; CHECK-NEXT: vmov.8  d16[2], r0
-  ; CHECK-NEXT: vmov.8  d16[3], r1
-  ; CHECK-NEXT: vmov.8  d16[4], r0
-  ; CHECK-NEXT: vmov.8  d16[5], r1
-  ; CHECK-NEXT: vmov.8  d16[6], r0
-  ; CHECK-NEXT: vmov.8  d16[7], r1
-  ; CHECK-NEXT: vmov    r0, r1, d16
+  ; CHECK: ldrb
+  ; CHECK-NEXT: ldrb
+  ; CHECK-NEXT: vmov.i16 d{{.*}}, #0x800
+  ; CHECK-NEXT: vmov.8
+  ; CHECK-NEXT: vmov.8
+  ; CHECK-NEXT: vtbl.8
+  ; CHECK-NEXT: vmov    r0, r1
   %0 = load i8, i8* %x, align 1
   %1 = insertelement <8 x i8> undef, i8 %0, i32 0
   %lane = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 undef, i32 undef, i32 undef, i32 undef>
   %2 = load i8, i8* %y, align 1
   %3 = insertelement <8 x i8> undef, i8 %2, i32 0
   %lane3 = shufflevector <8 x i8> %3, <8 x i8> undef, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 undef, i32 undef, i32 undef, i32 undef>
   %vzip.i = shufflevector <8 x i8> %lane, <8 x i8> %lane3, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
   ret <8 x i8> %vzip.i
 }

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

; AVX2-NEXT:    retq
   %r = shufflevector <4 x float> %a, <4 x float> %b, <5 x i32> <i32 0, i32 5, i32 1, i32 6, i32 3>
   store <5 x float> %r, <5 x float>* %p
   ret void
 }
 
 define void @v7i8(<4 x i8> %a, <4 x i8> %b, <7 x i8>* %p) nounwind {
 ; SSE2-LABEL: v7i8:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [0,255,0,255,0,255,255,255,255,255,255,255,255,255,255,255]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm1[0,1,2,3,4,6,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[0,1,2,0]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,4,6,5,7]
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8,8,9,9,10,10,11,11,12,12,13,13,14,14,15,15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[0,2,2,3,4,5,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[0,2,2,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[0,0,2,1,4,5,6,7]
-; SSE2-NEXT:    pand %xmm2, %xmm3
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,1,3]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [0,65535,0,65535,0,65535,65535,65535]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,0,2,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,4,7]
+; SSE2-NEXT:    pand %xmm2, %xmm1
 ; SSE2-NEXT:    pandn %xmm0, %xmm2
-; SSE2-NEXT:    por %xmm3, %xmm2
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm2[0,1,1,3]
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3]
-; SSE2-NEXT:    movdqa %xmm1, -{{[0-9]+}}(%rsp)
+; SSE2-NEXT:    por %xmm1, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [255,255,255,255,255,255,255,255]
+; SSE2-NEXT:    pand %xmm2, %xmm0
+; SSE2-NEXT:    packuswb %xmm0, %xmm0
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[0,1,1,3]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
+; SSE2-NEXT:    movdqa %xmm2, -{{[0-9]+}}(%rsp)
 ; SSE2-NEXT:    movb -{{[0-9]+}}(%rsp), %al
 ; SSE2-NEXT:    movb %al, 6(%rdi)
-; SSE2-NEXT:    movd %xmm0, (%rdi)
-; SSE2-NEXT:    pextrw $4, %xmm2, %eax
+; SSE2-NEXT:    movd %xmm1, (%rdi)
+; SSE2-NEXT:    pextrw $4, %xmm0, %eax
 ; SSE2-NEXT:    movw %ax, 4(%rdi)
 ; SSE2-NEXT:    retq
 ;
 ; SSE42-LABEL: v7i8:
 ; SSE42:       # BB#0:
-; SSE42-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0],zero,xmm0[12],zero,xmm0[4],zero,zero,xmm0[u,u,u,u,u,u,u,u,u]
-; SSE42-NEXT:    pextrb $0, %xmm1, 6(%rdi)
-; SSE42-NEXT:    pshufb {{.*#+}} xmm1 = zero,xmm1[8],zero,xmm1[8],zero,xmm1[12,0,u,u,u,u,u,u,u,u,u]
-; SSE42-NEXT:    por %xmm0, %xmm1
+; SSE42-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,1,3]
+; SSE42-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[8,9,8,9,4,5,8,9,0,1,12,13,0,1,14,15]
+; SSE42-NEXT:    pblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5,6,7]
+; SSE42-NEXT:    pextrb $12, %xmm1, 6(%rdi)
+; SSE42-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u]
 ; SSE42-NEXT:    pmovzxdq {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero
 ; SSE42-NEXT:    pmovzxwd {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
 ; SSE42-NEXT:    pextrw $4, %xmm1, 4(%rdi)
 ; SSE42-NEXT:    movd %xmm0, (%rdi)
 ; SSE42-NEXT:    retq
 ;
 ; AVX-LABEL: v7i8:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0],zero,xmm0[12],zero,xmm0[4],zero,zero,xmm0[u,u,u,u,u,u,u,u,u]
-; AVX-NEXT:    vpshufb {{.*#+}} xmm2 = zero,xmm1[8],zero,xmm1[8],zero,xmm1[12,0,u,u,u,u,u,u,u,u,u]
-; AVX-NEXT:    vpor %xmm0, %xmm2, %xmm0
-; AVX-NEXT:    vpmovzxdq {{.*#+}} xmm2 = xmm0[0],zero,xmm0[1],zero
-; AVX-NEXT:    vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
-; AVX-NEXT:    vpextrb $0, %xmm1, 6(%rdi)
-; AVX-NEXT:    vpextrw $4, %xmm0, 4(%rdi)
+; AVX-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,3,1,3]
+; AVX-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[8,9,8,9,4,5,8,9,0,1,12,13,0,1,14,15]
+; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5,6,7]
+; AVX-NEXT:    vpshufb {{.*#+}} xmm1 = xmm0[0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u]
+; AVX-NEXT:    vpmovzxdq {{.*#+}} xmm2 = xmm1[0],zero,xmm1[1],zero
+; AVX-NEXT:    vpmovzxwd {{.*#+}} xmm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
+; AVX-NEXT:    vpextrb $12, %xmm0, 6(%rdi)
+; AVX-NEXT:    vpextrw $4, %xmm1, 4(%rdi)
 ; AVX-NEXT:    vmovd %xmm2, (%rdi)
 ; AVX-NEXT:    retq
   %r = shufflevector <4 x i8> %a, <4 x i8> %b, <7 x i32> <i32 0, i32 6, i32 3, i32 6, i32 1, i32 7, i32 4>
   store <7 x i8> %r, <7 x i8>* %p
   ret void
 }
 
