From: Craig Topper <craig.topper@intel.com>
Date: Sun, 18 Nov 2018 17:59:28 +0000 (+0000)
Subject: [X86] Lower v16i16->v8i16 truncate using an 'and' with 255, an extract_subvector... 
X-Git-Tag: llvmorg-8.0.0-rc1~3997
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=bc8148f7b056f8cd1c6a2dfe59988b3e1ab4215d;p=platform%2Fupstream%2Fllvm.git

[X86] Lower v16i16->v8i16 truncate using an 'and' with 255, an extract_subvector, and a packuswb instruction.

Summary: This is an improvement over the two pshufbs and punpcklqdq we'd get otherwise.

Reviewers: RKSimon, spatel

Reviewed By: RKSimon

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D54671

llvm-svn: 347171
---

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 062c978..1c8f6aa 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -18047,6 +18047,22 @@ SDValue X86TargetLowering::LowerTRUNCATE(SDValue Op, SelectionDAG &DAG) const {
     return DAG.getBitcast(MVT::v8i16, res);
   }
 
+  if (VT == MVT::v16i8 && InVT == MVT::v16i16) {
+    // Use an AND to force a PACKUS.
+    In = DAG.getNode(ISD::AND, DL, InVT, In, DAG.getConstant(255, DL, InVT));
+
+    SDValue InLo = DAG.getNode(ISD::EXTRACT_SUBVECTOR, DL, MVT::v8i16, In,
+                               DAG.getIntPtrConstant(0, DL));
+    SDValue InHi = DAG.getNode(ISD::EXTRACT_SUBVECTOR, DL, MVT::v8i16, In,
+                               DAG.getIntPtrConstant(8, DL));
+    InLo = DAG.getBitcast(VT, InLo);
+    InHi = DAG.getBitcast(VT, InHi);
+
+    return DAG.getVectorShuffle(VT, DL, InLo, InHi,
+                                { 0,  2,  4,  6,  8, 10, 12, 14,
+                                 16, 18, 20, 22, 24, 26, 28, 30});
+  }
+
   // Handle truncation of V256 to V128 using shuffles.
   assert(VT.is128BitVector() && InVT.is256BitVector() && "Unexpected types!");
 
diff --git a/llvm/test/CodeGen/X86/avg.ll b/llvm/test/CodeGen/X86/avg.ll
index 02d8bd8..b1378b6 100644
--- a/llvm/test/CodeGen/X86/avg.ll
+++ b/llvm/test/CodeGen/X86/avg.ll
@@ -364,17 +364,17 @@ define void @avg_v48i8(<48 x i8>* %a, <48 x i8>* %b) nounwind {
 ; AVX1-NEXT:    vpsrld $1, %xmm9, %xmm5
 ; AVX1-NEXT:    vpsrld $1, %xmm10, %xmm6
 ; AVX1-NEXT:    vpackusdw %xmm5, %xmm6, %xmm5
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm6, %xmm5, %xmm5
-; AVX1-NEXT:    vpshufb %xmm6, %xmm4, %xmm4
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm5[0]
-; AVX1-NEXT:    vpshufb %xmm6, %xmm3, %xmm3
-; AVX1-NEXT:    vpshufb %xmm6, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm6, %xmm5, %xmm5
+; AVX1-NEXT:    vpand %xmm6, %xmm4, %xmm4
+; AVX1-NEXT:    vpackuswb %xmm5, %xmm4, %xmm4
+; AVX1-NEXT:    vpand %xmm6, %xmm3, %xmm3
+; AVX1-NEXT:    vpand %xmm6, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm1, %ymm1
-; AVX1-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX1-NEXT:    vpand %xmm6, %xmm2, %xmm2
+; AVX1-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX1-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX1-NEXT:    vmovups %ymm1, (%rax)
 ; AVX1-NEXT:    vzeroupper
@@ -421,27 +421,27 @@ define void @avg_v48i8(<48 x i8>* %a, <48 x i8>* %b) nounwind {
 ; AVX2-NEXT:    vpsrld $1, %ymm7, %ymm5
 ; AVX2-NEXT:    vextracti128 $1, %ymm5, %xmm6
 ; AVX2-NEXT:    vpackusdw %xmm6, %xmm5, %xmm5
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %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:    vpshufb %xmm6, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm5[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm6, %xmm1, %xmm1
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm5, %xmm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm5
 ; AVX2-NEXT:    vpackusdw %xmm5, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-NEXT:    vextracti128 $1, %ymm3, %xmm5
 ; AVX2-NEXT:    vpackusdw %xmm5, %xmm3, %xmm3
-; AVX2-NEXT:    vpshufb %xmm6, %xmm3, %xmm3
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
+; AVX2-NEXT:    vpand %xmm6, %xmm3, %xmm3
+; AVX2-NEXT:    vpackuswb %xmm3, %xmm2, %xmm2
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm4, %xmm2
 ; AVX2-NEXT:    vpackusdw %xmm2, %xmm4, %xmm2
-; AVX2-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm3
 ; AVX2-NEXT:    vpackusdw %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm2[0],xmm0[0]
+; AVX2-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm0, %xmm2, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    vmovdqu %ymm1, (%rax)
 ; AVX2-NEXT:    vzeroupper
@@ -2282,15 +2282,15 @@ define void @not_avg_v16i8_wide_constants(<16 x i8>* %a, <16 x i8>* %b) nounwind
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm12[0]
 ; AVX1-NEXT:    vshufps {{.*#+}} xmm2 = xmm3[0,2],xmm2[0,2]
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm3 = xmm5[0],xmm4[0]
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm4, %xmm0, %xmm0
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm2
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm7[0],xmm6[0]
 ; AVX1-NEXT:    vshufps {{.*#+}} xmm3 = xmm5[0,2],xmm3[0,2]
 ; AVX1-NEXT:    vpshufb %xmm1, %xmm3, %xmm1
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
-; AVX1-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vpand %xmm4, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX1-NEXT:    addq $24, %rsp
 ; AVX1-NEXT:    popq %rbx
@@ -2525,8 +2525,8 @@ define void @not_avg_v16i8_wide_constants(<16 x i8>* %a, <16 x i8>* %b) nounwind
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm8, %ymm2
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm3 = xmm5[0],xmm4[0]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm7[0],xmm6[0]
@@ -2536,8 +2536,8 @@ define void @not_avg_v16i8_wide_constants(<16 x i8>* %a, <16 x i8>* %b) nounwind
 ; AVX2-NEXT:    vinserti128 $1, %xmm3, %ymm0, %ymm0
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    addq $16, %rsp
 ; AVX2-NEXT:    popq %rbx
diff --git a/llvm/test/CodeGen/X86/avx-trunc.ll b/llvm/test/CodeGen/X86/avx-trunc.ll
index f1af384..49f57fa 100644
--- a/llvm/test/CodeGen/X86/avx-trunc.ll
+++ b/llvm/test/CodeGen/X86/avx-trunc.ll
@@ -29,11 +29,9 @@ define <8 x i16> @trunc_32_16(<8 x i32> %A) nounwind uwtable readnone ssp{
 define <16 x i8> @trunc_16_8(<16 x i16> %A) nounwind uwtable readnone ssp{
 ; CHECK-LABEL: trunc_16_8:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; CHECK-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; CHECK-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; CHECK-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; CHECK-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; CHECK-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; CHECK-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; CHECK-NEXT:    vzeroupper
 ; CHECK-NEXT:    retq
   %B = trunc <16 x i16> %A to <16 x i8>
diff --git a/llvm/test/CodeGen/X86/avx2-arith.ll b/llvm/test/CodeGen/X86/avx2-arith.ll
