[llvm] 4318b03 - [MC][X86] Merge lane/element broadcast comment printers. (#79020)
via llvm-commits
llvm-commits at lists.llvm.org
Tue Jan 23 04:34:00 PST 2024
Author: Simon Pilgrim
Date: 2024-01-23T12:33:52Z
New Revision: 4318b033bddc64d5654f3e368fddde859ff4d02e
URL: https://github.com/llvm/llvm-project/commit/4318b033bddc64d5654f3e368fddde859ff4d02e
DIFF: https://github.com/llvm/llvm-project/commit/4318b033bddc64d5654f3e368fddde859ff4d02e.diff
LOG: [MC][X86] Merge lane/element broadcast comment printers. (#79020)
This is /almost/ NFC - the only annoyance is that for some reason we were using "<C1,C2,..>" for ConstantVector types unlike all other cases - these now use the same "[C1,C2,..]" format as the other constant printers.
Added:
Modified:
llvm/lib/Target/X86/X86MCInstLower.cpp
llvm/test/CodeGen/X86/2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll
llvm/test/CodeGen/X86/2011-10-19-widen_vselect.ll
llvm/test/CodeGen/X86/addsub-constant-folding.ll
llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast.ll
llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast_from_memory.ll
llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll
llvm/test/CodeGen/X86/avx2-vperm.ll
llvm/test/CodeGen/X86/avx512-shuffles/partial_permute.ll
llvm/test/CodeGen/X86/bitreverse.ll
llvm/test/CodeGen/X86/combine-sdiv.ll
llvm/test/CodeGen/X86/combine-subo.ll
llvm/test/CodeGen/X86/combine-udiv.ll
llvm/test/CodeGen/X86/fma-fneg-combine-2.ll
llvm/test/CodeGen/X86/fold-int-pow2-with-fmul-or-fdiv.ll
llvm/test/CodeGen/X86/fpclamptosat_vec.ll
llvm/test/CodeGen/X86/hoist-and-by-const-from-lshr-in-eqcmp-zero.ll
llvm/test/CodeGen/X86/icmp-abs-C-vec.ll
llvm/test/CodeGen/X86/insert-into-constant-vector.ll
llvm/test/CodeGen/X86/masked_store_trunc.ll
llvm/test/CodeGen/X86/matrix-multiply.ll
llvm/test/CodeGen/X86/oddshuffles.ll
llvm/test/CodeGen/X86/omit-urem-of-power-of-two-or-zero-when-comparing-with-zero.ll
llvm/test/CodeGen/X86/peephole-fold-movsd.ll
llvm/test/CodeGen/X86/pr63108.ll
llvm/test/CodeGen/X86/psubus.ll
llvm/test/CodeGen/X86/sext-vsetcc.ll
llvm/test/CodeGen/X86/shrink_vmul.ll
llvm/test/CodeGen/X86/sink-addsub-of-const.ll
llvm/test/CodeGen/X86/slow-pmulld.ll
llvm/test/CodeGen/X86/srem-seteq-vec-nonsplat.ll
llvm/test/CodeGen/X86/sse2.ll
llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll
llvm/test/CodeGen/X86/urem-seteq-vec-nonsplat.ll
llvm/test/CodeGen/X86/vec_fp_to_int.ll
llvm/test/CodeGen/X86/vector-blend.ll
llvm/test/CodeGen/X86/vector-fshl-512.ll
llvm/test/CodeGen/X86/vector-fshr-128.ll
llvm/test/CodeGen/X86/vector-fshr-512.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-2.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-4.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-4.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-3.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-7.ll
llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll
llvm/test/CodeGen/X86/vector-mulfix-legalize.ll
llvm/test/CodeGen/X86/vector-partial-undef.ll
llvm/test/CodeGen/X86/vector-reduce-add-mask.ll
llvm/test/CodeGen/X86/vector-replicaton-i1-mask.ll
llvm/test/CodeGen/X86/vector-shift-ashr-128.ll
llvm/test/CodeGen/X86/vector-shift-ashr-sub128.ll
llvm/test/CodeGen/X86/vector-shift-lshr-128.ll
llvm/test/CodeGen/X86/vector-shift-lshr-512.ll
llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll
llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
llvm/test/CodeGen/X86/vector-shuffle-256-v8.ll
llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll
llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
llvm/test/CodeGen/X86/vector-shuffle-combining-sse41.ll
llvm/test/CodeGen/X86/vector-shuffle-combining-ssse3.ll
llvm/test/CodeGen/X86/vector-shuffle-combining.ll
llvm/test/CodeGen/X86/vector-shuffle-v1.ll
llvm/test/CodeGen/X86/vector-shuffle-v192.ll
llvm/test/CodeGen/X86/widen_arith-2.ll
llvm/test/CodeGen/X86/widen_arith-4.ll
llvm/test/CodeGen/X86/widen_arith-5.ll
llvm/test/CodeGen/X86/widen_arith-6.ll
llvm/test/CodeGen/X86/x86-interleaved-access.ll
llvm/test/CodeGen/X86/zero_extend_vector_inreg.ll
llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast.ll
llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast_from_memory.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/X86/X86MCInstLower.cpp b/llvm/lib/Target/X86/X86MCInstLower.cpp
index 6cf93dc51238c07..58ebe023cd61eca 100644
--- a/llvm/lib/Target/X86/X86MCInstLower.cpp
+++ b/llvm/lib/Target/X86/X86MCInstLower.cpp
@@ -1564,68 +1564,18 @@ static void printZeroUpperMove(const MachineInstr *MI, MCStreamer &OutStreamer,
OutStreamer.AddComment(CS.str());
}
-static void printLaneBroadcast(const MachineInstr *MI, MCStreamer &OutStreamer,
- int NumLanes, int BitWidth) {
- if (auto *C = X86::getConstantFromPool(*MI, 1)) {
- int CstEltSize = C->getType()->getScalarSizeInBits();
-
- std::string Comment;
- raw_string_ostream CS(Comment);
- const MachineOperand &DstOp = MI->getOperand(0);
- CS << X86ATTInstPrinter::getRegisterName(DstOp.getReg()) << " = ";
- if (auto *CDS = dyn_cast<ConstantDataSequential>(C)) {
- int NumElements = CDS->getNumElements();
- if ((BitWidth % CstEltSize) == 0)
- NumElements = std::min<int>(NumElements, BitWidth / CstEltSize);
- CS << "[";
- for (int l = 0; l != NumLanes; ++l) {
- for (int i = 0; i < NumElements; ++i) {
- if (i != 0 || l != 0)
- CS << ",";
- if (CDS->getElementType()->isIntegerTy())
- printConstant(CDS->getElementAsAPInt(i), CS);
- else if (CDS->getElementType()->isHalfTy() ||
- CDS->getElementType()->isFloatTy() ||
- CDS->getElementType()->isDoubleTy())
- printConstant(CDS->getElementAsAPFloat(i), CS);
- else
- CS << "?";
- }
- }
- CS << "]";
- OutStreamer.AddComment(CS.str());
- } else if (auto *CV = dyn_cast<ConstantVector>(C)) {
- int NumOperands = CV->getNumOperands();
- if ((BitWidth % CstEltSize) == 0)
- NumOperands = std::min<int>(NumOperands, BitWidth / CstEltSize);
- CS << "<";
- for (int l = 0; l != NumLanes; ++l) {
- for (int i = 0; i < NumOperands; ++i) {
- if (i != 0 || l != 0)
- CS << ",";
- printConstant(CV->getOperand(i),
- CV->getType()->getPrimitiveSizeInBits(), CS);
- }
- }
- CS << ">";
- OutStreamer.AddComment(CS.str());
- }
- }
-}
-
-static void printElementBroadcast(const MachineInstr *MI,
- MCStreamer &OutStreamer, int NumElts,
- int EltBits) {
+static void printBroadcast(const MachineInstr *MI, MCStreamer &OutStreamer,
+ int Repeats, int BitWidth) {
if (auto *C = X86::getConstantFromPool(*MI, 1)) {
std::string Comment;
raw_string_ostream CS(Comment);
const MachineOperand &DstOp = MI->getOperand(0);
CS << X86ATTInstPrinter::getRegisterName(DstOp.getReg()) << " = ";
CS << "[";
- for (int i = 0; i != NumElts; ++i) {
- if (i != 0)
+ for (int l = 0; l != Repeats; ++l) {
+ if (l != 0)
CS << ",";
- printConstant(C, EltBits, CS);
+ printConstant(C, BitWidth, CS);
}
CS << "]";
OutStreamer.AddComment(CS.str());
@@ -1943,13 +1893,13 @@ static void addConstantComments(const MachineInstr *MI,
// For loads from a constant pool to a vector register, print the constant
// loaded.
CASE_128_MOV_RM()
- printLaneBroadcast(MI, OutStreamer, 1, 128);
+ printBroadcast(MI, OutStreamer, 1, 128);
break;
CASE_256_MOV_RM()
- printLaneBroadcast(MI, OutStreamer, 1, 256);
+ printBroadcast(MI, OutStreamer, 1, 256);
break;
CASE_512_MOV_RM()
- printLaneBroadcast(MI, OutStreamer, 1, 512);
+ printBroadcast(MI, OutStreamer, 1, 512);
break;
case X86::VBROADCASTF128rm:
case X86::VBROADCASTI128rm:
@@ -1957,19 +1907,19 @@ static void addConstantComments(const MachineInstr *MI,
case X86::VBROADCASTF64X2Z128rm:
case X86::VBROADCASTI32X4Z256rm:
case X86::VBROADCASTI64X2Z128rm:
- printLaneBroadcast(MI, OutStreamer, 2, 128);
+ printBroadcast(MI, OutStreamer, 2, 128);
break;
case X86::VBROADCASTF32X4rm:
case X86::VBROADCASTF64X2rm:
case X86::VBROADCASTI32X4rm:
case X86::VBROADCASTI64X2rm:
- printLaneBroadcast(MI, OutStreamer, 4, 128);
+ printBroadcast(MI, OutStreamer, 4, 128);
break;
case X86::VBROADCASTF32X8rm:
case X86::VBROADCASTF64X4rm:
case X86::VBROADCASTI32X8rm:
case X86::VBROADCASTI64X4rm:
- printLaneBroadcast(MI, OutStreamer, 2, 256);
+ printBroadcast(MI, OutStreamer, 2, 256);
break;
// For broadcast loads from a constant pool to a vector register, repeatedly
@@ -1979,55 +1929,55 @@ static void addConstantComments(const MachineInstr *MI,
case X86::VMOVDDUPZ128rm:
case X86::VPBROADCASTQrm:
case X86::VPBROADCASTQZ128rm:
- printElementBroadcast(MI, OutStreamer, 2, 64);
+ printBroadcast(MI, OutStreamer, 2, 64);
break;
case X86::VBROADCASTSDYrm:
case X86::VBROADCASTSDZ256rm:
case X86::VPBROADCASTQYrm:
case X86::VPBROADCASTQZ256rm:
- printElementBroadcast(MI, OutStreamer, 4, 64);
+ printBroadcast(MI, OutStreamer, 4, 64);
break;
case X86::VBROADCASTSDZrm:
case X86::VPBROADCASTQZrm:
- printElementBroadcast(MI, OutStreamer, 8, 64);
+ printBroadcast(MI, OutStreamer, 8, 64);
break;
case X86::VBROADCASTSSrm:
case X86::VBROADCASTSSZ128rm:
case X86::VPBROADCASTDrm:
case X86::VPBROADCASTDZ128rm:
- printElementBroadcast(MI, OutStreamer, 4, 32);
+ printBroadcast(MI, OutStreamer, 4, 32);
break;
case X86::VBROADCASTSSYrm:
case X86::VBROADCASTSSZ256rm:
case X86::VPBROADCASTDYrm:
case X86::VPBROADCASTDZ256rm:
- printElementBroadcast(MI, OutStreamer, 8, 32);
+ printBroadcast(MI, OutStreamer, 8, 32);
break;
case X86::VBROADCASTSSZrm:
case X86::VPBROADCASTDZrm:
- printElementBroadcast(MI, OutStreamer, 16, 32);
+ printBroadcast(MI, OutStreamer, 16, 32);
break;
case X86::VPBROADCASTWrm:
case X86::VPBROADCASTWZ128rm:
- printElementBroadcast(MI, OutStreamer, 8, 16);
+ printBroadcast(MI, OutStreamer, 8, 16);
break;
case X86::VPBROADCASTWYrm:
case X86::VPBROADCASTWZ256rm:
- printElementBroadcast(MI, OutStreamer, 16, 16);
+ printBroadcast(MI, OutStreamer, 16, 16);
break;
case X86::VPBROADCASTWZrm:
- printElementBroadcast(MI, OutStreamer, 32, 16);
+ printBroadcast(MI, OutStreamer, 32, 16);
break;
case X86::VPBROADCASTBrm:
case X86::VPBROADCASTBZ128rm:
- printElementBroadcast(MI, OutStreamer, 16, 8);
+ printBroadcast(MI, OutStreamer, 16, 8);
break;
case X86::VPBROADCASTBYrm:
case X86::VPBROADCASTBZ256rm:
- printElementBroadcast(MI, OutStreamer, 32, 8);
+ printBroadcast(MI, OutStreamer, 32, 8);
break;
case X86::VPBROADCASTBZrm:
- printElementBroadcast(MI, OutStreamer, 64, 8);
+ printBroadcast(MI, OutStreamer, 64, 8);
break;
}
}
diff --git a/llvm/test/CodeGen/X86/2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll b/llvm/test/CodeGen/X86/2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll
index dd8efe9715d3afb..b5577c09c43207c 100644
--- a/llvm/test/CodeGen/X86/2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll
+++ b/llvm/test/CodeGen/X86/2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll
@@ -16,7 +16,7 @@ define fastcc double @tailcall() {
; CHECK-NEXT: movq %rax, {{[0-9]+}}(%rsp)
; CHECK-NEXT: fld1
; CHECK-NEXT: fstpt {{[0-9]+}}(%rsp)
-; CHECK-NEXT: movaps {{.*#+}} xmm0 = <1.0E+0,1.0E+0,u,u>
+; CHECK-NEXT: movaps {{.*#+}} xmm0 = [1.0E+0,1.0E+0,u,u]
; CHECK-NEXT: addq $24, %rsp
; CHECK-NEXT: jmp _tailcallee ## TAILCALL
entry:
diff --git a/llvm/test/CodeGen/X86/2011-10-19-widen_vselect.ll b/llvm/test/CodeGen/X86/2011-10-19-widen_vselect.ll
index 171e16e35fc2f0b..5a4cbac57eeebb7 100644
--- a/llvm/test/CodeGen/X86/2011-10-19-widen_vselect.ll
+++ b/llvm/test/CodeGen/X86/2011-10-19-widen_vselect.ll
@@ -73,7 +73,7 @@ define void @full_test() {
; X86-NEXT: cvtdq2ps %xmm0, %xmm1
; X86-NEXT: xorps %xmm0, %xmm0
; X86-NEXT: cmpltps %xmm2, %xmm0
-; X86-NEXT: movaps {{.*#+}} xmm3 = <1.0E+0,1.0E+0,u,u>
+; X86-NEXT: movaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,u,u]
; X86-NEXT: addps %xmm1, %xmm3
; X86-NEXT: movaps %xmm1, %xmm4
; X86-NEXT: blendvps %xmm0, %xmm3, %xmm4
@@ -93,7 +93,7 @@ define void @full_test() {
; X64-NEXT: cvtdq2ps %xmm0, %xmm1
; X64-NEXT: xorps %xmm0, %xmm0
; X64-NEXT: cmpltps %xmm2, %xmm0
-; X64-NEXT: movaps {{.*#+}} xmm3 = <1.0E+0,1.0E+0,u,u>
+; X64-NEXT: movaps {{.*#+}} xmm3 = [1.0E+0,1.0E+0,u,u]
; X64-NEXT: addps %xmm1, %xmm3
; X64-NEXT: movaps %xmm1, %xmm4
; X64-NEXT: blendvps %xmm0, %xmm3, %xmm4
diff --git a/llvm/test/CodeGen/X86/addsub-constant-folding.ll b/llvm/test/CodeGen/X86/addsub-constant-folding.ll
index 4dbaae5c1a74ae4..1cdc81223168f0a 100644
--- a/llvm/test/CodeGen/X86/addsub-constant-folding.ll
+++ b/llvm/test/CodeGen/X86/addsub-constant-folding.ll
@@ -367,14 +367,14 @@ define <4 x i32> @vec_add_const_const_sub_extrause(<4 x i32> %arg) {
define <4 x i32> @vec_add_const_const_sub_nonsplat(<4 x i32> %arg) {
; X86-LABEL: vec_add_const_const_sub_nonsplat:
; X86: # %bb.0:
-; X86-NEXT: movdqa {{.*#+}} xmm1 = <4294967277,u,u,4294967290>
+; X86-NEXT: movdqa {{.*#+}} xmm1 = [4294967277,u,u,4294967290]
; X86-NEXT: psubd %xmm0, %xmm1
; X86-NEXT: movdqa %xmm1, %xmm0
; X86-NEXT: retl
;
; X64-LABEL: vec_add_const_const_sub_nonsplat:
; X64: # %bb.0:
-; X64-NEXT: movdqa {{.*#+}} xmm1 = <4294967277,u,u,4294967290>
+; X64-NEXT: movdqa {{.*#+}} xmm1 = [4294967277,u,u,4294967290]
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
@@ -733,14 +733,14 @@ define <4 x i32> @vec_sub_const_const_sub_extrause(<4 x i32> %arg) {
define <4 x i32> @vec_sub_const_const_sub_nonsplat(<4 x i32> %arg) {
; X86-LABEL: vec_sub_const_const_sub_nonsplat:
; X86: # %bb.0:
-; X86-NEXT: movdqa {{.*#+}} xmm1 = <23,u,u,10>
+; X86-NEXT: movdqa {{.*#+}} xmm1 = [23,u,u,10]
; X86-NEXT: psubd %xmm0, %xmm1
; X86-NEXT: movdqa %xmm1, %xmm0
; X86-NEXT: retl
;
; X64-LABEL: vec_sub_const_const_sub_nonsplat:
; X64: # %bb.0:
-; X64-NEXT: movdqa {{.*#+}} xmm1 = <23,u,u,10>
+; X64-NEXT: movdqa {{.*#+}} xmm1 = [23,u,u,10]
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
@@ -867,14 +867,14 @@ define <4 x i32> @vec_const_sub_add_const_extrause(<4 x i32> %arg) {
define <4 x i32> @vec_const_sub_add_const_nonsplat(<4 x i32> %arg) {
; X86-LABEL: vec_const_sub_add_const_nonsplat:
; X86: # %bb.0:
-; X86-NEXT: movdqa {{.*#+}} xmm1 = <23,u,u,10>
+; X86-NEXT: movdqa {{.*#+}} xmm1 = [23,u,u,10]
; X86-NEXT: psubd %xmm0, %xmm1
; X86-NEXT: movdqa %xmm1, %xmm0
; X86-NEXT: retl
;
; X64-LABEL: vec_const_sub_add_const_nonsplat:
; X64: # %bb.0:
-; X64-NEXT: movdqa {{.*#+}} xmm1 = <23,u,u,10>
+; X64-NEXT: movdqa {{.*#+}} xmm1 = [23,u,u,10]
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
@@ -1001,14 +1001,14 @@ define <4 x i32> @vec_const_sub_sub_const_extrause(<4 x i32> %arg) {
define <4 x i32> @vec_const_sub_sub_const_nonsplat(<4 x i32> %arg) {
; X86-LABEL: vec_const_sub_sub_const_nonsplat:
; X86: # %bb.0:
-; X86-NEXT: movdqa {{.*#+}} xmm1 = <19,u,u,6>
+; X86-NEXT: movdqa {{.*#+}} xmm1 = [19,u,u,6]
; X86-NEXT: psubd %xmm0, %xmm1
; X86-NEXT: movdqa %xmm1, %xmm0
; X86-NEXT: retl
;
; X64-LABEL: vec_const_sub_sub_const_nonsplat:
; X64: # %bb.0:
-; X64-NEXT: movdqa {{.*#+}} xmm1 = <19,u,u,6>
+; X64-NEXT: movdqa {{.*#+}} xmm1 = [19,u,u,6]
; X64-NEXT: psubd %xmm0, %xmm1
; X64-NEXT: movdqa %xmm1, %xmm0
; X64-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast.ll b/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast.ll
index d61e33ccb22a957..fe48059f9d0e653 100644
--- a/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast.ll
+++ b/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast.ll
@@ -1053,7 +1053,7 @@ define void @vec256_i8_widen_to_i16_factor2_broadcast_to_v16i16_factor16(ptr %in
; SSE42-NEXT: paddb 48(%rsi), %xmm2
; SSE42-NEXT: paddb (%rsi), %xmm0
; SSE42-NEXT: paddb 32(%rsi), %xmm1
-; SSE42-NEXT: movdqa {{.*#+}} xmm3 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSE42-NEXT: movdqa {{.*#+}} xmm3 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pshufb %xmm3, %xmm1
; SSE42-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE42-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
@@ -3941,7 +3941,7 @@ define void @vec384_i16_widen_to_i96_factor6_broadcast_to_v4i96_factor4(ptr %in.
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpaddb (%rsi), %zmm0, %zmm0
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <16,9,10,11,12,13,16,15,u,u,u,u,16,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [16,9,10,11,12,13,16,15,u,u,u,u,16,u,u,u]
; AVX512BW-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512BW-NEXT: vpbroadcastw %xmm0, %ymm0
; AVX512BW-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm0
@@ -4181,7 +4181,7 @@ define void @vec384_i16_widen_to_i192_factor12_broadcast_to_v2i192_factor2(ptr %
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpaddb (%rsi), %zmm0, %zmm0
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <16,9,10,11,12,13,14,15,u,u,u,u,16,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [16,9,10,11,12,13,14,15,u,u,u,u,16,u,u,u]
; AVX512BW-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512BW-NEXT: vpaddb (%rdx), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rcx)
diff --git a/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast_from_memory.ll b/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast_from_memory.ll
index 99d9f6b41e70dd5..2f576fe67159046 100644
--- a/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast_from_memory.ll
+++ b/llvm/test/CodeGen/X86/any_extend_vector_inreg_of_broadcast_from_memory.ll
@@ -875,7 +875,7 @@ define void @vec256_i8_widen_to_i16_factor2_broadcast_to_v16i16_factor16(ptr %in
; SSE42-NEXT: movdqa (%rdi), %xmm0
; SSE42-NEXT: movdqa 32(%rdi), %xmm1
; SSE42-NEXT: movdqa 48(%rdi), %xmm2
-; SSE42-NEXT: movdqa {{.*#+}} xmm3 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSE42-NEXT: movdqa {{.*#+}} xmm3 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pshufb %xmm3, %xmm1
; SSE42-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE42-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll b/llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll
index d6001da849e31f1..ee504e30302b13b 100644
--- a/llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll
+++ b/llvm/test/CodeGen/X86/avx2-fma-fneg-combine.ll
@@ -131,7 +131,7 @@ define <4 x double> @test9(<4 x double> %a) {
define <4 x double> @test10(<4 x double> %a, <4 x double> %b) {
; CHECK-LABEL: test10:
; CHECK: # %bb.0:
-; CHECK-NEXT: vmovapd {{.*#+}} ymm2 = <-9.5E+0,u,-5.5E+0,-2.5E+0>
+; CHECK-NEXT: vmovapd {{.*#+}} ymm2 = [-9.5E+0,u,-5.5E+0,-2.5E+0]
; CHECK-NEXT: vmovapd %ymm2, %ymm3
; CHECK-NEXT: vfmadd213pd {{.*#+}} ymm3 = (ymm0 * ymm3) + ymm1
; CHECK-NEXT: vfnmadd213pd {{.*#+}} ymm2 = -(ymm0 * ymm2) + ymm1
diff --git a/llvm/test/CodeGen/X86/avx2-vperm.ll b/llvm/test/CodeGen/X86/avx2-vperm.ll
index 101fd1931278804..90430aa51dcdc0d 100644
--- a/llvm/test/CodeGen/X86/avx2-vperm.ll
+++ b/llvm/test/CodeGen/X86/avx2-vperm.ll
@@ -23,13 +23,13 @@ entry:
define <8 x float> @perm_cl_fp_8x32(<8 x float> %A) nounwind readnone {
; X86-LABEL: perm_cl_fp_8x32:
; X86: # %bb.0: # %entry
-; X86-NEXT: vmovaps {{.*#+}} ymm1 = <u,7,2,u,4,u,1,6>
+; X86-NEXT: vmovaps {{.*#+}} ymm1 = [u,7,2,u,4,u,1,6]
; X86-NEXT: vpermps %ymm0, %ymm1, %ymm0
; X86-NEXT: retl
;
; X64-LABEL: perm_cl_fp_8x32:
; X64: # %bb.0: # %entry
-; X64-NEXT: vmovaps {{.*#+}} ymm1 = <u,7,2,u,4,u,1,6>
+; X64-NEXT: vmovaps {{.*#+}} ymm1 = [u,7,2,u,4,u,1,6]
; X64-NEXT: vpermps %ymm0, %ymm1, %ymm0
; X64-NEXT: retq
entry:
diff --git a/llvm/test/CodeGen/X86/avx512-shuffles/partial_permute.ll b/llvm/test/CodeGen/X86/avx512-shuffles/partial_permute.ll
index ec0f14ae4e58e3e..b648b086a8b689f 100644
--- a/llvm/test/CodeGen/X86/avx512-shuffles/partial_permute.ll
+++ b/llvm/test/CodeGen/X86/avx512-shuffles/partial_permute.ll
@@ -1769,7 +1769,7 @@ define <4 x i32> @test_16xi32_to_4xi32_perm_mask9(<16 x i32> %vec) {
;
; CHECK-FAST-PERLANE-LABEL: test_16xi32_to_4xi32_perm_mask9:
; CHECK-FAST-PERLANE: # %bb.0:
-; CHECK-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <4,1,u,2>
+; CHECK-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [4,1,u,2]
; CHECK-FAST-PERLANE-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; CHECK-FAST-PERLANE-NEXT: vpermd %ymm2, %ymm1, %ymm1
; CHECK-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm2
diff --git a/llvm/test/CodeGen/X86/bitreverse.ll b/llvm/test/CodeGen/X86/bitreverse.ll
index 9daac1df1d97500..2d978b5e991c913 100644
--- a/llvm/test/CodeGen/X86/bitreverse.ll
+++ b/llvm/test/CodeGen/X86/bitreverse.ll
@@ -587,7 +587,7 @@ define <2 x i16> @fold_v2i16() {
;
; X64-LABEL: fold_v2i16:
; X64: # %bb.0:
-; X64-NEXT: movaps {{.*#+}} xmm0 = <61440,240,u,u,u,u,u,u>
+; X64-NEXT: movaps {{.*#+}} xmm0 = [61440,240,u,u,u,u,u,u]
; X64-NEXT: retq
;
; X86XOP-LABEL: fold_v2i16:
diff --git a/llvm/test/CodeGen/X86/combine-sdiv.ll b/llvm/test/CodeGen/X86/combine-sdiv.ll
index 549fe7262697301..e10d94c16696a21 100644
--- a/llvm/test/CodeGen/X86/combine-sdiv.ll
+++ b/llvm/test/CodeGen/X86/combine-sdiv.ll
@@ -581,7 +581,7 @@ define <16 x i16> @combine_vec_sdiv_by_pow2b_v16i16(<16 x i16> %x) {
; SSE2: # %bb.0:
; SSE2-NEXT: movdqa %xmm0, %xmm3
; SSE2-NEXT: psraw $15, %xmm0
-; SSE2-NEXT: movdqa {{.*#+}} xmm7 = <u,4,2,16,8,32,64,2>
+; SSE2-NEXT: movdqa {{.*#+}} xmm7 = [u,4,2,16,8,32,64,2]
; SSE2-NEXT: pmulhuw %xmm7, %xmm0
; SSE2-NEXT: paddw %xmm3, %xmm0
; SSE2-NEXT: movdqa {{.*#+}} xmm4 = [65535,65535,65535,0,65535,0,0,65535]
@@ -639,10 +639,10 @@ define <16 x i16> @combine_vec_sdiv_by_pow2b_v16i16(<16 x i16> %x) {
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psraw $15, %xmm2
-; SSE41-NEXT: movdqa {{.*#+}} xmm3 = <u,4,2,16,8,32,64,2>
+; SSE41-NEXT: movdqa {{.*#+}} xmm3 = [u,4,2,16,8,32,64,2]
; SSE41-NEXT: pmulhuw %xmm3, %xmm2
; SSE41-NEXT: paddw %xmm0, %xmm2
-; SSE41-NEXT: movdqa {{.*#+}} xmm4 = <u,16384,32768,4096,8192,2048,1024,32768>
+; SSE41-NEXT: movdqa {{.*#+}} xmm4 = [u,16384,32768,4096,8192,2048,1024,32768]
; SSE41-NEXT: movdqa %xmm2, %xmm5
; SSE41-NEXT: pmulhw %xmm4, %xmm5
; SSE41-NEXT: psraw $1, %xmm2
@@ -662,10 +662,10 @@ define <16 x i16> @combine_vec_sdiv_by_pow2b_v16i16(<16 x i16> %x) {
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-NEXT: vpsraw $15, %xmm1, %xmm2
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = <u,4,2,16,8,32,64,2>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [u,4,2,16,8,32,64,2]
; AVX1-NEXT: vpmulhuw %xmm3, %xmm2, %xmm2
; AVX1-NEXT: vpaddw %xmm2, %xmm1, %xmm1
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <u,16384,32768,4096,8192,2048,1024,32768>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [u,16384,32768,4096,8192,2048,1024,32768]
; AVX1-NEXT: vpmulhw %xmm2, %xmm1, %xmm4
; AVX1-NEXT: vpsraw $1, %xmm1, %xmm1
; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm4[0,1],xmm1[2],xmm4[3,4,5,6],xmm1[7]
@@ -718,10 +718,10 @@ define <16 x i16> @combine_vec_sdiv_by_pow2b_v16i16(<16 x i16> %x) {
; XOP: # %bb.0:
; XOP-NEXT: vextractf128 $1, %ymm0, %xmm1
; XOP-NEXT: vpsraw $15, %xmm1, %xmm2
-; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = <u,65522,65521,65524,65523,65525,65526,65521>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = [u,65522,65521,65524,65523,65525,65526,65521]
; XOP-NEXT: vpshlw %xmm3, %xmm2, %xmm2
; XOP-NEXT: vpaddw %xmm2, %xmm1, %xmm1
-; XOP-NEXT: vmovdqa {{.*#+}} xmm2 = <u,65534,65535,65532,65533,65531,65530,65535>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm2 = [u,65534,65535,65532,65533,65531,65530,65535]
; XOP-NEXT: vpshaw %xmm2, %xmm1, %xmm1
; XOP-NEXT: vpsraw $15, %xmm0, %xmm4
; XOP-NEXT: vpshlw %xmm3, %xmm4, %xmm3
@@ -742,7 +742,7 @@ define <32 x i16> @combine_vec_sdiv_by_pow2b_v32i16(<32 x i16> %x) {
; SSE2-NEXT: movdqa %xmm1, %xmm5
; SSE2-NEXT: movdqa %xmm0, %xmm1
; SSE2-NEXT: psraw $15, %xmm0
-; SSE2-NEXT: movdqa {{.*#+}} xmm9 = <u,4,2,16,8,32,64,2>
+; SSE2-NEXT: movdqa {{.*#+}} xmm9 = [u,4,2,16,8,32,64,2]
; SSE2-NEXT: pmulhuw %xmm9, %xmm0
; SSE2-NEXT: paddw %xmm1, %xmm0
; SSE2-NEXT: movdqa {{.*#+}} xmm6 = [65535,65535,65535,0,65535,0,0,65535]
@@ -853,10 +853,10 @@ define <32 x i16> @combine_vec_sdiv_by_pow2b_v32i16(<32 x i16> %x) {
; SSE41: # %bb.0:
; SSE41-NEXT: movdqa %xmm0, %xmm6
; SSE41-NEXT: psraw $15, %xmm6
-; SSE41-NEXT: movdqa {{.*#+}} xmm5 = <u,4,2,16,8,32,64,2>
+; SSE41-NEXT: movdqa {{.*#+}} xmm5 = [u,4,2,16,8,32,64,2]
; SSE41-NEXT: pmulhuw %xmm5, %xmm6
; SSE41-NEXT: paddw %xmm0, %xmm6
-; SSE41-NEXT: movdqa {{.*#+}} xmm4 = <u,16384,32768,4096,8192,2048,1024,32768>
+; SSE41-NEXT: movdqa {{.*#+}} xmm4 = [u,16384,32768,4096,8192,2048,1024,32768]
; SSE41-NEXT: movdqa %xmm6, %xmm7
; SSE41-NEXT: pmulhw %xmm4, %xmm7
; SSE41-NEXT: psraw $1, %xmm6
@@ -894,10 +894,10 @@ define <32 x i16> @combine_vec_sdiv_by_pow2b_v32i16(<32 x i16> %x) {
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm2
; AVX1-NEXT: vpsraw $15, %xmm2, %xmm3
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm4 = <u,4,2,16,8,32,64,2>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm4 = [u,4,2,16,8,32,64,2]
; AVX1-NEXT: vpmulhuw %xmm4, %xmm3, %xmm3
; AVX1-NEXT: vpaddw %xmm3, %xmm2, %xmm2
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = <u,16384,32768,4096,8192,2048,1024,32768>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [u,16384,32768,4096,8192,2048,1024,32768]
; AVX1-NEXT: vpmulhw %xmm3, %xmm2, %xmm5
; AVX1-NEXT: vpsraw $1, %xmm2, %xmm2
; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm5[0,1],xmm2[2],xmm5[3,4,5,6],xmm2[7]
@@ -995,10 +995,10 @@ define <32 x i16> @combine_vec_sdiv_by_pow2b_v32i16(<32 x i16> %x) {
; XOP: # %bb.0:
; XOP-NEXT: vextractf128 $1, %ymm0, %xmm2
; XOP-NEXT: vpsraw $15, %xmm2, %xmm3
-; XOP-NEXT: vmovdqa {{.*#+}} xmm4 = <u,65522,65521,65524,65523,65525,65526,65521>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm4 = [u,65522,65521,65524,65523,65525,65526,65521]
; XOP-NEXT: vpshlw %xmm4, %xmm3, %xmm3
; XOP-NEXT: vpaddw %xmm3, %xmm2, %xmm2
-; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = <u,65534,65535,65532,65533,65531,65530,65535>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = [u,65534,65535,65532,65533,65531,65530,65535]
; XOP-NEXT: vpshaw %xmm3, %xmm2, %xmm2
; XOP-NEXT: vpsraw $15, %xmm0, %xmm5
; XOP-NEXT: vpshlw %xmm4, %xmm5, %xmm5
@@ -1234,10 +1234,10 @@ define <8 x i32> @combine_vec_sdiv_by_pow2b_v8i32(<8 x i32> %x) {
; XOP: # %bb.0:
; XOP-NEXT: vextractf128 $1, %ymm0, %xmm1
; XOP-NEXT: vpsrad $31, %xmm1, %xmm2
-; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = <u,4294967266,4294967267,4294967268>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = [u,4294967266,4294967267,4294967268]
; XOP-NEXT: vpshld %xmm3, %xmm2, %xmm2
; XOP-NEXT: vpaddd %xmm2, %xmm1, %xmm1
-; XOP-NEXT: vmovdqa {{.*#+}} xmm2 = <u,4294967294,4294967293,4294967292>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm2 = [u,4294967294,4294967293,4294967292]
; XOP-NEXT: vpshad %xmm2, %xmm1, %xmm1
; XOP-NEXT: vpsrad $31, %xmm0, %xmm4
; XOP-NEXT: vpshld %xmm3, %xmm4, %xmm3
@@ -1510,10 +1510,10 @@ define <16 x i32> @combine_vec_sdiv_by_pow2b_v16i32(<16 x i32> %x) {
; XOP: # %bb.0:
; XOP-NEXT: vextractf128 $1, %ymm0, %xmm2
; XOP-NEXT: vpsrad $31, %xmm2, %xmm3
-; XOP-NEXT: vmovdqa {{.*#+}} xmm4 = <u,4294967266,4294967267,4294967268>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm4 = [u,4294967266,4294967267,4294967268]
; XOP-NEXT: vpshld %xmm4, %xmm3, %xmm3
; XOP-NEXT: vpaddd %xmm3, %xmm2, %xmm2
-; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = <u,4294967294,4294967293,4294967292>
+; XOP-NEXT: vmovdqa {{.*#+}} xmm3 = [u,4294967294,4294967293,4294967292]
; XOP-NEXT: vpshad %xmm3, %xmm2, %xmm2
; XOP-NEXT: vpsrad $31, %xmm0, %xmm5
; XOP-NEXT: vpshld %xmm4, %xmm5, %xmm5
@@ -1709,7 +1709,7 @@ define <4 x i64> @combine_vec_sdiv_by_pow2b_v4i64(<4 x i64> %x) {
; AVX2-NEXT: vpsrlvq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX2-NEXT: vpaddq %ymm1, %ymm0, %ymm1
; AVX2-NEXT: vpsrlvq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = <u,2305843009213693952,1152921504606846976,576460752303423488>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [u,2305843009213693952,1152921504606846976,576460752303423488]
; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm1
; AVX2-NEXT: vpsubq %ymm2, %ymm1, %ymm1
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5,6,7]
@@ -1718,7 +1718,7 @@ define <4 x i64> @combine_vec_sdiv_by_pow2b_v4i64(<4 x i64> %x) {
; AVX512F-LABEL: combine_vec_sdiv_by_pow2b_v4i64:
; AVX512F: # %bb.0:
; AVX512F-NEXT: # kill: def $ymm0 killed $ymm0 def $zmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = <u,2,3,4>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = [u,2,3,4]
; AVX512F-NEXT: vpsraq $63, %zmm0, %zmm2
; AVX512F-NEXT: vpsrlvq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm2
; AVX512F-NEXT: vpaddq %ymm2, %ymm0, %ymm2
@@ -1904,12 +1904,12 @@ define <8 x i64> @combine_vec_sdiv_by_pow2b_v8i64(<8 x i64> %x) {
; AVX2: # %bb.0:
; AVX2-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm2, %ymm3
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm4 = <u,62,61,60>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm4 = [u,62,61,60]
; AVX2-NEXT: vpsrlvq %ymm4, %ymm3, %ymm3
; AVX2-NEXT: vpaddq %ymm3, %ymm0, %ymm3
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm5 = <u,2,3,4>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm5 = [u,2,3,4]
; AVX2-NEXT: vpsrlvq %ymm5, %ymm3, %ymm3
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm6 = <u,2305843009213693952,1152921504606846976,576460752303423488>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm6 = [u,2305843009213693952,1152921504606846976,576460752303423488]
; AVX2-NEXT: vpxor %ymm6, %ymm3, %ymm3
; AVX2-NEXT: vpsubq %ymm6, %ymm3, %ymm3
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm3[2,3,4,5,6,7]
@@ -2701,7 +2701,7 @@ define <8 x i16> @combine_vec_sdiv_nonuniform5(<8 x i16> %x) {
; SSE41-NEXT: pmullw %xmm0, %xmm1
; SSE41-NEXT: pmulhw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: paddw %xmm1, %xmm0
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <256,16384,4096,u,u,u,512,256>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [256,16384,4096,u,u,u,512,256]
; SSE41-NEXT: pmulhw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6,7]
; SSE41-NEXT: movdqa %xmm0, %xmm2
@@ -2814,7 +2814,7 @@ define <8 x i16> @combine_vec_sdiv_nonuniform6(<8 x i16> %x) {
; SSE41-NEXT: pmullw %xmm0, %xmm1
; SSE41-NEXT: pmulhw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: paddw %xmm1, %xmm0
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <4,256,256,u,u,512,256,8>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [4,256,256,u,u,512,256,8]
; SSE41-NEXT: pmulhw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6,7]
; SSE41-NEXT: psrlw $15, %xmm0
diff --git a/llvm/test/CodeGen/X86/combine-subo.ll b/llvm/test/CodeGen/X86/combine-subo.ll
index 99f26525d49e5b5..235df0a666ee9f8 100644
--- a/llvm/test/CodeGen/X86/combine-subo.ll
+++ b/llvm/test/CodeGen/X86/combine-subo.ll
@@ -217,7 +217,7 @@ define { <4 x i8>, <4 x i1> } @always_usub_const_vector() nounwind {
define { <4 x i8>, <4 x i1> } @never_usub_const_vector() nounwind {
; SSE-LABEL: never_usub_const_vector:
; SSE: # %bb.0:
-; SSE-NEXT: movaps {{.*#+}} xmm0 = <127,255,0,254,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE-NEXT: movaps {{.*#+}} xmm0 = [127,255,0,254,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE-NEXT: xorps %xmm1, %xmm1
; SSE-NEXT: retq
;
diff --git a/llvm/test/CodeGen/X86/combine-udiv.ll b/llvm/test/CodeGen/X86/combine-udiv.ll
index 12ac819c9633965..f94ff72274d63e4 100644
--- a/llvm/test/CodeGen/X86/combine-udiv.ll
+++ b/llvm/test/CodeGen/X86/combine-udiv.ll
@@ -519,7 +519,7 @@ define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) {
; SSE41-NEXT: psubw %xmm1, %xmm0
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: paddw %xmm1, %xmm0
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <4096,2048,8,u,u,2,2,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [4096,2048,8,u,u,2,2,u]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6],xmm0[7]
; SSE41-NEXT: retq
@@ -683,7 +683,7 @@ define <16 x i8> @combine_vec_udiv_nonuniform4(<16 x i8> %x) {
define <8 x i16> @pr38477(<8 x i16> %a0) {
; SSE2-LABEL: pr38477:
; SSE2: # %bb.0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <u,4957,57457,4103,16385,35545,2048,2115>
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [u,4957,57457,4103,16385,35545,2048,2115]
; SSE2-NEXT: pmulhuw %xmm0, %xmm1
; SSE2-NEXT: movdqa %xmm0, %xmm2
; SSE2-NEXT: psubw %xmm1, %xmm2
@@ -702,13 +702,13 @@ define <8 x i16> @pr38477(<8 x i16> %a0) {
;
; SSE41-LABEL: pr38477:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,4957,57457,4103,16385,35545,2048,2115>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,4957,57457,4103,16385,35545,2048,2115]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: movdqa %xmm0, %xmm2
; SSE41-NEXT: psubw %xmm1, %xmm2
; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
; SSE41-NEXT: paddw %xmm1, %xmm2
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,1024,1024,16,4,1024,u,4096>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,1024,1024,16,4,1024,u,4096]
; SSE41-NEXT: pmulhuw %xmm2, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5],xmm2[6],xmm1[7]
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/fma-fneg-combine-2.ll b/llvm/test/CodeGen/X86/fma-fneg-combine-2.ll
index bb8ee2238a00431..b9a2772d3df70f5 100644
--- a/llvm/test/CodeGen/X86/fma-fneg-combine-2.ll
+++ b/llvm/test/CodeGen/X86/fma-fneg-combine-2.ll
@@ -145,7 +145,7 @@ define <4 x double> @negated_constant_v4f64(<4 x double> %a) {
define <4 x double> @negated_constant_v4f64_2fmas(<4 x double> %a, <4 x double> %b) {
; FMA3-LABEL: negated_constant_v4f64_2fmas:
; FMA3: # %bb.0:
-; FMA3-NEXT: vmovapd {{.*#+}} ymm2 = <-5.0E-1,u,-2.5E+0,-4.5E+0>
+; FMA3-NEXT: vmovapd {{.*#+}} ymm2 = [-5.0E-1,u,-2.5E+0,-4.5E+0]
; FMA3-NEXT: vmovapd %ymm2, %ymm3
; FMA3-NEXT: vfmadd213pd {{.*#+}} ymm3 = (ymm0 * ymm3) + ymm1
; FMA3-NEXT: vfnmadd213pd {{.*#+}} ymm2 = -(ymm0 * ymm2) + ymm1
@@ -154,7 +154,7 @@ define <4 x double> @negated_constant_v4f64_2fmas(<4 x double> %a, <4 x double>
;
; FMA4-LABEL: negated_constant_v4f64_2fmas:
; FMA4: # %bb.0:
-; FMA4-NEXT: vmovapd {{.*#+}} ymm2 = <-5.0E-1,u,-2.5E+0,-4.5E+0>
+; FMA4-NEXT: vmovapd {{.*#+}} ymm2 = [-5.0E-1,u,-2.5E+0,-4.5E+0]
; FMA4-NEXT: vfmaddpd {{.*#+}} ymm3 = (ymm0 * ymm2) + ymm1
; FMA4-NEXT: vfnmaddpd {{.*#+}} ymm0 = -(ymm0 * ymm2) + ymm1
; FMA4-NEXT: vaddpd %ymm0, %ymm3, %ymm0
diff --git a/llvm/test/CodeGen/X86/fold-int-pow2-with-fmul-or-fdiv.ll b/llvm/test/CodeGen/X86/fold-int-pow2-with-fmul-or-fdiv.ll
index abaefaee33ed607..7dbeba96a34cb57 100644
--- a/llvm/test/CodeGen/X86/fold-int-pow2-with-fmul-or-fdiv.ll
+++ b/llvm/test/CodeGen/X86/fold-int-pow2-with-fmul-or-fdiv.ll
@@ -1236,7 +1236,7 @@ define <2 x float> @fdiv_pow_shl_cnt_vec_with_expensive_cast(<2 x i64> %cnt) nou
; CHECK-SSE: # %bb.0:
; CHECK-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm0[0,2,2,3]
; CHECK-SSE-NEXT: pslld $23, %xmm1
-; CHECK-SSE-NEXT: movdqa {{.*#+}} xmm0 = <1065353216,1065353216,u,u>
+; CHECK-SSE-NEXT: movdqa {{.*#+}} xmm0 = [1065353216,1065353216,u,u]
; CHECK-SSE-NEXT: psubd %xmm1, %xmm0
; CHECK-SSE-NEXT: retq
;
diff --git a/llvm/test/CodeGen/X86/fpclamptosat_vec.ll b/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
index 42d6e8139c6aa32..933c546503048d1 100644
--- a/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
+++ b/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
@@ -1161,7 +1161,7 @@ define <2 x i16> @ustest_f64i16(<2 x double> %x) nounwind {
; SSE-LABEL: ustest_f64i16:
; SSE: # %bb.0: # %entry
; SSE-NEXT: cvttpd2dq %xmm0, %xmm0
-; SSE-NEXT: movdqa {{.*#+}} xmm1 = <65535,65535,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,u,u]
; SSE-NEXT: movdqa %xmm1, %xmm2
; SSE-NEXT: pcmpgtd %xmm0, %xmm2
; SSE-NEXT: pand %xmm2, %xmm0
@@ -3803,7 +3803,7 @@ define <2 x i16> @ustest_f64i16_mm(<2 x double> %x) nounwind {
; SSE-LABEL: ustest_f64i16_mm:
; SSE: # %bb.0: # %entry
; SSE-NEXT: cvttpd2dq %xmm0, %xmm0
-; SSE-NEXT: movdqa {{.*#+}} xmm1 = <65535,65535,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,u,u]
; SSE-NEXT: movdqa %xmm1, %xmm2
; SSE-NEXT: pcmpgtd %xmm0, %xmm2
; SSE-NEXT: pand %xmm2, %xmm0
diff --git a/llvm/test/CodeGen/X86/hoist-and-by-const-from-lshr-in-eqcmp-zero.ll b/llvm/test/CodeGen/X86/hoist-and-by-const-from-lshr-in-eqcmp-zero.ll
index d79aff5544b69d7..9121cf2d654a39f 100644
--- a/llvm/test/CodeGen/X86/hoist-and-by-const-from-lshr-in-eqcmp-zero.ll
+++ b/llvm/test/CodeGen/X86/hoist-and-by-const-from-lshr-in-eqcmp-zero.ll
@@ -690,7 +690,7 @@ define <4 x i1> @vec_4xi32_nonsplat_undef2_eq(<4 x i32> %x, <4 x i32> %y) nounwi
; X86-SSE2-LABEL: vec_4xi32_nonsplat_undef2_eq:
; X86-SSE2: # %bb.0:
; X86-SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[2,3,3,3,4,5,6,7]
-; X86-SSE2-NEXT: movdqa {{.*#+}} xmm2 = <1,1,u,1>
+; X86-SSE2-NEXT: movdqa {{.*#+}} xmm2 = [1,1,u,1]
; X86-SSE2-NEXT: movdqa %xmm2, %xmm4
; X86-SSE2-NEXT: psrld %xmm3, %xmm4
; X86-SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm1[0,1,1,1,4,5,6,7]
@@ -722,7 +722,7 @@ define <4 x i1> @vec_4xi32_nonsplat_undef2_eq(<4 x i32> %x, <4 x i32> %y) nounwi
; X64-SSE2-LABEL: vec_4xi32_nonsplat_undef2_eq:
; X64-SSE2: # %bb.0:
; X64-SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[2,3,3,3,4,5,6,7]
-; X64-SSE2-NEXT: movdqa {{.*#+}} xmm3 = <1,1,u,1>
+; X64-SSE2-NEXT: movdqa {{.*#+}} xmm3 = [1,1,u,1]
; X64-SSE2-NEXT: movdqa %xmm3, %xmm4
; X64-SSE2-NEXT: psrld %xmm2, %xmm4
; X64-SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm1[0,1,1,1,4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/icmp-abs-C-vec.ll b/llvm/test/CodeGen/X86/icmp-abs-C-vec.ll
index 1e50e5b81123381..7ad2bb712ae9369 100644
--- a/llvm/test/CodeGen/X86/icmp-abs-C-vec.ll
+++ b/llvm/test/CodeGen/X86/icmp-abs-C-vec.ll
@@ -944,7 +944,7 @@ define <4 x i1> @eq_or_to_abs_vec4x16(<4 x i16> %x) {
;
; SSE41-LABEL: eq_or_to_abs_vec4x16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u]
; SSE41-NEXT: pcmpeqw %xmm0, %xmm1
; SSE41-NEXT: pcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: por %xmm1, %xmm0
@@ -953,7 +953,7 @@ define <4 x i1> @eq_or_to_abs_vec4x16(<4 x i16> %x) {
;
; SSE2-LABEL: eq_or_to_abs_vec4x16:
; SSE2: # %bb.0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u]
; SSE2-NEXT: pcmpeqw %xmm0, %xmm1
; SSE2-NEXT: pcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
@@ -989,7 +989,7 @@ define <4 x i8> @eq_or_to_abs_vec4x8_sext(<4 x i8> %x) {
;
; SSE41-LABEL: eq_or_to_abs_vec4x8_sext:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pcmpeqb %xmm0, %xmm1
; SSE41-NEXT: pcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE41-NEXT: por %xmm1, %xmm0
@@ -997,7 +997,7 @@ define <4 x i8> @eq_or_to_abs_vec4x8_sext(<4 x i8> %x) {
;
; SSE2-LABEL: eq_or_to_abs_vec4x8_sext:
; SSE2: # %bb.0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE2-NEXT: pcmpeqb %xmm0, %xmm1
; SSE2-NEXT: pcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
@@ -1038,7 +1038,7 @@ define <4 x i1> @ne_and_to_abs_vec4x8(<4 x i8> %x) {
;
; SSE41-LABEL: ne_and_to_abs_vec4x8:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pcmpeqb %xmm0, %xmm1
; SSE41-NEXT: pcmpeqd %xmm2, %xmm2
; SSE41-NEXT: pcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
@@ -1049,7 +1049,7 @@ define <4 x i1> @ne_and_to_abs_vec4x8(<4 x i8> %x) {
;
; SSE2-LABEL: ne_and_to_abs_vec4x8:
; SSE2: # %bb.0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE2-NEXT: pcmpeqb %xmm0, %xmm1
; SSE2-NEXT: pcmpeqb {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT: por %xmm1, %xmm0
@@ -1093,7 +1093,7 @@ define <4 x i16> @ne_and_to_abs_vec4x16_sext(<4 x i16> %x) {
;
; SSE41-LABEL: ne_and_to_abs_vec4x16_sext:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u]
; SSE41-NEXT: pcmpeqw %xmm0, %xmm1
; SSE41-NEXT: pcmpeqd %xmm2, %xmm2
; SSE41-NEXT: pcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
@@ -1103,7 +1103,7 @@ define <4 x i16> @ne_and_to_abs_vec4x16_sext(<4 x i16> %x) {
;
; SSE2-LABEL: ne_and_to_abs_vec4x16_sext:
; SSE2: # %bb.0:
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <88,88,88,88,u,u,u,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [88,88,88,88,u,u,u,u]
; SSE2-NEXT: pcmpeqw %xmm0, %xmm1
; SSE2-NEXT: pcmpeqd %xmm2, %xmm2
; SSE2-NEXT: pcmpeqw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
diff --git a/llvm/test/CodeGen/X86/insert-into-constant-vector.ll b/llvm/test/CodeGen/X86/insert-into-constant-vector.ll
index 07787f3851bd907..418d632480328fb 100644
--- a/llvm/test/CodeGen/X86/insert-into-constant-vector.ll
+++ b/llvm/test/CodeGen/X86/insert-into-constant-vector.ll
@@ -27,25 +27,25 @@ define <16 x i8> @elt0_v16i8(i8 %x) {
;
; X86-SSE4-LABEL: elt0_v16i8:
; X86-SSE4: # %bb.0:
-; X86-SSE4-NEXT: movdqa {{.*#+}} xmm0 = <u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15>
+; X86-SSE4-NEXT: movdqa {{.*#+}} xmm0 = [u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
; X86-SSE4-NEXT: pinsrb $0, {{[0-9]+}}(%esp), %xmm0
; X86-SSE4-NEXT: retl
;
; X64-SSE4-LABEL: elt0_v16i8:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = <u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15>
+; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = [u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
; X64-SSE4-NEXT: pinsrb $0, %edi, %xmm0
; X64-SSE4-NEXT: retq
;
; X86-AVX-LABEL: elt0_v16i8:
; X86-AVX: # %bb.0:
-; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15>
+; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
; X86-AVX-NEXT: vpinsrb $0, {{[0-9]+}}(%esp), %xmm0, %xmm0
; X86-AVX-NEXT: retl
;
; X64-AVX-LABEL: elt0_v16i8:
; X64-AVX: # %bb.0:
-; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15>
+; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [u,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
; X64-AVX-NEXT: vpinsrb $0, %edi, %xmm0, %xmm0
; X64-AVX-NEXT: retq
%ins = insertelement <16 x i8> <i8 42, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, i8 %x, i32 0
@@ -55,25 +55,25 @@ define <16 x i8> @elt0_v16i8(i8 %x) {
define <8 x i16> @elt5_v8i16(i16 %x) {
; X86-SSE-LABEL: elt5_v8i16:
; X86-SSE: # %bb.0:
-; X86-SSE-NEXT: movdqa {{.*#+}} xmm0 = <42,1,2,3,4,u,6,7>
+; X86-SSE-NEXT: movdqa {{.*#+}} xmm0 = [42,1,2,3,4,u,6,7]
; X86-SSE-NEXT: pinsrw $5, {{[0-9]+}}(%esp), %xmm0
; X86-SSE-NEXT: retl
;
; X64-SSE-LABEL: elt5_v8i16:
; X64-SSE: # %bb.0:
-; X64-SSE-NEXT: movdqa {{.*#+}} xmm0 = <42,1,2,3,4,u,6,7>
+; X64-SSE-NEXT: movdqa {{.*#+}} xmm0 = [42,1,2,3,4,u,6,7]
; X64-SSE-NEXT: pinsrw $5, %edi, %xmm0
; X64-SSE-NEXT: retq
;
; X86-AVX-LABEL: elt5_v8i16:
; X86-AVX: # %bb.0:
-; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <42,1,2,3,4,u,6,7>
+; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [42,1,2,3,4,u,6,7]
; X86-AVX-NEXT: vpinsrw $5, {{[0-9]+}}(%esp), %xmm0, %xmm0
; X86-AVX-NEXT: retl
;
; X64-AVX-LABEL: elt5_v8i16:
; X64-AVX: # %bb.0:
-; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <42,1,2,3,4,u,6,7>
+; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [42,1,2,3,4,u,6,7]
; X64-AVX-NEXT: vpinsrw $5, %edi, %xmm0, %xmm0
; X64-AVX-NEXT: retq
%ins = insertelement <8 x i16> <i16 42, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, i16 %x, i32 5
@@ -84,7 +84,7 @@ define <4 x i32> @elt3_v4i32(i32 %x) {
; X86-SSE2-LABEL: elt3_v4i32:
; X86-SSE2: # %bb.0:
; X86-SSE2-NEXT: movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
-; X86-SSE2-NEXT: movaps {{.*#+}} xmm0 = <42,1,2,u>
+; X86-SSE2-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,u]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,3]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,0]
; X86-SSE2-NEXT: retl
@@ -92,32 +92,32 @@ define <4 x i32> @elt3_v4i32(i32 %x) {
; X64-SSE2-LABEL: elt3_v4i32:
; X64-SSE2: # %bb.0:
; X64-SSE2-NEXT: movd %edi, %xmm1
-; X64-SSE2-NEXT: movaps {{.*#+}} xmm0 = <42,1,2,u>
+; X64-SSE2-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,u]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,3]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,0]
; X64-SSE2-NEXT: retq
;
; X86-SSE4-LABEL: elt3_v4i32:
; X86-SSE4: # %bb.0:
-; X86-SSE4-NEXT: movdqa {{.*#+}} xmm0 = <42,1,2,u>
+; X86-SSE4-NEXT: movdqa {{.*#+}} xmm0 = [42,1,2,u]
; X86-SSE4-NEXT: pinsrd $3, {{[0-9]+}}(%esp), %xmm0
; X86-SSE4-NEXT: retl
;