 define void @v7i16(<4 x i16> %a, <4 x i16> %b, <7 x i16>* %p) nounwind {

llvm/trunk/test/CodeGen/X86/vector-shuffle-128-v16.ll

; AVX512VL-NEXT:    retq
   %tmp3 = bitcast <4 x i32> %tmp2 to <16 x i8>
   %tmp4 = shufflevector <16 x i8> %tmp3, <16 x i8> undef, <16 x i32> <i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2>
   ret <16 x i8> %tmp4
 }
 
 define <16 x i8> @PR31364(i8* nocapture readonly %a, i8* nocapture readonly %b) {
 ; SSE2-LABEL: PR31364:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    xorl %eax, %eax
-; SSE2-NEXT:    movd %eax, %xmm1
 ; SSE2-NEXT:    movzbl (%rdi), %eax
-; SSE2-NEXT:    movd %eax, %xmm2
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3],xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
-; SSE2-NEXT:    movzbl (%rsi), %eax
-; SSE2-NEXT:    movd %eax, %xmm3
-; SSE2-NEXT:    movdqa %xmm3, %xmm0
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1],xmm4[2],xmm1[2],xmm4[3],xmm1[3],xmm4[4],xmm1[4],xmm4[5],xmm1[5],xmm4[6],xmm1[6],xmm4[7],xmm1[7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3],xmm3[4],xmm2[4],xmm3[5],xmm2[5],xmm3[6],xmm2[6],xmm3[7],xmm2[7]
-; SSE2-NEXT:    movdqa %xmm0, %xmm1
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3],xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSE2-NEXT:    movzbl (%rsi), %ecx
+; SSE2-NEXT:    shll $8, %ecx
+; SSE2-NEXT:    orl %eax, %ecx
+; SSE2-NEXT:    pxor %xmm0, %xmm0
+; SSE2-NEXT:    pxor %xmm1, %xmm1
+; SSE2-NEXT:    pinsrw $0, %ecx, %xmm1
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm0[0,1,2,3,5,4,4,4]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[0,1,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,7]
+; SSE2-NEXT:    packuswb %xmm2, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: PR31364:
 ; SSSE3:       # BB#0:
-; SSSE3-NEXT:    xorl %eax, %eax
-; SSSE3-NEXT:    movd %eax, %xmm1
 ; SSSE3-NEXT:    movzbl (%rdi), %eax
-; SSSE3-NEXT:    movd %eax, %xmm2
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3],xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
-; SSSE3-NEXT:    movzbl (%rsi), %eax
-; SSSE3-NEXT:    movd %eax, %xmm3
-; SSSE3-NEXT:    movdqa %xmm3, %xmm0
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSSE3-NEXT:    movdqa %xmm0, %xmm4
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1],xmm4[2],xmm1[2],xmm4[3],xmm1[3],xmm4[4],xmm1[4],xmm4[5],xmm1[5],xmm4[6],xmm1[6],xmm4[7],xmm1[7]
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3],xmm3[4],xmm2[4],xmm3[5],xmm2[5],xmm3[6],xmm2[6],xmm3[7],xmm2[7]
-; SSSE3-NEXT:    movdqa %xmm0, %xmm1
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3],xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7]
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
-; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSSE3-NEXT:    movzbl (%rsi), %ecx
+; SSSE3-NEXT:    shll $8, %ecx
+; SSSE3-NEXT:    orl %eax, %ecx
+; SSSE3-NEXT:    pxor %xmm0, %xmm0
+; SSSE3-NEXT:    pinsrw $0, %ecx, %xmm0
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1],zero,xmm0[1,1,1,1,1,0,0,0]
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: PR31364:
 ; SSE41:       # BB#0:
-; SSE41-NEXT:    movzbl (%rsi), %eax
-; SSE41-NEXT:    movd %eax, %xmm0