index c11337b..e54ca61 100644
--- a/llvm/test/CodeGen/X86/avx2-arith.ll
+++ b/llvm/test/CodeGen/X86/avx2-arith.ll
@@ -148,11 +148,9 @@ define <16 x i8> @mul_v16i8(<16 x i8> %i, <16 x i8> %j) nounwind readnone {
 ; X32-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; X32-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X32-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; X32-NEXT:    vpand {{\.LCPI.*}}, %ymm0, %ymm0
 ; X32-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X32-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X32-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X32-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X32-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X32-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X32-NEXT:    vzeroupper
 ; X32-NEXT:    retl
 ;
@@ -161,11 +159,9 @@ define <16 x i8> @mul_v16i8(<16 x i8> %i, <16 x i8> %j) nounwind readnone {
 ; X64-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; X64-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X64-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; X64-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; X64-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X64-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-NEXT:    vzeroupper
 ; X64-NEXT:    retq
   %x = mul <16 x i8> %i, %j
@@ -180,18 +176,16 @@ define <32 x i8> @mul_v32i8(<32 x i8> %i, <32 x i8> %j) nounwind readnone {
 ; X32-NEXT:    vextracti128 $1, %ymm0, %xmm3
 ; X32-NEXT:    vpmovzxbw {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; X32-NEXT:    vpmullw %ymm2, %ymm3, %ymm2
-; X32-NEXT:    vextracti128 $1, %ymm2, %xmm3
-; X32-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X32-NEXT:    vpshufb %xmm4, %xmm3, %xmm3
-; X32-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; X32-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
+; X32-NEXT:    vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; X32-NEXT:    vpand %ymm3, %ymm2, %ymm2
+; X32-NEXT:    vextracti128 $1, %ymm2, %xmm4
+; X32-NEXT:    vpackuswb %xmm4, %xmm2, %xmm2
 ; X32-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; X32-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X32-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; X32-NEXT:    vpand %ymm3, %ymm0, %ymm0
 ; X32-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X32-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; X32-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; X32-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X32-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X32-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
 ; X32-NEXT:    retl
 ;
@@ -202,18 +196,16 @@ define <32 x i8> @mul_v32i8(<32 x i8> %i, <32 x i8> %j) nounwind readnone {
 ; X64-NEXT:    vextracti128 $1, %ymm0, %xmm3
 ; X64-NEXT:    vpmovzxbw {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; X64-NEXT:    vpmullw %ymm2, %ymm3, %ymm2
-; X64-NEXT:    vextracti128 $1, %ymm2, %xmm3
-; X64-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-NEXT:    vpshufb %xmm4, %xmm3, %xmm3
-; X64-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; X64-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
+; X64-NEXT:    vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; X64-NEXT:    vpand %ymm3, %ymm2, %ymm2
+; X64-NEXT:    vextracti128 $1, %ymm2, %xmm4
+; X64-NEXT:    vpackuswb %xmm4, %xmm2, %xmm2
 ; X64-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; X64-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X64-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; X64-NEXT:    vpand %ymm3, %ymm0, %ymm0
 ; X64-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; X64-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; X64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
 ; X64-NEXT:    retq
   %x = mul <32 x i8> %i, %j
diff --git a/llvm/test/CodeGen/X86/avx2-conversions.ll b/llvm/test/CodeGen/X86/avx2-conversions.ll
index d4c2330..1060d70 100644
--- a/llvm/test/CodeGen/X86/avx2-conversions.ll
+++ b/llvm/test/CodeGen/X86/avx2-conversions.ll
@@ -163,21 +163,17 @@ define <16 x i16> @sext_16i8_16i16(<16 x i8> %z) {
 define <16 x i8> @trunc_16i16_16i8(<16 x i16> %z) {
 ; X32-LABEL: trunc_16i16_16i8:
 ; X32:       # %bb.0:
+; X32-NEXT:    vpand {{\.LCPI.*}}, %ymm0, %ymm0
 ; X32-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X32-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X32-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X32-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X32-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X32-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X32-NEXT:    vzeroupper
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: trunc_16i16_16i8:
 ; X64:       # %bb.0:
+; X64-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; X64-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X64-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X64-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-NEXT:    vzeroupper
 ; X64-NEXT:    retq
   %t = trunc <16 x i16> %z to <16 x i8>
diff --git a/llvm/test/CodeGen/X86/combine-mul.ll b/llvm/test/CodeGen/X86/combine-mul.ll
index e47c598..d88b659 100644
--- a/llvm/test/CodeGen/X86/combine-mul.ll
+++ b/llvm/test/CodeGen/X86/combine-mul.ll
@@ -306,11 +306,9 @@ define <16 x i8> @PR35579(<16 x i8> %x) {
 ; AVX:       # %bb.0:
 ; AVX-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; AVX-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    vzeroupper
 ; AVX-NEXT:    retq
   %r = mul <16 x i8> %x, <i8 0, i8 1, i8 2, i8 1, i8 4, i8 1, i8 2, i8 1, i8 8, i8 1, i8 2, i8 1, i8 4, i8 1, i8 2, i8 1>
diff --git a/llvm/test/CodeGen/X86/pmul.ll b/llvm/test/CodeGen/X86/pmul.ll
index f5f7019..1526325 100644
--- a/llvm/test/CodeGen/X86/pmul.ll
+++ b/llvm/test/CodeGen/X86/pmul.ll
@@ -38,11 +38,9 @@ define <16 x i8> @mul_v16i8c(<16 x i8> %i) nounwind  {
 ; AVX2:       # %bb.0: # %entry
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -173,11 +171,9 @@ define <16 x i8> @mul_v16i8(<16 x i8> %i, <16 x i8> %j) nounwind  {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -424,17 +420,15 @@ define <32 x i8> @mul_v32i8c(<32 x i8> %i) nounwind  {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [117,117,117,117,117,117,117,117,117,117,117,117,117,117,117,117]
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm1, %ymm1
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm4, %xmm3, %xmm3
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm3, %ymm1, %ymm1
+; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm4
+; AVX2-NEXT:    vpackuswb %xmm4, %xmm1, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm3, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
 ;
@@ -608,18 +602,16 @@ define <32 x i8> @mul_v32i8(<32 x i8> %i, <32 x i8> %j) nounwind  {
