[X86][AVX] Fold concat(packus(),packus()) -> packus(concat(),concat()) (PR34773)

Basic "revectorization" combine, we can probably do more opcodes here but it can be a tricky cost-benefit depending on where the subvectors came from - but this case helps shuffle combining.

llvm-svn: 360134

8 files changed, 106 insertions, 92 deletions

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 9659ae5a74a..e2abb30653c 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -42611,6 +42611,30 @@ static SDValue combineConcatVectorOps(const SDLoc &DL, MVT VT,
       return DAG.getNode(X86ISD::VBROADCAST, DL, VT, Op0.getOperand(0));
   }
 
+  // Repeated opcode.
+  if (llvm::all_of(Ops, [Op0](SDValue Op) {
+        return Op.getOpcode() == Op0.getOpcode();
+      })) {
+    unsigned NumOps = Ops.size();
+    switch (Op0.getOpcode()) {
+    case X86ISD::PACKUS:
+      if (NumOps == 2 && VT.is256BitVector() && Subtarget.hasInt256()) {
+        SmallVector<SDValue, 2> LHS, RHS;
+        for (unsigned i = 0; i != NumOps; ++i) {
+          LHS.push_back(Ops[i].getOperand(0));
+          RHS.push_back(Ops[i].getOperand(1));
+        }
+        MVT SrcVT = Op0.getOperand(0).getSimpleValueType();
+        SrcVT = MVT::getVectorVT(SrcVT.getScalarType(),
+                                 NumOps * SrcVT.getVectorNumElements());
+        return DAG.getNode(Op0.getOpcode(), DL, VT,
+                           DAG.getNode(ISD::CONCAT_VECTORS, DL, SrcVT, LHS),
+                           DAG.getNode(ISD::CONCAT_VECTORS, DL, SrcVT, RHS));
+      }
+      break;
+    }
+  }
+
   // If we're inserting all zeros into the upper half, change this to
   // an insert into an all zeros vector. We will match this to a move
   // with implicit upper bit zeroing during isel.
diff --git a/llvm/test/CodeGen/X86/avg.ll b/llvm/test/CodeGen/X86/avg.ll
index 011640c0d3f..cfa9f11a9c7 100644
--- a/llvm/test/CodeGen/X86/avg.ll
+++ b/llvm/test/CodeGen/X86/avg.ll
@@ -379,72 +379,68 @@ define void @avg_v48i8(<48 x i8>* %a, <48 x i8>* %b) nounwind {
 ;
 ; AVX2-LABEL: avg_v48i8:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
-; AVX2-NEXT:    vpbroadcastq 24(%rdi), %ymm1
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
-; AVX2-NEXT:    vmovdqa (%rdi), %xmm2
-; AVX2-NEXT:    vmovdqa 32(%rdi), %xmm3
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm4 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,3,0,1]
+; AVX2-NEXT:    vmovdqa (%rdi), %xmm0
+; AVX2-NEXT:    vmovdqa 32(%rdi), %xmm1
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
 ; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm5 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,3,0,1]
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX2-NEXT:    vpbroadcastq 24(%rdi), %ymm3
 ; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm6 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm6, %ymm0, %ymm0
-; AVX2-NEXT:    vpbroadcastq 24(%rsi), %ymm6
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm6 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm6, %ymm1, %ymm1
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm4 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm5 = xmm1[2,3,0,1]
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm5 = xmm5[0],zero,zero,zero,xmm5[1],zero,zero,zero,xmm5[2],zero,zero,zero,xmm5[3],zero,zero,zero,xmm5[4],zero,zero,zero,xmm5[5],zero,zero,zero,xmm5[6],zero,zero,zero,xmm5[7],zero,zero,zero
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm8 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
 ; AVX2-NEXT:    vmovdqa (%rsi), %xmm6
 ; AVX2-NEXT:    vmovdqa 32(%rsi), %xmm7
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm8 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm8, %ymm4, %ymm4
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm6 = xmm6[2,3,0,1]
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm6 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm6, %ymm2, %ymm2
