Index: lib/Target/X86/InstPrinter/X86InstComments.cpp
===================================================================
--- lib/Target/X86/InstPrinter/X86InstComments.cpp
+++ lib/Target/X86/InstPrinter/X86InstComments.cpp
@@ -46,6 +46,16 @@
   CASE_AVX_INS_COMMON(Inst, Y, r##src)            \
   CASE_SSE_INS_COMMON(Inst, r##src)
 
+#define CASE_MASK_MOVDUP(Inst, src)               \
+  CASE_MASK_INS_COMMON(Inst, Z, r##src)           \
+  CASE_MASK_INS_COMMON(Inst, Z256, r##src)        \
+  CASE_MASK_INS_COMMON(Inst, Z128, r##src)
+
+#define CASE_MASKZ_MOVDUP(Inst, src)              \
+  CASE_MASKZ_INS_COMMON(Inst, Z, r##src)          \
+  CASE_MASKZ_INS_COMMON(Inst, Z256, r##src)       \
+  CASE_MASKZ_INS_COMMON(Inst, Z128, r##src)
+
 #define CASE_PMOVZX(Inst, src)                    \
   CASE_AVX512_INS_COMMON(Inst, Z, r##src)         \
   CASE_AVX512_INS_COMMON(Inst, Z256, r##src)      \
@@ -129,6 +139,48 @@
   }
 }
 
+/// Wraps the destination register name with AVX512 mask/maskz filtering.
+static std::string getMaskName(const MCInst *MI, const char *DestName,
+                               const char *(*getRegName)(unsigned)) {
+  std::string OpMaskName(DestName);
+
+  bool MaskWithZero = false;
+  const char *MaskRegName = nullptr;
+
+  switch (MI->getOpcode()) {
+  default:
+    return OpMaskName;
+  CASE_MASKZ_MOVDUP(MOVDDUP, m)
+  CASE_MASKZ_MOVDUP(MOVDDUP, r)
+  CASE_MASKZ_MOVDUP(MOVSHDUP, m)
+  CASE_MASKZ_MOVDUP(MOVSHDUP, r)
+  CASE_MASKZ_MOVDUP(MOVSLDUP, m)
+  CASE_MASKZ_MOVDUP(MOVSLDUP, r)
+    MaskWithZero = true;
+    MaskRegName = getRegName(MI->getOperand(1).getReg());
+    break;
+  CASE_MASK_MOVDUP(MOVDDUP, m)
+  CASE_MASK_MOVDUP(MOVDDUP, r)
+  CASE_MASK_MOVDUP(MOVSHDUP, m)
+  CASE_MASK_MOVDUP(MOVSHDUP, r)
+  CASE_MASK_MOVDUP(MOVSLDUP, m)
+  CASE_MASK_MOVDUP(MOVSLDUP, r)
+    MaskRegName = getRegName(MI->getOperand(2).getReg());
+    break;
+  }
+
+  // MASK: zmmX {%kY}
+  OpMaskName += " {%";
+  OpMaskName += MaskRegName;
+  OpMaskName += "}";
+
+  // MASKZ: zmmX {%kY} {z}
+  if (MaskWithZero)
+    OpMaskName += " {z}";
+
+  return OpMaskName;
+}
+
 //===----------------------------------------------------------------------===//
 // Top Level Entrypoint
 //===----------------------------------------------------------------------===//
@@ -753,9 +805,8 @@
   if (ShuffleMask.empty())
     return false;
 
-  // TODO: Add support for specifying an AVX512 style mask register in the comment.
   if (!DestName) DestName = Src1Name;
-  OS << (DestName ? DestName : "mem") << " = ";
+  OS << (DestName ? getMaskName(MI, DestName, getRegName) : "mem") << " = ";
 
   // If the two sources are the same, canonicalize the input elements to be
   // from the first src so that we get larger element spans.
Index: test/CodeGen/X86/avx512-intrinsics-fast-isel.ll
===================================================================
--- test/CodeGen/X86/avx512-intrinsics-fast-isel.ll
+++ test/CodeGen/X86/avx512-intrinsics-fast-isel.ll
@@ -23,13 +23,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} zmm0 = zmm1[0,0,2,2,4,4,6,6]
+; X32-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} = zmm1[0,0,2,2,4,4,6,6]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_mask_movddup_pd:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovddup {{.*#+}} zmm0 = zmm1[0,0,2,2,4,4,6,6]
+; X64-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} = zmm1[0,0,2,2,4,4,6,6]
 ; X64-NEXT:    retq
   %arg1 = bitcast i8 %a1 to <8 x i1>
   %res0 = shufflevector <8 x double> %a2, <8 x double> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
@@ -42,13 +42,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6]
+; X32-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_maskz_movddup_pd:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovddup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6]
+; X64-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6]
 ; X64-NEXT:    retq
   %arg0 = bitcast i8 %a0 to <8 x i1>
   %res0 = shufflevector <8 x double> %a1, <8 x double> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
@@ -75,13 +75,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movw {{[0-9]+}}(%esp), %ax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm1[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; X32-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} = zmm1[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_mask_movehdup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm1[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; X64-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} = zmm1[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; X64-NEXT:    retq
   %arg1 = bitcast i16 %a1 to <16 x i1>
   %res0 = shufflevector <16 x float> %a2, <16 x float> undef, <16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>
@@ -94,13 +94,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movw {{[0-9]+}}(%esp), %ax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; X32-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} {z} = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_maskz_movehdup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; X64-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} {z} = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; X64-NEXT:    retq
   %arg0 = bitcast i16 %a0 to <16 x i1>
   %res0 = shufflevector <16 x float> %a1, <16 x float> undef, <16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>
@@ -127,13 +127,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movw {{[0-9]+}}(%esp), %ax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm1[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; X32-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} = zmm1[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_mask_moveldup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm1[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; X64-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} = zmm1[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; X64-NEXT:    retq
   %arg1 = bitcast i16 %a1 to <16 x i1>
   %res0 = shufflevector <16 x float> %a2, <16 x float> undef, <16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>
@@ -146,13 +146,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movw {{[0-9]+}}(%esp), %ax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; X32-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm512_maskz_moveldup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; X64-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; X64-NEXT:    retq
   %arg0 = bitcast i16 %a0 to <16 x i1>
   %res0 = shufflevector <16 x float> %a1, <16 x float> undef, <16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>
Index: test/CodeGen/X86/avx512-intrinsics-upgrade.ll