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm3
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm3, %ymm2
-; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm3
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm4, %xmm3, %xmm3
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm3, %ymm2, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm4
+; AVX2-NEXT:    vpackuswb %xmm4, %xmm2, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm3, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
 ;
@@ -826,31 +818,27 @@ define <64 x i8> @mul_v64i8c(<64 x i8> %i) nounwind  {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm3 = [117,117,117,117,117,117,117,117,117,117,117,117,117,117,117,117]
 ; AVX2-NEXT:    vpmullw %ymm3, %ymm2, %ymm2
-; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm4
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm5, %xmm4, %xmm4
-; AVX2-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm4[0]
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm4, %ymm2, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm5
+; AVX2-NEXT:    vpackuswb %xmm5, %xmm2, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm3, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm4
-; AVX2-NEXT:    vpshufb %xmm5, %xmm4, %xmm4
-; AVX2-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm4[0]
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm5
+; AVX2-NEXT:    vpackuswb %xmm5, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
 ; AVX2-NEXT:    vpmullw %ymm3, %ymm2, %ymm2
-; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm4
-; AVX2-NEXT:    vpshufb %xmm5, %xmm4, %xmm4
-; AVX2-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm4[0]
+; AVX2-NEXT:    vpand %ymm4, %ymm2, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm5
+; AVX2-NEXT:    vpackuswb %xmm5, %xmm2, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm3, %ymm1, %ymm1
+; AVX2-NEXT:    vpand %ymm4, %ymm1, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpshufb %xmm5, %xmm3, %xmm3
-; AVX2-NEXT:    vpshufb %xmm5, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
+; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
 ; AVX2-NEXT:    retq
 ;
@@ -997,36 +985,32 @@ define <64 x i8> @mul_v64i8(<64 x i8> %i, <64 x i8> %j) nounwind  {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm4 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero,xmm4[4],zero,xmm4[5],zero,xmm4[6],zero,xmm4[7],zero,xmm4[8],zero,xmm4[9],zero,xmm4[10],zero,xmm4[11],zero,xmm4[12],zero,xmm4[13],zero,xmm4[14],zero,xmm4[15],zero
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm5
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm5 = xmm5[0],zero,xmm5[1],zero,xmm5[2],zero,xmm5[3],zero,xmm5[4],zero,xmm5[5],zero,xmm5[6],zero,xmm5[7],zero,xmm5[8],zero,xmm5[9],zero,xmm5[10],zero,xmm5[11],zero,xmm5[12],zero,xmm5[13],zero,xmm5[14],zero,xmm5[15],zero
-; AVX2-NEXT:    vpmullw %ymm4, %ymm5, %ymm5
-; AVX2-NEXT:    vextracti128 $1, %ymm5, %xmm6
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm4, %xmm6, %xmm6
-; AVX2-NEXT:    vpshufb %xmm4, %xmm5, %xmm5
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm6[0]
+; AVX2-NEXT:    vpmullw %ymm4, %ymm5, %ymm4
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm5 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm5, %ymm4, %ymm4
+; AVX2-NEXT:    vextracti128 $1, %ymm4, %xmm6
+; AVX2-NEXT:    vpackuswb %xmm6, %xmm4, %xmm4
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm5, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
-; AVX2-NEXT:    vinserti128 $1, %xmm5, %ymm0, %ymm0
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vinserti128 $1, %xmm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm3, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm5
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm5 = xmm5[0],zero,xmm5[1],zero,xmm5[2],zero,xmm5[3],zero,xmm5[4],zero,xmm5[5],zero,xmm5[6],zero,xmm5[7],zero,xmm5[8],zero,xmm5[9],zero,xmm5[10],zero,xmm5[11],zero,xmm5[12],zero,xmm5[13],zero,xmm5[14],zero,xmm5[15],zero
-; AVX2-NEXT:    vpmullw %ymm2, %ymm5, %ymm2
-; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm5
-; AVX2-NEXT:    vpshufb %xmm4, %xmm5, %xmm5
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm5[0]
+; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm4
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm4 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero,xmm4[4],zero,xmm4[5],zero,xmm4[6],zero,xmm4[7],zero,xmm4[8],zero,xmm4[9],zero,xmm4[10],zero,xmm4[11],zero,xmm4[12],zero,xmm4[13],zero,xmm4[14],zero,xmm4[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm4, %ymm2
+; AVX2-NEXT:    vpand %ymm5, %ymm2, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm2, %xmm4
+; AVX2-NEXT:    vpackuswb %xmm4, %xmm2, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm3, %ymm1, %ymm1
+; AVX2-NEXT:    vpand %ymm5, %ymm1, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpshufb %xmm4, %xmm3, %xmm3
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
+; AVX2-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm1, %ymm1
 ; AVX2-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/prefer-avx256-mask-extend.ll b/llvm/test/CodeGen/X86/prefer-avx256-mask-extend.ll
index 829dd42..b4f8e5b 100644
--- a/llvm/test/CodeGen/X86/prefer-avx256-mask-extend.ll
+++ b/llvm/test/CodeGen/X86/prefer-avx256-mask-extend.ll
@@ -48,11 +48,11 @@ define <16 x i8> @testv16i1_sext_v16i8(<8 x i32>* %p, <8 x i32>* %q) {
 ; AVX256-NEXT:    vpcmpeqd %ymm0, %ymm0, %ymm0
 ; AVX256-NEXT:    vmovdqa32 %ymm0, %ymm1 {%k2} {z}
 ; AVX256-NEXT:    vpmovdw %ymm1, %xmm1
-; AVX256-NEXT:    vpacksswb %xmm0, %xmm1, %xmm1
+; AVX256-NEXT:    vpsrlw $8, %xmm1, %xmm1
 ; AVX256-NEXT:    vmovdqa32 %ymm0, %ymm0 {%k1} {z}
 ; AVX256-NEXT:    vpmovdw %ymm0, %xmm0
-; AVX256-NEXT:    vpacksswb %xmm0, %xmm0, %xmm0
-; AVX256-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX256-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX256-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX256-NEXT:    vzeroupper
 ; AVX256-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll b/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
index 120e448..92f6e27 100644
--- a/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
+++ b/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
@@ -34,11 +34,11 @@ define <16 x i1> @shuf16i1_3_6_22_12_3_7_7_0_3_6_1_13_3_21_7_0(<8 x i32>* %a, <8
 ; AVX256VL-NEXT:    kshiftrw $8, %k0, %k2
 ; AVX256VL-NEXT:    vmovdqa32 %ymm0, %ymm1 {%k2} {z}
 ; AVX256VL-NEXT:    vpmovdw %ymm1, %xmm1
-; AVX256VL-NEXT:    vpacksswb %xmm0, %xmm1, %xmm1
+; AVX256VL-NEXT:    vpsrlw $8, %xmm1, %xmm1
 ; AVX256VL-NEXT:    vmovdqa32 %ymm0, %ymm0 {%k1} {z}
 ; AVX256VL-NEXT:    vpmovdw %ymm0, %xmm0
-; AVX256VL-NEXT:    vpacksswb %xmm0, %xmm0, %xmm0
