From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Tue, 25 Jun 2019 10:51:15 +0000 (+0000)
Subject: [TargetLowering] SimplifyDemandedBits - legal checks for SIGN/ZERO_EXTEND -> ZERO... 
X-Git-Tag: android-x86-9.0-r1~1397
X-Git-Url: http://git.osdn.net/view?a=commitdiff_plain;h=691701d2f8aea47c36cad4769b906bd06c48c86b;p=android-x86%2Fexternal-llvm.git

[TargetLowering] SimplifyDemandedBits - legal checks for SIGN/ZERO_EXTEND -> ZERO/ANY_EXTEND

As part of the fix for rL364264 + rL364272 - limit the *_EXTEND conversion to !TLO.LegalOperations || isOperationLegal cases.

We'll improve X86 legality in future commits.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@364290 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
index 64fc3b894a5..2608d06ff90 100644
--- a/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -1390,8 +1390,11 @@ bool TargetLowering::SimplifyDemandedBits(
           TLO.DAG.getDataLayout().isLittleEndian())
         return TLO.CombineTo(Op, TLO.DAG.getBitcast(VT, Src));
 
-      if (!IsVecInReg)
-        return TLO.CombineTo(Op, TLO.DAG.getNode(ISD::ANY_EXTEND, dl, VT, Src));
+      if (!IsVecInReg) {
+        unsigned Opc = ISD::ANY_EXTEND;
+        if (!TLO.LegalOperations() || isOperationLegal(Opc, VT))
+          return TLO.CombineTo(Op, TLO.DAG.getNode(Opc, dl, VT, Src));
+      }
     }
 
     APInt InDemandedBits = DemandedBits.trunc(InBits);
@@ -1414,8 +1417,11 @@ bool TargetLowering::SimplifyDemandedBits(
 
     // If none of the top bits are demanded, convert this into an any_extend.
     // TODO: Add SIGN_EXTEND_VECTOR_INREG - ANY_EXTEND_VECTOR_INREG fold.
-    if (DemandedBits.getActiveBits() <= InBits && !IsVecInReg)
-      return TLO.CombineTo(Op, TLO.DAG.getNode(ISD::ANY_EXTEND, dl, VT, Src));
+    if (DemandedBits.getActiveBits() <= InBits && !IsVecInReg) {
+      unsigned Opc = ISD::ANY_EXTEND;
+      if (!TLO.LegalOperations() || isOperationLegal(Opc, VT))
+        return TLO.CombineTo(Op, TLO.DAG.getNode(Opc, dl, VT, Src));
+    }
 
     APInt InDemandedBits = DemandedBits.trunc(InBits);
     APInt InDemandedElts = DemandedElts.zextOrSelf(InElts);
@@ -1435,8 +1441,11 @@ bool TargetLowering::SimplifyDemandedBits(
 
     // If the sign bit is known zero, convert this to a zero extend.
     // TODO: Add SIGN_EXTEND_VECTOR_INREG - ZERO_EXTEND_VECTOR_INREG fold.
-    if (Known.isNonNegative() && !IsVecInReg)
-      return TLO.CombineTo(Op, TLO.DAG.getNode(ISD::ZERO_EXTEND, dl, VT, Src));
+    if (Known.isNonNegative() && !IsVecInReg) {
+      unsigned Opc = ISD::ZERO_EXTEND;
+      if (!TLO.LegalOperations() || isOperationLegal(Opc, VT))
+        return TLO.CombineTo(Op, TLO.DAG.getNode(Opc, dl, VT, Src));
+    }
     break;
   }
   case ISD::ANY_EXTEND: {
diff --git a/test/CodeGen/X86/avx512-ext.ll b/test/CodeGen/X86/avx512-ext.ll
index 072e3c8bdae..d2c0c7d66da 100644
--- a/test/CodeGen/X86/avx512-ext.ll
+++ b/test/CodeGen/X86/avx512-ext.ll
@@ -475,7 +475,7 @@ define <4 x i32> @sext_4x8mem_to_4x32(<4 x i8> *%i , <4 x i1> %mask) nounwind re
 define <8 x i32> @zext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x8mem_to_8x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
@@ -492,7 +492,7 @@ define <8 x i32> @zext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 ;
 ; AVX512DQNOBW-LABEL: zext_8x8mem_to_8x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} ymm0 {%k1} {z} = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
@@ -506,7 +506,7 @@ define <8 x i32> @zext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 define <8 x i32> @sext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_8x8mem_to_8x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxbd (%rdi), %ymm0
@@ -523,7 +523,7 @@ define <8 x i32> @sext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 ;
 ; AVX512DQNOBW-LABEL: sext_8x8mem_to_8x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxbd (%rdi), %ymm0 {%k1} {z}
@@ -537,7 +537,7 @@ define <8 x i32> @sext_8x8mem_to_8x32(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 define <16 x i32> @zext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_16x8mem_to_16x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
@@ -552,7 +552,7 @@ define <16 x i32> @zext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwi
 ;
 ; AVX512DQNOBW-LABEL: zext_16x8mem_to_16x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
@@ -566,7 +566,7 @@ define <16 x i32> @zext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwi
 define <16 x i32> @sext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_16x8mem_to_16x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxbd (%rdi), %zmm0 {%k1} {z}
@@ -581,7 +581,7 @@ define <16 x i32> @sext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwi
 ;
 ; AVX512DQNOBW-LABEL: sext_16x8mem_to_16x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxbd (%rdi), %zmm0 {%k1} {z}
@@ -595,7 +595,7 @@ define <16 x i32> @sext_16x8mem_to_16x32(<16 x i8> *%i , <16 x i1> %mask) nounwi
 define <16 x i32> @zext_16x8_to_16x32_mask(<16 x i8> %a , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_16x8_to_16x32_mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 {%k1} {z} = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
@@ -610,7 +610,7 @@ define <16 x i32> @zext_16x8_to_16x32_mask(<16 x i8> %a , <16 x i1> %mask) nounw
 ;
 ; AVX512DQNOBW-LABEL: zext_16x8_to_16x32_mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 {%k1} {z} = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
@@ -623,7 +623,7 @@ define <16 x i32> @zext_16x8_to_16x32_mask(<16 x i8> %a , <16 x i1> %mask) nounw
 define <16 x i32> @sext_16x8_to_16x32_mask(<16 x i8> %a , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_16x8_to_16x32_mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0 {%k1} {z}
@@ -638,7 +638,7 @@ define <16 x i32> @sext_16x8_to_16x32_mask(<16 x i8> %a , <16 x i1> %mask) nounw
 ;
 ; AVX512DQNOBW-LABEL: sext_16x8_to_16x32_mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0 {%k1} {z}
@@ -777,7 +777,7 @@ define <4 x i64> @sext_4x8mem_to_4x64(<4 x i8> *%i) nounwind readnone {
 define <8 x i64> @zext_8x8mem_to_8x64(<8 x i8> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x8mem_to_8x64:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxbq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero,mem[4],zero,zero,zero,zero,zero,zero,zero,mem[5],zero,zero,zero,zero,zero,zero,zero,mem[6],zero,zero,zero,zero,zero,zero,zero,mem[7],zero,zero,zero,zero,zero,zero,zero
@@ -792,7 +792,7 @@ define <8 x i64> @zext_8x8mem_to_8x64(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 ;
 ; AVX512DQNOBW-LABEL: zext_8x8mem_to_8x64:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxbq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero,mem[4],zero,zero,zero,zero,zero,zero,zero,mem[5],zero,zero,zero,zero,zero,zero,zero,mem[6],zero,zero,zero,zero,zero,zero,zero,mem[7],zero,zero,zero,zero,zero,zero,zero
@@ -806,7 +806,7 @@ define <8 x i64> @zext_8x8mem_to_8x64(<8 x i8> *%i , <8 x i1> %mask) nounwind re
 define <8 x i64> @sext_8x8mem_to_8x64mask(<8 x i8> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_8x8mem_to_8x64mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxbq (%rdi), %zmm0 {%k1} {z}
@@ -821,7 +821,7 @@ define <8 x i64> @sext_8x8mem_to_8x64mask(<8 x i8> *%i , <8 x i1> %mask) nounwin
 ;
 ; AVX512DQNOBW-LABEL: sext_8x8mem_to_8x64mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxbq (%rdi), %zmm0 {%k1} {z}
@@ -902,7 +902,7 @@ define <4 x i32> @sext_4x16mem_to_4x32(<4 x i16> *%i) nounwind readnone {
 define <8 x i32> @zext_8x16mem_to_8x32(<8 x i16> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x16mem_to_8x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxwd {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
@@ -919,7 +919,7 @@ define <8 x i32> @zext_8x16mem_to_8x32(<8 x i16> *%i , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x16mem_to_8x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} ymm0 {%k1} {z} = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
@@ -933,7 +933,7 @@ define <8 x i32> @zext_8x16mem_to_8x32(<8 x i16> *%i , <8 x i1> %mask) nounwind
 define <8 x i32> @sext_8x16mem_to_8x32mask(<8 x i16> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_8x16mem_to_8x32mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxwd (%rdi), %ymm0
@@ -950,7 +950,7 @@ define <8 x i32> @sext_8x16mem_to_8x32mask(<8 x i16> *%i , <8 x i1> %mask) nounw
 ;
 ; AVX512DQNOBW-LABEL: sext_8x16mem_to_8x32mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxwd (%rdi), %ymm0 {%k1} {z}
@@ -974,7 +974,7 @@ define <8 x i32> @sext_8x16mem_to_8x32(<8 x i16> *%i) nounwind readnone {
 define <8 x i32> @zext_8x16_to_8x32mask(<8 x i16> %a , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x16_to_8x32mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovzxwd {{.*#+}} 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
@@ -991,7 +991,7 @@ define <8 x i32> @zext_8x16_to_8x32mask(<8 x i16> %a , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x16_to_8x32mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} ymm0 {%k1} {z} = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
@@ -1013,7 +1013,7 @@ define <8 x i32> @zext_8x16_to_8x32(<8 x i16> %a ) nounwind readnone {
 define <16 x i32> @zext_16x16mem_to_16x32(<16 x i16> *%i , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_16x16mem_to_16x32:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxwd {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
@@ -1028,7 +1028,7 @@ define <16 x i32> @zext_16x16mem_to_16x32(<16 x i16> *%i , <16 x i1> %mask) noun
 ;
 ; AVX512DQNOBW-LABEL: zext_16x16mem_to_16x32:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
@@ -1042,7 +1042,7 @@ define <16 x i32> @zext_16x16mem_to_16x32(<16 x i16> *%i , <16 x i1> %mask) noun
 define <16 x i32> @sext_16x16mem_to_16x32mask(<16 x i16> *%i , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_16x16mem_to_16x32mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxwd (%rdi), %zmm0 {%k1} {z}
@@ -1057,7 +1057,7 @@ define <16 x i32> @sext_16x16mem_to_16x32mask(<16 x i16> *%i , <16 x i1> %mask)
 ;
 ; AVX512DQNOBW-LABEL: sext_16x16mem_to_16x32mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxwd (%rdi), %zmm0 {%k1} {z}
@@ -1080,7 +1080,7 @@ define <16 x i32> @sext_16x16mem_to_16x32(<16 x i16> *%i) nounwind readnone {
 define <16 x i32> @zext_16x16_to_16x32mask(<16 x i16> %a , <16 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_16x16_to_16x32mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovzxwd {{.*#+}} zmm0 {%k1} {z} = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
@@ -1095,7 +1095,7 @@ define <16 x i32> @zext_16x16_to_16x32mask(<16 x i16> %a , <16 x i1> %mask) noun
 ;
 ; AVX512DQNOBW-LABEL: zext_16x16_to_16x32mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} zmm0 {%k1} {z} = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
@@ -1227,7 +1227,7 @@ define <4 x i64> @sext_4x16mem_to_4x64(<4 x i16> *%i) nounwind readnone {
 define <8 x i64> @zext_8x16mem_to_8x64(<8 x i16> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x16mem_to_8x64:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
@@ -1242,7 +1242,7 @@ define <8 x i64> @zext_8x16mem_to_8x64(<8 x i16> *%i , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x16mem_to_8x64:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
@@ -1256,7 +1256,7 @@ define <8 x i64> @zext_8x16mem_to_8x64(<8 x i16> *%i , <8 x i1> %mask) nounwind
 define <8 x i64> @sext_8x16mem_to_8x64mask(<8 x i16> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_8x16mem_to_8x64mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxwq (%rdi), %zmm0 {%k1} {z}
@@ -1271,7 +1271,7 @@ define <8 x i64> @sext_8x16mem_to_8x64mask(<8 x i16> *%i , <8 x i1> %mask) nounw
 ;
 ; AVX512DQNOBW-LABEL: sext_8x16mem_to_8x64mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxwq (%rdi), %zmm0 {%k1} {z}
@@ -1295,7 +1295,7 @@ define <8 x i64> @sext_8x16mem_to_8x64(<8 x i16> *%i) nounwind readnone {
 define <8 x i64> @zext_8x16_to_8x64mask(<8 x i16> %a , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x16_to_8x64mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 {%k1} {z} = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
@@ -1310,7 +1310,7 @@ define <8 x i64> @zext_8x16_to_8x64mask(<8 x i16> %a , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x16_to_8x64mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxwq {{.*#+}} zmm0 {%k1} {z} = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
@@ -1472,7 +1472,7 @@ define <4 x i64> @zext_4x32_to_4x64mask(<4 x i32> %a , <4 x i1> %mask) nounwind
 define <8 x i64> @zext_8x32mem_to_8x64(<8 x i32> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x32mem_to_8x64:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovzxdq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
@@ -1487,7 +1487,7 @@ define <8 x i64> @zext_8x32mem_to_8x64(<8 x i32> *%i , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x32mem_to_8x64:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxdq {{.*#+}} zmm0 {%k1} {z} = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
@@ -1501,7 +1501,7 @@ define <8 x i64> @zext_8x32mem_to_8x64(<8 x i32> *%i , <8 x i1> %mask) nounwind
 define <8 x i64> @sext_8x32mem_to_8x64mask(<8 x i32> *%i , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: sext_8x32mem_to_8x64mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; KNL-NEXT:    vpmovsxdq (%rdi), %zmm0 {%k1} {z}
@@ -1516,7 +1516,7 @@ define <8 x i64> @sext_8x32mem_to_8x64mask(<8 x i32> *%i , <8 x i1> %mask) nounw
 ;
 ; AVX512DQNOBW-LABEL: sext_8x32mem_to_8x64mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512DQNOBW-NEXT:    vpmovsxdq (%rdi), %zmm0 {%k1} {z}
@@ -1549,7 +1549,7 @@ define <8 x i64> @sext_8x32_to_8x64(<8 x i32> %a) nounwind readnone {
 define <8 x i64> @zext_8x32_to_8x64mask(<8 x i32> %a , <8 x i1> %mask) nounwind readnone {
 ; KNL-LABEL: zext_8x32_to_8x64mask:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpmovzxdq {{.*#+}} zmm0 {%k1} {z} = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero
@@ -1564,7 +1564,7 @@ define <8 x i64> @zext_8x32_to_8x64mask(<8 x i32> %a , <8 x i1> %mask) nounwind
 ;
 ; AVX512DQNOBW-LABEL: zext_8x32_to_8x64mask:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512DQNOBW-NEXT:    vpmovzxdq {{.*#+}} zmm0 {%k1} {z} = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero
@@ -1705,7 +1705,7 @@ define <4 x i32> @trunc_4i32_to_4i1(<4 x i32> %a, <4 x i32> %b) {
 define i8 @trunc_8i16_to_8i1(<8 x i16> %a) {
 ; KNL-LABEL: trunc_8i16_to_8i1:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1723,7 +1723,7 @@ define i8 @trunc_8i16_to_8i1(<8 x i16> %a) {
 ;
 ; AVX512DQNOBW-LABEL: trunc_8i16_to_8i1:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k0
 ; AVX512DQNOBW-NEXT:    kmovw %k0, %eax
diff --git a/test/CodeGen/X86/avx512-fma.ll b/test/CodeGen/X86/avx512-fma.ll
index e9a70f3bfe2..07eeeb4c662 100644
--- a/test/CodeGen/X86/avx512-fma.ll
+++ b/test/CodeGen/X86/avx512-fma.ll
@@ -122,7 +122,7 @@ define <16 x float> @test213_br(<16 x float> %a1, <16 x float> %a2) nounwind {
 define <16 x float> @test_x86_fmadd132_ps(<16 x float> %a0, <16 x float> %a1, <16 x float> *%a2_ptrt, <16 x i1> %mask) {
 ; KNL-LABEL: test_x86_fmadd132_ps:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL-NEXT:    vfmadd132ps {{.*#+}} zmm0 = (zmm0 * mem) + zmm1
@@ -145,7 +145,7 @@ define <16 x float> @test_x86_fmadd132_ps(<16 x float> %a0, <16 x float> %a1, <1
 define <16 x float> @test_x86_fmadd231_ps(<16 x float> %a0, <16 x float> %a1, <16 x float> *%a2_ptrt, <16 x i1> %mask) {
 ; KNL-LABEL: test_x86_fmadd231_ps:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL-NEXT:    vfmadd231ps {{.*#+}} zmm1 = (zmm0 * mem) + zmm1
@@ -170,7 +170,7 @@ define <16 x float> @test_x86_fmadd231_ps(<16 x float> %a0, <16 x float> %a1, <1
 define <16 x float> @test_x86_fmadd213_ps(<16 x float> %a0, <16 x float> %a1, <16 x float> *%a2_ptrt, <16 x i1> %mask) {
 ; KNL-LABEL: test_x86_fmadd213_ps:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL-NEXT:    vfmadd213ps {{.*#+}} zmm1 = (zmm0 * zmm1) + mem
diff --git a/test/CodeGen/X86/avx512-insert-extract.ll b/test/CodeGen/X86/avx512-insert-extract.ll
index 8bdd7dc2c1d..02c67ba079c 100644
--- a/test/CodeGen/X86/avx512-insert-extract.ll
+++ b/test/CodeGen/X86/avx512-insert-extract.ll
@@ -1710,11 +1710,11 @@ define i32 @test_insertelement_variable_v32i1(<32 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    testb %dil, %dil
 ; KNL-NEXT:    vmovdqa %ymm0, (%rsp)
 ; KNL-NEXT:    setne (%rsp,%rsi)
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd (%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1776,21 +1776,21 @@ define i64 @test_insertelement_variable_v64i1(<64 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    vmovdqa %ymm1, {{[0-9]+}}(%rsp)
 ; KNL-NEXT:    vmovdqa %ymm0, (%rsp)
 ; KNL-NEXT:    setne (%rsp,%rsi)
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd (%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    shll $16, %ecx
 ; KNL-NEXT:    orl %eax, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1956,21 +1956,21 @@ define i96 @test_insertelement_variable_v96i1(<96 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    vmovdqa %ymm1, {{[0-9]+}}(%rsp)
 ; KNL-NEXT:    vmovdqa %ymm2, (%rsp)
 ; KNL-NEXT:    setne (%rsp,%rax)
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd (%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    shll $16, %ecx
 ; KNL-NEXT:    orl %eax, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1978,21 +1978,21 @@ define i96 @test_insertelement_variable_v96i1(<96 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    orl %edx, %eax
 ; KNL-NEXT:    shlq $32, %rax
 ; KNL-NEXT:    orq %rcx, %rax
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %esi
 ; KNL-NEXT:    shll $16, %esi
 ; KNL-NEXT:    orl %ecx, %esi
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
@@ -2168,21 +2168,21 @@ define i128 @test_insertelement_variable_v128i1(<128 x i8> %a, i8 %b, i32 %index
 ; KNL-NEXT:    vmovdqa %ymm1, {{[0-9]+}}(%rsp)
 ; KNL-NEXT:    vmovdqa %ymm0, (%rsp)
 ; KNL-NEXT:    setne (%rsp,%rsi)
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd (%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    shll $16, %ecx
 ; KNL-NEXT:    orl %eax, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2190,21 +2190,21 @@ define i128 @test_insertelement_variable_v128i1(<128 x i8> %a, i8 %b, i32 %index
 ; KNL-NEXT:    orl %edx, %eax
 ; KNL-NEXT:    shlq $32, %rax
 ; KNL-NEXT:    orq %rcx, %rax
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %esi
 ; KNL-NEXT:    shll $16, %esi
 ; KNL-NEXT:    orl %ecx, %esi
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
diff --git a/test/CodeGen/X86/avx512-mask-op.ll b/test/CodeGen/X86/avx512-mask-op.ll
index 96bbadba695..02cf7544e0c 100644
--- a/test/CodeGen/X86/avx512-mask-op.ll
+++ b/test/CodeGen/X86/avx512-mask-op.ll
@@ -613,7 +613,7 @@ false:
 define void @test7(<8 x i1> %mask)  {
 ; KNL-LABEL: test7:
 ; KNL:       ## %bb.0: ## %allocas
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -640,7 +640,7 @@ define void @test7(<8 x i1> %mask)  {
 ;
 ; AVX512DQ-LABEL: test7:
 ; AVX512DQ:       ## %bb.0: ## %allocas
-; AVX512DQ-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovw %k0, %eax
@@ -763,10 +763,10 @@ define <16 x i1> @test9(<16 x i1>%a, <16 x i1>%b, i32 %a1, i32 %b1) {
 ; KNL-NEXT:    cmpl %esi, %edi
 ; KNL-NEXT:    jg LBB18_1
 ; KNL-NEXT:  ## %bb.2:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm1, %zmm0
 ; KNL-NEXT:    jmp LBB18_3
 ; KNL-NEXT:  LBB18_1:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:  LBB18_3:
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
@@ -810,10 +810,10 @@ define <16 x i1> @test9(<16 x i1>%a, <16 x i1>%b, i32 %a1, i32 %b1) {
 ; AVX512DQ-NEXT:    cmpl %esi, %edi
 ; AVX512DQ-NEXT:    jg LBB18_1
 ; AVX512DQ-NEXT:  ## %bb.2:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm1, %zmm0
 ; AVX512DQ-NEXT:    jmp LBB18_3
 ; AVX512DQ-NEXT:  LBB18_1:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQ-NEXT:  LBB18_3:
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
@@ -1666,7 +1666,7 @@ define void @store_v4i1(<4 x i1> %c , <4 x i1>* %ptr) {
 define void @store_v8i1(<8 x i1> %c , <8 x i1>* %ptr) {
 ; KNL-LABEL: store_v8i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestnmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1694,7 +1694,7 @@ define void @store_v8i1(<8 x i1> %c , <8 x i1>* %ptr) {
 ;
 ; AVX512DQ-LABEL: store_v8i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    knotb %k0, %k0
@@ -1718,7 +1718,7 @@ define void @store_v8i1(<8 x i1> %c , <8 x i1>* %ptr) {
 define void @store_v16i1(<16 x i1> %c , <16 x i1>* %ptr) {
 ; KNL-LABEL: store_v16i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestnmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, (%rdi)
@@ -1744,7 +1744,7 @@ define void @store_v16i1(<16 x i1> %c , <16 x i1>* %ptr) {
 ;
 ; AVX512DQ-LABEL: store_v16i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
 ; AVX512DQ-NEXT:    knotw %k0, %k0
@@ -2483,7 +2483,7 @@ define <64 x i8> @load_64i1(<64 x i1>* %a) {
 define void @store_8i1(<8 x i1>* %a, <8 x i1> %v) {
 ; KNL-LABEL: store_8i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2509,7 +2509,7 @@ define void @store_8i1(<8 x i1>* %a, <8 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_8i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovb %k0, (%rdi)
@@ -2530,7 +2530,7 @@ define void @store_8i1(<8 x i1>* %a, <8 x i1> %v) {
 define void @store_8i1_1(<8 x i1>* %a, <8 x i16> %v) {
 ; KNL-LABEL: store_8i1_1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; KNL-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2556,7 +2556,7 @@ define void @store_8i1_1(<8 x i1>* %a, <8 x i16> %v) {
 ;
 ; AVX512DQ-LABEL: store_8i1_1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovb %k0, (%rdi)
@@ -2578,7 +2578,7 @@ define void @store_8i1_1(<8 x i1>* %a, <8 x i16> %v) {
 define void @store_16i1(<16 x i1>* %a, <16 x i1> %v) {
 ; KNL-LABEL: store_16i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, (%rdi)
@@ -2602,7 +2602,7 @@ define void @store_16i1(<16 x i1>* %a, <16 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_16i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovw %k0, (%rdi)
@@ -2623,11 +2623,11 @@ define void @store_16i1(<16 x i1>* %a, <16 x i1> %v) {
 define void @store_32i1(<32 x i1>* %a, <32 x i1> %v) {
 ; KNL-LABEL: store_32i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm1
 ; KNL-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    kmovw %k1, 2(%rdi)
@@ -2653,11 +2653,11 @@ define void @store_32i1(<32 x i1>* %a, <32 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_32i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm1
 ; AVX512DQ-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512DQ-NEXT:    vpmovd2m %zmm1, %k0
 ; AVX512DQ-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQ-NEXT:    kmovw %k1, 2(%rdi)
@@ -2680,10 +2680,10 @@ define void @store_32i1(<32 x i1>* %a, <32 x i1> %v) {
 define void @store_32i1_1(<32 x i1>* %a, <32 x i16> %v) {
 ; KNL-LABEL: store_32i1_1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
-; KNL-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm1[0],zero,ymm1[1],zero,ymm1[2],zero,ymm1[3],zero,ymm1[4],zero,ymm1[5],zero,ymm1[6],zero,ymm1[7],zero,ymm1[8],zero,ymm1[9],zero,ymm1[10],zero,ymm1[11],zero,ymm1[12],zero,ymm1[13],zero,ymm1[14],zero,ymm1[15],zero
+; KNL-NEXT:    vpmovsxwd %ymm1, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    kmovw %k1, 2(%rdi)
@@ -2709,10 +2709,10 @@ define void @store_32i1_1(<32 x i1>* %a, <32 x i16> %v) {
 ;
 ; AVX512DQ-LABEL: store_32i1_1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
-; AVX512DQ-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm1[0],zero,ymm1[1],zero,ymm1[2],zero,ymm1[3],zero,ymm1[4],zero,ymm1[5],zero,ymm1[6],zero,ymm1[7],zero,ymm1[8],zero,ymm1[9],zero,ymm1[10],zero,ymm1[11],zero,ymm1[12],zero,ymm1[13],zero,ymm1[14],zero,ymm1[15],zero
+; AVX512DQ-NEXT:    vpmovsxwd %ymm1, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQ-NEXT:    kmovw %k1, 2(%rdi)
@@ -2738,16 +2738,16 @@ define void @store_64i1(<64 x i1>* %a, <64 x i1> %v) {
 ;
 ; KNL-LABEL: store_64i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm1, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm2, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k2
-; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero,xmm3[8],zero,zero,zero,xmm3[9],zero,zero,zero,xmm3[10],zero,zero,zero,xmm3[11],zero,zero,zero,xmm3[12],zero,zero,zero,xmm3[13],zero,zero,zero,xmm3[14],zero,zero,zero,xmm3[15],zero,zero,zero
+; KNL-NEXT:    vpmovsxbd %xmm3, %zmm0
 ; KNL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k3
 ; KNL-NEXT:    kmovw %k3, 6(%rdi)
@@ -2775,16 +2775,16 @@ define void @store_64i1(<64 x i1>* %a, <64 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_64i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm1, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm2, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k2
-; AVX512DQ-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero,xmm3[8],zero,zero,zero,xmm3[9],zero,zero,zero,xmm3[10],zero,zero,zero,xmm3[11],zero,zero,zero,xmm3[12],zero,zero,zero,xmm3[13],zero,zero,zero,xmm3[14],zero,zero,zero,xmm3[15],zero,zero,zero
+; AVX512DQ-NEXT:    vpmovsxbd %xmm3, %zmm0
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k3
 ; AVX512DQ-NEXT:    kmovw %k3, 6(%rdi)
diff --git a/test/CodeGen/X86/avx512-masked-memop-64-32.ll b/test/CodeGen/X86/avx512-masked-memop-64-32.ll
index 89dfc69fcd5..f199cb097aa 100644
--- a/test/CodeGen/X86/avx512-masked-memop-64-32.ll
+++ b/test/CodeGen/X86/avx512-masked-memop-64-32.ll
@@ -111,7 +111,7 @@ declare <16 x %mystruct*> @llvm.masked.load.v16p0mystruct.p0v16p0mystruct(<16 x
 define <16 x %mystruct*> @test24(<16 x i1> %mask, <16 x %mystruct*>* %addr) {
 ; AVX512F-LABEL: test24:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovdqu64 (%rdi), %zmm0 {%k1} {z}
@@ -134,7 +134,7 @@ define <16 x %mystruct*> @test24(<16 x i1> %mask, <16 x %mystruct*>* %addr) {
 define void @test_store_16i64(<16 x i64>* %ptrs, <16 x i1> %mask, <16 x i64> %src0)  {
 ; AVX512F-LABEL: test_store_16i64:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovdqu64 %zmm1, (%rdi) {%k1}
@@ -160,7 +160,7 @@ declare void @llvm.masked.store.v16i64.p0v16i64(<16 x i64> %src0, <16 x i64>* %p
 define void @test_store_16f64(<16 x double>* %ptrs, <16 x i1> %mask, <16 x double> %src0)  {
 ; AVX512F-LABEL: test_store_16f64:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovupd %zmm1, (%rdi) {%k1}
@@ -186,7 +186,7 @@ declare void @llvm.masked.store.v16f64.p0v16f64(<16 x double> %src0, <16 x doubl
 define <16 x i64> @test_load_16i64(<16 x i64>* %ptrs, <16 x i1> %mask, <16 x i64> %src0)  {
 ; AVX512F-LABEL: test_load_16i64:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vpblendmq (%rdi), %zmm1, %zmm0 {%k1}
@@ -210,7 +210,7 @@ declare <16 x i64> @llvm.masked.load.v16i64.p0v16i64(<16 x i64>* %ptrs, i32, <16
 define <16 x double> @test_load_16f64(<16 x double>* %ptrs, <16 x i1> %mask, <16 x double> %src0)  {
 ; AVX512F-LABEL: test_load_16f64:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vblendmpd (%rdi), %zmm1, %zmm0 {%k1}
@@ -235,10 +235,10 @@ define <32 x double> @test_load_32f64(<32 x double>* %ptrs, <32 x i1> %mask, <32
 ; AVX512F-LABEL: test_load_32f64:
 ; AVX512F:       ## %bb.0:
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm5
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm5 = xmm5[0],zero,zero,zero,xmm5[1],zero,zero,zero,xmm5[2],zero,zero,zero,xmm5[3],zero,zero,zero,xmm5[4],zero,zero,zero,xmm5[5],zero,zero,zero,xmm5[6],zero,zero,zero,xmm5[7],zero,zero,zero,xmm5[8],zero,zero,zero,xmm5[9],zero,zero,zero,xmm5[10],zero,zero,zero,xmm5[11],zero,zero,zero,xmm5[12],zero,zero,zero,xmm5[13],zero,zero,zero,xmm5[14],zero,zero,zero,xmm5[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm5, %zmm5
 ; AVX512F-NEXT:    vpslld $31, %zmm5, %zmm5
 ; AVX512F-NEXT:    vptestmd %zmm5, %zmm5, %k1
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512F-NEXT:    vblendmpd (%rdi), %zmm1, %zmm0 {%k2}
diff --git a/test/CodeGen/X86/avx512-vec-cmp.ll b/test/CodeGen/X86/avx512-vec-cmp.ll
index a823e1b275b..1799ebd53fa 100644
--- a/test/CodeGen/X86/avx512-vec-cmp.ll
+++ b/test/CodeGen/X86/avx512-vec-cmp.ll
@@ -850,8 +850,7 @@ define <4  x float> @test42(<4  x float> %x, <4  x float> %x1, float* %ptr) noun
 define <8 x double> @test43(<8 x double> %x, <8 x double> %x1, double* %ptr,<8 x i1> %mask_in) nounwind {
 ; KNL-LABEL: test43:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovzxwq %xmm2, %zmm2 ## encoding: [0x62,0xf2,0x7d,0x48,0x34,0xd2]
-; KNL-NEXT:    ## zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
+; KNL-NEXT:    vpmovsxwq %xmm2, %zmm2 ## encoding: [0x62,0xf2,0x7d,0x48,0x24,0xd2]
 ; KNL-NEXT:    vpsllq $63, %zmm2, %zmm2 ## encoding: [0x62,0xf1,0xed,0x48,0x73,0xf2,0x3f]
 ; KNL-NEXT:    vcmpltpd (%rdi){1to8}, %zmm0, %k1 ## encoding: [0x62,0xf1,0xfd,0x58,0xc2,0x0f,0x01]
 ; KNL-NEXT:    vptestmq %zmm2, %zmm2, %k1 {%k1} ## encoding: [0x62,0xf2,0xed,0x49,0x27,0xca]
diff --git a/test/CodeGen/X86/masked_compressstore.ll b/test/CodeGen/X86/masked_compressstore.ll
index 0806ccd86ef..2b3ee63fc23 100644
--- a/test/CodeGen/X86/masked_compressstore.ll
+++ b/test/CodeGen/X86/masked_compressstore.ll
@@ -194,7 +194,7 @@ define void @compressstore_v8f64_v8i1(double* %base, <8 x double> %V, <8 x i1> %
 ;
 ; AVX512F-LABEL: compressstore_v8f64_v8i1:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
@@ -203,7 +203,7 @@ define void @compressstore_v8f64_v8i1(double* %base, <8 x double> %V, <8 x i1> %
 ;
 ; AVX512VLDQ-LABEL: compressstore_v8f64_v8i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512VLDQ-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
@@ -579,7 +579,7 @@ define void @compressstore_v16f64_v16i1(double* %base, <16 x double> %V, <16 x i
 ;
 ; AVX512F-LABEL: compressstore_v16f64_v16i1:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; AVX512F-NEXT:    vpslld $31, %zmm2, %zmm2
 ; AVX512F-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; AVX512F-NEXT:    kshiftrw $8, %k1, %k2
@@ -607,7 +607,7 @@ define void @compressstore_v16f64_v16i1(double* %base, <16 x double> %V, <16 x i
 ;
 ; AVX512VLDQ-LABEL: compressstore_v16f64_v16i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; AVX512VLDQ-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; AVX512VLDQ-NEXT:    vpslld $31, %zmm2, %zmm2
 ; AVX512VLDQ-NEXT:    vpmovd2m %zmm2, %k1
 ; AVX512VLDQ-NEXT:    kmovb %k1, %eax
@@ -1094,7 +1094,7 @@ define void @compressstore_v8f32_v8i1(float* %base, <8 x float> %V, <8 x i1> %ma
 ; AVX512F-LABEL: compressstore_v8f32_v8i1:
 ; AVX512F:       ## %bb.0:
 ; AVX512F-NEXT:    ## kill: def $ymm0 killed $ymm0 def $zmm0
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
@@ -1103,7 +1103,7 @@ define void @compressstore_v8f32_v8i1(float* %base, <8 x float> %V, <8 x i1> %ma
 ;
 ; AVX512VLDQ-LABEL: compressstore_v8f32_v8i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512VLDQ-NEXT:    vcompressps %ymm0, (%rdi) {%k1}
@@ -2997,7 +2997,7 @@ define void @compressstore_v8i64_v8i1(i64* %base, <8 x i64> %V, <8 x i1> %mask)
 ;
 ; AVX512F-LABEL: compressstore_v8i64_v8i1:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
@@ -3006,7 +3006,7 @@ define void @compressstore_v8i64_v8i1(i64* %base, <8 x i64> %V, <8 x i1> %mask)
 ;
 ; AVX512VLDQ-LABEL: compressstore_v8i64_v8i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512VLDQ-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
diff --git a/test/CodeGen/X86/masked_expandload.ll b/test/CodeGen/X86/masked_expandload.ll
index 351e0409f3f..62b012ca2ab 100644
--- a/test/CodeGen/X86/masked_expandload.ll
+++ b/test/CodeGen/X86/masked_expandload.ll
@@ -466,7 +466,7 @@ define <8 x double> @expandload_v8f64_v8i1(double* %base, <8 x double> %src0, <8
 ;
 ; AVX512F-LABEL: expandload_v8f64_v8i1:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vexpandpd (%rdi), %zmm0 {%k1}
@@ -474,7 +474,7 @@ define <8 x double> @expandload_v8f64_v8i1(double* %base, <8 x double> %src0, <8
 ;
 ; AVX512VLDQ-LABEL: expandload_v8f64_v8i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm1, %k1
 ; AVX512VLDQ-NEXT:    vexpandpd (%rdi), %zmm0 {%k1}
diff --git a/test/CodeGen/X86/masked_gather_scatter.ll b/test/CodeGen/X86/masked_gather_scatter.ll
index 2c4294e07db..610fca0949a 100644
--- a/test/CodeGen/X86/masked_gather_scatter.ll
+++ b/test/CodeGen/X86/masked_gather_scatter.ll
@@ -1850,7 +1850,7 @@ define <16 x float*> @test31(<16 x float**> %ptrs) {
 define <16 x i32> @test_gather_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %src0)  {
 ; KNL_64-LABEL: test_gather_16i32:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    vextracti64x4 $1, %zmm3, %ymm2
@@ -1862,7 +1862,7 @@ define <16 x i32> @test_gather_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i
 ;
 ; KNL_32-LABEL: test_gather_16i32:
 ; KNL_32:       # %bb.0:
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vpgatherdd (,%zmm0), %zmm2 {%k1}
@@ -1871,7 +1871,7 @@ define <16 x i32> @test_gather_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i
 ;
 ; SKX-LABEL: test_gather_16i32:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    vextracti64x4 $1, %zmm3, %ymm2
@@ -1883,7 +1883,7 @@ define <16 x i32> @test_gather_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i
 ;
 ; SKX_32-LABEL: test_gather_16i32:
 ; SKX_32:       # %bb.0:
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vpgatherdd (,%zmm0), %zmm2 {%k1}
@@ -1895,7 +1895,7 @@ define <16 x i32> @test_gather_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i
 define <16 x i64> @test_gather_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i64> %src0)  {
 ; KNL_64-LABEL: test_gather_16i64:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -1914,7 +1914,7 @@ define <16 x i64> @test_gather_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i
 ; KNL_32-NEXT:    .cfi_def_cfa_register %ebp
 ; KNL_32-NEXT:    andl $-64, %esp
 ; KNL_32-NEXT:    subl $64, %esp
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vmovdqa64 8(%ebp), %zmm1
@@ -1930,7 +1930,7 @@ define <16 x i64> @test_gather_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i
 ;
 ; SKX-LABEL: test_gather_16i64:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -1949,7 +1949,7 @@ define <16 x i64> @test_gather_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i
 ; SKX_32-NEXT:    .cfi_def_cfa_register %ebp
 ; SKX_32-NEXT:    andl $-64, %esp
 ; SKX_32-NEXT:    subl $64, %esp
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vmovdqa64 8(%ebp), %zmm1
@@ -1969,7 +1969,7 @@ declare <16 x i64> @llvm.masked.gather.v16i64.v16p0i64(<16 x i64*> %ptrs, i32, <
 define <16 x float> @test_gather_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16 x float> %src0)  {
 ; KNL_64-LABEL: test_gather_16f32:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    vextractf64x4 $1, %zmm3, %ymm2
@@ -1981,7 +1981,7 @@ define <16 x float> @test_gather_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16
 ;
 ; KNL_32-LABEL: test_gather_16f32:
 ; KNL_32:       # %bb.0:
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vgatherdps (,%zmm0), %zmm2 {%k1}
@@ -1990,7 +1990,7 @@ define <16 x float> @test_gather_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16
 ;
 ; SKX-LABEL: test_gather_16f32:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    vextractf64x4 $1, %zmm3, %ymm2
@@ -2002,7 +2002,7 @@ define <16 x float> @test_gather_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16
 ;
 ; SKX_32-LABEL: test_gather_16f32:
 ; SKX_32:       # %bb.0:
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vgatherdps (,%zmm0), %zmm2 {%k1}
@@ -2014,7 +2014,7 @@ define <16 x float> @test_gather_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16
 define <16 x double> @test_gather_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <16 x double> %src0)  {
 ; KNL_64-LABEL: test_gather_16f64:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -2033,7 +2033,7 @@ define <16 x double> @test_gather_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <
 ; KNL_32-NEXT:    .cfi_def_cfa_register %ebp
 ; KNL_32-NEXT:    andl $-64, %esp
 ; KNL_32-NEXT:    subl $64, %esp
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vmovapd 8(%ebp), %zmm1
@@ -2049,7 +2049,7 @@ define <16 x double> @test_gather_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <
 ;
 ; SKX-LABEL: test_gather_16f64:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -2068,7 +2068,7 @@ define <16 x double> @test_gather_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <
 ; SKX_32-NEXT:    .cfi_def_cfa_register %ebp
 ; SKX_32-NEXT:    andl $-64, %esp
 ; SKX_32-NEXT:    subl $64, %esp
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vmovapd 8(%ebp), %zmm1
@@ -2088,7 +2088,7 @@ declare <16 x double> @llvm.masked.gather.v16f64.v16p0f64(<16 x double*> %ptrs,
 define void @test_scatter_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %src0)  {
 ; KNL_64-LABEL: test_scatter_16i32:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -2100,7 +2100,7 @@ define void @test_scatter_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %
 ;
 ; KNL_32-LABEL: test_scatter_16i32:
 ; KNL_32:       # %bb.0:
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vpscatterdd %zmm2, (,%zmm0) {%k1}
@@ -2109,7 +2109,7 @@ define void @test_scatter_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %
 ;
 ; SKX-LABEL: test_scatter_16i32:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -2121,7 +2121,7 @@ define void @test_scatter_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %
 ;
 ; SKX_32-LABEL: test_scatter_16i32:
 ; SKX_32:       # %bb.0:
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vpscatterdd %zmm2, (,%zmm0) {%k1}
@@ -2133,7 +2133,7 @@ define void @test_scatter_16i32(<16 x i32*> %ptrs, <16 x i1> %mask, <16 x i32> %
 define void @test_scatter_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i64> %src0)  {
 ; KNL_64-LABEL: test_scatter_16i64:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -2151,7 +2151,7 @@ define void @test_scatter_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i64> %
 ; KNL_32-NEXT:    .cfi_def_cfa_register %ebp
 ; KNL_32-NEXT:    andl $-64, %esp
 ; KNL_32-NEXT:    subl $64, %esp
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vmovdqa64 8(%ebp), %zmm1
@@ -2167,7 +2167,7 @@ define void @test_scatter_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i64> %
 ;
 ; SKX-LABEL: test_scatter_16i64:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -2185,7 +2185,7 @@ define void @test_scatter_16i64(<16 x i64*> %ptrs, <16 x i1> %mask, <16 x i64> %
 ; SKX_32-NEXT:    .cfi_def_cfa_register %ebp
 ; SKX_32-NEXT:    andl $-64, %esp
 ; SKX_32-NEXT:    subl $64, %esp
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vmovdqa64 8(%ebp), %zmm1
@@ -2205,7 +2205,7 @@ declare void @llvm.masked.scatter.v16i64.v16p0i64(<16 x i64> %src0, <16 x i64*>
 define void @test_scatter_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16 x float> %src0)  {
 ; KNL_64-LABEL: test_scatter_16f32:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -2217,7 +2217,7 @@ define void @test_scatter_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16 x floa
 ;
 ; KNL_32-LABEL: test_scatter_16f32:
 ; KNL_32:       # %bb.0:
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vscatterdps %zmm2, (,%zmm0) {%k1}
@@ -2226,7 +2226,7 @@ define void @test_scatter_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16 x floa
 ;
 ; SKX-LABEL: test_scatter_16f32:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -2238,7 +2238,7 @@ define void @test_scatter_16f32(<16 x float*> %ptrs, <16 x i1> %mask, <16 x floa
 ;
 ; SKX_32-LABEL: test_scatter_16f32:
 ; SKX_32:       # %bb.0:
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vscatterdps %zmm2, (,%zmm0) {%k1}
@@ -2251,7 +2251,7 @@ declare void @llvm.masked.scatter.v16f32.v16p0f32(<16 x float> %src0, <16 x floa
 define void @test_scatter_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <16 x double> %src0)  {
 ; KNL_64-LABEL: test_scatter_16f64:
 ; KNL_64:       # %bb.0:
-; KNL_64-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; KNL_64-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; KNL_64-NEXT:    vpslld $31, %zmm2, %zmm2
 ; KNL_64-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL_64-NEXT:    kshiftrw $8, %k1, %k2
@@ -2269,7 +2269,7 @@ define void @test_scatter_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <16 x dou
 ; KNL_32-NEXT:    .cfi_def_cfa_register %ebp
 ; KNL_32-NEXT:    andl $-64, %esp
 ; KNL_32-NEXT:    subl $64, %esp
-; KNL_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; KNL_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; KNL_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; KNL_32-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; KNL_32-NEXT:    vmovapd 8(%ebp), %zmm1
@@ -2285,7 +2285,7 @@ define void @test_scatter_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <16 x dou
 ;
 ; SKX-LABEL: test_scatter_16f64:
 ; SKX:       # %bb.0:
-; SKX-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
+; SKX-NEXT:    vpmovsxbd %xmm2, %zmm2
 ; SKX-NEXT:    vpslld $31, %zmm2, %zmm2
 ; SKX-NEXT:    vpmovd2m %zmm2, %k1
 ; SKX-NEXT:    kshiftrw $8, %k1, %k2
@@ -2303,7 +2303,7 @@ define void @test_scatter_16f64(<16 x double*> %ptrs, <16 x i1> %mask, <16 x dou
 ; SKX_32-NEXT:    .cfi_def_cfa_register %ebp
 ; SKX_32-NEXT:    andl $-64, %esp
 ; SKX_32-NEXT:    subl $64, %esp
-; SKX_32-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero
+; SKX_32-NEXT:    vpmovsxbd %xmm1, %zmm1
 ; SKX_32-NEXT:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vmovapd 8(%ebp), %zmm1
diff --git a/test/CodeGen/X86/masked_load.ll b/test/CodeGen/X86/masked_load.ll
index b5efae10536..81ebf448914 100644
--- a/test/CodeGen/X86/masked_load.ll
+++ b/test/CodeGen/X86/masked_load.ll
@@ -1256,7 +1256,7 @@ define <8 x float> @load_v8f32_v8i1_zero(<8 x i1> %mask, <8 x float>* %addr) {
 ;
 ; AVX512F-LABEL: load_v8f32_v8i1_zero:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovups (%rdi), %zmm0 {%k1} {z}
@@ -1265,7 +1265,7 @@ define <8 x float> @load_v8f32_v8i1_zero(<8 x i1> %mask, <8 x float>* %addr) {
 ;
 ; AVX512VLDQ-LABEL: load_v8f32_v8i1_zero:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512VLDQ-NEXT:    vmovaps (%rdi), %ymm0 {%k1} {z}
@@ -2442,7 +2442,7 @@ define <8 x i32> @load_v8i32_v8i1(<8 x i1> %mask, <8 x i32>* %addr, <8 x i32> %d
 ; AVX512F-LABEL: load_v8i32_v8i1:
 ; AVX512F:       ## %bb.0:
 ; AVX512F-NEXT:    ## kill: def $ymm1 killed $ymm1 def $zmm1
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vpblendmd (%rdi), %zmm1, %zmm0 {%k1}
@@ -2451,7 +2451,7 @@ define <8 x i32> @load_v8i32_v8i1(<8 x i1> %mask, <8 x i32>* %addr, <8 x i32> %d
 ;
 ; AVX512VLDQ-LABEL: load_v8i32_v8i1:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512VLDQ-NEXT:    vpblendmd (%rdi), %ymm1, %ymm0 {%k1}
@@ -2617,7 +2617,7 @@ define <8 x i32> @load_v8i32_v8i1_zero(<8 x i1> %mask, <8 x i32>* %addr) {
 ;
 ; AVX512F-LABEL: load_v8i32_v8i1_zero:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovdqu32 (%rdi), %zmm0 {%k1} {z}
@@ -2626,7 +2626,7 @@ define <8 x i32> @load_v8i32_v8i1_zero(<8 x i1> %mask, <8 x i32>* %addr) {
 ;
 ; AVX512VLDQ-LABEL: load_v8i32_v8i1_zero:
 ; AVX512VLDQ:       ## %bb.0:
-; AVX512VLDQ-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VLDQ-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512VLDQ-NEXT:    vpmovd2m %ymm0, %k1
 ; AVX512VLDQ-NEXT:    vmovdqu32 (%rdi), %ymm0 {%k1} {z}
diff --git a/test/CodeGen/X86/pmul.ll b/test/CodeGen/X86/pmul.ll
index 1960e790925..2768bf0dc9d 100644
--- a/test/CodeGen/X86/pmul.ll
+++ b/test/CodeGen/X86/pmul.ll
@@ -1345,9 +1345,9 @@ define <8 x i64> @mul_v8i64_sext(<8 x i16> %val1, <8 x i32> %val2) {
 ; AVX2-NEXT:    vpmovsxwq %xmm2, %ymm2
 ; AVX2-NEXT:    vpmovsxwq %xmm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm3
-; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero
+; AVX2-NEXT:    vpmovsxdq %xmm3, %ymm3
 ; AVX2-NEXT:    vpmuldq %ymm3, %ymm2, %ymm2
-; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
+; AVX2-NEXT:    vpmovsxdq %xmm1, %ymm1
 ; AVX2-NEXT:    vpmuldq %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vmovdqa %ymm2, %ymm1
 ; AVX2-NEXT:    retq
diff --git a/test/CodeGen/X86/pr30284.ll b/test/CodeGen/X86/pr30284.ll
index 763bd1ba7b3..be1a8d44184 100644
--- a/test/CodeGen/X86/pr30284.ll
+++ b/test/CodeGen/X86/pr30284.ll
@@ -16,7 +16,7 @@ define void @undef_cond() {
 define void @f_f___un_3C_unf_3E_un_3C_unf_3E_(<16 x i1> %x) {
 ; CHECK-LABEL: f_f___un_3C_unf_3E_un_3C_unf_3E_:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; CHECK-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; CHECK-NEXT:    vpslld $31, %zmm0, %zmm0
 ; CHECK-NEXT:    vpmovd2m %zmm0, %k1
 ; CHECK-NEXT:    vmovapd 0, %zmm0
diff --git a/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll b/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
index 3a5d073759e..7abb3a043a6 100644
--- a/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
+++ b/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
@@ -22,13 +22,13 @@ 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:    vpmovdw %ymm2, %xmm2
 ; AVX256VL-NEXT:    vpblendw {{.*#+}} xmm3 = xmm2[0,1],xmm1[2],xmm2[3],xmm1[4],xmm2[5,6,7]
 ; AVX256VL-NEXT:    vpshufb {{.*#+}} xmm3 = xmm3[6,7,12,13,4,5,8,9,6,7,14,15,14,15,0,1]
-; AVX256VL-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX256VL-NEXT:    vpmovsxwd %xmm3, %ymm3
 ; AVX256VL-NEXT:    vpslld $31, %ymm3, %ymm3
 ; AVX256VL-NEXT:    vptestmd %ymm3, %ymm3, %k1
 ; AVX256VL-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,2,1,3]
 ; AVX256VL-NEXT:    vpshufb {{.*#+}} xmm2 = xmm2[6,7,12,13,2,3,14,15,6,7,6,7,14,15,0,1]
 ; AVX256VL-NEXT:    vpblendw {{.*#+}} xmm1 = xmm2[0,1,2],xmm1[3],xmm2[4],xmm1[5],xmm2[6,7]
-; AVX256VL-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX256VL-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX256VL-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX256VL-NEXT:    vptestmd %ymm1, %ymm1, %k0
 ; AVX256VL-NEXT:    kunpckbw %k1, %k0, %k0
@@ -160,11 +160,11 @@ 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:    vpermq {{.*#+}} ymm2 = ymm2[1,1,2,1]
 ; AVX256VL-NEXT:    vmovdqa {{.*#+}} ymm3 = [255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
 ; AVX256VL-NEXT:    vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
-; AVX256VL-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX256VL-NEXT:    vpmovsxwd %xmm1, %ymm2
 ; AVX256VL-NEXT:    vpslld $31, %ymm2, %ymm2
 ; AVX256VL-NEXT:    vptestmd %ymm2, %ymm2, %k1
 ; AVX256VL-NEXT:    vextracti128 $1, %ymm1, %xmm1
-; AVX256VL-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX256VL-NEXT:    vpmovsxwd %xmm1, %ymm1
 ; AVX256VL-NEXT:    vpslld $31, %ymm1, %ymm1
 ; AVX256VL-NEXT:    vptestmd %ymm1, %ymm1, %k0
 ; AVX256VL-NEXT:    kunpckbw %k1, %k0, %k0
diff --git a/test/CodeGen/X86/setcc-lowering.ll b/test/CodeGen/X86/setcc-lowering.ll
index 1cf20bf62e4..aade54b0b02 100644
--- a/test/CodeGen/X86/setcc-lowering.ll
+++ b/test/CodeGen/X86/setcc-lowering.ll
@@ -65,7 +65,7 @@ define void @pr26232(i64 %a, <16 x i1> %b) {
 ; KNL-32-NEXT:    pushl %esi
 ; KNL-32-NEXT:    .cfi_def_cfa_offset 8
 ; KNL-32-NEXT:    .cfi_offset %esi, -8
-; KNL-32-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; KNL-32-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; KNL-32-NEXT:    vpslld $31, %zmm0, %zmm0
 ; KNL-32-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-32-NEXT:    movl {{[0-9]+}}(%esp), %eax
diff --git a/test/CodeGen/X86/vector-reduce-and-bool.ll b/test/CodeGen/X86/vector-reduce-and-bool.ll
index b7ebe04cefb..b8d57f6954d 100644
--- a/test/CodeGen/X86/vector-reduce-and-bool.ll
+++ b/test/CodeGen/X86/vector-reduce-and-bool.ll
@@ -138,7 +138,7 @@ define i1 @trunc_v8i16_v8i1(<8 x i8>) {
 ;
 ; AVX512F-LABEL: trunc_v8i16_v8i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kmovw %k0, %eax
@@ -397,7 +397,7 @@ define i1 @trunc_v16i16_v16i1(<16 x i16>) {
 ;
 ; AVX512F-LABEL: trunc_v16i16_v16i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kortestw %k0, %k0
@@ -461,7 +461,7 @@ define i1 @trunc_v32i8_v32i1(<32 x i8>) {
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm1
 ; AVX512F-NEXT:    vpand %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -752,7 +752,7 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX512F-LABEL: trunc_v32i16_v32i1:
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vpand %ymm1, %ymm0, %ymm0
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -872,7 +872,7 @@ define i1 @trunc_v64i8_v64i1(<64 x i8>) {
 ; AVX512F-NEXT:    vpand %xmm2, %xmm3, %xmm2
 ; AVX512F-NEXT:    vpand %xmm2, %xmm1, %xmm1
 ; AVX512F-NEXT:    vpand %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
diff --git a/test/CodeGen/X86/vector-reduce-or-bool.ll b/test/CodeGen/X86/vector-reduce-or-bool.ll
index d41403e6794..de5f52ffc92 100644
--- a/test/CodeGen/X86/vector-reduce-or-bool.ll
+++ b/test/CodeGen/X86/vector-reduce-or-bool.ll
@@ -132,7 +132,7 @@ define i1 @trunc_v8i16_v8i1(<8 x i8>) {
 ;
 ; AVX512F-LABEL: trunc_v8i16_v8i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kmovw %k0, %eax
@@ -388,7 +388,7 @@ define i1 @trunc_v16i16_v16i1(<16 x i16>) {
 ;
 ; AVX512F-LABEL: trunc_v16i16_v16i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kortestw %k0, %k0
@@ -452,7 +452,7 @@ define i1 @trunc_v32i8_v32i1(<32 x i8>) {
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm1
 ; AVX512F-NEXT:    vpor %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -743,7 +743,7 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX512F-LABEL: trunc_v32i16_v32i1:
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vpor %ymm1, %ymm0, %ymm0
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -863,7 +863,7 @@ define i1 @trunc_v64i8_v64i1(<64 x i8>) {
 ; AVX512F-NEXT:    vpor %xmm2, %xmm3, %xmm2
 ; AVX512F-NEXT:    vpor %xmm2, %xmm1, %xmm1
 ; AVX512F-NEXT:    vpor %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
diff --git a/test/CodeGen/X86/vector-reduce-xor-bool.ll b/test/CodeGen/X86/vector-reduce-xor-bool.ll
index 29d032b4b44..eb0ba63fc12 100644
--- a/test/CodeGen/X86/vector-reduce-xor-bool.ll
+++ b/test/CodeGen/X86/vector-reduce-xor-bool.ll
@@ -138,7 +138,7 @@ define i1 @trunc_v8i16_v8i1(<8 x i8>) {
 ;
 ; AVX512F-LABEL: trunc_v8i16_v8i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kmovw %k0, %eax
@@ -411,7 +411,7 @@ define i1 @trunc_v16i16_v16i1(<16 x i16>) {
 ;
 ; AVX512F-LABEL: trunc_v16i16_v16i1:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kmovw %k0, %eax
@@ -493,7 +493,7 @@ define i1 @trunc_v32i8_v32i1(<32 x i8>) {
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm1
 ; AVX512F-NEXT:    vpxor %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -840,7 +840,7 @@ define i1 @trunc_v32i16_v32i1(<32 x i16>) {
 ; AVX512F-LABEL: trunc_v32i16_v32i1:
 ; AVX512F:       # %bb.0:
 ; AVX512F-NEXT:    vpxor %ymm1, %ymm0, %ymm0
-; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
@@ -972,7 +972,7 @@ define i1 @trunc_v64i8_v64i1(<64 x i8>) {
 ; AVX512F-NEXT:    vpxor %xmm2, %xmm3, %xmm2
 ; AVX512F-NEXT:    vpxor %xmm2, %xmm1, %xmm1
 ; AVX512F-NEXT:    vpxor %xmm1, %xmm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; AVX512F-NEXT:    kshiftrw $8, %k0, %k1
diff --git a/test/CodeGen/X86/vector-shuffle-512-v16.ll b/test/CodeGen/X86/vector-shuffle-512-v16.ll
index 62c06a8cd3b..eb096ff1de7 100644
--- a/test/CodeGen/X86/vector-shuffle-512-v16.ll
+++ b/test/CodeGen/X86/vector-shuffle-512-v16.ll
@@ -593,7 +593,7 @@ define <16 x i32> @test_vshufi32x4_512(<16 x i32> %x, <16 x i32> %x1) nounwind {
 define <16 x float> @test_vshuff32x4_512_mask(<16 x float> %x, <16 x float> %x1, <16 x float> %y, <16 x i1> %mask) nounwind {
 ; AVX512F-LABEL: test_vshuff32x4_512_mask:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero,xmm3[8],zero,zero,zero,xmm3[9],zero,zero,zero,xmm3[10],zero,zero,zero,xmm3[11],zero,zero,zero,xmm3[12],zero,zero,zero,xmm3[13],zero,zero,zero,xmm3[14],zero,zero,zero,xmm3[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm3, %zmm3
 ; AVX512F-NEXT:    vpslld $31, %zmm3, %zmm3
 ; AVX512F-NEXT:    vpmovd2m %zmm3, %k1
 ; AVX512F-NEXT:    vshuff32x4 {{.*#+}} zmm2 {%k1} = zmm0[0,1,2,3,4,5,6,7],zmm1[4,5,6,7,0,1,2,3]
@@ -615,7 +615,7 @@ define <16 x float> @test_vshuff32x4_512_mask(<16 x float> %x, <16 x float> %x1,
 define <16 x i32> @test_vshufi32x4_512_mask(<16 x i32> %x, <16 x i32> %x1, <16 x i32> %y, <16 x i1> %mask) nounwind {
 ; AVX512F-LABEL: test_vshufi32x4_512_mask:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero,xmm3[8],zero,zero,zero,xmm3[9],zero,zero,zero,xmm3[10],zero,zero,zero,xmm3[11],zero,zero,zero,xmm3[12],zero,zero,zero,xmm3[13],zero,zero,zero,xmm3[14],zero,zero,zero,xmm3[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm3, %zmm3
 ; AVX512F-NEXT:    vpslld $31, %zmm3, %zmm3
 ; AVX512F-NEXT:    vpmovd2m %zmm3, %k1
 ; AVX512F-NEXT:    vshufi32x4 {{.*#+}} zmm2 {%k1} = zmm0[0,1,2,3,4,5,6,7],zmm1[4,5,6,7,0,1,2,3]
diff --git a/test/CodeGen/X86/vector-shuffle-512-v8.ll b/test/CodeGen/X86/vector-shuffle-512-v8.ll
index ff3a3106337..52b04fdc8dd 100644
--- a/test/CodeGen/X86/vector-shuffle-512-v8.ll
+++ b/test/CodeGen/X86/vector-shuffle-512-v8.ll
@@ -1849,7 +1849,7 @@ define <8 x double> @test_vshuff64x2_512(<8 x double> %x, <8 x double> %x1) noun
 define <8 x double> @test_vshuff64x2_512_maskz(<8 x double> %x, <8 x double> %x1, <8 x i1> %mask) nounwind {
 ; ALL-LABEL: test_vshuff64x2_512_maskz:
 ; ALL:       # %bb.0:
-; ALL-NEXT:    vpmovzxwq {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
+; ALL-NEXT:    vpmovsxwq %xmm2, %zmm2
 ; ALL-NEXT:    vpsllq $63, %zmm2, %zmm2
 ; ALL-NEXT:    vptestmq %zmm2, %zmm2, %k1
 ; ALL-NEXT:    vshuff64x2 {{.*#+}} zmm0 {%k1} {z} = zmm0[0,1,4,5],zmm1[2,3,0,1]
@@ -1862,7 +1862,7 @@ define <8 x double> @test_vshuff64x2_512_maskz(<8 x double> %x, <8 x double> %x1
 define <8 x i64> @test_vshufi64x2_512_mask(<8 x i64> %x, <8 x i64> %x1, <8 x i1> %mask) nounwind {
 ; ALL-LABEL: test_vshufi64x2_512_mask:
 ; ALL:       # %bb.0:
-; ALL-NEXT:    vpmovzxwq {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero
+; ALL-NEXT:    vpmovsxwq %xmm2, %zmm2
 ; ALL-NEXT:    vpsllq $63, %zmm2, %zmm2
 ; ALL-NEXT:    vptestmq %zmm2, %zmm2, %k1
 ; ALL-NEXT:    vshufi64x2 {{.*#+}} zmm0 {%k1} = zmm0[0,1,4,5],zmm1[2,3,0,1]
@@ -1891,7 +1891,7 @@ define <8 x double> @test_vshuff64x2_512_mem(<8 x double> %x, <8 x double> *%ptr
 define <8 x double> @test_vshuff64x2_512_mem_mask(<8 x double> %x, <8 x double> *%ptr, <8 x i1> %mask) nounwind {
 ; AVX512F-LABEL: test_vshuff64x2_512_mem_mask:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vshuff64x2 {{.*#+}} zmm0 {%k1} = zmm0[0,1,4,5],mem[2,3,0,1]
@@ -1899,7 +1899,7 @@ define <8 x double> @test_vshuff64x2_512_mem_mask(<8 x double> %x, <8 x double>
 ;
 ; AVX512F-32-LABEL: test_vshuff64x2_512_mem_mask:
 ; AVX512F-32:       # %bb.0:
-; AVX512F-32-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-32-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-32-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-32-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-32-NEXT:    movl {{[0-9]+}}(%esp), %eax
@@ -1914,7 +1914,7 @@ define <8 x double> @test_vshuff64x2_512_mem_mask(<8 x double> %x, <8 x double>
 define <8 x double> @test_vshuff64x2_512_mem_maskz(<8 x double> %x, <8 x double> *%ptr, <8 x i1> %mask) nounwind {
 ; AVX512F-LABEL: test_vshuff64x2_512_mem_maskz:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vshuff64x2 {{.*#+}} zmm0 {%k1} {z} = zmm0[0,1,4,5],mem[2,3,0,1]
@@ -1922,7 +1922,7 @@ define <8 x double> @test_vshuff64x2_512_mem_maskz(<8 x double> %x, <8 x double>
 ;
 ; AVX512F-32-LABEL: test_vshuff64x2_512_mem_maskz:
 ; AVX512F-32:       # %bb.0:
-; AVX512F-32-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
+; AVX512F-32-NEXT:    vpmovsxwq %xmm1, %zmm1
 ; AVX512F-32-NEXT:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-32-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-32-NEXT:    movl {{[0-9]+}}(%esp), %eax
diff --git a/test/CodeGen/X86/vector-shuffle-v1.ll b/test/CodeGen/X86/vector-shuffle-v1.ll
index 3897e55a1ad..d926e5427dd 100644
--- a/test/CodeGen/X86/vector-shuffle-v1.ll
+++ b/test/CodeGen/X86/vector-shuffle-v1.ll
@@ -213,11 +213,11 @@ define <16 x i1> @shuf16i1_3_6_22_12_3_7_7_0_3_6_1_13_3_21_7_0(<16 x i32> %a, <1
 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_3_6_1_13_3_21_7_0(<32 x i1> %a) {
 ; AVX512F-LABEL: 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_3_6_1_13_3_21_7_0:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm1
 ; AVX512F-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512F-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512F-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k2} {z}
@@ -232,11 +232,11 @@ 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
 ;
 ; AVX512VL-LABEL: 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_3_6_1_13_3_21_7_0:
 ; AVX512VL:       # %bb.0:
-; AVX512VL-NEXT:    vpmovzxbd {{.*#+}} zmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512VL-NEXT:    vpmovsxbd %xmm0, %zmm1
 ; AVX512VL-NEXT:    vpslld $31, %zmm1, %zmm1
 ; AVX512VL-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; AVX512VL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512VL-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512VL-NEXT:    vpmovsxbd %xmm0, %zmm0
 ; AVX512VL-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512VL-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512VL-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k2} {z}
@@ -749,7 +749,7 @@ define i8 @shuf8i1__9_6_1_10_3_7_7_1(i8 %a) {
 define i8 @shuf8i1_9_6_1_10_3_7_7_0_all_ones(<8 x i1> %a) {
 ; AVX512F-LABEL: shuf8i1_9_6_1_10_3_7_7_0_all_ones:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
 ; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vpternlogq $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
@@ -764,7 +764,7 @@ define i8 @shuf8i1_9_6_1_10_3_7_7_0_all_ones(<8 x i1> %a) {
 ;
 ; AVX512VL-LABEL: shuf8i1_9_6_1_10_3_7_7_0_all_ones:
 ; AVX512VL:       # %bb.0:
-; AVX512VL-NEXT:    vpmovzxwd {{.*#+}} 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
+; AVX512VL-NEXT:    vpmovsxwd %xmm0, %ymm0
 ; AVX512VL-NEXT:    vpslld $31, %ymm0, %ymm0
 ; AVX512VL-NEXT:    vptestmd %ymm0, %ymm0, %k1
 ; AVX512VL-NEXT:    vpcmpeqd %ymm0, %ymm0, %ymm0