===================================================================
--- test/CodeGen/X86/avx512-intrinsics-upgrade.ll
+++ test/CodeGen/X86/avx512-intrinsics-upgrade.ll
@@ -8,8 +8,8 @@
 ; CHECK:       ## BB#0:
 ; CHECK-NEXT:    vmovsldup {{.*#+}} zmm2 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovsldup {{.*#+}} zmm1 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
-; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vmovsldup {{.*#+}} zmm1 {%k1} = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; CHECK-NEXT:    vaddps %zmm2, %zmm1, %zmm1
 ; CHECK-NEXT:    vaddps %zmm1, %zmm0, %zmm0
 ; CHECK-NEXT:    retq
@@ -28,8 +28,8 @@
 ; CHECK:       ## BB#0:
 ; CHECK-NEXT:    vmovshdup {{.*#+}} zmm2 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovshdup {{.*#+}} zmm1 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
-; CHECK-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; CHECK-NEXT:    vmovshdup {{.*#+}} zmm1 {%k1} = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; CHECK-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} {z} = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; CHECK-NEXT:    vaddps %zmm2, %zmm1, %zmm1
 ; CHECK-NEXT:    vaddps %zmm1, %zmm0, %zmm0
 ; CHECK-NEXT:    retq
@@ -48,8 +48,8 @@
 ; CHECK:       ## BB#0:
 ; CHECK-NEXT:    vmovddup {{.*#+}} zmm2 = zmm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovddup {{.*#+}} zmm1 = zmm0[0,0,2,2,4,4,6,6]
-; CHECK-NEXT:    vmovddup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    vmovddup {{.*#+}} zmm1 {%k1} = zmm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    vaddpd %zmm2, %zmm1, %zmm1
 ; CHECK-NEXT:    vaddpd %zmm1, %zmm0, %zmm0
 ; CHECK-NEXT:    retq
Index: test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll
===================================================================
--- test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll
+++ test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll
@@ -29,7 +29,7 @@
 ; X32-NEXT:    movb %al, {{[0-9]+}}(%esp)
 ; X32-NEXT:    movzbl {{[0-9]+}}(%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} xmm0 = xmm1[0,0]
+; X32-NEXT:    vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -39,7 +39,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovddup {{.*#+}} xmm0 = xmm1[0,0]
+; X64-NEXT:    vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a1 to i2
   %arg1 = bitcast i2 %trn1 to <2 x i1>
@@ -59,7 +59,7 @@
 ; X32-NEXT:    movb %al, {{[0-9]+}}(%esp)
 ; X32-NEXT:    movzbl {{[0-9]+}}(%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; X32-NEXT:    vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -69,7 +69,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; X64-NEXT:    vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a0 to i2
   %arg0 = bitcast i2 %trn1 to <2 x i1>
@@ -103,7 +103,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} ymm0 = ymm1[0,0,2,2]
+; X32-NEXT:    vmovddup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -113,7 +113,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovddup {{.*#+}} ymm0 = ymm1[0,0,2,2]
+; X64-NEXT:    vmovddup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a1 to i4
   %arg1 = bitcast i4 %trn1 to <4 x i1>
@@ -133,7 +133,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovddup {{.*#+}} ymm0 = ymm0[0,0,2,2]
+; X32-NEXT:    vmovddup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -143,7 +143,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovddup {{.*#+}} ymm0 = ymm0[0,0,2,2]
+; X64-NEXT:    vmovddup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a0 to i4
   %arg0 = bitcast i4 %trn1 to <4 x i1>
@@ -177,7 +177,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
+; X32-NEXT:    vmovshdup {{.*#+}} xmm0 {%k1} = xmm1[1,1,3,3]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -187,7 +187,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
+; X64-NEXT:    vmovshdup {{.*#+}} xmm0 {%k1} = xmm1[1,1,3,3]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a1 to i4
   %arg1 = bitcast i4 %trn1 to <4 x i1>
@@ -207,7 +207,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
+; X32-NEXT:    vmovshdup {{.*#+}} xmm0 {%k1} {z} = xmm0[1,1,3,3]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -217,7 +217,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
+; X64-NEXT:    vmovshdup {{.*#+}} xmm0 {%k1} {z} = xmm0[1,1,3,3]
 ; X64-NEXT:    retq
   %trn0 = trunc i8 %a0 to i4
   %arg0 = bitcast i4 %trn0 to <4 x i1>
@@ -245,13 +245,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} ymm0 = ymm1[1,1,3,3,5,5,7,7]
+; X32-NEXT:    vmovshdup {{.*#+}} ymm0 {%k1} = ymm1[1,1,3,3,5,5,7,7]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm256_mask_movehdup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} ymm0 = ymm1[1,1,3,3,5,5,7,7]
+; X64-NEXT:    vmovshdup {{.*#+}} ymm0 {%k1} = ymm1[1,1,3,3,5,5,7,7]
 ; X64-NEXT:    retq
   %arg1 = bitcast i8 %a1 to <8 x i1>
   %res0 = shufflevector <8 x float> %a2, <8 x float> undef, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7>
@@ -264,13 +264,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovshdup {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
+; X32-NEXT:    vmovshdup {{.*#+}} ymm0 {%k1} {z} = ymm0[1,1,3,3,5,5,7,7]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm256_maskz_movehdup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovshdup {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
+; X64-NEXT:    vmovshdup {{.*#+}} ymm0 {%k1} {z} = ymm0[1,1,3,3,5,5,7,7]
 ; X64-NEXT:    retq
   %arg0 = bitcast i8 %a0 to <8 x i1>
   %res0 = shufflevector <8 x float> %a1, <8 x float> undef, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7>
@@ -303,7 +303,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} xmm0 = xmm1[0,0,2,2]
+; X32-NEXT:    vmovsldup {{.*#+}} xmm0 {%k1} = xmm1[0,0,2,2]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -313,7 +313,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} xmm0 = xmm1[0,0,2,2]
+; X64-NEXT:    vmovsldup {{.*#+}} xmm0 {%k1} = xmm1[0,0,2,2]
 ; X64-NEXT:    retq
   %trn1 = trunc i8 %a1 to i4
   %arg1 = bitcast i4 %trn1 to <4 x i1>
@@ -333,7 +333,7 @@
 ; X32-NEXT:    movb %al, (%esp)
 ; X32-NEXT:    movzbl (%esp), %eax
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} xmm0 = xmm0[0,0,2,2]
+; X32-NEXT:    vmovsldup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0,2,2]
 ; X32-NEXT:    popl %eax
 ; X32-NEXT:    retl
 ;
@@ -343,7 +343,7 @@
 ; X64-NEXT:    movb %dil, -{{[0-9]+}}(%rsp)
 ; X64-NEXT:    movzbl -{{[0-9]+}}(%rsp), %eax
 ; X64-NEXT:    kmovw %eax, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} xmm0 = xmm0[0,0,2,2]
+; X64-NEXT:    vmovsldup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0,2,2]
 ; X64-NEXT:    retq
   %trn0 = trunc i8 %a0 to i4
   %arg0 = bitcast i4 %trn0 to <4 x i1>
@@ -371,13 +371,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} ymm0 = ymm1[0,0,2,2,4,4,6,6]
+; X32-NEXT:    vmovsldup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2,4,4,6,6]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm256_mask_moveldup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} ymm0 = ymm1[0,0,2,2,4,4,6,6]
+; X64-NEXT:    vmovsldup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2,4,4,6,6]
 ; X64-NEXT:    retq
   %arg1 = bitcast i8 %a1 to <8 x i1>
   %res0 = shufflevector <8 x float> %a2, <8 x float> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
@@ -390,13 +390,13 @@
 ; X32:       # BB#0:
 ; X32-NEXT:    movb {{[0-9]+}}(%esp), %al
 ; X32-NEXT:    kmovw %eax, %k1
-; X32-NEXT:    vmovsldup {{.*#+}} ymm0 = ymm0[0,0,2,2,4,4,6,6]
+; X32-NEXT:    vmovsldup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2,4,4,6,6]
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: test_mm256_maskz_moveldup_ps:
 ; X64:       # BB#0:
 ; X64-NEXT:    kmovw %edi, %k1
-; X64-NEXT:    vmovsldup {{.*#+}} ymm0 = ymm0[0,0,2,2,4,4,6,6]
+; X64-NEXT:    vmovsldup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2,4,4,6,6]
 ; X64-NEXT:    retq
   %arg0 = bitcast i8 %a0 to <8 x i1>
   %res0 = shufflevector <8 x float> %a1, <8 x float> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
Index: test/CodeGen/X86/avx512vl-intrinsics-upgrade.ll
===================================================================
--- test/CodeGen/X86/avx512vl-intrinsics-upgrade.ll
+++ test/CodeGen/X86/avx512vl-intrinsics-upgrade.ll
@@ -10,9 +10,9 @@
 ; CHECK-NEXT:    ## xmm2 = xmm0[0,0,2,2]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovsldup %xmm0, %xmm1 {%k1} ## encoding: [0x62,0xf1,0x7e,0x09,0x12,0xc8]
-; CHECK-NEXT:    ## xmm1 = xmm0[0,0,2,2]
+; CHECK-NEXT:    ## xmm1 {%k1} = xmm0[0,0,2,2]
 ; CHECK-NEXT:    vmovsldup %xmm0, %xmm0 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0x89,0x12,0xc0]
-; CHECK-NEXT:    ## xmm0 = xmm0[0,0,2,2]
+; CHECK-NEXT:    ## xmm0 {%k1} {z} = xmm0[0,0,2,2]
 ; CHECK-NEXT:    vaddps %xmm2, %xmm1, %xmm1 ## encoding: [0x62,0xf1,0x74,0x08,0x58,0xca]
 ; CHECK-NEXT:    vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0x62,0xf1,0x7c,0x08,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
@@ -33,9 +33,9 @@
 ; CHECK-NEXT:    ## ymm2 = ymm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovsldup %ymm0, %ymm1 {%k1} ## encoding: [0x62,0xf1,0x7e,0x29,0x12,0xc8]
-; CHECK-NEXT:    ## ymm1 = ymm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    ## ymm1 {%k1} = ymm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    vmovsldup %ymm0, %ymm0 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0xa9,0x12,0xc0]
-; CHECK-NEXT:    ## ymm0 = ymm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    ## ymm0 {%k1} {z} = ymm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    vaddps %ymm2, %ymm1, %ymm1 ## encoding: [0x62,0xf1,0x74,0x28,0x58,0xca]
 ; CHECK-NEXT:    vaddps %ymm1, %ymm0, %ymm0 ## encoding: [0x62,0xf1,0x7c,0x28,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
@@ -56,9 +56,9 @@
 ; CHECK-NEXT:    ## xmm2 = xmm0[1,1,3,3]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovshdup %xmm0, %xmm1 {%k1} ## encoding: [0x62,0xf1,0x7e,0x09,0x16,0xc8]
-; CHECK-NEXT:    ## xmm1 = xmm0[1,1,3,3]
+; CHECK-NEXT:    ## xmm1 {%k1} = xmm0[1,1,3,3]
 ; CHECK-NEXT:    vmovshdup %xmm0, %xmm0 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0x89,0x16,0xc0]
-; CHECK-NEXT:    ## xmm0 = xmm0[1,1,3,3]
+; CHECK-NEXT:    ## xmm0 {%k1} {z} = xmm0[1,1,3,3]
 ; CHECK-NEXT:    vaddps %xmm2, %xmm1, %xmm1 ## encoding: [0x62,0xf1,0x74,0x08,0x58,0xca]
 ; CHECK-NEXT:    vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0x62,0xf1,0x7c,0x08,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
@@ -79,9 +79,9 @@
 ; CHECK-NEXT:    ## ymm2 = ymm0[1,1,3,3,5,5,7,7]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovshdup %ymm0, %ymm1 {%k1} ## encoding: [0x62,0xf1,0x7e,0x29,0x16,0xc8]
-; CHECK-NEXT:    ## ymm1 = ymm0[1,1,3,3,5,5,7,7]
+; CHECK-NEXT:    ## ymm1 {%k1} = ymm0[1,1,3,3,5,5,7,7]
 ; CHECK-NEXT:    vmovshdup %ymm0, %ymm0 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0xa9,0x16,0xc0]
-; CHECK-NEXT:    ## ymm0 = ymm0[1,1,3,3,5,5,7,7]
+; CHECK-NEXT:    ## ymm0 {%k1} {z} = ymm0[1,1,3,3,5,5,7,7]
 ; CHECK-NEXT:    vaddps %ymm2, %ymm1, %ymm1 ## encoding: [0x62,0xf1,0x74,0x28,0x58,0xca]
 ; CHECK-NEXT:    vaddps %ymm1, %ymm0, %ymm0 ## encoding: [0x62,0xf1,0x7c,0x28,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
@@ -101,9 +101,9 @@
 ; CHECK-NEXT:    ## xmm2 = xmm0[0,0]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovddup %xmm0, %xmm1 {%k1} ## encoding: [0x62,0xf1,0xff,0x09,0x12,0xc8]
-; CHECK-NEXT:    ## xmm1 = xmm0[0,0]
+; CHECK-NEXT:    ## xmm1 {%k1} = xmm0[0,0]
 ; CHECK-NEXT:    vmovddup %xmm0, %xmm0 {%k1} {z} ## encoding: [0x62,0xf1,0xff,0x89,0x12,0xc0]
-; CHECK-NEXT:    ## xmm0 = xmm0[0,0]
+; CHECK-NEXT:    ## xmm0 {%k1} {z} = xmm0[0,0]
 ; CHECK-NEXT:    vaddpd %xmm2, %xmm1, %xmm1 ## encoding: [0x62,0xf1,0xf5,0x08,0x58,0xca]
 ; CHECK-NEXT:    vaddpd %xmm1, %xmm0, %xmm0 ## encoding: [0x62,0xf1,0xfd,0x08,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
@@ -124,9 +124,9 @@
 ; CHECK-NEXT:    ## ymm2 = ymm0[0,0,2,2]
 ; CHECK-NEXT:    kmovw %edi, %k1 ## encoding: [0xc5,0xf8,0x92,0xcf]
 ; CHECK-NEXT:    vmovddup %ymm0, %ymm1 {%k1} ## encoding: [0x62,0xf1,0xff,0x29,0x12,0xc8]
-; CHECK-NEXT:    ## ymm1 = ymm0[0,0,2,2]
+; CHECK-NEXT:    ## ymm1 {%k1} = ymm0[0,0,2,2]
 ; CHECK-NEXT:    vmovddup %ymm0, %ymm0 {%k1} {z} ## encoding: [0x62,0xf1,0xff,0xa9,0x12,0xc0]
-; CHECK-NEXT:    ## ymm0 = ymm0[0,0,2,2]
+; CHECK-NEXT:    ## ymm0 {%k1} {z} = ymm0[0,0,2,2]
 ; CHECK-NEXT:    vaddpd %ymm2, %ymm1, %ymm1 ## encoding: [0x62,0xf1,0xf5,0x28,0x58,0xca]
 ; CHECK-NEXT:    vaddpd %ymm1, %ymm0, %ymm0 ## encoding: [0x62,0xf1,0xfd,0x28,0x58,0xc1]
 ; CHECK-NEXT:    retq ## encoding: [0xc3]
Index: test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
===================================================================
--- test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
+++ test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
@@ -53,7 +53,7 @@
 ; CHECK-LABEL: combine_vpermt2var_8f64_movddup_mask:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovddup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    vmovddup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6]
 ; CHECK-NEXT:    retq
   %res0 = call <8 x double> @llvm.x86.avx512.maskz.vpermt2var.pd.512(<8 x i64> <i64 0, i64 0, i64 2, i64 2, i64 4, i64 4, i64 6, i64 6>, <8 x double> %x0, <8 x double> %x1, i8 %m)
   ret <8 x double> %res0
@@ -169,7 +169,7 @@
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovshdup_mask:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovshdup {{.*#+}} zmm0 = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; CHECK-NEXT:    vmovshdup {{.*#+}} zmm0 {%k1} {z} = zmm0[1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
 ; CHECK-NEXT:    retq
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>, <16 x float> %x0, <16 x float> %x1, i16 %m)
   ret <16 x float> %res0
@@ -196,7 +196,7 @@
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovsldup_mask:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    kmovw %edi, %k1
-; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} {z} = zmm0[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; CHECK-NEXT:    retq
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 undef, i32 0, i32 undef, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 %m)
   ret <16 x float> %res0
@@ -205,7 +205,7 @@
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovsldup_mask_load:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    kmovw %esi, %k1
-; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 = mem[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vmovsldup {{.*#+}} zmm0 {%k1} {z} = mem[0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
 ; CHECK-NEXT:    retq
   %x0 = load <16 x float>, <16 x float> *%p0
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 undef, i32 0, i32 undef, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 %m)
Index: utils/update_llc_test_checks.py
===================================================================
--- utils/update_llc_test_checks.py
+++ utils/update_llc_test_checks.py
@@ -32,7 +32,7 @@
 SCRUB_TRAILING_WHITESPACE_RE = re.compile(r'[ \t]+$', flags=re.M)
 SCRUB_X86_SHUFFLES_RE = (
     re.compile(
-        r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem) = .*)$',
+        r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem)( \{%k\d+\}( \{z\})?)? = .*)$',
         flags=re.M))
 SCRUB_X86_SP_RE = re.compile(r'\d+\(%(esp|rsp)\)')
 SCRUB_X86_RIP_RE = re.compile(r'[.\w]+\(%rip\)')
Index: utils/update_test_checks.py
===================================================================
--- utils/update_test_checks.py
+++ utils/update_test_checks.py
@@ -47,7 +47,7 @@
 SCRUB_TRAILING_WHITESPACE_RE = re.compile(r'[ \t]+$', flags=re.M)
 SCRUB_X86_SHUFFLES_RE = (
     re.compile(
-        r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem) = .*)$',
+        r'^(\s*\w+) [^#\n]+#+ ((?:[xyz]mm\d+|mem)( \{%k\d+\}( \{z\})?)? = .*)$',
         flags=re.M))
 SCRUB_X86_SP_RE = re.compile(r'\d+\(%(esp|rsp)\)')
 SCRUB_X86_RIP_RE = re.compile(r'[.\w]+\(%rip\)')