-; AVX256VL-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX256VL-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX256VL-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX256VL-NEXT:    vzeroupper
 ; AVX256VL-NEXT:    retq
 ;
@@ -169,11 +169,11 @@ define <32 x i1> @shuf32i1_3_6_22_12_3_7_7_0_3_6_1_13_3_21_7_0_3_6_22_12_3_7_7_0
 ; AVX256VL-NEXT:    kshiftrw $8, %k0, %k2
 ; AVX256VL-NEXT:    vmovdqa32 %ymm0, %ymm1 {%k2} {z}
 ; AVX256VL-NEXT:    vpmovdw %ymm1, %xmm1
-; AVX256VL-NEXT:    vpacksswb %xmm0, %xmm1, %xmm1
+; AVX256VL-NEXT:    vpsrlw $8, %xmm1, %xmm1
 ; AVX256VL-NEXT:    vmovdqa32 %ymm0, %ymm0 {%k1} {z}
 ; AVX256VL-NEXT:    vpmovdw %ymm0, %xmm0
-; AVX256VL-NEXT:    vpacksswb %xmm0, %xmm0, %xmm0
-; AVX256VL-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX256VL-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX256VL-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX256VL-NEXT:    vinserti128 $1, %xmm0, %ymm0, %ymm0
 ; AVX256VL-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/prefer-avx256-trunc.ll b/llvm/test/CodeGen/X86/prefer-avx256-trunc.ll
index 6e21584..3b1ff44 100644
--- a/llvm/test/CodeGen/X86/prefer-avx256-trunc.ll
+++ b/llvm/test/CodeGen/X86/prefer-avx256-trunc.ll
@@ -11,11 +11,9 @@
 define <16 x i8> @testv16i16_trunc_v16i8(<16 x i16> %x) {
 ; AVX256NOBW-LABEL: testv16i16_trunc_v16i8:
 ; AVX256NOBW:       # %bb.0:
+; AVX256NOBW-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX256NOBW-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX256NOBW-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX256NOBW-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX256NOBW-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX256NOBW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX256NOBW-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX256NOBW-NEXT:    vzeroupper
 ; AVX256NOBW-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/psubus.ll b/llvm/test/CodeGen/X86/psubus.ll
index 6e2e979..6e3c137 100644
--- a/llvm/test/CodeGen/X86/psubus.ll
+++ b/llvm/test/CodeGen/X86/psubus.ll
@@ -824,12 +824,12 @@ define <16 x i8> @test14(<16 x i8> %x, <16 x i32> %y) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm2, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpandn %xmm0, %xmm4, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256-widen.ll b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256-widen.ll
index 940f475..629d37b 100644
--- a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256-widen.ll
+++ b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256-widen.ll
@@ -41,16 +41,25 @@ define void @shuffle_v32i8_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
 }
 
 define void @trunc_v16i16_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
-; AVX-LABEL: trunc_v16i16_to_v16i8:
-; AVX:       # %bb.0:
-; AVX-NEXT:    vmovdqa (%rdi), %xmm0
-; AVX-NEXT:    vmovdqa 16(%rdi), %xmm1
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT:    vmovdqa %xmm0, (%rsi)
-; AVX-NEXT:    retq
+; AVX1-LABEL: trunc_v16i16_to_v16i8:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vmovaps (%rdi), %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: trunc_v16i16_to_v16i8:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vmovdqa (%rdi), %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: trunc_v16i16_to_v16i8:
 ; AVX512F:       # %bb.0:
diff --git a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
index 7ef9db5..fc6a2cd 100644
--- a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
+++ b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
@@ -41,16 +41,25 @@ define void @shuffle_v32i8_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
 }
 
 define void @trunc_v16i16_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
-; AVX-LABEL: trunc_v16i16_to_v16i8:
-; AVX:       # %bb.0:
-; AVX-NEXT:    vmovdqa (%rdi), %xmm0
-; AVX-NEXT:    vmovdqa 16(%rdi), %xmm1
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT:    vmovdqa %xmm0, (%rsi)
-; AVX-NEXT:    retq
+; AVX1-LABEL: trunc_v16i16_to_v16i8:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vmovaps (%rdi), %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: trunc_v16i16_to_v16i8:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vmovdqa (%rdi), %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: trunc_v16i16_to_v16i8:
 ; AVX512F:       # %bb.0:
diff --git a/llvm/test/CodeGen/X86/vector-mul.ll b/llvm/test/CodeGen/X86/vector-mul.ll
index 4e4c0a0..59f1abb 100644
--- a/llvm/test/CodeGen/X86/vector-mul.ll
+++ b/llvm/test/CodeGen/X86/vector-mul.ll
@@ -214,11 +214,9 @@ define <16 x i8> @mul_v16i8_1_2_4_8_1_2_4_8_1_2_4_8_1_2_4_8(<16 x i8> %a0) nounw
 ; X64-AVX2:       # %bb.0:
 ; X64-AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X64-AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; X64-AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; X64-AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X64-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-AVX2-NEXT:    vzeroupper
 ; X64-AVX2-NEXT:    retq
   %1 = mul <16 x i8> %a0, <i8 1, i8 2, i8 4, i8 8, i8 1, i8 2, i8 4, i8 8, i8 1, i8 2, i8 4, i8 8, i8 1, i8 2, i8 4, i8 8>
@@ -577,11 +575,9 @@ define <16 x i8> @mul_v16i8_2_3_9_17_33_65_129_2_3_9_17_33_65_129_2_3(<16 x i8>
 ; X64-AVX2:       # %bb.0:
 ; X64-AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X64-AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; X64-AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; X64-AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X64-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-AVX2-NEXT:    vzeroupper
 ; X64-AVX2-NEXT:    retq
   %1 = mul <16 x i8> %a0, <i8 2, i8 3, i8 9, i8 17, i8 33, i8 65, i8 129, i8 2, i8 3, i8 9, i8 17, i8 33, i8 65, i8 129, i8 2, i8 3>
@@ -1158,11 +1154,9 @@ define <16 x i8> @mul_v16i8_0_1_3_7_15_31_63_127_0_1_3_7_15_31_63_127(<16 x i8>
 ; X64-AVX2:       # %bb.0:
 ; X64-AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; X64-AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; X64-AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; X64-AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; X64-AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; X64-AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; X64-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; X64-AVX2-NEXT:    vzeroupper
 ; X64-AVX2-NEXT:    retq
   %1 = mul <16 x i8> %a0, <i8 0, i8 1, i8 3, i8 7, i8 15, i8 31, i8 63, i8 127, i8 0, i8 1, i8 3, i8 7, i8 15, i8 31, i8 63, i8 127>
diff --git a/llvm/test/CodeGen/X86/vector-reduce-mul.ll b/llvm/test/CodeGen/X86/vector-reduce-mul.ll
index 50d4a74..d7718ba 100644
--- a/llvm/test/CodeGen/X86/vector-reduce-mul.ll
+++ b/llvm/test/CodeGen/X86/vector-reduce-mul.ll
@@ -1557,31 +1557,29 @@ define i8 @test_v16i8(<16 x i8> %a0) {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; 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:    vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrlw $8, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX2-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX2-NEXT:    vzeroupper
@@ -1843,39 +1841,37 @@ define i8 @test_v32i8(<32 x i8> %a0) {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; 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:    vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
-; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm2
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
-; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
+; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrlw $8, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX2-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX2-NEXT:    vzeroupper
@@ -2213,55 +2209,51 @@ define i8 @test_v64i8(<64 x i8> %a0) {
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm3 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm3, %ymm3
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm2, %ymm3, %ymm3
 ; AVX2-NEXT:    vextracti128 $1, %ymm3, %xmm4
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm4, %xmm4
-; AVX2-NEXT:    vpshufb %xmm2, %xmm3, %xmm3
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm4[0]
+; AVX2-NEXT:    vpackuswb %xmm4, %xmm3, %xmm3
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; AVX2-NEXT:    vpmullw %ymm0, %ymm1, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrlw $8, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX2-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX2-NEXT:    vzeroupper
@@ -2737,83 +2729,75 @@ define i8 @test_v128i8(<128 x i8> %a0) {
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm5
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm5 = xmm5[0],zero,xmm5[1],zero,xmm5[2],zero,xmm5[3],zero,xmm5[4],zero,xmm5[5],zero,xmm5[6],zero,xmm5[7],zero,xmm5[8],zero,xmm5[9],zero,xmm5[10],zero,xmm5[11],zero,xmm5[12],zero,xmm5[13],zero,xmm5[14],zero,xmm5[15],zero
 ; AVX2-NEXT:    vpmullw %ymm4, %ymm5, %ymm5
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %ymm4, %ymm5, %ymm5
 ; AVX2-NEXT:    vextracti128 $1, %ymm5, %xmm6
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm4, %xmm6, %xmm6
-; AVX2-NEXT:    vpshufb %xmm4, %xmm5, %xmm5
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm6[0]
+; AVX2-NEXT:    vpackuswb %xmm6, %xmm5, %xmm5
 ; AVX2-NEXT:    vextracti128 $1, %ymm3, %xmm6
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm6 = xmm6[0],zero,xmm6[1],zero,xmm6[2],zero,xmm6[3],zero,xmm6[4],zero,xmm6[5],zero,xmm6[6],zero,xmm6[7],zero,xmm6[8],zero,xmm6[9],zero,xmm6[10],zero,xmm6[11],zero,xmm6[12],zero,xmm6[13],zero,xmm6[14],zero,xmm6[15],zero
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm7
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm7 = xmm7[0],zero,xmm7[1],zero,xmm7[2],zero,xmm7[3],zero,xmm7[4],zero,xmm7[5],zero,xmm7[6],zero,xmm7[7],zero,xmm7[8],zero,xmm7[9],zero,xmm7[10],zero,xmm7[11],zero,xmm7[12],zero,xmm7[13],zero,xmm7[14],zero,xmm7[15],zero
 ; AVX2-NEXT:    vpmullw %ymm6, %ymm7, %ymm6
+; AVX2-NEXT:    vpand %ymm4, %ymm6, %ymm6
 ; AVX2-NEXT:    vextracti128 $1, %ymm6, %xmm7
-; AVX2-NEXT:    vpshufb %xmm4, %xmm7, %xmm7
-; AVX2-NEXT:    vpshufb %xmm4, %xmm6, %xmm6
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm6 = xmm6[0],xmm7[0]
+; AVX2-NEXT:    vpackuswb %xmm7, %xmm6, %xmm6
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero,xmm2[8],zero,xmm2[9],zero,xmm2[10],zero,xmm2[11],zero,xmm2[12],zero,xmm2[13],zero,xmm2[14],zero,xmm2[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero,xmm3[8],zero,xmm3[9],zero,xmm3[10],zero,xmm3[11],zero,xmm3[12],zero,xmm3[13],zero,xmm3[14],zero,xmm3[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm2, %ymm1, %ymm1
+; AVX2-NEXT:    vpand %ymm4, %ymm1, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm1, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm6[0],zero,xmm6[1],zero,xmm6[2],zero,xmm6[3],zero,xmm6[4],zero,xmm6[5],zero,xmm6[6],zero,xmm6[7],zero,xmm6[8],zero,xmm6[9],zero,xmm6[10],zero,xmm6[11],zero,xmm6[12],zero,xmm6[13],zero,xmm6[14],zero,xmm6[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm5[0],zero,xmm5[1],zero,xmm5[2],zero,xmm5[3],zero,xmm5[4],zero,xmm5[5],zero,xmm5[6],zero,xmm5[7],zero,xmm5[8],zero,xmm5[9],zero,xmm5[10],zero,xmm5[11],zero,xmm5[12],zero,xmm5[13],zero,xmm5[14],zero,xmm5[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm2, %ymm1
+; AVX2-NEXT:    vpand %ymm4, %ymm1, %ymm1
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm1, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrld $16, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand %ymm4, %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpsrlw $8, %xmm0, %xmm1
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero,xmm1[8],zero,xmm1[9],zero,xmm1[10],zero,xmm1[11],zero,xmm1[12],zero,xmm1[13],zero,xmm1[14],zero,xmm1[15],zero
-; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vpextrb $0, %xmm0, %eax
 ; AVX2-NEXT:    # kill: def $al killed $al killed $eax
 ; AVX2-NEXT:    vzeroupper
diff --git a/llvm/test/CodeGen/X86/vector-shift-shl-128.ll b/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
index 1c52b4b..4e8e785 100644
--- a/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
@@ -909,11 +909,9 @@ define <16 x i8> @constant_shift_v16i8(<16 x i8> %a) nounwind {
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/vector-trunc-math-widen.ll b/llvm/test/CodeGen/X86/vector-trunc-math-widen.ll
index 8f6a1fe..fb5407b 100644
--- a/llvm/test/CodeGen/X86/vector-trunc-math-widen.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-math-widen.ll
@@ -267,8 +267,8 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -276,8 +276,8 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -294,15 +294,15 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -366,12 +366,12 @@ define <16 x i8> @trunc_add_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -403,21 +403,19 @@ define <16 x i8> @trunc_add_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_add_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpaddw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -715,8 +713,8 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -724,8 +722,8 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -739,15 +737,15 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -801,12 +799,12 @@ define <16 x i8> @trunc_add_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -834,22 +832,18 @@ define <16 x i8> @trunc_add_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1142,8 +1136,8 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -1151,8 +1145,8 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -1169,15 +1163,15 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -1241,12 +1235,12 @@ define <16 x i8> @trunc_sub_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1278,21 +1272,19 @@ define <16 x i8> @trunc_sub_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpsubw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_sub_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpsubw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1612,8 +1604,8 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -1621,8 +1613,8 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -1639,15 +1631,15 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -1709,12 +1701,12 @@ define <16 x i8> @trunc_sub_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1745,21 +1737,19 @@ define <16 x i8> @trunc_sub_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ; AVX1-NEXT:    vpsubw {{.*}}(%rip), %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpsubw {{.*}}(%rip), %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_sub_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpsubw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2315,8 +2305,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm5 = ymm5[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm5 = ymm5[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -2330,8 +2320,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -2348,8 +2338,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm5, %ymm8, %ymm5
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm8, %ymm1
 ; AVX2-FAST-NEXT:    vpmulld %xmm5, %xmm1, %xmm1
@@ -2359,8 +2349,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -2474,12 +2464,12 @@ define <16 x i8> @trunc_mul_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2511,21 +2501,19 @@ define <16 x i8> @trunc_mul_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_mul_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2946,8 +2934,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpmulld {{.*}}(%rip), %xmm0, %xmm0
@@ -2957,8 +2945,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -2973,8 +2961,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpmulld {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
@@ -2982,8 +2970,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -3072,13 +3060,13 @@ define <16 x i8> @trunc_mul_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %xmm1, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm0
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3109,21 +3097,19 @@ define <16 x i8> @trunc_mul_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ; AVX1-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_mul_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3399,8 +3385,8 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -3408,8 +3394,8 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -3426,15 +3412,15 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -3494,12 +3480,12 @@ define <16 x i8> @trunc_and_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3528,22 +3514,18 @@ define <16 x i8> @trunc_and_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-LABEL: trunc_and_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vandps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_and_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3792,8 +3774,8 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -3801,8 +3783,8 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -3816,15 +3798,15 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -3878,12 +3860,12 @@ define <16 x i8> @trunc_and_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -3911,22 +3893,18 @@ define <16 x i8> @trunc_and_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -4203,8 +4181,8 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -4212,8 +4190,8 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -4230,15 +4208,15 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -4298,12 +4276,12 @@ define <16 x i8> @trunc_xor_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -4332,22 +4310,18 @@ define <16 x i8> @trunc_xor_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-LABEL: trunc_xor_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vxorps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_xor_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpxor %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -4596,8 +4570,8 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -4605,8 +4579,8 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -4620,15 +4594,15 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -4682,12 +4656,12 @@ define <16 x i8> @trunc_xor_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -4715,22 +4689,18 @@ define <16 x i8> @trunc_xor_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -5007,8 +4977,8 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -5016,8 +4986,8 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -5034,15 +5004,15 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -5102,12 +5072,12 @@ define <16 x i8> @trunc_or_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwind
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -5136,22 +5106,18 @@ define <16 x i8> @trunc_or_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwind
 ; AVX1-LABEL: trunc_or_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vorps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_or_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpor %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -5400,8 +5366,8 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -5409,8 +5375,8 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -5424,15 +5390,15 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -5486,12 +5452,12 @@ define <16 x i8> @trunc_or_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -5519,22 +5485,18 @@ define <16 x i8> @trunc_or_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vector-trunc-math.ll b/llvm/test/CodeGen/X86/vector-trunc-math.ll
index 97c4f3d..c8ccab0 100644
--- a/llvm/test/CodeGen/X86/vector-trunc-math.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-math.ll
@@ -267,8 +267,8 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -276,8 +276,8 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -294,15 +294,15 @@ define <16 x i8> @trunc_add_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -366,12 +366,12 @@ define <16 x i8> @trunc_add_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -403,21 +403,19 @@ define <16 x i8> @trunc_add_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_add_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpaddw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -715,8 +713,8 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -724,8 +722,8 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -739,15 +737,15 @@ define <16 x i8> @trunc_add_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -801,12 +799,12 @@ define <16 x i8> @trunc_add_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -834,22 +832,18 @@ define <16 x i8> @trunc_add_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1142,8 +1136,8 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -1151,8 +1145,8 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -1169,15 +1163,15 @@ define <16 x i8> @trunc_sub_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -1241,12 +1235,12 @@ define <16 x i8> @trunc_sub_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1278,21 +1272,19 @@ define <16 x i8> @trunc_sub_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpsubw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_sub_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpsubw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1612,8 +1604,8 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -1621,8 +1613,8 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -1639,15 +1631,15 @@ define <16 x i8> @trunc_sub_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -1709,12 +1701,12 @@ define <16 x i8> @trunc_sub_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -1745,21 +1737,19 @@ define <16 x i8> @trunc_sub_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ; AVX1-NEXT:    vpsubw {{.*}}(%rip), %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpsubw {{.*}}(%rip), %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_sub_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpsubw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2315,8 +2305,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm5 = ymm5[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm5 = ymm5[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -2330,8 +2320,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -2348,8 +2338,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm6 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm6, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm6, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm5, %ymm8, %ymm5
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm8, %ymm1
 ; AVX2-FAST-NEXT:    vpmulld %xmm5, %xmm1, %xmm1
@@ -2359,8 +2349,8 @@ define <16 x i8> @trunc_mul_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm6, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -2474,12 +2464,12 @@ define <16 x i8> @trunc_mul_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2511,21 +2501,19 @@ define <16 x i8> @trunc_mul_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpmullw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm1, %xmm2, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm1 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm1, %xmm2, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_mul_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpmullw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -2946,8 +2934,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpmulld {{.*}}(%rip), %xmm0, %xmm0
@@ -2957,8 +2945,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -2973,8 +2961,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpmulld {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
@@ -2982,8 +2970,8 @@ define <16 x i8> @trunc_mul_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -3072,13 +3060,13 @@ define <16 x i8> @trunc_mul_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %xmm1, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm0
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3109,21 +3097,19 @@ define <16 x i8> @trunc_mul_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ; AVX1-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm1
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
 ; AVX1-NEXT:    vpmullw {{.*}}(%rip), %xmm0, %xmm0
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vpand %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpand %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpackuswb %xmm0, %xmm1, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_mul_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpmullw {{.*}}(%rip), %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3399,8 +3385,8 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -3408,8 +3394,8 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -3426,15 +3412,15 @@ define <16 x i8> @trunc_and_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -3494,12 +3480,12 @@ define <16 x i8> @trunc_and_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3528,22 +3514,18 @@ define <16 x i8> @trunc_and_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-LABEL: trunc_and_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vandps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_and_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpand %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -3792,8 +3774,8 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -3801,8 +3783,8 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -3816,15 +3798,15 @@ define <16 x i8> @trunc_and_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -3878,12 +3860,12 @@ define <16 x i8> @trunc_and_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -3911,22 +3893,18 @@ define <16 x i8> @trunc_and_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -4203,8 +4181,8 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -4212,8 +4190,8 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -4230,15 +4208,15 @@ define <16 x i8> @trunc_xor_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwin
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -4298,12 +4276,12 @@ define <16 x i8> @trunc_xor_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwin
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -4332,22 +4310,18 @@ define <16 x i8> @trunc_xor_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwin
 ; AVX1-LABEL: trunc_xor_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vxorps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_xor_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpxor %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -4596,8 +4570,8 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -4605,8 +4579,8 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -4620,15 +4594,15 @@ define <16 x i8> @trunc_xor_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -4682,12 +4656,12 @@ define <16 x i8> @trunc_xor_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -4715,22 +4689,18 @@ define <16 x i8> @trunc_xor_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -5007,8 +4977,8 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -5016,8 +4986,8 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
 ;
@@ -5034,15 +5004,15 @@ define <16 x i8> @trunc_or_v16i64_v16i8(<16 x i64> %a0, <16 x i64> %a1) nounwind
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
 ;
@@ -5102,12 +5072,12 @@ define <16 x i8> @trunc_or_v16i32_v16i8(<16 x i32> %a0, <16 x i32> %a1) nounwind
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -5136,22 +5106,18 @@ define <16 x i8> @trunc_or_v16i16_v16i8(<16 x i16> %a0, <16 x i16> %a1) nounwind
 ; AVX1-LABEL: trunc_or_v16i16_v16i8:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vorps %ymm1, %ymm0, %ymm0
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_or_v16i16_v16i8:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vpor %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
 ;
@@ -5400,8 +5366,8 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm2, %xmm2
+; AVX2-SLOW-NEXT:    vmovdqa {{.*#+}} xmm4 = [255,255,255,255,255,255,255,255]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm2, %xmm2
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm0 = ymm0[0,2,2,3,4,6,6,7]
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
 ; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} ymm1 = ymm1[0,2,2,3,4,6,6,7]
@@ -5409,8 +5375,8 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-SLOW-NEXT:    vpshufb %xmm4, %xmm0, %xmm0
-; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-SLOW-NEXT:    vpand %xmm4, %xmm0, %xmm0
+; AVX2-SLOW-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-SLOW-NEXT:    vzeroupper
 ; AVX2-SLOW-NEXT:    retq
@@ -5424,15 +5390,15 @@ define <16 x i8> @trunc_or_const_v16i64_v16i8(<16 x i64> %a0) nounwind {
 ; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} ymm3 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm2, %xmm2
+; AVX2-FAST-NEXT:    vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm2, %xmm2
 ; AVX2-FAST-NEXT:    vpermd %ymm0, %ymm4, %ymm0
 ; AVX2-FAST-NEXT:    vpermd %ymm1, %ymm4, %ymm1
 ; AVX2-FAST-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpshufb %ymm3, %ymm0, %ymm0
 ; AVX2-FAST-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-FAST-NEXT:    vpshufb %xmm5, %xmm0, %xmm0
-; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-FAST-NEXT:    vpand %xmm5, %xmm0, %xmm0
+; AVX2-FAST-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-FAST-NEXT:    vzeroupper
 ; AVX2-FAST-NEXT:    retq
@@ -5486,12 +5452,12 @@ define <16 x i8> @trunc_or_const_v16i32_v16i8(<16 x i32> %a0) nounwind {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -5519,22 +5485,18 @@ define <16 x i8> @trunc_or_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX1-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX1:       # %bb.0:
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vector-trunc-widen.ll b/llvm/test/CodeGen/X86/vector-trunc-widen.ll
index a15b2aa..4973937 100644
--- a/llvm/test/CodeGen/X86/vector-trunc-widen.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-widen.ll
@@ -916,12 +916,12 @@ define void @trunc16i32_16i8(<16 x i32> %a) {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1098,22 +1098,18 @@ define void @trunc16i16_16i8(<16 x i16> %a) {
 ;
 ; AVX1-LABEL: trunc16i16_16i8:
 ; AVX1:       # %bb.0: # %entry
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc16i16_16i8:
 ; AVX2:       # %bb.0: # %entry
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1323,15 +1319,13 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ;
 ; AVX1-LABEL: trunc32i16_32i8:
 ; AVX1:       # %bb.0: # %entry
-; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
+; AVX1-NEXT:    vmovaps {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vandps %ymm2, %ymm1, %ymm1
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm3
+; AVX1-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
+; AVX1-NEXT:    vandps %ymm2, %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX1-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    vmovups %ymm0, (%rax)
 ; AVX1-NEXT:    vzeroupper
@@ -1339,15 +1333,13 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ;
 ; AVX2-LABEL: trunc32i16_32i8:
 ; AVX2:       # %bb.0: # %entry
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[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:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vmovdqu %ymm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
diff --git a/llvm/test/CodeGen/X86/vector-trunc.ll b/llvm/test/CodeGen/X86/vector-trunc.ll
index fb52636..d80e9b0 100644
--- a/llvm/test/CodeGen/X86/vector-trunc.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc.ll
@@ -926,12 +926,12 @@ define void @trunc16i32_16i8(<16 x i32> %a) {
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm1, %ymm1
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
+; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,255,255,255]
+; AVX2-NEXT:    vpand %xmm3, %xmm1, %xmm1
 ; AVX2-NEXT:    vpshufb %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpand %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1108,22 +1108,18 @@ define void @trunc16i16_16i8(<16 x i16> %a) {
 ;
 ; AVX1-LABEL: trunc16i16_16i8:
 ; AVX1:       # %bb.0: # %entry
+; AVX1-NEXT:    vandps {{.*}}(%rip), %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX1-NEXT:    vzeroupper
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: trunc16i16_16i8:
 ; AVX2:       # %bb.0: # %entry
+; AVX2-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX2-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    vmovdqu %xmm0, (%rax)
 ; AVX2-NEXT:    vzeroupper
 ; AVX2-NEXT:    retq
@@ -1333,15 +1329,13 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ;
 ; AVX1-LABEL: trunc32i16_32i8:
 ; AVX1:       # %bb.0: # %entry
-; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX1-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0]
+; AVX1-NEXT:    vmovaps {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
+; AVX1-NEXT:    vandps %ymm2, %ymm1, %ymm1
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm3
+; AVX1-NEXT:    vpackuswb %xmm3, %xmm1, %xmm1
+; AVX1-NEXT:    vandps %ymm2, %ymm0, %ymm0
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX1-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    vmovups %ymm0, (%rax)
 ; AVX1-NEXT:    vzeroupper
@@ -1349,15 +1343,13 @@ define void @trunc32i16_32i8(<32 x i16> %a) {
 ;
 ; AVX2-LABEL: trunc32i16_32i8:
 ; AVX2:       # %bb.0: # %entry
-; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm2
-; AVX2-NEXT:    vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX2-NEXT:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[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:    vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpshufb %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
+; AVX2-NEXT:    vpackuswb %xmm2, %xmm0, %xmm0
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vmovdqu %ymm0, (%rax)
 ; AVX2-NEXT:    vzeroupper