; X64-SSE4-LABEL: elt3_v4i32:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = <42,1,2,u>
+; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = [42,1,2,u]
; X64-SSE4-NEXT: pinsrd $3, %edi, %xmm0
; X64-SSE4-NEXT: retq
;
; X86-AVX-LABEL: elt3_v4i32:
; X86-AVX: # %bb.0:
-; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <42,1,2,u>
+; X86-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [42,1,2,u]
; X86-AVX-NEXT: vpinsrd $3, {{[0-9]+}}(%esp), %xmm0, %xmm0
; X86-AVX-NEXT: retl
;
; X64-AVX-LABEL: elt3_v4i32:
; X64-AVX: # %bb.0:
-; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <42,1,2,u>
+; X64-AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [42,1,2,u]
; X64-AVX-NEXT: vpinsrd $3, %edi, %xmm0, %xmm0
; X64-AVX-NEXT: retq
%ins = insertelement <4 x i32> <i32 42, i32 1, i32 2, i32 3>, i32 %x, i32 3
@@ -134,13 +134,13 @@ define <2 x i64> @elt0_v2i64(i64 %x) {
; X64-SSE2-LABEL: elt0_v2i64:
; X64-SSE2: # %bb.0:
; X64-SSE2-NEXT: movq %rdi, %xmm1
-; X64-SSE2-NEXT: movapd {{.*#+}} xmm0 = <u,1>
+; X64-SSE2-NEXT: movapd {{.*#+}} xmm0 = [u,1]
; X64-SSE2-NEXT: movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
; X64-SSE2-NEXT: retq
;
; X64-SSE4-LABEL: elt0_v2i64:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = <u,1>
+; X64-SSE4-NEXT: movdqa {{.*#+}} xmm0 = [u,1]
; X64-SSE4-NEXT: pinsrq $0, %rdi, %xmm0
; X64-SSE4-NEXT: retq
;
@@ -175,7 +175,7 @@ define <2 x i64> @elt0_v2i64(i64 %x) {
define <4 x float> @elt1_v4f32(float %x) {
; X86-SSE2-LABEL: elt1_v4f32:
; X86-SSE2: # %bb.0:
-; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X86-SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-SSE2-NEXT: movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[2,0],xmm1[2,3]
@@ -183,33 +183,33 @@ define <4 x float> @elt1_v4f32(float %x) {
;
; X64-SSE2-LABEL: elt1_v4f32:
; X64-SSE2: # %bb.0:
-; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X64-SSE2-NEXT: movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[2,0],xmm1[2,3]
; X64-SSE2-NEXT: retq
;
; X86-SSE4-LABEL: elt1_v4f32:
; X86-SSE4: # %bb.0:
-; X86-SSE4-NEXT: movaps {{.*#+}} xmm0 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X86-SSE4-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,u,2.0E+0,3.0E+0]
; X86-SSE4-NEXT: insertps {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[2,3]
; X86-SSE4-NEXT: retl
;
; X64-SSE4-LABEL: elt1_v4f32:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movaps {{.*#+}} xmm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X64-SSE4-NEXT: movaps {{.*#+}} xmm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X64-SSE4-NEXT: insertps {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[2,3]
; X64-SSE4-NEXT: movaps %xmm1, %xmm0
; X64-SSE4-NEXT: retq
;
; X86-AVX-LABEL: elt1_v4f32:
; X86-AVX: # %bb.0:
-; X86-AVX-NEXT: vmovaps {{.*#+}} xmm0 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X86-AVX-NEXT: vmovaps {{.*#+}} xmm0 = [4.2E+1,u,2.0E+0,3.0E+0]
; X86-AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[2,3]
; X86-AVX-NEXT: retl
;
; X64-AVX-LABEL: elt1_v4f32:
; X64-AVX: # %bb.0:
-; X64-AVX-NEXT: vmovaps {{.*#+}} xmm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X64-AVX-NEXT: vmovaps {{.*#+}} xmm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X64-AVX-NEXT: vinsertps {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[2,3]
; X64-AVX-NEXT: retq
%ins = insertelement <4 x float> <float 42.0, float 1.0, float 2.0, float 3.0>, float %x, i32 1
@@ -219,13 +219,13 @@ define <4 x float> @elt1_v4f32(float %x) {
define <2 x double> @elt1_v2f64(double %x) {
; X86-SSE-LABEL: elt1_v2f64:
; X86-SSE: # %bb.0:
-; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = <4.2E+1,u>
+; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,u]
; X86-SSE-NEXT: movhps {{.*#+}} xmm0 = xmm0[0,1],mem[0,1]
; X86-SSE-NEXT: retl
;
; X64-SSE-LABEL: elt1_v2f64:
; X64-SSE: # %bb.0:
-; X64-SSE-NEXT: movaps {{.*#+}} xmm1 = <4.2E+1,u>
+; X64-SSE-NEXT: movaps {{.*#+}} xmm1 = [4.2E+1,u]
; X64-SSE-NEXT: movlhps {{.*#+}} xmm1 = xmm1[0],xmm0[0]
; X64-SSE-NEXT: movaps %xmm1, %xmm0
; X64-SSE-NEXT: retq
@@ -251,7 +251,7 @@ define <8 x i32> @elt7_v8i32(i32 %x) {
; X86-SSE2-LABEL: elt7_v8i32:
; X86-SSE2: # %bb.0:
; X86-SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4,5,6,u>
+; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4,5,6,u]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,0]
; X86-SSE2-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,3]
@@ -260,7 +260,7 @@ define <8 x i32> @elt7_v8i32(i32 %x) {
; X64-SSE2-LABEL: elt7_v8i32:
; X64-SSE2: # %bb.0:
; X64-SSE2-NEXT: movd %edi, %xmm0
-; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4,5,6,u>
+; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4,5,6,u]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,0]
; X64-SSE2-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,3]
@@ -268,14 +268,14 @@ define <8 x i32> @elt7_v8i32(i32 %x) {
;
; X86-SSE4-LABEL: elt7_v8i32:
; X86-SSE4: # %bb.0:
-; X86-SSE4-NEXT: movdqa {{.*#+}} xmm1 = <4,5,6,u>
+; X86-SSE4-NEXT: movdqa {{.*#+}} xmm1 = [4,5,6,u]
; X86-SSE4-NEXT: pinsrd $3, {{[0-9]+}}(%esp), %xmm1
; X86-SSE4-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,3]
; X86-SSE4-NEXT: retl
;
; X64-SSE4-LABEL: elt7_v8i32:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movdqa {{.*#+}} xmm1 = <4,5,6,u>
+; X64-SSE4-NEXT: movdqa {{.*#+}} xmm1 = [4,5,6,u]
; X64-SSE4-NEXT: pinsrd $3, %edi, %xmm1
; X64-SSE4-NEXT: movaps {{.*#+}} xmm0 = [42,1,2,3]
; X64-SSE4-NEXT: retq
@@ -314,7 +314,7 @@ define <8 x i32> @elt7_v8i32(i32 %x) {
define <8 x float> @elt6_v8f32(float %x) {
; X86-SSE2-LABEL: elt6_v8f32:
; X86-SSE2: # %bb.0:
-; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4.0E+0,5.0E+0,u,7.0E+0>
+; X86-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4.0E+0,5.0E+0,u,7.0E+0]
; X86-SSE2-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,0],xmm1[3,0]
; X86-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[0,2]
@@ -323,7 +323,7 @@ define <8 x float> @elt6_v8f32(float %x) {
;
; X64-SSE2-LABEL: elt6_v8f32:
; X64-SSE2: # %bb.0:
-; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = <4.0E+0,5.0E+0,u,7.0E+0>
+; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = [4.0E+0,5.0E+0,u,7.0E+0]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,0],xmm1[3,0]
; X64-SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[0,2]
; X64-SSE2-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,1.0E+0,2.0E+0,3.0E+0]
@@ -331,14 +331,14 @@ define <8 x float> @elt6_v8f32(float %x) {
;
; X86-SSE4-LABEL: elt6_v8f32:
; X86-SSE4: # %bb.0:
-; X86-SSE4-NEXT: movaps {{.*#+}} xmm1 = <4.0E+0,5.0E+0,u,7.0E+0>
+; X86-SSE4-NEXT: movaps {{.*#+}} xmm1 = [4.0E+0,5.0E+0,u,7.0E+0]
; X86-SSE4-NEXT: insertps {{.*#+}} xmm1 = xmm1[0,1],mem[0],xmm1[3]
; X86-SSE4-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,1.0E+0,2.0E+0,3.0E+0]
; X86-SSE4-NEXT: retl
;
; X64-SSE4-LABEL: elt6_v8f32:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movaps {{.*#+}} xmm1 = <4.0E+0,5.0E+0,u,7.0E+0>
+; X64-SSE4-NEXT: movaps {{.*#+}} xmm1 = [4.0E+0,5.0E+0,u,7.0E+0]
; X64-SSE4-NEXT: insertps {{.*#+}} xmm1 = xmm1[0,1],xmm0[0],xmm1[3]
; X64-SSE4-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,1.0E+0,2.0E+0,3.0E+0]
; X64-SSE4-NEXT: retq
@@ -385,7 +385,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
; X64-SSE2-LABEL: elt5_v8i64:
; X64-SSE2: # %bb.0:
; X64-SSE2-NEXT: movq %rdi, %xmm0
-; X64-SSE2-NEXT: movdqa {{.*#+}} xmm2 = <4,u>
+; X64-SSE2-NEXT: movdqa {{.*#+}} xmm2 = [4,u]
; X64-SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm0[0]
; X64-SSE2-NEXT: movaps {{.*#+}} xmm0 = [42,1]
; X64-SSE2-NEXT: movaps {{.*#+}} xmm1 = [2,3]
@@ -394,7 +394,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
;
; X64-SSE4-LABEL: elt5_v8i64:
; X64-SSE4: # %bb.0:
-; X64-SSE4-NEXT: movdqa {{.*#+}} xmm2 = <4,u>
+; X64-SSE4-NEXT: movdqa {{.*#+}} xmm2 = [4,u]
; X64-SSE4-NEXT: pinsrq $1, %rdi, %xmm2
; X64-SSE4-NEXT: movaps {{.*#+}} xmm0 = [42,1]
; X64-SSE4-NEXT: movaps {{.*#+}} xmm1 = [2,3]
@@ -412,7 +412,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
;
; X64-AVX1-LABEL: elt5_v8i64:
; X64-AVX1: # %bb.0:
-; X64-AVX1-NEXT: vmovdqa {{.*#+}} ymm0 = <4,u,6,7>
+; X64-AVX1-NEXT: vmovdqa {{.*#+}} ymm0 = [4,u,6,7]
; X64-AVX1-NEXT: vpinsrq $1, %rdi, %xmm0, %xmm1
; X64-AVX1-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm0[4,5,6,7]
; X64-AVX1-NEXT: vmovaps {{.*#+}} ymm0 = [42,1,2,3]
@@ -429,7 +429,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
;
; X64-AVX2-LABEL: elt5_v8i64:
; X64-AVX2: # %bb.0:
-; X64-AVX2-NEXT: vmovdqa {{.*#+}} ymm0 = <4,u,6,7>
+; X64-AVX2-NEXT: vmovdqa {{.*#+}} ymm0 = [4,u,6,7]
; X64-AVX2-NEXT: vpinsrq $1, %rdi, %xmm0, %xmm1
; X64-AVX2-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm0[4,5,6,7]
; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm0 = [42,1,2,3]
@@ -449,7 +449,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
; X64-AVX512F: # %bb.0:
; X64-AVX512F-NEXT: vmovq %rdi, %xmm1
; X64-AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,4,8,6,7]
-; X64-AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <42,1,2,3,4,u,6,7>
+; X64-AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [42,1,2,3,4,u,6,7]
; X64-AVX512F-NEXT: vpermt2q %zmm1, %zmm2, %zmm0
; X64-AVX512F-NEXT: retq
%ins = insertelement <8 x i64> <i64 42, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7>, i64 %x, i32 5
@@ -459,7 +459,7 @@ define <8 x i64> @elt5_v8i64(i64 %x) {
define <8 x double> @elt1_v8f64(double %x) {
; X86-SSE-LABEL: elt1_v8f64:
; X86-SSE: # %bb.0:
-; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = <4.2E+1,u>
+; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = [4.2E+1,u]
; X86-SSE-NEXT: movhps {{.*#+}} xmm0 = xmm0[0,1],mem[0,1]
; X86-SSE-NEXT: movaps {{.*#+}} xmm1 = [2.0E+0,3.0E+0]
; X86-SSE-NEXT: movaps {{.*#+}} xmm2 = [4.0E+0,5.0E+0]
@@ -468,7 +468,7 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X64-SSE-LABEL: elt1_v8f64:
; X64-SSE: # %bb.0:
-; X64-SSE-NEXT: movaps {{.*#+}} xmm4 = <4.2E+1,u>
+; X64-SSE-NEXT: movaps {{.*#+}} xmm4 = [4.2E+1,u]
; X64-SSE-NEXT: movlhps {{.*#+}} xmm4 = xmm4[0],xmm0[0]
; X64-SSE-NEXT: movaps {{.*#+}} xmm1 = [2.0E+0,3.0E+0]
; X64-SSE-NEXT: movaps {{.*#+}} xmm2 = [4.0E+0,5.0E+0]
@@ -478,7 +478,7 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X86-AVX1-LABEL: elt1_v8f64:
; X86-AVX1: # %bb.0:
-; X86-AVX1-NEXT: vmovaps {{.*#+}} ymm0 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X86-AVX1-NEXT: vmovaps {{.*#+}} ymm0 = [4.2E+1,u,2.0E+0,3.0E+0]
; X86-AVX1-NEXT: vmovhps {{.*#+}} xmm1 = xmm0[0,1],mem[0,1]
; X86-AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
; X86-AVX1-NEXT: vmovaps {{.*#+}} ymm1 = [4.0E+0,5.0E+0,6.0E+0,7.0E+0]
@@ -486,7 +486,7 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X64-AVX1-LABEL: elt1_v8f64:
; X64-AVX1: # %bb.0:
-; X64-AVX1-NEXT: vmovaps {{.*#+}} ymm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X64-AVX1-NEXT: vmovaps {{.*#+}} ymm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X64-AVX1-NEXT: vmovlhps {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; X64-AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; X64-AVX1-NEXT: vmovaps {{.*#+}} ymm1 = [4.0E+0,5.0E+0,6.0E+0,7.0E+0]
@@ -494,7 +494,7 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X86-AVX2-LABEL: elt1_v8f64:
; X86-AVX2: # %bb.0:
-; X86-AVX2-NEXT: vmovaps {{.*#+}} ymm0 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X86-AVX2-NEXT: vmovaps {{.*#+}} ymm0 = [4.2E+1,u,2.0E+0,3.0E+0]
; X86-AVX2-NEXT: vmovhps {{.*#+}} xmm1 = xmm0[0,1],mem[0,1]
; X86-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
; X86-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [4.0E+0,5.0E+0,6.0E+0,7.0E+0]
@@ -502,7 +502,7 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X64-AVX2-LABEL: elt1_v8f64:
; X64-AVX2: # %bb.0:
-; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = <4.2E+1,u,2.0E+0,3.0E+0>
+; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [4.2E+1,u,2.0E+0,3.0E+0]
; X64-AVX2-NEXT: vmovlhps {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; X64-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [4.0E+0,5.0E+0,6.0E+0,7.0E+0]
@@ -510,14 +510,14 @@ define <8 x double> @elt1_v8f64(double %x) {
;
; X86-AVX512F-LABEL: elt1_v8f64:
; X86-AVX512F: # %bb.0:
-; X86-AVX512F-NEXT: vmovaps {{.*#+}} zmm0 = <4.2E+1,u,2.0E+0,3.0E+0,4.0E+0,5.0E+0,6.0E+0,7.0E+0>
+; X86-AVX512F-NEXT: vmovaps {{.*#+}} zmm0 = [4.2E+1,u,2.0E+0,3.0E+0,4.0E+0,5.0E+0,6.0E+0,7.0E+0]
; X86-AVX512F-NEXT: vmovhps {{.*#+}} xmm1 = xmm0[0,1],mem[0,1]
; X86-AVX512F-NEXT: vinsertf32x4 $0, %xmm1, %zmm0, %zmm0
; X86-AVX512F-NEXT: retl
;
; X64-AVX512F-LABEL: elt1_v8f64:
; X64-AVX512F: # %bb.0:
-; X64-AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <4.2E+1,u,2.0E+0,3.0E+0,4.0E+0,5.0E+0,6.0E+0,7.0E+0>
+; X64-AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [4.2E+1,u,2.0E+0,3.0E+0,4.0E+0,5.0E+0,6.0E+0,7.0E+0]
; X64-AVX512F-NEXT: vmovlhps {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; X64-AVX512F-NEXT: vinsertf32x4 $0, %xmm0, %zmm1, %zmm0
; X64-AVX512F-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/masked_store_trunc.ll b/llvm/test/CodeGen/X86/masked_store_trunc.ll
index 11803e32ad437ed..f78646afccb3453 100644
--- a/llvm/test/CodeGen/X86/masked_store_trunc.ll
+++ b/llvm/test/CodeGen/X86/masked_store_trunc.ll
@@ -1395,7 +1395,7 @@ define void @truncstore_v4i64_v4i8(<4 x i64> %x, ptr %p, <4 x i32> %mask) {
; SSE4-LABEL: truncstore_v4i64_v4i8:
; SSE4: # %bb.0:
; SSE4-NEXT: pxor %xmm3, %xmm3
-; SSE4-NEXT: movdqa {{.*#+}} xmm4 = <0,8,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE4-NEXT: movdqa {{.*#+}} xmm4 = [0,8,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE4-NEXT: pshufb %xmm4, %xmm1
; SSE4-NEXT: pshufb %xmm4, %xmm0
; SSE4-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
diff --git a/llvm/test/CodeGen/X86/matrix-multiply.ll b/llvm/test/CodeGen/X86/matrix-multiply.ll
index c8c5afbd579df4d..17a82ce5f6209f4 100644
--- a/llvm/test/CodeGen/X86/matrix-multiply.ll
+++ b/llvm/test/CodeGen/X86/matrix-multiply.ll
@@ -330,7 +330,7 @@ define <9 x float> @test_mul3x3_f32(<9 x float> %a0, <9 x float> %a1) nounwind {
; AVX2-NEXT: vmulss %xmm3, %xmm8, %xmm3
; AVX2-NEXT: vaddss %xmm3, %xmm2, %xmm2
; AVX2-NEXT: vinsertf128 $1, %xmm4, %ymm1, %ymm1
-; AVX2-NEXT: vmovaps {{.*#+}} ymm3 = <0,1,2,4,5,6,u,u>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm3 = [0,1,2,4,5,6,u,u]
; AVX2-NEXT: vpermps %ymm1, %ymm3, %ymm1
; AVX2-NEXT: vbroadcastsd %xmm0, %ymm0
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
@@ -394,7 +394,7 @@ define <9 x float> @test_mul3x3_f32(<9 x float> %a0, <9 x float> %a1) nounwind {
; AVX512F-NEXT: vaddss %xmm1, %xmm2, %xmm1
; AVX512F-NEXT: vinsertps {{.*#+}} xmm1 = xmm0[0,1],xmm1[0],xmm0[3]
; AVX512F-NEXT: vinsertf32x4 $1, %xmm7, %zmm6, %zmm2
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm0 = <0,1,2,4,5,6,16,17,18,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm0 = [0,1,2,4,5,6,16,17,18,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2ps %zmm1, %zmm2, %zmm0
; AVX512F-NEXT: retq
;
@@ -453,7 +453,7 @@ define <9 x float> @test_mul3x3_f32(<9 x float> %a0, <9 x float> %a1) nounwind {
; AVX512VL-NEXT: vaddss %xmm1, %xmm2, %xmm1
; AVX512VL-NEXT: vinsertps {{.*#+}} xmm1 = xmm0[0,1],xmm1[0],xmm0[3]
; AVX512VL-NEXT: vinsertf32x4 $1, %xmm5, %zmm3, %zmm2
-; AVX512VL-NEXT: vmovaps {{.*#+}} zmm0 = <0,1,2,4,5,6,16,17,18,u,u,u,u,u,u,u>
+; AVX512VL-NEXT: vmovaps {{.*#+}} zmm0 = [0,1,2,4,5,6,16,17,18,u,u,u,u,u,u,u]
; AVX512VL-NEXT: vpermi2ps %zmm1, %zmm2, %zmm0
; AVX512VL-NEXT: retq
entry:
diff --git a/llvm/test/CodeGen/X86/oddshuffles.ll b/llvm/test/CodeGen/X86/oddshuffles.ll
index f0fb89496aa1ed2..de192aa038e5b31 100644
--- a/llvm/test/CodeGen/X86/oddshuffles.ll
+++ b/llvm/test/CodeGen/X86/oddshuffles.ll
@@ -527,7 +527,7 @@ define void @v12i32(<8 x i32> %a, <8 x i32> %b, ptr %p) nounwind {
;
; AVX2-SLOW-LABEL: v12i32:
; AVX2-SLOW: # %bb.0:
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = <0,4,u,1,5,u,2,6>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = [0,4,u,1,5,u,2,6]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vbroadcastsd %xmm1, %ymm3
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
@@ -542,7 +542,7 @@ define void @v12i32(<8 x i32> %a, <8 x i32> %b, ptr %p) nounwind {
;
; AVX2-FAST-ALL-LABEL: v12i32:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <0,4,u,1,5,u,2,6>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [0,4,u,1,5,u,2,6]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm2
; AVX2-FAST-ALL-NEXT: vbroadcastsd %xmm1, %ymm3
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
@@ -557,7 +557,7 @@ define void @v12i32(<8 x i32> %a, <8 x i32> %b, ptr %p) nounwind {
;
; AVX2-FAST-PERLANE-LABEL: v12i32:
; AVX2-FAST-PERLANE: # %bb.0:
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = <0,4,u,1,5,u,2,6>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = [0,4,u,1,5,u,2,6]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd %xmm1, %ymm3
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
@@ -1308,7 +1308,7 @@ define void @interleave_24i16_in(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,0,1,u,u,u,u,2,3,u,u,u,u,4,5,u,u,22,23,u,u,u,u,24,25,u,u,u,u,26,27]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6],ymm3[7],ymm4[8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14],ymm3[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-SLOW-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -1327,7 +1327,7 @@ define void @interleave_24i16_in(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-FAST-ALL-NEXT: vmovdqu (%rdx), %xmm1
; AVX2-FAST-ALL-NEXT: vmovdqu (%rcx), %xmm2
; AVX2-FAST-ALL-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm3
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-ALL-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm5 = [0,4,1,5,1,5,2,6]
; AVX2-FAST-ALL-NEXT: vpermd %ymm3, %ymm5, %ymm3
@@ -1353,7 +1353,7 @@ define void @interleave_24i16_in(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,0,1,u,u,u,u,2,3,u,u,u,u,4,5,u,u,22,23,u,u,u,u,24,25,u,u,u,u,26,27]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6],ymm3[7],ymm4[8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14],ymm3[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -1507,18 +1507,18 @@ define void @interleave_24i32_out(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-SLOW-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-SLOW-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-SLOW-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm1 = ymm2[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
@@ -1537,21 +1537,21 @@ define void @interleave_24i32_out(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-FAST-ALL-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-ALL-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-ALL-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-ALL-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-FAST-ALL-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-ALL-NEXT: vbroadcastf128 {{.*#+}} ymm5 = [0,1,4,7,0,1,4,7]
; AVX2-FAST-ALL-NEXT: # ymm5 = mem[0,1,0,1]
; AVX2-FAST-ALL-NEXT: vpermps %ymm2, %ymm5, %ymm2
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4],ymm2[5,6,7]
; AVX2-FAST-ALL-NEXT: vmovups %ymm3, (%rsi)
@@ -1568,18 +1568,18 @@ define void @interleave_24i32_out(ptr %p, ptr %q1, ptr %q2, ptr %q3) nounwind {
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-PERLANE-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-PERLANE-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm1 = ymm2[0,1,0,3,4,5,4,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
diff --git a/llvm/test/CodeGen/X86/omit-urem-of-power-of-two-or-zero-when-comparing-with-zero.ll b/llvm/test/CodeGen/X86/omit-urem-of-power-of-two-or-zero-when-comparing-with-zero.ll
index 83dd0b5b9dbfd91..5052a08177d7fc3 100644
--- a/llvm/test/CodeGen/X86/omit-urem-of-power-of-two-or-zero-when-comparing-with-zero.ll
+++ b/llvm/test/CodeGen/X86/omit-urem-of-power-of-two-or-zero-when-comparing-with-zero.ll
@@ -149,7 +149,7 @@ define <4 x i1> @p5_vector_urem_by_const__nonsplat(<4 x i32> %x, <4 x i32> %y) {
; SSE4: # %bb.0:
; SSE4-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE4-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; SSE4-NEXT: movdqa {{.*#+}} xmm1 = <1,u,2147483648,u>
+; SSE4-NEXT: movdqa {{.*#+}} xmm1 = [1,u,2147483648,u]
; SSE4-NEXT: pmuludq %xmm0, %xmm1
; SSE4-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; SSE4-NEXT: psrlq $32, %xmm1
diff --git a/llvm/test/CodeGen/X86/peephole-fold-movsd.ll b/llvm/test/CodeGen/X86/peephole-fold-movsd.ll
index b44d0ff6dafd15a..c0a6e00ec695e4c 100644
--- a/llvm/test/CodeGen/X86/peephole-fold-movsd.ll
+++ b/llvm/test/CodeGen/X86/peephole-fold-movsd.ll
@@ -18,7 +18,7 @@ define dso_local void @foo1(double %a.coerce0, double %a.coerce1, double %b.coer
; CHECK-NEXT: movq %rsp, %rdi
; CHECK-NEXT: callq foo3 at PLT
; CHECK-NEXT: movsd {{.*#+}} xmm0 = mem[0],zero
-; CHECK-NEXT: movapd {{.*#+}} xmm1 = <1.0E+0,u>
+; CHECK-NEXT: movapd {{.*#+}} xmm1 = [1.0E+0,u]
; CHECK-NEXT: movhpd {{.*#+}} xmm1 = xmm1[0],mem[0]
; CHECK-NEXT: addpd %xmm0, %xmm1
; CHECK-NEXT: movapd %xmm1, g(%rip)
diff --git a/llvm/test/CodeGen/X86/pr63108.ll b/llvm/test/CodeGen/X86/pr63108.ll
index 6b0543315d8670f..38e45595a8846e4 100644
--- a/llvm/test/CodeGen/X86/pr63108.ll
+++ b/llvm/test/CodeGen/X86/pr63108.ll
@@ -11,7 +11,7 @@ define i32 @PR63108() {
; SSE-NEXT: testb %al, %al
; SSE-NEXT: je .LBB0_2
; SSE-NEXT: # %bb.1:
-; SSE-NEXT: movdqa {{.*#+}} xmm0 = <251,223,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm0 = [251,223,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE-NEXT: jmp .LBB0_5
; SSE-NEXT: .LBB0_2: # %vector.body.preheader
; SSE-NEXT: pxor %xmm0, %xmm0
diff --git a/llvm/test/CodeGen/X86/psubus.ll b/llvm/test/CodeGen/X86/psubus.ll
index d2da4c0f3e86ad9..4f7f44d1d7c149b 100644
--- a/llvm/test/CodeGen/X86/psubus.ll
+++ b/llvm/test/CodeGen/X86/psubus.ll
@@ -208,7 +208,7 @@ define <4 x i32> @usubsat_custom(<4 x i32> %x) nounwind {
;
; SSE41-LABEL: usubsat_custom:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <2147483648,2147483648,2147483648,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [2147483648,2147483648,2147483648,u]
; SSE41-NEXT: pmaxud %xmm1, %xmm0
; SSE41-NEXT: psubd %xmm1, %xmm0
; SSE41-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/sext-vsetcc.ll b/llvm/test/CodeGen/X86/sext-vsetcc.ll
index acf123bde41f083..839c972b23dbab3 100644
--- a/llvm/test/CodeGen/X86/sext-vsetcc.ll
+++ b/llvm/test/CodeGen/X86/sext-vsetcc.ll
@@ -216,7 +216,7 @@ define <4 x i32> @cmp_ult_load_const(ptr %x) nounwind {
; SSE-LABEL: cmp_ult_load_const:
; SSE: # %bb.0:
; SSE-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; SSE-NEXT: movdqa {{.*#+}} xmm1 = <42,214,0,255,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm1 = [42,214,0,255,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE-NEXT: pmaxub %xmm0, %xmm1
; SSE-NEXT: pcmpeqb %xmm0, %xmm1
; SSE-NEXT: punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
@@ -243,7 +243,7 @@ define <3 x i32> @cmp_ult_load_const_bad_type(ptr %x) nounwind {
; SSE-LABEL: cmp_ult_load_const_bad_type:
; SSE: # %bb.0:
; SSE-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; SSE-NEXT: movdqa {{.*#+}} xmm1 = <42,214,0,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm1 = [42,214,0,u,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE-NEXT: pmaxub %xmm0, %xmm1
; SSE-NEXT: pcmpeqb %xmm0, %xmm1
; SSE-NEXT: punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
diff --git a/llvm/test/CodeGen/X86/shrink_vmul.ll b/llvm/test/CodeGen/X86/shrink_vmul.ll
index 3d80c1554b6c3c2..8dd06e0d848d328 100644
--- a/llvm/test/CodeGen/X86/shrink_vmul.ll
+++ b/llvm/test/CodeGen/X86/shrink_vmul.ll
@@ -1745,7 +1745,7 @@ define void @mul_2xi16_varconst1(ptr nocapture readonly %a, i64 %index) {
; X86-SSE-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X86-SSE-NEXT: movl c, %edx
; X86-SSE-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = <0,65535,u,u,u,u,u,u>
+; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,u,u,u,u,u,u]
; X86-SSE-NEXT: movdqa %xmm0, %xmm2
; X86-SSE-NEXT: pmulhuw %xmm1, %xmm2
; X86-SSE-NEXT: pmullw %xmm1, %xmm0
@@ -1768,7 +1768,7 @@ define void @mul_2xi16_varconst1(ptr nocapture readonly %a, i64 %index) {
; X64-SSE: # %bb.0: # %entry
; X64-SSE-NEXT: movq c(%rip), %rax
; X64-SSE-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; X64-SSE-NEXT: movdqa {{.*#+}} xmm1 = <0,65535,u,u,u,u,u,u>
+; X64-SSE-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,u,u,u,u,u,u]
; X64-SSE-NEXT: movdqa %xmm0, %xmm2
; X64-SSE-NEXT: pmulhuw %xmm1, %xmm2
; X64-SSE-NEXT: pmullw %xmm1, %xmm0
diff --git a/llvm/test/CodeGen/X86/sink-addsub-of-const.ll b/llvm/test/CodeGen/X86/sink-addsub-of-const.ll
index 9f08cc4cba32c93..cc46f04b8dabeee 100644
--- a/llvm/test/CodeGen/X86/sink-addsub-of-const.ll
+++ b/llvm/test/CodeGen/X86/sink-addsub-of-const.ll
@@ -448,7 +448,7 @@ define <4 x i32> @vec_sink_sub_of_const_to_sub2(<4 x i32> %a, <4 x i32> %b) {
define <4 x i32> @vec_sink_sub_from_const_to_sub(<4 x i32> %a, <4 x i32> %b) {
; ALL-LABEL: vec_sink_sub_from_const_to_sub:
; ALL: # %bb.0:
-; ALL-NEXT: movdqa {{.*#+}} xmm2 = <42,24,u,46>
+; ALL-NEXT: movdqa {{.*#+}} xmm2 = [42,24,u,46]
; ALL-NEXT: paddd %xmm1, %xmm0
; ALL-NEXT: psubd %xmm0, %xmm2
; ALL-NEXT: movdqa %xmm2, %xmm0
diff --git a/llvm/test/CodeGen/X86/slow-pmulld.ll b/llvm/test/CodeGen/X86/slow-pmulld.ll
index 8e330c3bfc676ac..913ae195c36319d 100644
--- a/llvm/test/CodeGen/X86/slow-pmulld.ll
+++ b/llvm/test/CodeGen/X86/slow-pmulld.ll
@@ -317,7 +317,7 @@ define <16 x i32> @test_mul_v16i32_v16i8(<16 x i8> %A) {
define <4 x i32> @test_mul_v4i32_v4i16(<4 x i16> %A) {
; SLM-LABEL: test_mul_v4i32_v4i16:
; SLM: # %bb.0:
-; SLM-NEXT: movdqa {{.*#+}} xmm1 = <18778,18778,18778,18778,u,u,u,u>
+; SLM-NEXT: movdqa {{.*#+}} xmm1 = [18778,18778,18778,18778,u,u,u,u]
; SLM-NEXT: movdqa %xmm0, %xmm2
; SLM-NEXT: pmulhuw %xmm1, %xmm2
; SLM-NEXT: pmullw %xmm1, %xmm0
@@ -326,7 +326,7 @@ define <4 x i32> @test_mul_v4i32_v4i16(<4 x i16> %A) {
;
; SLOW-LABEL: test_mul_v4i32_v4i16:
; SLOW: # %bb.0:
-; SLOW-NEXT: movdqa {{.*#+}} xmm1 = <18778,18778,18778,18778,u,u,u,u>
+; SLOW-NEXT: movdqa {{.*#+}} xmm1 = [18778,18778,18778,18778,u,u,u,u]
; SLOW-NEXT: movdqa %xmm0, %xmm2
; SLOW-NEXT: pmulhuw %xmm1, %xmm2
; SLOW-NEXT: pmullw %xmm1, %xmm0
diff --git a/llvm/test/CodeGen/X86/srem-seteq-vec-nonsplat.ll b/llvm/test/CodeGen/X86/srem-seteq-vec-nonsplat.ll
index 0d28dfc68f5bf04..56d1fef5f6d6d8c 100644
--- a/llvm/test/CodeGen/X86/srem-seteq-vec-nonsplat.ll
+++ b/llvm/test/CodeGen/X86/srem-seteq-vec-nonsplat.ll
@@ -555,7 +555,7 @@ define <4 x i32> @test_srem_odd_poweroftwo(<4 x i32> %X) nounwind {
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; CHECK-SSE41-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
@@ -1098,7 +1098,7 @@ define <4 x i32> @test_srem_even_INT_MIN(<4 x i32> %X) nounwind {
; CHECK-SSE2-LABEL: test_srem_even_INT_MIN:
; CHECK-SSE2: # %bb.0:
; CHECK-SSE2-NEXT: pxor %xmm2, %xmm2
-; CHECK-SSE2-NEXT: movdqa {{.*#+}} xmm1 = <3067833783,u,1,u>
+; CHECK-SSE2-NEXT: movdqa {{.*#+}} xmm1 = [3067833783,u,1,u]
; CHECK-SSE2-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
; CHECK-SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3]
@@ -1357,7 +1357,7 @@ define <4 x i32> @test_srem_odd_allones_and_poweroftwo(<4 x i32> %X) nounwind {
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; CHECK-SSE41-NEXT: paddd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
diff --git a/llvm/test/CodeGen/X86/sse2.ll b/llvm/test/CodeGen/X86/sse2.ll
index 231f274db83cd46..cf5f527b16114fa 100644
--- a/llvm/test/CodeGen/X86/sse2.ll
+++ b/llvm/test/CodeGen/X86/sse2.ll
@@ -597,7 +597,7 @@ define <2 x double> @test16(ptr nocapture %srcA, ptr nocapture %dst) {
define fastcc void @test17() nounwind {
; X86-SSE-LABEL: test17:
; X86-SSE: # %bb.0: # %entry
-; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = <u,u,32768,32768>
+; X86-SSE-NEXT: movaps {{.*#+}} xmm0 = [u,u,32768,32768]
; X86-SSE-NEXT: movaps %xmm0, (%eax)
; X86-SSE-NEXT: retl
;
@@ -609,7 +609,7 @@ define fastcc void @test17() nounwind {
;
; X64-SSE-LABEL: test17:
; X64-SSE: # %bb.0: # %entry
-; X64-SSE-NEXT: movaps {{.*#+}} xmm0 = <u,u,32768,32768>
+; X64-SSE-NEXT: movaps {{.*#+}} xmm0 = [u,u,32768,32768]
; X64-SSE-NEXT: movaps %xmm0, (%rax)
; X64-SSE-NEXT: retq
;
diff --git a/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll b/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll
index f2819bcbfe8a050..f677e6737987e64 100644
--- a/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll
+++ b/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll
@@ -137,7 +137,7 @@ define <3 x i1> @test_urem_vec(<3 x i11> %X) nounwind {
; SSE2-NEXT: movd %edx, %xmm0
; SSE2-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0]
; SSE2-NEXT: psubd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
-; SSE2-NEXT: movdqa {{.*#+}} xmm0 = <683,u,819,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm0 = [683,u,819,u]
; SSE2-NEXT: pmuludq %xmm1, %xmm0
; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[0,2,2,3]
; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,1,1,1]
diff --git a/llvm/test/CodeGen/X86/urem-seteq-vec-nonsplat.ll b/llvm/test/CodeGen/X86/urem-seteq-vec-nonsplat.ll
index 7846f0e8f6e037b..484ae2bf95602da 100644
--- a/llvm/test/CodeGen/X86/urem-seteq-vec-nonsplat.ll
+++ b/llvm/test/CodeGen/X86/urem-seteq-vec-nonsplat.ll
@@ -483,7 +483,7 @@ define <4 x i32> @test_urem_odd_poweroftwo(<4 x i32> %X) nounwind {
; CHECK-SSE41-LABEL: test_urem_odd_poweroftwo:
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
@@ -915,7 +915,7 @@ define <4 x i32> @test_urem_odd_INT_MIN(<4 x i32> %X) nounwind {
; CHECK-SSE41-LABEL: test_urem_odd_INT_MIN:
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,2,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,2,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
@@ -1145,7 +1145,7 @@ define <4 x i32> @test_urem_odd_allones_and_poweroftwo(<4 x i32> %X) nounwind {
; CHECK-SSE41-LABEL: test_urem_odd_allones_and_poweroftwo:
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
@@ -1797,7 +1797,7 @@ define <4 x i32> @test_urem_odd_allones_and_poweroftwo_and_one(<4 x i32> %X) nou
; CHECK-SSE41-LABEL: test_urem_odd_allones_and_poweroftwo_and_one:
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <1,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [1,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
@@ -1865,7 +1865,7 @@ define <4 x i32> @test_urem_even_allones_and_poweroftwo_and_one(<4 x i32> %X) no
; CHECK-SSE41-LABEL: test_urem_even_allones_and_poweroftwo_and_one:
; CHECK-SSE41: # %bb.0:
; CHECK-SSE41-NEXT: pmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = <2147483648,u,268435456,u>
+; CHECK-SSE41-NEXT: movdqa {{.*#+}} xmm1 = [2147483648,u,268435456,u]
; CHECK-SSE41-NEXT: pmuludq %xmm0, %xmm1
; CHECK-SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3],xmm1[4,5],xmm0[6,7]
; CHECK-SSE41-NEXT: psrlq $32, %xmm1
diff --git a/llvm/test/CodeGen/X86/vec_fp_to_int.ll b/llvm/test/CodeGen/X86/vec_fp_to_int.ll
index cb19e202e79473c..1d81cb377308007 100644
--- a/llvm/test/CodeGen/X86/vec_fp_to_int.ll
+++ b/llvm/test/CodeGen/X86/vec_fp_to_int.ll
@@ -1907,7 +1907,7 @@ define <2 x i64> @fptosi_2f64_to_2i64_const() {
define <4 x i32> @fptosi_2f64_to_2i32_const() {
; SSE-LABEL: fptosi_2f64_to_2i32_const:
; SSE: # %bb.0:
-; SSE-NEXT: movaps {{.*#+}} xmm0 = <4294967295,1,u,u>
+; SSE-NEXT: movaps {{.*#+}} xmm0 = [4294967295,1,u,u]
; SSE-NEXT: retq
;
; AVX-LABEL: fptosi_2f64_to_2i32_const:
@@ -1966,7 +1966,7 @@ define <2 x i64> @fptoui_2f64_to_2i64_const() {
define <4 x i32> @fptoui_2f64_to_2i32_const(<2 x double> %a) {
; SSE-LABEL: fptoui_2f64_to_2i32_const:
; SSE: # %bb.0:
-; SSE-NEXT: movaps {{.*#+}} xmm0 = <2,4,u,u>
+; SSE-NEXT: movaps {{.*#+}} xmm0 = [2,4,u,u]
; SSE-NEXT: retq
;
; AVX-LABEL: fptoui_2f64_to_2i32_const:
diff --git a/llvm/test/CodeGen/X86/vector-blend.ll b/llvm/test/CodeGen/X86/vector-blend.ll
index 502dc9c0b918bf4..73c2f4ca5b10bf1 100644
--- a/llvm/test/CodeGen/X86/vector-blend.ll
+++ b/llvm/test/CodeGen/X86/vector-blend.ll
@@ -79,7 +79,7 @@ define <4 x i8> @vsel_4xi8(<4 x i8> %v1, <4 x i8> %v2) {
; SSE41-LABEL: vsel_4xi8:
; SSE41: # %bb.0: # %entry
; SSE41-NEXT: movdqa %xmm0, %xmm2
-; SSE41-NEXT: movaps {{.*#+}} xmm0 = <255,255,0,255,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movaps {{.*#+}} xmm0 = [255,255,0,255,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/vector-fshl-512.ll b/llvm/test/CodeGen/X86/vector-fshl-512.ll
index 0724d879f557ddd..558e77ede0186ac 100644
--- a/llvm/test/CodeGen/X86/vector-fshl-512.ll
+++ b/llvm/test/CodeGen/X86/vector-fshl-512.ll
@@ -806,7 +806,7 @@ define <32 x i16> @constant_funnnel_v32i16(<32 x i16> %x, <32 x i16> %y) nounwin
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti64x4 $1, %zmm1, %ymm2
; AVX512F-NEXT: vpsrlw $1, %ymm2, %ymm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,u]
; AVX512F-NEXT: vpmulhuw %ymm3, %ymm2, %ymm4
; AVX512F-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0,1,2,3,4,5,6],ymm2[7],ymm4[8,9,10,11,12,13,14],ymm2[15]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0,1,2,3],ymm2[4,5,6,7]
@@ -827,7 +827,7 @@ define <32 x i16> @constant_funnnel_v32i16(<32 x i16> %x, <32 x i16> %y) nounwin
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vextracti64x4 $1, %zmm1, %ymm2
; AVX512VL-NEXT: vpsrlw $1, %ymm2, %ymm2
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm3 = <2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm3 = [2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,u]
; AVX512VL-NEXT: vpmulhuw %ymm3, %ymm2, %ymm4
; AVX512VL-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0,1,2,3,4,5,6],ymm2[7],ymm4[8,9,10,11,12,13,14],ymm2[15]
; AVX512VL-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0,1,2,3],ymm2[4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/vector-fshr-128.ll b/llvm/test/CodeGen/X86/vector-fshr-128.ll
index eb2df0dcda98a35..1f51f02a197ac40 100644
--- a/llvm/test/CodeGen/X86/vector-fshr-128.ll
+++ b/llvm/test/CodeGen/X86/vector-fshr-128.ll
@@ -1868,7 +1868,7 @@ define <8 x i16> @constant_funnnel_v8i16(<8 x i16> %x, <8 x i16> %y) nounwind {
;
; SSE41-LABEL: constant_funnnel_v8i16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm2 = <u,32768,16384,8192,4096,2048,1024,512>
+; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [u,32768,16384,8192,4096,2048,1024,512]
; SSE41-NEXT: pmulhuw %xmm1, %xmm2
; SSE41-NEXT: pblendw {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3,4,5,6,7]
; SSE41-NEXT: paddw %xmm0, %xmm0
diff --git a/llvm/test/CodeGen/X86/vector-fshr-512.ll b/llvm/test/CodeGen/X86/vector-fshr-512.ll
index 1c6646152ab1f7d..29c258b857eec99 100644
--- a/llvm/test/CodeGen/X86/vector-fshr-512.ll
+++ b/llvm/test/CodeGen/X86/vector-fshr-512.ll
@@ -815,7 +815,7 @@ define <32 x i16> @constant_funnnel_v32i16(<32 x i16> %x, <32 x i16> %y) nounwin
; AVX512F-LABEL: constant_funnnel_v32i16:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2]
; AVX512F-NEXT: vpmulhuw %ymm3, %ymm2, %ymm4
; AVX512F-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0],xmm4[1,2,3,4,5,6,7]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm4[4,5,6,7]
@@ -834,7 +834,7 @@ define <32 x i16> @constant_funnnel_v32i16(<32 x i16> %x, <32 x i16> %y) nounwin
; AVX512VL-LABEL: constant_funnnel_v32i16:
; AVX512VL: # %bb.0:
; AVX512VL-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm3 = <u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm3 = [u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2]
; AVX512VL-NEXT: vpmulhuw %ymm3, %ymm2, %ymm4
; AVX512VL-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0],xmm4[1,2,3,4,5,6,7]
; AVX512VL-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm4[4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-2.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-2.ll
index 04fd6d9300c18d4..8883b3b87df7a3c 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-2.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-2.ll
@@ -211,7 +211,7 @@ define void @load_i16_stride2_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST: # %bb.0:
; AVX2-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm3
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm0, %ymm2
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm2 = ymm2[0,2],ymm3[0,2],ymm2[4,6],ymm3[4,6]
@@ -229,7 +229,7 @@ define void @load_i16_stride2_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-PERLANE: # %bb.0:
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm1, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm0, %ymm2
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm2 = ymm2[0,2],ymm3[0,2],ymm2[4,6],ymm3[4,6]
@@ -383,9 +383,9 @@ define void @load_i16_stride2_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm6 = ymm6[0,1,2,3,4,6,6,7,8,9,10,11,12,14,14,15]
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm5 = ymm6[0,2],ymm5[0,2],ymm6[4,6],ymm5[4,6]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm5 = ymm5[0,2,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm3, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm7, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1],ymm3[2,3],ymm2[4,5],ymm3[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,2,1,3]
@@ -406,7 +406,7 @@ define void @load_i16_stride2_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %ymm1
; AVX2-FAST-NEXT: vmovdqa 64(%rdi), %ymm2
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm5
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm2, %ymm6
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm5 = ymm6[0,2],ymm5[0,2],ymm6[4,6],ymm5[4,6]
@@ -415,9 +415,9 @@ define void @load_i16_stride2_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm0, %ymm4
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm4 = ymm4[0,2],ymm6[0,2],ymm4[4,6],ymm6[4,6]
; AVX2-FAST-NEXT: vpermpd {{.*#+}} ymm4 = ymm4[0,2,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm3, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1],ymm3[2,3],ymm2[4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,2,1,3]
@@ -438,7 +438,7 @@ define void @load_i16_stride2_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa 64(%rdi), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm3, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm2, %ymm6
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm5 = ymm6[0,2],ymm5[0,2],ymm6[4,6],ymm5[4,6]
@@ -447,9 +447,9 @@ define void @load_i16_stride2_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm0, %ymm4
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm4 = ymm4[0,2],ymm6[0,2],ymm4[4,6],ymm6[4,6]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm4 = ymm4[0,2,1,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm3, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1],ymm3[2,3],ymm2[4,5],ymm3[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,2,1,3]
@@ -775,9 +775,9 @@ define void @load_i16_stride2_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm12 = ymm12[0,1,2,3,4,6,6,7,8,9,10,11,12,14,14,15]
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm11 = ymm12[0,2],ymm11[0,2],ymm12[4,6],ymm11[4,6]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm11 = ymm11[0,2,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-SLOW-NEXT: vpshufb %ymm12, %ymm9, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm13, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm8 = ymm8[0,1],ymm9[2,3],ymm8[4,5],ymm9[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,2,1,3]
@@ -814,7 +814,7 @@ define void @load_i16_stride2_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %ymm6
; AVX2-FAST-NEXT: vmovdqa 64(%rdi), %ymm7
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm8, %ymm2
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm7, %ymm10
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm2 = ymm10[0,2],ymm2[0,2],ymm10[4,6],ymm2[4,6]
@@ -831,9 +831,9 @@ define void @load_i16_stride2_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm0, %ymm9
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm9 = ymm9[0,2],ymm12[0,2],ymm9[4,6],ymm12[4,6]
; AVX2-FAST-NEXT: vpermpd {{.*#+}} ymm9 = ymm9[0,2,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm8, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1],ymm8[2,3],ymm7[4,5],ymm8[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,2,1,3]
@@ -870,7 +870,7 @@ define void @load_i16_stride2_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa 64(%rdi), %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [0,1,4,5,u,u,u,u,8,9,12,13,u,u,u,u,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm8, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm7, %ymm10
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm2 = ymm10[0,2],ymm2[0,2],ymm10[4,6],ymm2[4,6]
@@ -887,9 +887,9 @@ define void @load_i16_stride2_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1) no
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm0, %ymm9
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm9 = ymm9[0,2],ymm12[0,2],ymm9[4,6],ymm12[4,6]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm9 = ymm9[0,2,1,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm12, %ymm8, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm13, %ymm7, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1],ymm8[2,3],ymm7[4,5],ymm8[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,2,1,3]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll
index 384e9cf36cf85d2..d2bef6b234e3833 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-3.ll
@@ -575,7 +575,7 @@ define void @load_i16_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY: # %bb.0:
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm2
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm3 = ymm2[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5,6],ymm3[7],ymm2[8],ymm3[9],ymm2[10,11],ymm3[12],ymm2[13,14],ymm3[15]
@@ -588,7 +588,7 @@ define void @load_i16_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm5 = ymm2[0,1,2],ymm5[3,4,5,6,7],ymm2[8,9,10],ymm5[11,12,13,14,15]