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm6 = xmm7[0],zero,zero,zero,xmm7[1],zero,zero,zero,xmm7[2],zero,zero,zero,xmm7[3],zero,zero,zero,xmm7[4],zero,zero,zero,xmm7[5],zero,zero,zero,xmm7[6],zero,zero,zero,xmm7[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm6, %ymm5, %ymm5
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm6 = xmm7[2,3,0,1]
-; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm6 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero
-; AVX2-NEXT:    vpaddd %ymm6, %ymm3, %ymm3
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm6[2,3,0,1]
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm1, %ymm2, %ymm1
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm2 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpbroadcastq 24(%rsi), %ymm2
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm2, %ymm3, %ymm2
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm3 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm3, %ymm4, %ymm3
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm4 = xmm7[2,3,0,1]
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm4 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero,xmm4[2],zero,zero,zero,xmm4[3],zero,zero,zero,xmm4[4],zero,zero,zero,xmm4[5],zero,zero,zero,xmm4[6],zero,zero,zero,xmm4[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm4, %ymm5, %ymm4
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm5 = xmm7[0],zero,zero,zero,xmm7[1],zero,zero,zero,xmm7[2],zero,zero,zero,xmm7[3],zero,zero,zero,xmm7[4],zero,zero,zero,xmm7[5],zero,zero,zero,xmm7[6],zero,zero,zero,xmm7[7],zero,zero,zero
+; AVX2-NEXT:    vpaddd %ymm5, %ymm8, %ymm5
 ; AVX2-NEXT:    vpcmpeqd %ymm6, %ymm6, %ymm6
-; AVX2-NEXT:    vpsubd %ymm6, %ymm0, %ymm7
 ; AVX2-NEXT:    vpsubd %ymm6, %ymm1, %ymm1
-; AVX2-NEXT:    vpsubd %ymm6, %ymm4, %ymm4
+; AVX2-NEXT:    vpsubd %ymm6, %ymm0, %ymm0
 ; AVX2-NEXT:    vpsubd %ymm6, %ymm2, %ymm2
+; AVX2-NEXT:    vpsubd %ymm6, %ymm3, %ymm3
+; AVX2-NEXT:    vpsubd %ymm6, %ymm4, %ymm4
 ; AVX2-NEXT:    vpsubd %ymm6, %ymm5, %ymm5
-; AVX2-NEXT:    vpsubd %ymm6, %ymm3, %ymm0
-; AVX2-NEXT:    vpsrld $1, %ymm0, %ymm0
-; AVX2-NEXT:    vpsrld $1, %ymm5, %ymm3
-; AVX2-NEXT:    vpsrld $1, %ymm2, %ymm2
+; AVX2-NEXT:    vpsrld $1, %ymm5, %ymm5
 ; AVX2-NEXT:    vpsrld $1, %ymm4, %ymm4
+; AVX2-NEXT:    vpsrld $1, %ymm3, %ymm3
+; AVX2-NEXT:    vpsrld $1, %ymm2, %ymm2
+; AVX2-NEXT:    vpsrld $1, %ymm0, %ymm0
 ; AVX2-NEXT:    vpsrld $1, %ymm1, %ymm1
-; AVX2-NEXT:    vpsrld $1, %ymm7, %ymm5
-; AVX2-NEXT:    vextracti128 $1, %ymm5, %xmm6
-; AVX2-NEXT:    vpackusdw %xmm6, %xmm5, %xmm5
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %xmm6, %xmm5, %xmm5
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm7
-; AVX2-NEXT:    vpackusdw %xmm7, %xmm1, %xmm1
-; AVX2-NEXT:    vpand %xmm6, %xmm1, %xmm1
-; AVX2-NEXT:    vpackuswb %xmm1, %xmm5, %xmm1
-; AVX2-NEXT:    vextracti128 $1, %ymm4, %xmm5
-; AVX2-NEXT:    vpackusdw %xmm5, %xmm4, %xmm4
-; AVX2-NEXT:    vpand %xmm6, %xmm4, %xmm4
-; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm5
-; AVX2-NEXT:    vpackusdw %xmm5, %xmm2, %xmm2
-; AVX2-NEXT:    vpand %xmm6, %xmm2, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm4, %xmm2
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm3, %xmm2
-; AVX2-NEXT:    vpackusdw %xmm2, %xmm3, %xmm2
-; AVX2-NEXT:    vpand %xmm6, %xmm2, %xmm2
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm3
-; AVX2-NEXT:    vpackusdw %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpand %xmm6, %xmm0, %xmm0
-; AVX2-NEXT:    vpackuswb %xmm0, %xmm2, %xmm0
-; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
-; AVX2-NEXT:    vmovdqu %ymm1, (%rax)
+; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm6 = ymm0[2,3],ymm1[2,3]
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpackusdw %ymm6, %ymm0, %ymm0
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm1 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm6 = ymm3[2,3],ymm2[2,3]
+; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm3, %ymm2
+; AVX2-NEXT:    vpackusdw %ymm6, %ymm2, %ymm2
+; AVX2-NEXT:    vpand %ymm1, %ymm2, %ymm2
+; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm3