-; SSE41-NEXT:    pinsrb $1, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $2, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $3, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $4, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $5, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $6, %eax, %xmm0
-; SSE41-NEXT:    xorl %ecx, %ecx
-; SSE41-NEXT:    pinsrb $7, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $8, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $9, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $10, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $11, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $12, %eax, %xmm0
-; SSE41-NEXT:    movzbl (%rdi), %eax
-; SSE41-NEXT:    pinsrb $13, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $14, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $15, %eax, %xmm0
+; SSE41-NEXT:    pxor %xmm0, %xmm0
+; SSE41-NEXT:    pinsrb $0, (%rdi), %xmm0
+; SSE41-NEXT:    pinsrb $1, (%rsi), %xmm0
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1],zero,xmm0[1,1,1,1,1,0,0,0]
 ; SSE41-NEXT:    retq
 ;
-; AVX-LABEL: PR31364:
-; AVX:       # BB#0:
-; AVX-NEXT:    movzbl (%rsi), %eax
-; AVX-NEXT:    vmovd %eax, %xmm0
-; AVX-NEXT:    vpinsrb $1, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $2, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $3, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $4, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $5, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $6, %eax, %xmm0, %xmm0
-; AVX-NEXT:    xorl %ecx, %ecx
-; AVX-NEXT:    vpinsrb $7, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $8, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $9, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $10, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $11, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $12, %eax, %xmm0, %xmm0
-; AVX-NEXT:    movzbl (%rdi), %eax
-; AVX-NEXT:    vpinsrb $13, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $14, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $15, %eax, %xmm0, %xmm0
-; AVX-NEXT:    retq
+; AVX1OR2-LABEL: PR31364:
+; AVX1OR2:       # BB#0:
+; AVX1OR2-NEXT:    vpxor %xmm0, %xmm0, %xmm0
+; AVX1OR2-NEXT:    vpinsrb $0, (%rdi), %xmm0, %xmm0
+; AVX1OR2-NEXT:    vpinsrb $1, (%rsi), %xmm0, %xmm0
+; AVX1OR2-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1],zero,xmm0[1,1,1,1,1,0,0,0]
+; AVX1OR2-NEXT:    retq
+;
+; AVX512VL-LABEL: PR31364:
+; AVX512VL:       # BB#0:
+; AVX512VL-NEXT:    vpxord %xmm0, %xmm0, %xmm0
+; AVX512VL-NEXT:    vpinsrb $0, (%rdi), %xmm0, %xmm0
+; AVX512VL-NEXT:    vpinsrb $1, (%rsi), %xmm0, %xmm0
+; AVX512VL-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1],zero,xmm0[1,1,1,1,1,0,0,0]
+; AVX512VL-NEXT:    retq
   %v0 = load i8, i8* %a, align 1
   %vecins = insertelement <16 x i8> <i8 undef, i8 undef, i8 undef, i8 0, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef, i8 undef>, i8 %v0, i32 0
   %v1 = load i8, i8* %b, align 1
   %vecins2 = insertelement <16 x i8> %vecins, i8 %v1, i32 1
   %result = shufflevector <16 x i8> %vecins2, <16 x i8> undef, <16 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 3, i32 1, i32 1, i32 1, i32 1, i32 1, i32 0, i32 0, i32 0>
   ret <16 x i8> %result
 }
 