; AVX2-ONLY-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,6,5,4,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm5[4,5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm5, %ymm1, %ymm0, %ymm5
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm6 = ymm5[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1],ymm6[2],ymm5[3,4],ymm6[5],ymm5[6,7,8,9],ymm6[10],ymm5[11,12],ymm6[13],ymm5[14,15]
@@ -599,11 +599,11 @@ define void @load_i16_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm6 = ymm5[0,1,2],ymm6[3,4,5,6,7],ymm5[8,9,10],ymm6[11,12,13,14,15]
; AVX2-ONLY-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,5,6,7,4]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3],ymm6[4,5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm6, %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0],ymm0[1,2],ymm1[3],ymm0[4,5],ymm1[6],ymm0[7],ymm1[8],ymm0[9,10],ymm1[11],ymm0[12,13],ymm1[14],ymm0[15]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <4,5,10,11,0,1,6,7,12,13,2,3,8,9,14,15,20,21,26,27,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [4,5,10,11,0,1,6,7,12,13,2,3,8,9,14,15,20,21,26,27,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm4[0],xmm3[1],xmm4[2,3],xmm3[4],xmm4[5,6],xmm3[7]
; AVX2-ONLY-NEXT: vpshufb %xmm1, %xmm3, %xmm1
@@ -647,7 +647,7 @@ define void @load_i16_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpternlogq $202, %ymm2, %ymm1, %ymm0
; AVX512F-NEXT: vmovdqa 16(%rdi), %xmm1
; AVX512F-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0],ymm0[1,2],ymm1[3],ymm0[4,5],ymm1[6],ymm0[7],ymm1[8],ymm0[9,10],ymm1[11],ymm0[12,13],ymm1[14],ymm0[15]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = <4,5,10,11,0,1,6,7,12,13,2,3,8,9,14,15,20,21,26,27,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = [4,5,10,11,0,1,6,7,12,13,2,3,8,9,14,15,20,21,26,27,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm1, %ymm0, %ymm0
; AVX512F-NEXT: vpblendw {{.*#+}} xmm2 = xmm5[0],xmm4[1],xmm5[2,3],xmm4[4],xmm5[5,6],xmm4[7]
; AVX512F-NEXT: vpshufb %xmm1, %xmm2, %xmm1
@@ -939,7 +939,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[0,3,2,3]
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm4[0,1,2],xmm3[3,4,5,6,7]
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm4 = xmm5[0],xmm8[1],xmm5[2,3],xmm8[4],xmm5[5,6],xmm8[7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,6,7,12,13,2,3,8,9,14,15,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,6,7,12,13,2,3,8,9,14,15,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm4, %xmm4
; AVX1-ONLY-NEXT: vpshufd {{.*#+}} xmm7 = xmm6[0,1,2,1]
; AVX1-ONLY-NEXT: vpshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,4,5,6,5]
@@ -973,7 +973,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm15[0,1,2],xmm11[3,4,5,6,7]
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm15 = xmm5[0,1],xmm8[2],xmm5[3,4],xmm8[5],xmm5[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <2,3,8,9,14,15,4,5,10,11,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,8,9,14,15,4,5,10,11,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm15, %xmm15
; AVX1-ONLY-NEXT: vmovddup {{.*#+}} xmm0 = [0,0,0,1,6,7,12,13,0,0,0,1,6,7,12,13]
; AVX1-ONLY-NEXT: # xmm0 = mem[0,0]
@@ -991,7 +991,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovddup {{.*#+}} xmm5 = [0,0,2,3,8,9,14,15,0,0,2,3,8,9,14,15]
; AVX1-ONLY-NEXT: # xmm5 = mem[0,0]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm6, %xmm6
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = <4,5,10,11,0,1,6,7,12,13,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = [4,5,10,11,0,1,6,7,12,13,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3,4],xmm6[5,6,7]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm14, %xmm5
@@ -1001,7 +1001,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm6 # 16-byte Reload
; AVX1-ONLY-NEXT: vpblendw $146, {{[-0-9]+}}(%r{{[sb]}}p), %xmm6, %xmm6 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: # xmm6 = xmm6[0],mem[1],xmm6[2,3],mem[4],xmm6[5,6],mem[7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,u,0,1,6,7,12,13,2,3,8,9,14,15>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,u,0,1,6,7,12,13,2,3,8,9,14,15]
; AVX1-ONLY-NEXT: vpshufb %xmm8, %xmm6, %xmm6
; AVX1-ONLY-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[2,1,2,3]
; AVX1-ONLY-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[2,1,2,3,4,5,6,7]
@@ -1033,7 +1033,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm1
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm2
; AVX2-ONLY-NEXT: vmovdqa 128(%rdi), %ymm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm2, %ymm4, %ymm3
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm5 = ymm3[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0],ymm5[1],ymm3[2,3],ymm5[4],ymm3[5,6],ymm5[7],ymm3[8],ymm5[9],ymm3[10,11],ymm5[12],ymm3[13,14],ymm5[15]
@@ -1060,7 +1060,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm10 = ymm9[0,1,2],ymm10[3,4,5,6,7],ymm9[8,9,10],ymm10[11,12,13,14,15]
; AVX2-ONLY-NEXT: vpshufhw {{.*#+}} xmm9 = xmm9[0,1,2,3,6,5,4,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3],ymm10[4,5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm4, %ymm2, %ymm10
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm12 = ymm10[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm10 = ymm10[0,1],ymm12[2],ymm10[3,4],ymm12[5],ymm10[6,7,8,9],ymm12[10],ymm10[11,12],ymm12[13],ymm10[14,15]
@@ -1083,7 +1083,7 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm12 = ymm11[0,1,2],ymm12[3,4,5,6,7],ymm11[8,9,10],ymm12[11,12,13,14,15]
; AVX2-ONLY-NEXT: vpshufhw {{.*#+}} xmm11 = xmm11[0,1,2,3,5,6,7,4]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm11 = ymm11[0,1,2,3],ymm12[4,5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm4, %ymm2, %ymm2
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm4 = ymm2[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0],ymm2[1,2],ymm4[3],ymm2[4,5],ymm4[6],ymm2[7],ymm4[8],ymm2[9,10],ymm4[11],ymm2[12,13],ymm4[14],ymm2[15]
@@ -1196,15 +1196,15 @@ define void @load_i16_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512BW-NEXT: vmovdqa64 128(%rdi), %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,3,6,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,3,6,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,34,37,40,43,46,49,52,55,58,61]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <1,4,7,10,13,16,19,22,25,28,31,34,37,40,43,46,49,52,55,58,61,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [1,4,7,10,13,16,19,22,25,28,31,34,37,40,43,46,49,52,55,58,61,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,32,35,38,41,44,47,50,53,56,59,62]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <34,37,40,43,46,49,52,55,58,61,0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [34,37,40,43,46,49,52,55,58,61,0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,33,36,39,42,45,48,51,54,57,60,63]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm0
@@ -1725,7 +1725,7 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovdqa 208(%rdi), %xmm10
; AVX1-ONLY-NEXT: vmovdqa 192(%rdi), %xmm9
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm9[0],xmm10[1],xmm9[2,3],xmm10[4],xmm9[5,6],xmm10[7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,6,7,12,13,2,3,8,9,14,15,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,6,7,12,13,2,3,8,9,14,15,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm0, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vmovdqa 224(%rdi), %xmm4
; AVX1-ONLY-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -1820,7 +1820,7 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm14[0,1,2],xmm13[3,4,5,6,7]
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm13 = xmm9[0,1],xmm10[2],xmm9[3,4],xmm10[5],xmm9[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <2,3,8,9,14,15,4,5,10,11,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [2,3,8,9,14,15,4,5,10,11,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm13, %xmm13
; AVX1-ONLY-NEXT: vmovddup {{.*#+}} xmm0 = [0,0,0,1,6,7,12,13,0,0,0,1,6,7,12,13]
; AVX1-ONLY-NEXT: # xmm0 = mem[0,0]
@@ -1872,7 +1872,7 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovddup {{.*#+}} xmm13 = [0,0,2,3,8,9,14,15,0,0,2,3,8,9,14,15]
; AVX1-ONLY-NEXT: # xmm13 = mem[0,0]
; AVX1-ONLY-NEXT: vpshufb %xmm13, %xmm11, %xmm3
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <4,5,10,11,0,1,6,7,12,13,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [4,5,10,11,0,1,6,7,12,13,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm0[0,1,2,3,4],xmm3[5,6,7]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
@@ -1890,7 +1890,7 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm2, %xmm2
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3,4],xmm1[5,6,7]
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm1 = xmm15[0],xmm5[1],xmm15[2,3],xmm5[4],xmm15[5,6],xmm5[7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,u,u,0,1,6,7,12,13,2,3,8,9,14,15>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,u,u,0,1,6,7,12,13,2,3,8,9,14,15]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vpshufd {{.*#+}} xmm6 = xmm6[2,1,2,3]
; AVX1-ONLY-NEXT: vpshuflw {{.*#+}} xmm6 = xmm6[2,1,2,3,4,5,6,7]
@@ -1966,17 +1966,17 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vmovdqa 320(%rdi), %ymm10
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm12
; AVX2-ONLY-NEXT: vmovdqa 128(%rdi), %ymm13
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm12, %ymm13, %ymm0
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm3 = ymm0[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2,3],ymm3[4],ymm0[5,6],ymm3[7],ymm0[8],ymm3[9],ymm0[10,11],ymm3[12],ymm0[13,14],ymm3[15]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm15 = ymm0[0,1,6,7,12,13,2,3,4,5,14,15,8,9,10,11,16,17,22,23,28,29,18,19,20,21,30,31,24,25,26,27]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm9, %ymm10, %ymm3
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm4, %ymm5, %ymm8
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm0, %ymm13, %ymm12, %ymm6
; AVX2-ONLY-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm13, %ymm12, %ymm6
; AVX2-ONLY-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vpblendvb %ymm0, %ymm10, %ymm9, %ymm13
@@ -2308,21 +2308,21 @@ define void @load_i16_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm3
; AVX512BW-NEXT: vmovdqa64 128(%rdi), %zmm4
; AVX512BW-NEXT: vmovdqa64 192(%rdi), %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,3,6,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,3,6,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm7
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,34,37,40,43,46,49,52,55,58,61]
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm8, %zmm7
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
; AVX512BW-NEXT: vpermt2w %zmm4, %zmm8, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <1,4,7,10,13,16,19,22,25,28,31,34,37,40,43,46,49,52,55,58,61,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [1,4,7,10,13,16,19,22,25,28,31,34,37,40,43,46,49,52,55,58,61,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm8, %zmm9
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,32,35,38,41,44,47,50,53,56,59,62]
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm10, %zmm9
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm8
; AVX512BW-NEXT: vpermt2w %zmm4, %zmm10, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <34,37,40,43,46,49,52,55,58,61,0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [34,37,40,43,46,49,52,55,58,61,0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm5, %zmm10, %zmm1
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,33,36,39,42,45,48,51,54,57,60,63]
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm5, %zmm1
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll
index 44f2d58cb4e9681..241e3d6a1a3e426 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-4.ll
@@ -832,7 +832,7 @@ define void @load_i16_stride4_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm5 = [16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm4, %ymm6
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm7, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5],ymm6[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm3[4,5,6,7]
@@ -851,7 +851,7 @@ define void @load_i16_stride4_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm11 = xmm12[0,1],xmm11[2,3]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm12 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm7[0,1,2,3,4,5],ymm4[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm11[0,1,2,3],ymm4[4,5,6,7]
@@ -1065,7 +1065,7 @@ define void @load_i16_stride4_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa 96(%rdi), %ymm2
; AVX512F-FAST-NEXT: vpermd %ymm2, %ymm1, %ymm3
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm3, %ymm4
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX512F-FAST-NEXT: vmovdqa 64(%rdi), %ymm6
; AVX512F-FAST-NEXT: vpermd %ymm6, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm1, %ymm7
@@ -1076,7 +1076,7 @@ define void @load_i16_stride4_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm9[0,1,2,3],ymm7[4,5,6,7]
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} ymm9 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX512F-FAST-NEXT: vpshufb %ymm9, %ymm3, %ymm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm10, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5],ymm3[6,7]
; AVX512F-FAST-NEXT: vpsrlq $16, %zmm4, %zmm3
@@ -1884,7 +1884,7 @@ define void @load_i16_stride4_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm10, %ymm3
; AVX2-FAST-NEXT: vmovdqa %ymm4, %ymm9
; AVX2-FAST-NEXT: vpermd %ymm7, %ymm2, %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm11, %ymm4
; AVX2-FAST-NEXT: vmovdqa %ymm7, %ymm12
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
@@ -1920,7 +1920,7 @@ define void @load_i16_stride4_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm0 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm10, %ymm13
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm11, %ymm11