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm2[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm3, %ymm0
+; AVX2-NEXT:    vperm2i128 {{.*#+}} ymm2 = ymm5[2,3],ymm4[2,3]
+; AVX2-NEXT:    vinserti128 $1, %xmm4, %ymm5, %ymm3
+; AVX2-NEXT:    vpackusdw %ymm2, %ymm3, %ymm2
+; AVX2-NEXT:    vpand %ymm1, %ymm2, %ymm1
+; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqu %xmm1, (%rax)
+; AVX2-NEXT:    vmovdqu %ymm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/masked_store_trunc.ll b/llvm/test/CodeGen/X86/masked_store_trunc.ll
index b0479ef6be4..b9a3dc2b298 100644
--- a/llvm/test/CodeGen/X86/masked_store_trunc.ll
+++ b/llvm/test/CodeGen/X86/masked_store_trunc.ll
@@ -5587,13 +5587,12 @@ define void @truncstore_v32i16_v32i8(<32 x i16> %x, <32 x i8>* %p, <32 x i8> %ma
 ; AVX2-NEXT:    vpxor %xmm3, %xmm3, %xmm3
 ; AVX2-NEXT:    vpcmpeqb %ymm3, %ymm2, %ymm3
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm4, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm5
-; AVX2-NEXT:    vpackuswb %xmm5, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm4
-; AVX2-NEXT:    vpackuswb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm4
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm4, %ymm0
 ; AVX2-NEXT:    vpextrb $0, %xmm3, %eax
 ; AVX2-NEXT:    notb %al
 ; AVX2-NEXT:    testb $1, %al
diff --git a/llvm/test/CodeGen/X86/vector-reduce-and-bool.ll b/llvm/test/CodeGen/X86/vector-reduce-and-bool.ll
index 5f8eb82407a..7c432a36397 100644
--- a/llvm/test/CodeGen/X86/vector-reduce-and-bool.ll
+++ b/llvm/test/CodeGen/X86/vector-reduce-and-bool.ll
@@ -622,13 +622,12 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX2-LABEL: trunc_v32i16_v32i1:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm2, %ymm0
 ; AVX2-NEXT:    vpsllw $7, %ymm0, %ymm0
 ; AVX2-NEXT:    vpmovmskb %ymm0, %eax
 ; AVX2-NEXT:    cmpl $-1, %eax
diff --git a/llvm/test/CodeGen/X86/vector-reduce-or-bool.ll b/llvm/test/CodeGen/X86/vector-reduce-or-bool.ll
index f501a8be0c1..e887ee28ce0 100644
--- a/llvm/test/CodeGen/X86/vector-reduce-or-bool.ll
+++ b/llvm/test/CodeGen/X86/vector-reduce-or-bool.ll
@@ -616,13 +616,12 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX2-LABEL: trunc_v32i16_v32i1:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm2, %ymm0
 ; AVX2-NEXT:    vpsllw $7, %ymm0, %ymm0
 ; AVX2-NEXT:    vpmovmskb %ymm0, %eax
 ; AVX2-NEXT:    testl %eax, %eax
diff --git a/llvm/test/CodeGen/X86/vector-reduce-xor-bool.ll b/llvm/test/CodeGen/X86/vector-reduce-xor-bool.ll
index a5009cf4744..34d5e3f988f 100644
--- a/llvm/test/CodeGen/X86/vector-reduce-xor-bool.ll
+++ b/llvm/test/CodeGen/X86/vector-reduce-xor-bool.ll
@@ -673,13 +673,12 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX2-LABEL: trunc_v32i16_v32i1:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm2, %ymm0
 ; AVX2-NEXT:    vpsllw $7, %ymm0, %ymm0
 ; AVX2-NEXT:    vpmovmskb %ymm0, %eax
 ; AVX2-NEXT:    movl %eax, %ecx
diff --git a/llvm/test/CodeGen/X86/vector-trunc-widen.ll b/llvm/test/CodeGen/X86/vector-trunc-widen.ll
index 6a240ab092d..c6b36e4aae7 100644
--- a/llvm/test/CodeGen/X86/vector-trunc-widen.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-widen.ll
@@ -1305,13 +1305,12 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ; AVX2-LABEL: trunc32i16_32i8:
 ; AVX2:       # %bb.0: # %entry
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm2, %ymm0
 ; AVX2-NEXT:    vmovdqu %ymm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vector-trunc.ll b/llvm/test/CodeGen/X86/vector-trunc.ll
index 6e8d2b4659f..56e86a6bc95 100644
--- a/llvm/test/CodeGen/X86/vector-trunc.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc.ll
@@ -1315,13 +1315,12 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ; AVX2-LABEL: trunc32i16_32i8:
 ; AVX2:       # %bb.0: # %entry
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
-; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT:    vpackuswb %ymm0, %ymm2, %ymm0
 ; AVX2-NEXT:    vmovdqu %ymm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq