[X86] Turn X86ISD::VSEXT into X86ISD::VZEXT if the upper bits aren't demanded.

This makes X86ISD::VSEXT more similar to ISD::SIGN_EXTEND and ISD::ZERO_EXTEND.

I'm hoping to replace X86ISD::VSEXT/VZEXT with target independent nodes. Making the target specific nodes similar to the target independent nodes helps minimize test diffs in that patch.

llvm-svn: 346539

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 6c844fa6d8f2f0cb0e5abde17e347bed39372977..2fca8e742b3e253b1984920ef1c409876c51bdb4 100644 (file)
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -31941,6 +31941,18 @@ bool X86TargetLowering::SimplifyDemandedBitsForTargetNode(
     }
     break;
   }
+  case X86ISD::VSEXT: {
+    SDValue Src = Op.getOperand(0);
+    unsigned InBits = Src.getScalarValueSizeInBits();
+
+    // If none of the top bits are demanded, convert this into an any_extend.
+    if (OriginalDemandedBits.getActiveBits() <= InBits)
+      return TLO.CombineTo(Op,
+                           TLO.DAG.getNode(X86ISD::VZEXT, SDLoc(Op),
+                                           Op.getValueType(), Src));
+
+    break;
+  }
   }
 
   return TargetLowering::SimplifyDemandedBitsForTargetNode(

diff --git a/llvm/test/CodeGen/X86/avx512-ext.ll b/llvm/test/CodeGen/X86/avx512-ext.ll
index d56cf0fe09ee9c8999a62f68f818c4bffeac8355..2381180af763a234b18745a907758110f6126037 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-ext.ll
+++ b/llvm/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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm1, %zmm1
+; 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:    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:    vpmovsxwd %xmm1, %ymm1
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm1, %zmm1
+; 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:    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:    vpmovsxwd %xmm1, %ymm1
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwd %xmm0, %ymm0
+; 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:    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:    vpmovsxwq %xmm1, %zmm1
+; 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:    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:    vpmovsxwd %xmm1, %ymm1
+; 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:    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
@@ -1646,7 +1646,7 @@ define   <8 x i64> @zext_8i1_to_8xi64(i8 %b) {
 define i16 @trunc_16i8_to_16i1(<16 x i8> %a) {
 ; KNL-LABEL: trunc_16i8_to_16i1:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1664,7 +1664,7 @@ define i16 @trunc_16i8_to_16i1(<16 x i8> %a) {
 ;
 ; AVX512DQNOBW-LABEL: trunc_16i8_to_16i1:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %zmm0, %k0
 ; AVX512DQNOBW-NEXT:    kmovw %k0, %eax
@@ -1726,7 +1726,7 @@ 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:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1744,7 +1744,7 @@ define i8 @trunc_8i16_to_8i1(<8 x i16> %a) {
 ;
 ; AVX512DQNOBW-LABEL: trunc_8i16_to_8i1:
 ; AVX512DQNOBW:       # %bb.0:
-; AVX512DQNOBW-NEXT:    vpmovsxwd %xmm0, %ymm0
+; 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:    vpslld $31, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpmovd2m %ymm0, %k0
 ; AVX512DQNOBW-NEXT:    kmovw %k0, %eax
@@ -2123,10 +2123,10 @@ define <32 x i8> @zext_32xi1_to_32xi8(<32 x i16> %x, <32 x i16> %y) #0 {
 ; KNL-LABEL: zext_32xi1_to_32xi8:
 ; KNL:       # %bb.0:
 ; KNL-NEXT:    vpcmpeqw %ymm2, %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; KNL-NEXT:    vpcmpeqw %ymm3, %ymm1, %ymm1
-; KNL-NEXT:    vpmovsxwd %ymm1, %zmm1
+; KNL-NEXT:    vpmovzxwd {{.*#+}} zmm1 = 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:    vpmovdb %zmm1, %xmm1
 ; KNL-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; KNL-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0
@@ -2141,10 +2141,10 @@ define <32 x i8> @zext_32xi1_to_32xi8(<32 x i16> %x, <32 x i16> %y) #0 {
 ; AVX512DQNOBW-LABEL: zext_32xi1_to_32xi8:
 ; AVX512DQNOBW:       # %bb.0:
 ; AVX512DQNOBW-NEXT:    vpcmpeqw %ymm2, %ymm0, %ymm0
-; AVX512DQNOBW-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512DQNOBW-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
 ; AVX512DQNOBW-NEXT:    vpmovdb %zmm0, %xmm0
 ; AVX512DQNOBW-NEXT:    vpcmpeqw %ymm3, %ymm1, %ymm1
-; AVX512DQNOBW-NEXT:    vpmovsxwd %ymm1, %zmm1
+; AVX512DQNOBW-NEXT:    vpmovzxwd {{.*#+}} zmm1 = 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
 ; AVX512DQNOBW-NEXT:    vpmovdb %zmm1, %xmm1
 ; AVX512DQNOBW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
 ; AVX512DQNOBW-NEXT:    vpand {{.*}}(%rip), %ymm0, %ymm0