 define <16 x i8> @PR31301(i8* nocapture readonly %x, i8* nocapture readonly %y)  {
 ; SSE2-LABEL: PR31301:
 ; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movzbl (%rdi), %eax
+; SSE2-NEXT:    movd %eax, %xmm0
 ; SSE2-NEXT:    movzbl (%rsi), %eax
 ; SSE2-NEXT:    movd %eax, %xmm1
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[0,0,1,1]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [255,0,255,0,255,0,255,0,255,0,255,0,255,0,255,0]
+; SSE2-NEXT:    pand %xmm0, %xmm2
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    movzbl (%rdi), %eax
-; SSE2-NEXT:    movd %eax, %xmm0
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,0,1,1]
+; SSE2-NEXT:    pandn %xmm1, %xmm0
+; SSE2-NEXT:    por %xmm2, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: PR31301:
 ; SSSE3:       # BB#0: # %entry
-; SSSE3-NEXT:    movzbl (%rsi), %eax
-; SSSE3-NEXT:    movd %eax, %xmm1
-; SSSE3-NEXT:    movdqa {{.*#+}} xmm2 = [0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1]
-; SSSE3-NEXT:    pshufb %xmm2, %xmm1
 ; SSSE3-NEXT:    movzbl (%rdi), %eax
 ; SSSE3-NEXT:    movd %eax, %xmm0
-; SSSE3-NEXT:    pshufb %xmm2, %xmm0
+; SSSE3-NEXT:    movzbl (%rsi), %eax
+; SSSE3-NEXT:    movd %eax, %xmm1
 ; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSSE3-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: PR31301:
 ; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    movzbl (%rdi), %eax
+; SSE41-NEXT:    movd %eax, %xmm0
 ; SSE41-NEXT:    movzbl (%rsi), %eax
-; SSE41-NEXT:    movzbl (%rdi), %ecx
-; SSE41-NEXT:    movd %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $1, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $2, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $3, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $4, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $5, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $6, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $7, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $8, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $9, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $10, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $11, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $12, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $13, %eax, %xmm0
-; SSE41-NEXT:    pinsrb $14, %ecx, %xmm0
-; SSE41-NEXT:    pinsrb $15, %eax, %xmm0
-; SSE41-NEXT:    retq
-;
-; AVX-LABEL: PR31301:
-; AVX:       # BB#0: # %entry
-; AVX-NEXT:    movzbl (%rsi), %eax
-; AVX-NEXT:    movzbl (%rdi), %ecx
-; AVX-NEXT:    vmovd %ecx, %xmm0
-; AVX-NEXT:    vpinsrb $1, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $2, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $3, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $4, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $5, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $6, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $7, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $8, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $9, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $10, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $11, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $12, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $13, %eax, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $14, %ecx, %xmm0, %xmm0
-; AVX-NEXT:    vpinsrb $15, %eax, %xmm0, %xmm0
-; AVX-NEXT:    retq
+; SSE41-NEXT:    movd %eax, %xmm1
+; SSE41-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSE41-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: PR31301:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    movzbl (%rdi), %eax
+; AVX1-NEXT:    vmovd %eax, %xmm0
+; AVX1-NEXT:    movzbl (%rsi), %eax
+; AVX1-NEXT:    vmovd %eax, %xmm1
+; AVX1-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
+; AVX1-NEXT:    retq
+;
+; AVX2OR512VL-LABEL: PR31301:
+; AVX2OR512VL:       # BB#0: # %entry
+; AVX2OR512VL-NEXT:    movzbl (%rdi), %eax
+; AVX2OR512VL-NEXT:    vmovd %eax, %xmm0
+; AVX2OR512VL-NEXT:    movzbl (%rsi), %eax
+; AVX2OR512VL-NEXT:    vmovd %eax, %xmm1
+; AVX2OR512VL-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; AVX2OR512VL-NEXT:    vpbroadcastw %xmm0, %xmm0
+; AVX2OR512VL-NEXT:    retq
 entry:
   %0 = load i8, i8* %x, align 1
   %1 = insertelement <16 x i8> undef, i8 %0, i32 0
   %lane = shufflevector <16 x i8> %1, <16 x i8> undef, <16 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
   %2 = load i8, i8* %y, align 1
   %3 = insertelement <16 x i8> undef, i8 %2, i32 0
   %lane3 = shufflevector <16 x i8> %3, <16 x i8> undef, <16 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
   %vzip.i = shufflevector <16 x i8> %lane, <16 x i8> %lane3, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23>
   ret <16 x i8> %vzip.i
 }