diff --git a/llvm/test/CodeGen/X86/avx512-fma.ll b/llvm/test/CodeGen/X86/avx512-fma.ll
index 29ab76d4d37255cac8385e29a25e5696f63ad53a..e9a70f3bfe23c08d552da4fac9417c3503630844 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-fma.ll
+++ b/llvm/test/CodeGen/X86/avx512-fma.ll
@@ -5,7 +5,7 @@
 define <16 x float> @test_x86_fmadd_ps_z(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
 ; ALL-LABEL: test_x86_fmadd_ps_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmadd213ps %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfmadd213ps {{.*#+}} zmm0 = (zmm1 * zmm0) + zmm2
 ; ALL-NEXT:    retq
   %x = fmul <16 x float> %a0, %a1
   %res = fadd <16 x float> %x, %a2
@@ -15,7 +15,7 @@ define <16 x float> @test_x86_fmadd_ps_z(<16 x float> %a0, <16 x float> %a1, <16
 define <16 x float> @test_x86_fmsub_ps_z(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
 ; ALL-LABEL: test_x86_fmsub_ps_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmsub213ps %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfmsub213ps {{.*#+}} zmm0 = (zmm1 * zmm0) - zmm2
 ; ALL-NEXT:    retq
   %x = fmul <16 x float> %a0, %a1
   %res = fsub <16 x float> %x, %a2
@@ -25,7 +25,7 @@ define <16 x float> @test_x86_fmsub_ps_z(<16 x float> %a0, <16 x float> %a1, <16
 define <16 x float> @test_x86_fnmadd_ps_z(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
 ; ALL-LABEL: test_x86_fnmadd_ps_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfnmadd213ps %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfnmadd213ps {{.*#+}} zmm0 = -(zmm1 * zmm0) + zmm2
 ; ALL-NEXT:    retq
   %x = fmul <16 x float> %a0, %a1
   %res = fsub <16 x float> %a2, %x
@@ -35,7 +35,7 @@ define <16 x float> @test_x86_fnmadd_ps_z(<16 x float> %a0, <16 x float> %a1, <1
 define <16 x float> @test_x86_fnmsub_ps_z(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
 ; ALL-LABEL: test_x86_fnmsub_ps_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfnmsub213ps %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfnmsub213ps {{.*#+}} zmm0 = -(zmm1 * zmm0) - zmm2
 ; ALL-NEXT:    retq
   %x = fmul <16 x float> %a0, %a1
   %y = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00,
@@ -49,7 +49,7 @@ define <16 x float> @test_x86_fnmsub_ps_z(<16 x float> %a0, <16 x float> %a1, <1
 define <8 x double> @test_x86_fmadd_pd_z(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
 ; ALL-LABEL: test_x86_fmadd_pd_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmadd213pd %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfmadd213pd {{.*#+}} zmm0 = (zmm1 * zmm0) + zmm2
 ; ALL-NEXT:    retq
   %x = fmul <8 x double> %a0, %a1
   %res = fadd <8 x double> %x, %a2
@@ -59,7 +59,7 @@ define <8 x double> @test_x86_fmadd_pd_z(<8 x double> %a0, <8 x double> %a1, <8
 define <8 x double> @test_x86_fmsub_pd_z(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
 ; ALL-LABEL: test_x86_fmsub_pd_z:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmsub213pd %zmm2, %zmm1, %zmm0
+; ALL-NEXT:    vfmsub213pd {{.*#+}} zmm0 = (zmm1 * zmm0) - zmm2
 ; ALL-NEXT:    retq
   %x = fmul <8 x double> %a0, %a1
   %res = fsub <8 x double> %x, %a2
@@ -69,7 +69,7 @@ define <8 x double> @test_x86_fmsub_pd_z(<8 x double> %a0, <8 x double> %a1, <8
 define double @test_x86_fmsub_213(double %a0, double %a1, double %a2) {
 ; ALL-LABEL: test_x86_fmsub_213:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmsub213sd %xmm2, %xmm1, %xmm0
+; ALL-NEXT:    vfmsub213sd {{.*#+}} xmm0 = (xmm1 * xmm0) - xmm2
 ; ALL-NEXT:    retq
   %x = fmul double %a0, %a1
   %res = fsub double %x, %a2
@@ -79,7 +79,7 @@ define double @test_x86_fmsub_213(double %a0, double %a1, double %a2) {
 define double @test_x86_fmsub_213_m(double %a0, double %a1, double * %a2_ptr) {
 ; ALL-LABEL: test_x86_fmsub_213_m:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmsub213sd (%rdi), %xmm1, %xmm0
+; ALL-NEXT:    vfmsub213sd {{.*#+}} xmm0 = (xmm1 * xmm0) - mem
 ; ALL-NEXT:    retq
   %a2 = load double , double *%a2_ptr
   %x = fmul double %a0, %a1
@@ -90,7 +90,7 @@ define double @test_x86_fmsub_213_m(double %a0, double %a1, double * %a2_ptr) {
 define double @test_x86_fmsub_231_m(double %a0, double %a1, double * %a2_ptr) {
 ; ALL-LABEL: test_x86_fmsub_231_m:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmsub132sd (%rdi), %xmm1, %xmm0
+; ALL-NEXT:    vfmsub132sd {{.*#+}} xmm0 = (xmm0 * mem) - xmm1
 ; ALL-NEXT:    retq
   %a2 = load double , double *%a2_ptr
   %x = fmul double %a0, %a2
@@ -101,7 +101,7 @@ define double @test_x86_fmsub_231_m(double %a0, double %a1, double * %a2_ptr) {
 define <16 x float> @test231_br(<16 x float> %a1, <16 x float> %a2) nounwind {
 ; ALL-LABEL: test231_br:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmadd132ps {{.*}}(%rip){1to16}, %zmm1, %zmm0
+; ALL-NEXT:    vfmadd132ps {{.*#+}} zmm0 = (zmm0 * mem) + zmm1
 ; ALL-NEXT:    retq
   %b1 = fmul <16 x float> %a1, <float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000>
   %b2 = fadd <16 x float> %b1, %a2
@@ -111,7 +111,7 @@ define <16 x float> @test231_br(<16 x float> %a1, <16 x float> %a2) nounwind {
 define <16 x float> @test213_br(<16 x float> %a1, <16 x float> %a2) nounwind {
 ; ALL-LABEL: test213_br:
 ; ALL:       ## %bb.0:
-; ALL-NEXT:    vfmadd213ps {{.*}}(%rip){1to16}, %zmm1, %zmm0
+; ALL-NEXT:    vfmadd213ps {{.*#+}} zmm0 = (zmm1 * zmm0) + mem
 ; ALL-NEXT:    retq
   %b1 = fmul <16 x float> %a1, %a2
   %b2 = fadd <16 x float> %b1, <float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000, float 0x3FB99999A0000000>
@@ -122,17 +122,17 @@ 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:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
-; KNL-NEXT:    vfmadd132ps (%rdi), %zmm1, %zmm0 {%k1}
+; KNL-NEXT:    vfmadd132ps {{.*#+}} zmm0 = (zmm0 * mem) + zmm1
 ; KNL-NEXT:    retq
 ;
 ; SKX-LABEL: test_x86_fmadd132_ps:
 ; SKX:       ## %bb.0:
 ; SKX-NEXT:    vpsllw $7, %xmm2, %xmm2
 ; SKX-NEXT:    vpmovb2m %xmm2, %k1
-; SKX-NEXT:    vfmadd132ps (%rdi), %zmm1, %zmm0 {%k1}
+; SKX-NEXT:    vfmadd132ps {{.*#+}} zmm0 = (zmm0 * mem) + zmm1
 ; SKX-NEXT:    retq
   %a2   = load <16 x float>,<16 x float> *%a2_ptrt,align 1
   %x = fmul <16 x float> %a0, %a2
@@ -145,10 +145,10 @@ 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:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
-; KNL-NEXT:    vfmadd231ps (%rdi), %zmm0, %zmm1 {%k1}
+; KNL-NEXT:    vfmadd231ps {{.*#+}} zmm1 = (zmm0 * mem) + zmm1
 ; KNL-NEXT:    vmovaps %zmm1, %zmm0
 ; KNL-NEXT:    retq
 ;
@@ -156,7 +156,7 @@ define <16 x float> @test_x86_fmadd231_ps(<16 x float> %a0, <16 x float> %a1, <1
 ; SKX:       ## %bb.0:
 ; SKX-NEXT:    vpsllw $7, %xmm2, %xmm2
 ; SKX-NEXT:    vpmovb2m %xmm2, %k1
-; SKX-NEXT:    vfmadd231ps (%rdi), %zmm0, %zmm1 {%k1}
+; SKX-NEXT:    vfmadd231ps {{.*#+}} zmm1 = (zmm0 * mem) + zmm1
 ; SKX-NEXT:    vmovaps %zmm1, %zmm0
 ; SKX-NEXT:    retq
   %a2   = load <16 x float>,<16 x float> *%a2_ptrt,align 1
@@ -170,10 +170,10 @@ 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:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
-; KNL-NEXT:    vfmadd213ps (%rdi), %zmm0, %zmm1 {%k1}
+; KNL-NEXT:    vfmadd213ps {{.*#+}} zmm1 = (zmm0 * zmm1) + mem
 ; KNL-NEXT:    vmovaps %zmm1, %zmm0
 ; KNL-NEXT:    retq
 ;
@@ -181,7 +181,7 @@ define <16 x float> @test_x86_fmadd213_ps(<16 x float> %a0, <16 x float> %a1, <1
 ; SKX:       ## %bb.0:
 ; SKX-NEXT:    vpsllw $7, %xmm2, %xmm2
 ; SKX-NEXT:    vpmovb2m %xmm2, %k1
-; SKX-NEXT:    vfmadd213ps (%rdi), %zmm0, %zmm1 {%k1}
+; SKX-NEXT:    vfmadd213ps {{.*#+}} zmm1 = (zmm0 * zmm1) + mem
 ; SKX-NEXT:    vmovaps %zmm1, %zmm0
 ; SKX-NEXT:    retq
   %a2   = load <16 x float>,<16 x float> *%a2_ptrt,align 1

diff --git a/llvm/test/CodeGen/X86/avx512-insert-extract.ll b/llvm/test/CodeGen/X86/avx512-insert-extract.ll
index 6944d3ea27b64d566a50b1b7bcce6a7219f9c3ef..4a7e301da99a6cc9ee4ae487a02621f2cff9988c 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-insert-extract.ll
+++ b/llvm/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:    vpmovsxbd (%rsp), %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
-; KNL-NEXT:    vpmovsxbd {{[0-9]+}}(%rsp), %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1778,23 +1778,23 @@ define i64 @test_insertelement_variable_v64i1(<64 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    setne (%rsp,%rsi)
 ; KNL-NEXT:    vmovdqa (%rsp), %ymm0
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm1
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm2
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
 ; KNL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm1, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
 ; KNL-NEXT:    vextracti128 $1, %ymm1, %xmm0
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1965,23 +1965,23 @@ define i96 @test_insertelement_variable_v96i1(<96 x i8> %a, i8 %b, i32 %index) {
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm3
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm1
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm0
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm4
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm4 = 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:    vpslld $31, %zmm4, %zmm4
 ; KNL-NEXT:    vptestmd %zmm4, %zmm4, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
 ; KNL-NEXT:    vextracti128 $1, %ymm2, %xmm2
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    shll $16, %ecx
 ; KNL-NEXT:    orl %eax, %ecx
-; KNL-NEXT:    vpmovsxbd %xmm3, %zmm2
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
 ; KNL-NEXT:    vextracti128 $1, %ymm3, %xmm2
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1989,23 +1989,23 @@ 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:    vpmovsxbd %xmm1, %zmm2
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    vextracti128 $1, %ymm1, %xmm1
-; KNL-NEXT:    vpmovsxbd %xmm1, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    kmovw %k0, %esi
 ; KNL-NEXT:    shll $16, %esi
 ; KNL-NEXT:    orl %ecx, %esi
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
@@ -2185,23 +2185,23 @@ define i128 @test_insertelement_variable_v128i1(<128 x i8> %a, i8 %b, i32 %index
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm3
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm1
 ; KNL-NEXT:    vmovdqa {{[0-9]+}}(%rsp), %ymm0
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm4
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm4 = 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:    vpslld $31, %zmm4, %zmm4
 ; KNL-NEXT:    vptestmd %zmm4, %zmm4, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
 ; KNL-NEXT:    vextracti128 $1, %ymm2, %xmm2
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    shll $16, %ecx
 ; KNL-NEXT:    orl %eax, %ecx
-; KNL-NEXT:    vpmovsxbd %xmm3, %zmm2
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %edx
 ; KNL-NEXT:    vextracti128 $1, %ymm3, %xmm2
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2209,23 +2209,23 @@ 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:    vpmovsxbd %xmm1, %zmm2
+; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    vextracti128 $1, %ymm1, %xmm1
-; KNL-NEXT:    vpmovsxbd %xmm1, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    kmovw %k0, %esi
 ; KNL-NEXT:    shll $16, %esi
 ; KNL-NEXT:    orl %ecx, %esi
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    kmovw %k0, %ecx
 ; KNL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %edx

diff --git a/llvm/test/CodeGen/X86/avx512-mask-op.ll b/llvm/test/CodeGen/X86/avx512-mask-op.ll
index 13ffb9f65bf76f9d809d70987b22342bfcb13039..28ce5bb9c43ef4b0ce03db051b74f6fd59397b36 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-mask-op.ll
+++ b/llvm/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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxwq %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm1, %zmm0
+; 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:    jmp LBB18_3
 ; KNL-NEXT:  LBB18_1:
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:  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:    vpmovsxbd %xmm1, %zmm0
+; 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:    jmp LBB18_3
 ; AVX512DQ-NEXT:  LBB18_1:
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:  LBB18_3:
 ; AVX512DQ-NEXT:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
@@ -1671,7 +1671,7 @@ 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:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestnmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -1699,7 +1699,7 @@ define void @store_v8i1(<8 x i1> %c , <8 x i1>* %ptr) {
 ;
 ; AVX512DQ-LABEL: store_v8i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    knotb %k0, %k0
@@ -1723,7 +1723,7 @@ 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:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestnmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, (%rdi)
@@ -1749,7 +1749,7 @@ define void @store_v16i1(<16 x i1> %c , <16 x i1>* %ptr) {
 ;
 ; AVX512DQ-LABEL: store_v16i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
 ; AVX512DQ-NEXT:    knotw %k0, %k0
@@ -2463,7 +2463,7 @@ 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:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2489,7 +2489,7 @@ define void @store_8i1(<8 x i1>* %a, <8 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_8i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovb %k0, (%rdi)
@@ -2510,7 +2510,7 @@ 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:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmq %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, %eax
@@ -2536,7 +2536,7 @@ define void @store_8i1_1(<8 x i1>* %a, <8 x i16> %v) {
 ;
 ; AVX512DQ-LABEL: store_8i1_1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovq2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovb %k0, (%rdi)
@@ -2558,7 +2558,7 @@ 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:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-NEXT:    kmovw %k0, (%rdi)
@@ -2582,7 +2582,7 @@ define void @store_16i1(<16 x i1>* %a, <16 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_16i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
 ; AVX512DQ-NEXT:    kmovw %k0, (%rdi)
@@ -2603,11 +2603,11 @@ 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:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
 ; KNL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    kmovw %k1, 2(%rdi)
@@ -2633,11 +2633,11 @@ define void @store_32i1(<32 x i1>* %a, <32 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_32i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; AVX512DQ-NEXT:    vpmovd2m %zmm1, %k0
 ; AVX512DQ-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQ-NEXT:    kmovw %k1, 2(%rdi)
@@ -2660,10 +2660,10 @@ 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:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
-; KNL-NEXT:    vpmovsxwd %ymm1, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; KNL-NEXT:    kmovw %k1, 2(%rdi)
@@ -2689,10 +2689,10 @@ define void @store_32i1_1(<32 x i1>* %a, <32 x i16> %v) {
 ;
 ; AVX512DQ-LABEL: store_32i1_1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
-; AVX512DQ-NEXT:    vpmovsxwd %ymm1, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
 ; AVX512DQ-NEXT:    kmovw %k1, 2(%rdi)
@@ -2718,16 +2718,16 @@ define void @store_64i1(<64 x i1>* %a, <64 x i1> %v) {
 ;
 ; KNL-LABEL: store_64i1:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k0
-; KNL-NEXT:    vpmovsxbd %xmm1, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k1
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k2
-; KNL-NEXT:    vpmovsxbd %xmm3, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-NEXT:    vptestmd %zmm0, %zmm0, %k3
 ; KNL-NEXT:    kmovw %k3, 6(%rdi)
@@ -2755,16 +2755,16 @@ define void @store_64i1(<64 x i1>* %a, <64 x i1> %v) {
 ;
 ; AVX512DQ-LABEL: store_64i1:
 ; AVX512DQ:       ## %bb.0:
-; AVX512DQ-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k0
-; AVX512DQ-NEXT:    vpmovsxbd %xmm1, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k1
-; AVX512DQ-NEXT:    vpmovsxbd %xmm2, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k2
-; AVX512DQ-NEXT:    vpmovsxbd %xmm3, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512DQ-NEXT:    vpmovd2m %zmm0, %k3
 ; AVX512DQ-NEXT:    kmovw %k3, 6(%rdi)

diff --git a/llvm/test/CodeGen/X86/avx512-masked-memop-64-32.ll b/llvm/test/CodeGen/X86/avx512-masked-memop-64-32.ll
index f199cb097aa472949876bf4c1f43bc3998ef7d59..89dfc69fcd5558b6fce1c7b68b71962835e028c7 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-masked-memop-64-32.ll
+++ b/llvm/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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm0, %zmm0
+; 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:    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:    vpmovsxbd %xmm5, %zmm5
+; 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:    vpslld $31, %zmm5, %zmm5
 ; AVX512F-NEXT:    vptestmd %zmm5, %zmm5, %k1
-; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512F-NEXT:    vblendmpd (%rdi), %zmm1, %zmm0 {%k2}

diff --git a/llvm/test/CodeGen/X86/avx512-select.ll b/llvm/test/CodeGen/X86/avx512-select.ll
index 90e533c09b7f6c31db8f611ac0b7ea696d546a81..e714a728230c74e81d955ab84f5796407e4f6767 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-select.ll
+++ b/llvm/test/CodeGen/X86/avx512-select.ll
@@ -318,9 +318,9 @@ define float @pr30561_f32(float %b, float %a, i1 %c) {
 define <16 x i16> @pr31515(<16 x i1> %a, <16 x i1> %b, <16 x i16> %c) nounwind {
 ; X86-LABEL: pr31515:
 ; X86:       # %bb.0:
-; X86-NEXT:    vpmovsxbd %xmm1, %zmm1
+; X86-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
 ; X86-NEXT:    vpslld $31, %zmm1, %zmm1
-; X86-NEXT:    vpmovsxbd %xmm0, %zmm0
+; X86-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
 ; X86-NEXT:    vpslld $31, %zmm0, %zmm0
 ; X86-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; X86-NEXT:    vptestmd %zmm1, %zmm1, %k1 {%k1}
@@ -331,9 +331,9 @@ define <16 x i16> @pr31515(<16 x i1> %a, <16 x i1> %b, <16 x i16> %c) nounwind {
 ;
 ; X64-LABEL: pr31515:
 ; X64:       # %bb.0:
-; X64-NEXT:    vpmovsxbd %xmm1, %zmm1
+; X64-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
 ; X64-NEXT:    vpslld $31, %zmm1, %zmm1
-; X64-NEXT:    vpmovsxbd %xmm0, %zmm0
+; X64-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
 ; X64-NEXT:    vpslld $31, %zmm0, %zmm0
 ; X64-NEXT:    vptestmd %zmm0, %zmm0, %k1
 ; X64-NEXT:    vptestmd %zmm1, %zmm1, %k1 {%k1}

diff --git a/llvm/test/CodeGen/X86/avx512-trunc.ll b/llvm/test/CodeGen/X86/avx512-trunc.ll
index ca7e22316fb887dba160a74748d58b6ed1460ea9..efcd249d6dd7eb5f4cd7c53191c606ccb30c7540 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-trunc.ll
+++ b/llvm/test/CodeGen/X86/avx512-trunc.ll
@@ -497,7 +497,7 @@ define <16 x i8> @trunc_wb_256(<16 x i16> %i) #0 {
 define void @trunc_wb_256_mem(<16 x i16> %i, <16 x i8>* %res) #0 {
 ; KNL-LABEL: trunc_wb_256_mem:
 ; KNL:       ## %bb.0:
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq
@@ -541,7 +541,7 @@ define void @usat_trunc_wb_256_mem(<16 x i16> %i, <16 x i8>* %res) {
 ; KNL-LABEL: usat_trunc_wb_256_mem:
 ; KNL:       ## %bb.0:
 ; KNL-NEXT:    vpminuw {{.*}}(%rip), %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq
@@ -781,7 +781,7 @@ define void @smax_usat_trunc_wb_256_mem1(<16 x i16> %i, <16 x i8>* %res) {
 ; KNL-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; KNL-NEXT:    vpmaxsw %ymm1, %ymm0, %ymm0
 ; KNL-NEXT:    vpminsw {{.*}}(%rip), %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq
@@ -809,7 +809,7 @@ define void @smax_usat_trunc_wb_256_mem2(<16 x i16> %i, <16 x i8>* %res) {
 ; KNL-NEXT:    vpminsw {{.*}}(%rip), %ymm0, %ymm0
 ; KNL-NEXT:    vpxor %xmm1, %xmm1, %xmm1
 ; KNL-NEXT:    vpmaxsw %ymm1, %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq
@@ -1039,7 +1039,7 @@ define void @negative_test1_smax_usat_trunc_wb_256_mem(<16 x i16> %i, <16 x i8>*
 ; KNL-NEXT:    vpmaxsw %ymm1, %ymm0, %ymm0
 ; KNL-NEXT:    vpcmpeqd %ymm1, %ymm1, %ymm1
 ; KNL-NEXT:    vpminsw %ymm1, %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq
@@ -1067,7 +1067,7 @@ define void @negative_test2_smax_usat_trunc_wb_256_mem(<16 x i16> %i, <16 x i8>*
 ; KNL:       ## %bb.0:
 ; KNL-NEXT:    vpmaxsw {{.*}}(%rip), %ymm0, %ymm0
 ; KNL-NEXT:    vpminsw {{.*}}(%rip), %ymm0, %ymm0
-; KNL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rdi)
 ; KNL-NEXT:    vzeroupper
 ; KNL-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/avx512-vec-cmp.ll b/llvm/test/CodeGen/X86/avx512-vec-cmp.ll
index 42647da95273d87769ba551e608e091f56ec5b89..cae7d418f91f0d0ef82a2be2cb2ab4991973895d 100644 (file)
--- a/llvm/test/CodeGen/X86/avx512-vec-cmp.ll
+++ b/llvm/test/CodeGen/X86/avx512-vec-cmp.ll
@@ -850,7 +850,8 @@ 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:    vpmovsxwq %xmm2, %zmm2 ## encoding: [0x62,0xf2,0x7d,0x48,0x24,0xd2]
+; 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:    vpsllq $63, %zmm2, %zmm2 ## encoding: [0x62,0xf1,0xed,0x48,0x73,0xf2,0x3f]
 ; KNL-NEXT:    vptestmq %zmm2, %zmm2, %k1 ## encoding: [0x62,0xf2,0xed,0x48,0x27,0xca]
 ; KNL-NEXT:    vcmpltpd (%rdi){1to8}, %zmm0, %k1 {%k1} ## encoding: [0x62,0xf1,0xfd,0x59,0xc2,0x0f,0x01]

diff --git a/llvm/test/CodeGen/X86/compress_expand.ll b/llvm/test/CodeGen/X86/compress_expand.ll
index 209991506da4d17a592fd268981b785fe24fdedb..db407a753388326cd01a748de074b2f3f84c1c9f 100644 (file)
--- a/llvm/test/CodeGen/X86/compress_expand.ll
+++ b/llvm/test/CodeGen/X86/compress_expand.ll
@@ -53,7 +53,7 @@ define <8 x double> @test3(double* %base, <8 x double> %src0, <8 x i1> %mask) {
 ;
 ; KNL-LABEL: test3:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vexpandpd (%rdi), %zmm0 {%k1}
@@ -138,7 +138,7 @@ define void @test7(float* %base, <8 x float> %V, <8 x i1> %mask) {
 ; KNL-LABEL: test7:
 ; KNL:       # %bb.0:
 ; KNL-NEXT:    # kill: def $ymm0 killed $ymm0 def $zmm0
-; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
@@ -158,7 +158,7 @@ define void @test8(double* %base, <8 x double> %V, <8 x i1> %mask) {
 ;
 ; KNL-LABEL: test8:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
@@ -178,7 +178,7 @@ define void @test9(i64* %base, <8 x i64> %V, <8 x i1> %mask) {
 ;
 ; KNL-LABEL: test9:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; KNL-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
@@ -384,7 +384,7 @@ define void @test18(double* %base, <16 x double> %V, <16 x i1> %mask) {
 ;
 ; KNL-LABEL: test18:
 ; KNL:       # %bb.0:
-; KNL-NEXT:    vpmovsxbd %xmm2, %zmm2
+; 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:    vpslld $31, %zmm2, %zmm2
 ; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
 ; KNL-NEXT:    kshiftrw $8, %k1, %k2

diff --git a/llvm/test/CodeGen/X86/masked_gather_scatter.ll b/llvm/test/CodeGen/X86/masked_gather_scatter.ll
index 0aef996e54e15b11456b2a88f6782d2afeec2021..82120537f913d530c2f3ba832d63c8a81f467978 100644 (file)
--- a/llvm/test/CodeGen/X86/masked_gather_scatter.ll
+++ b/llvm/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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    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:    vpmovsxbd %xmm2, %zmm2
+; 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:    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:    vpmovsxbd %xmm1, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; SKX_32-NEXT:    vpmovd2m %zmm1, %k1
 ; SKX_32-NEXT:    vmovapd 8(%ebp), %zmm1

diff --git a/llvm/test/CodeGen/X86/masked_memop.ll b/llvm/test/CodeGen/X86/masked_memop.ll
index f1ab2aeefa818f5030a4ef87638793d88f2656b6..c057809863505a5512c157b2d630ab3633ee3cde 100644 (file)
--- a/llvm/test/CodeGen/X86/masked_memop.ll
+++ b/llvm/test/CodeGen/X86/masked_memop.ll
@@ -343,7 +343,7 @@ define <8 x i32> @test11b(<8 x i1> %mask, <8 x i32>* %addr, <8 x i32> %dst) {
 ; AVX512F-LABEL: test11b:
 ; AVX512F:       ## %bb.0:
 ; AVX512F-NEXT:    ## kill: def $ymm1 killed $ymm1 def $zmm1
-; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vpblendmd (%rdi), %zmm1, %zmm0 {%k1}
@@ -383,7 +383,7 @@ define <8 x float> @test11c(<8 x i1> %mask, <8 x float>* %addr) {
 ;
 ; AVX512F-LABEL: test11c:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovups (%rdi), %zmm0 {%k1} {z}
@@ -423,7 +423,7 @@ define <8 x i32> @test11d(<8 x i1> %mask, <8 x i32>* %addr) {
 ;
 ; AVX512F-LABEL: test11d:
 ; AVX512F:       ## %bb.0:
-; AVX512F-NEXT:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vmovdqu32 (%rdi), %zmm0 {%k1} {z}

diff --git a/llvm/test/CodeGen/X86/pmul.ll b/llvm/test/CodeGen/X86/pmul.ll
index 36f06a64a65400f0e4df4697c5f1fb232f4c1240..2417e20e7654103bdbbf07a98f307bec2abfa4e5 100644 (file)
--- a/llvm/test/CodeGen/X86/pmul.ll
+++ b/llvm/test/CodeGen/X86/pmul.ll
@@ -1418,9 +1418,9 @@ 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:    vpmovsxdq %xmm3, %ymm3
+; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero
 ; AVX2-NEXT:    vpmuldq %ymm3, %ymm2, %ymm2
-; AVX2-NEXT:    vpmovsxdq %xmm1, %ymm1
+; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
 ; AVX2-NEXT:    vpmuldq %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vmovdqa %ymm2, %ymm1
 ; AVX2-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll b/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
index 8bd0f1c98ae96074621a7915368113d28965edc5..120e4482c3e660e420c292cd7ed51da02242333b 100644 (file)
--- a/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
+++ b/llvm/test/CodeGen/X86/prefer-avx256-mask-shuffle.ll
@@ -21,13 +21,13 @@ 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:    vpmovsxwd %xmm3, %ymm3
+; 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:    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:    vpmovsxwd %xmm1, %ymm1
+; 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:    vpslld $31, %ymm1, %ymm1
 ; AVX256VL-NEXT:    vptestmd %ymm1, %ymm1, %k0
 ; AVX256VL-NEXT:    kunpckbw %k1, %k0, %k0
@@ -158,11 +158,11 @@ 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:    vpmovsxwd %xmm1, %ymm2
+; 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:    vpslld $31, %ymm2, %ymm2
 ; AVX256VL-NEXT:    vptestmd %ymm2, %ymm2, %k1
 ; AVX256VL-NEXT:    vextracti128 $1, %ymm1, %xmm1
-; AVX256VL-NEXT:    vpmovsxwd %xmm1, %ymm1
+; 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:    vpslld $31, %ymm1, %ymm1
 ; AVX256VL-NEXT:    vptestmd %ymm1, %ymm1, %k0
 ; AVX256VL-NEXT:    kunpckbw %k1, %k0, %k0

diff --git a/llvm/test/CodeGen/X86/setcc-lowering.ll b/llvm/test/CodeGen/X86/setcc-lowering.ll
index 100461d22c995ffccb3229071ae7a8c888b1a9b8..9bcb4fd1e63ea4289b30df76c470bb0d3c16e8be 100644 (file)
--- a/llvm/test/CodeGen/X86/setcc-lowering.ll
+++ b/llvm/test/CodeGen/X86/setcc-lowering.ll
@@ -66,7 +66,7 @@ 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:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; KNL-32-NEXT:    vptestmd %zmm0, %zmm0, %k0
 ; KNL-32-NEXT:    movl {{[0-9]+}}(%esp), %eax

diff --git a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
index 152742c80163e1c914cc079e54c5a9f28aee6cd0..f21a2b71cf792c6303103a73f00f24a4eab7eeb0 100644 (file)
--- a/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
+++ b/llvm/test/CodeGen/X86/shuffle-vs-trunc-256.ll
@@ -81,14 +81,14 @@ define void @trunc_v16i16_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
 ;
 ; AVX512F-LABEL: trunc_v16i16_to_v16i8:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovsxwd (%rdi), %zmm0
+; AVX512F-NEXT:    vpmovzxwd {{.*#+}} zmm0 = 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
 ; AVX512F-NEXT:    vpmovdb %zmm0, (%rsi)
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
 ;
 ; AVX512VL-LABEL: trunc_v16i16_to_v16i8:
 ; AVX512VL:       # %bb.0:
-; AVX512VL-NEXT:    vpmovsxwd (%rdi), %zmm0
+; AVX512VL-NEXT:    vpmovzxwd {{.*#+}} zmm0 = 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
 ; AVX512VL-NEXT:    vpmovdb %zmm0, (%rsi)
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
index e35a0e99efe4a1d8fa676a3c96cb2b4131e519e9..b7ea04d23df96502840550457f88b475cc3a9060 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
@@ -582,7 +582,7 @@ 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:    vpmovsxbd %xmm3, %zmm3
+; 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:    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]
@@ -604,7 +604,7 @@ 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:    vpmovsxbd %xmm3, %zmm3
+; 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:    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/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
index 97b89a0be37ba9abdaab680fe288493cad321f41..dce7f02605207a63f54fb2d48268802ce50ac3c1 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
@@ -1845,7 +1845,7 @@ 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:    vpmovsxwq %xmm2, %zmm2
+; 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:    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]
@@ -1858,7 +1858,7 @@ 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:    vpmovsxwq %xmm2, %zmm2
+; 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:    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]
@@ -1887,7 +1887,7 @@ 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:    vpmovsxwq %xmm1, %zmm1
+; 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:    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]
@@ -1895,7 +1895,7 @@ 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:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-32-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-32-NEXT:    movl {{[0-9]+}}(%esp), %eax
@@ -1910,7 +1910,7 @@ 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:    vpmovsxwq %xmm1, %zmm1
+; 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:    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]
@@ -1918,7 +1918,7 @@ 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:    vpmovsxwq %xmm1, %zmm1
+; 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:    vpsllq $63, %zmm1, %zmm1
 ; AVX512F-32-NEXT:    vptestmq %zmm1, %zmm1, %k1
 ; AVX512F-32-NEXT:    movl {{[0-9]+}}(%esp), %eax

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-v1.ll b/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
index 19c3deb31e78d7c840b49976a74e7eba22687a09..621058c29e6f7432c2e1cf1820917d04fbcc0e8f 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
@@ -214,11 +214,11 @@ 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:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; AVX512F-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; AVX512F-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512F-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512F-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k2} {z}
@@ -233,11 +233,11 @@ 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:    vpmovsxbd %xmm0, %zmm1
+; 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:    vpslld $31, %zmm1, %zmm1
 ; AVX512VL-NEXT:    vptestmd %zmm1, %zmm1, %k1
 ; AVX512VL-NEXT:    vextracti128 $1, %ymm0, %xmm0
-; AVX512VL-NEXT:    vpmovsxbd %xmm0, %zmm0
+; 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:    vpslld $31, %zmm0, %zmm0
 ; AVX512VL-NEXT:    vptestmd %zmm0, %zmm0, %k2
 ; AVX512VL-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k2} {z}
@@ -745,7 +745,7 @@ 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:    vpmovsxwq %xmm0, %zmm0
+; 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:    vpsllq $63, %zmm0, %zmm0
 ; AVX512F-NEXT:    vptestmq %zmm0, %zmm0, %k1
 ; AVX512F-NEXT:    vpternlogq $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
@@ -760,7 +760,7 @@ 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:    vpmovsxwd %xmm0, %ymm0
+; 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:    vpslld $31, %ymm0, %ymm0
 ; AVX512VL-NEXT:    vptestmd %ymm0, %ymm0, %k1
 ; AVX512VL-NEXT:    vpcmpeqd %ymm0, %ymm0, %ymm0

diff --git a/llvm/test/CodeGen/X86/vector-trunc-math.ll b/llvm/test/CodeGen/X86/vector-trunc-math.ll
index 7dc850391bca4f95b704fc866b26d4883a83d4a0..e3a742bccead788713b567a9bf25acd7a0ce6054 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-trunc-math.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc-math.ll
@@ -856,7 +856,7 @@ define <16 x i8> @trunc_add_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512F-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512F-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512F-NEXT:    vzeroupper
@@ -872,7 +872,7 @@ define <16 x i8> @trunc_add_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512DQ-LABEL: trunc_add_const_v16i16_v16i8:
 ; AVX512DQ:       # %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512DQ-NEXT:    vpaddb {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512DQ-NEXT:    vzeroupper
@@ -3933,7 +3933,7 @@ define <16 x i8> @trunc_and_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512F-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512F-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512F-NEXT:    vzeroupper
@@ -3949,7 +3949,7 @@ define <16 x i8> @trunc_and_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512DQ-LABEL: trunc_and_const_v16i16_v16i8:
 ; AVX512DQ:       # %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512DQ-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512DQ-NEXT:    vzeroupper
@@ -4737,7 +4737,7 @@ define <16 x i8> @trunc_xor_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512F-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512F-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512F-NEXT:    vzeroupper
@@ -4753,7 +4753,7 @@ define <16 x i8> @trunc_xor_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512DQ-LABEL: trunc_xor_const_v16i16_v16i8:
 ; AVX512DQ:       # %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512DQ-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512DQ-NEXT:    vzeroupper
@@ -5541,7 +5541,7 @@ define <16 x i8> @trunc_or_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512F-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX512F:       # %bb.0:
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512F-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512F-NEXT:    vzeroupper
@@ -5557,7 +5557,7 @@ define <16 x i8> @trunc_or_const_v16i16_v16i8(<16 x i16> %a0) nounwind {
 ;
 ; AVX512DQ-LABEL: trunc_or_const_v16i16_v16i8:
 ; AVX512DQ:       # %bb.0:
-; AVX512DQ-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, %xmm0
 ; AVX512DQ-NEXT:    vpor {{.*}}(%rip), %xmm0, %xmm0
 ; AVX512DQ-NEXT:    vzeroupper

diff --git a/llvm/test/CodeGen/X86/vector-trunc.ll b/llvm/test/CodeGen/X86/vector-trunc.ll
index 79cbb8cc924fdefba8bc92ea82de8524af58e439..a62cb30638fdd1e8cbcb769d79f7194aa4af845b 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-trunc.ll
+++ b/llvm/test/CodeGen/X86/vector-trunc.ll
@@ -1130,14 +1130,14 @@ define void @trunc16i16_16i8(<16 x i16> %a) {
 ;
 ; AVX512F-LABEL: trunc16i16_16i8:
 ; AVX512F:       # %bb.0: # %entry
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rax)
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
 ;
 ; AVX512VL-LABEL: trunc16i16_16i8:
 ; AVX512VL:       # %bb.0: # %entry
-; AVX512VL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512VL-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
 ; AVX512VL-NEXT:    vpmovdb %zmm0, (%rax)
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
@@ -1192,7 +1192,7 @@ define void @trunc16i16_16i8_ashr(<16 x i16> %a) {
 ; AVX512F-LABEL: trunc16i16_16i8_ashr:
 ; AVX512F:       # %bb.0: # %entry
 ; AVX512F-NEXT:    vpsraw $8, %ymm0, %ymm0
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rax)
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
@@ -1200,7 +1200,7 @@ define void @trunc16i16_16i8_ashr(<16 x i16> %a) {
 ; AVX512VL-LABEL: trunc16i16_16i8_ashr:
 ; AVX512VL:       # %bb.0: # %entry
 ; AVX512VL-NEXT:    vpsraw $8, %ymm0, %ymm0
-; AVX512VL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512VL-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
 ; AVX512VL-NEXT:    vpmovdb %zmm0, (%rax)
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
@@ -1257,7 +1257,7 @@ define void @trunc16i16_16i8_lshr(<16 x i16> %a) {
 ; AVX512F-LABEL: trunc16i16_16i8_lshr:
 ; AVX512F:       # %bb.0: # %entry
 ; AVX512F-NEXT:    vpsrlw $8, %ymm0, %ymm0
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
+; 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:    vpmovdb %zmm0, (%rax)
 ; AVX512F-NEXT:    vzeroupper
 ; AVX512F-NEXT:    retq
@@ -1265,7 +1265,7 @@ define void @trunc16i16_16i8_lshr(<16 x i16> %a) {
 ; AVX512VL-LABEL: trunc16i16_16i8_lshr:
 ; AVX512VL:       # %bb.0: # %entry
 ; AVX512VL-NEXT:    vpsrlw $8, %ymm0, %ymm0
-; AVX512VL-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512VL-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
 ; AVX512VL-NEXT:    vpmovdb %zmm0, (%rax)
 ; AVX512VL-NEXT:    vzeroupper
 ; AVX512VL-NEXT:    retq
@@ -2024,8 +2024,8 @@ define void @PR34773(i16* %a0, i8* %a1) {
 ; AVX512F-NEXT:    vmovdqu 32(%rdi), %ymm1
 ; AVX512F-NEXT:    vpsrlw $8, %ymm0, %ymm0
 ; AVX512F-NEXT:    vpsrlw $8, %ymm1, %ymm1
-; AVX512F-NEXT:    vpmovsxwd %ymm0, %zmm0
-; AVX512F-NEXT:    vpmovsxwd %ymm1, %zmm1
+; 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:    vpmovzxwd {{.*#+}} zmm1 = 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
 ; AVX512F-NEXT:    vpmovdb %zmm0, (%rsi)
 ; AVX512F-NEXT:    vpmovdb %zmm1, 16(%rsi)
 ; AVX512F-NEXT:    vzeroupper
@@ -2037,8 +2037,8 @@ define void @PR34773(i16* %a0, i8* %a1) {
 ; AVX512VL-NEXT:    vmovdqu 32(%rdi), %ymm1
 ; AVX512VL-NEXT:    vpsrlw $8, %ymm0, %ymm0
 ; AVX512VL-NEXT:    vpsrlw $8, %ymm1, %ymm1
-; AVX512VL-NEXT:    vpmovsxwd %ymm0, %zmm0
-; AVX512VL-NEXT:    vpmovsxwd %ymm1, %zmm1
+; AVX512VL-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
+; AVX512VL-NEXT:    vpmovzxwd {{.*#+}} zmm1 = 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
 ; AVX512VL-NEXT:    vpmovdb %zmm0, (%rsi)
 ; AVX512VL-NEXT:    vpmovdb %zmm1, 16(%rsi)
 ; AVX512VL-NEXT:    vzeroupper

diff --git a/llvm/test/CodeGen/X86/vselect-packss.ll b/llvm/test/CodeGen/X86/vselect-packss.ll
index 8d517151181eb1eb66a4c0572ae8a574381d121a..34b336c8820f45a9267932a3cd47565d6127f574 100644 (file)
--- a/llvm/test/CodeGen/X86/vselect-packss.ll
+++ b/llvm/test/CodeGen/X86/vselect-packss.ll
@@ -55,7 +55,7 @@ define <16 x i8> @vselect_packss_v16i16(<16 x i16> %a0, <16 x i16> %a1, <16 x i8
 ; AVX512NOBW-LABEL: vselect_packss_v16i16:
 ; AVX512NOBW:       # %bb.0:
 ; AVX512NOBW-NEXT:    vpcmpeqw %ymm1, %ymm0, %ymm0
-; AVX512NOBW-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512NOBW-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
 ; AVX512NOBW-NEXT:    vpmovdb %zmm0, %xmm0
 ; AVX512NOBW-NEXT:    vpblendvb %xmm0, %xmm2, %xmm3, %xmm0
 ; AVX512NOBW-NEXT:    vzeroupper