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm10
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm11[0,1,2,3,4,5],ymm13[6,7]
@@ -1981,7 +1981,7 @@ define void @load_i16_stride4_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm7[4,5,6,7]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm10 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm10, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm7[0,1,2,3,4,5],ymm4[6,7]
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,1,3,1,4,5,6,7]
@@ -2381,7 +2381,7 @@ define void @load_i16_stride4_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa 224(%rdi), %ymm6
; AVX512F-FAST-NEXT: vpermd %ymm6, %ymm4, %ymm5
; AVX512F-FAST-NEXT: vpshufb %ymm2, %ymm5, %ymm7
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX512F-FAST-NEXT: vmovdqa 192(%rdi), %ymm8
; AVX512F-FAST-NEXT: vpermd %ymm8, %ymm4, %ymm9
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm9, %ymm10
@@ -2401,7 +2401,7 @@ define void @load_i16_stride4_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vshufi64x2 {{.*#+}} zmm7 = zmm4[0,1,2,3],zmm7[0,1,2,3]
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} ymm4 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX512F-FAST-NEXT: vpshufb %ymm4, %ymm5, %ymm13
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm9, %ymm9
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],ymm13[6,7]
; AVX512F-FAST-NEXT: vpsrlq $16, %zmm1, %zmm13
@@ -4096,7 +4096,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX2-FAST-NEXT: vmovdqa %ymm4, %ymm8
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm2, %ymm10
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm10, %ymm4
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm3[4,5,6,7]
@@ -4171,7 +4171,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm13 = xmm14[0,1],xmm13[2,3]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm6 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm3, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm7[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm13[0,1,2,3],ymm3[4,5,6,7]
@@ -4237,7 +4237,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm12 = [16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29,16,17,20,21,24,25,28,29]
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm15, %ymm6, %ymm11
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm11[0,1,2,3,4,5],ymm1[6,7]
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm6 = xmm7[3,1,2,3]
@@ -4331,7 +4331,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm10 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm10, %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
@@ -5186,7 +5186,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa64 224(%rdi), %ymm24
; AVX512F-FAST-NEXT: vpermd %ymm24, %ymm1, %ymm10
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm10, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,4,5,u,u,u,u,8,9,12,13,12,13,14,15,16,17,20,21,u,u,u,u,24,25,28,29,u,u,u,u]
; AVX512F-FAST-NEXT: vmovdqa64 192(%rdi), %ymm25
; AVX512F-FAST-NEXT: vpermd %ymm25, %ymm1, %ymm11
; AVX512F-FAST-NEXT: vpshufb %ymm6, %ymm11, %ymm3
@@ -5225,7 +5225,7 @@ define void @load_i16_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vshufi64x2 {{.*#+}} zmm22 = zmm14[0,1,2,3],zmm9[0,1,2,3]
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} ymm9 = [18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31,18,19,22,23,26,27,30,31]
; AVX512F-FAST-NEXT: vpshufb %ymm9, %ymm10, %ymm14
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,3,6,7,10,11,14,15,14,15,10,11,12,13,14,15,18,19,22,23,26,27,30,31,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm10, %ymm11, %ymm11
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm11[0,1,2,3,4,5],ymm14[6,7]
; AVX512F-FAST-NEXT: vpsrlq $16, %zmm4, %zmm14
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-5.ll
index 30f96e0b3e1492d..904f8ddd2e7f443 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-5.ll
@@ -1325,7 +1325,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0,1],xmm9[2,3],xmm6[4,5,6],xmm9[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[2,3,12,13,6,7,0,1,10,11,4,5,14,15,10,11]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm9 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5],ymm1[6],ymm2[7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13],ymm1[14],ymm2[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,u,u,u,4,7,1,6>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,u,u,u,4,7,1,6]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm10, %ymm9
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm9[2,3,4,5,10,11,0,1,14,15,2,3,12,13,0,1,18,19,20,21,26,27,16,17,30,31,18,19,28,29,16,17]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm6, %ymm9, %ymm6
@@ -1341,7 +1341,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0,1,2],xmm10[3,4],xmm9[5,6,7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm10 = ymm1[0],ymm2[1],ymm1[2,3],ymm2[4],ymm1[5],ymm2[6],ymm1[7,8],ymm2[9],ymm1[10,11],ymm2[12],ymm1[13],ymm2[14],ymm1[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,2,u,u,5,7,2,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,2,u,u,5,7,2,4]
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm11, %ymm10
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2],ymm10[3,4,5,6,7]
@@ -1356,7 +1356,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm9 = xmm10[0],xmm9[1],xmm10[2],xmm9[3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm10 = ymm2[0],ymm1[1],ymm2[2],ymm1[3],ymm2[4,5],ymm1[6],ymm2[7,8],ymm1[9],ymm2[10],ymm1[11],ymm2[12,13],ymm1[14],ymm2[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,3,u,u,5,0,2,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,3,u,u,5,0,2,7]
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm11, %ymm10
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2],ymm10[3,4,5,6,7]
@@ -1371,7 +1371,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm4[0,1,2],xmm3[3,4],xmm4[5,6,7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[8,9,2,3,12,13,6,7,0,1,10,11,u,u,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1],ymm1[2],ymm2[3],ymm1[4],ymm2[5,6],ymm1[7],ymm2[8,9],ymm1[10],ymm2[11],ymm1[12],ymm2[13,14],ymm1[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <1,3,u,u,6,0,3,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [1,3,u,u,6,0,3,5]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2],ymm1[3,4,5,6,7]
@@ -1570,7 +1570,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX512F-FAST-NEXT: vmovdqa 96(%rdi), %ymm1
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm0[0],ymm1[1,2],ymm0[3],ymm1[4],ymm0[5],ymm1[6,7],ymm0[8],ymm1[9,10],ymm0[11],ymm1[12],ymm0[13],ymm1[14,15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <1,u,u,u,4,6,1,3>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [1,u,u,u,4,6,1,3]
; AVX512F-FAST-NEXT: vpermd %ymm4, %ymm5, %ymm4
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm4 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[2,3,16,17,22,23,24,25,30,31,20,21],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm5 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5],ymm3[6],ymm2[7,8],ymm3[9],ymm2[10,11],ymm3[12],ymm2[13],ymm3[14],ymm2[15]
@@ -1587,7 +1587,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm6 = ymm5[0,1,2,3,4],ymm6[5,6,7],ymm5[8,9,10,11,12],ymm6[13,14,15]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3],ymm6[4,5,6,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm6 = ymm1[0],ymm0[1],ymm1[2,3],ymm0[4],ymm1[5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10,11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <2,u,u,u,4,7,1,6>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [2,u,u,u,4,7,1,6]
; AVX512F-FAST-NEXT: vpermd %ymm6, %ymm8, %ymm6
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm6 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm6[0,1,18,19,20,21,26,27,16,17,30,31],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm3[0],ymm2[1],ymm3[2],ymm2[3],ymm3[4,5],ymm2[6],ymm3[7,8],ymm2[9],ymm3[10],ymm2[11],ymm3[12,13],ymm2[14],ymm3[15]
@@ -1607,7 +1607,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0,1,2],xmm10[3,4],xmm9[5,6,7]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm10 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5],ymm1[6],ymm0[7,8],ymm1[9],ymm0[10,11],ymm1[12],ymm0[13],ymm1[14],ymm0[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,2,u,u,5,7,2,4>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,2,u,u,5,7,2,4]
; AVX512F-FAST-NEXT: vpermd %ymm10, %ymm11, %ymm10
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2],ymm10[3,4,5,6,7]
@@ -1622,7 +1622,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm9 = xmm10[0],xmm9[1],xmm10[2],xmm9[3]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm10 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10],ymm0[11],ymm1[12,13],ymm0[14],ymm1[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,3,u,u,5,0,2,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,3,u,u,5,0,2,7]
; AVX512F-FAST-NEXT: vpermd %ymm10, %ymm11, %ymm10
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2],ymm10[3,4,5,6,7]
@@ -1637,7 +1637,7 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2],xmm2[3,4],xmm3[5,6,7]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[8,9,2,3,12,13,6,7,0,1,10,11,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3],ymm0[4],ymm1[5,6],ymm0[7],ymm1[8,9],ymm0[10],ymm1[11],ymm0[12],ymm1[13,14],ymm0[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <1,3,u,u,6,0,3,5>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [1,3,u,u,6,0,3,5]
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7]
@@ -1659,24 +1659,24 @@ define void @load_i16_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,2,3,4,5,6,7,8,9,10,11,12,17,22,27]
; AVX512BW-NEXT: vmovdqa 128(%rdi), %ymm4
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm2, %ymm3
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,2,3,4,5,6,7,8,9,10,11,12,18,23,28]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm2, %ymm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,2,3,4,5,6,7,8,9,10,11,12,19,24,29]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm2, %ymm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,2,3,4,5,6,7,8,9,10,11,12,20,25,30]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm2, %ymm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,2,3,4,5,6,7,8,9,10,11,16,21,26,31]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm2, %ymm0
@@ -2357,7 +2357,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpblendw $207, {{[-0-9]+}}(%r{{[sb]}}p), %xmm6, %xmm0 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: # xmm0 = mem[0,1,2,3],xmm6[4,5],mem[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,8,9,2,3,12,13,12,13,12,13,12,13>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,8,9,2,3,12,13,12,13,12,13,12,13]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vpshufd {{.*#+}} xmm3 = xmm11[3,1,2,3]
; AVX1-ONLY-NEXT: vpshuflw {{.*#+}} xmm3 = xmm3[2,1,2,3,4,5,6,7]
@@ -2409,7 +2409,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm10, %ymm1, %ymm1
; AVX1-ONLY-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm5[0,1],xmm0[2,3],xmm5[4,5],xmm0[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,0,1,10,11,4,5,14,15,14,15,14,15,14,15>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,0,1,10,11,4,5,14,15,14,15,14,15,14,15]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm3, %xmm0
; AVX1-ONLY-NEXT: vpsrlq $48, %xmm2, %xmm3
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm3[0],xmm0[1,2,3,4,5,6,7]
@@ -2623,7 +2623,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm4[0,1],ymm2[2],ymm4[3],ymm2[4],ymm4[5,6],ymm2[7],ymm4[8,9],ymm2[10],ymm4[11],ymm2[12],ymm4[13,14],ymm2[15]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm8
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2],xmm8[3,4],xmm1[5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27]
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7]
@@ -2659,9 +2659,9 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm7
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm8
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm1 = xmm8[0],xmm1[1],xmm8[2],xmm1[3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29]
; AVX2-SLOW-NEXT: vpshufb %ymm12, %ymm0, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm1 = xmm13[0,1],xmm14[2],xmm13[3]
@@ -2788,7 +2788,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %ymm11, %ymm10, %ymm10
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm12, %ymm10, %ymm11
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm10 = ymm7[0],ymm6[1],ymm7[2,3],ymm6[4],ymm7[5],ymm6[6],ymm7[7,8],ymm6[9],ymm7[10,11],ymm6[12],ymm7[13],ymm6[14],ymm7[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,u,u,u,4,7,1,6>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,u,u,u,4,7,1,6]
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm12, %ymm10
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [2,3,4,5,10,11,0,1,14,15,2,3,12,13,0,1,18,19,20,21,26,27,16,17,30,31,18,19,28,29,16,17]
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm10, %ymm10
@@ -2843,11 +2843,11 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm5[0],ymm4[1],ymm5[2,3],ymm4[4],ymm5[5],ymm4[6],ymm5[7,8],ymm4[9],ymm5[10,11],ymm4[12],ymm5[13],ymm4[14],ymm5[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,2,u,u,5,7,2,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,2,u,u,5,7,2,4]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm14, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23]
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm8, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm8[3,4,5,6,7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [1,4,6,0,1,4,6,0]
@@ -2878,11 +2878,11 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2],xmm1[3]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0],ymm5[1],ymm4[2],ymm5[3],ymm4[4,5],ymm5[6],ymm4[7,8],ymm5[9],ymm4[10],ymm5[11],ymm4[12,13],ymm5[14],ymm4[15]
; AVX2-FAST-NEXT: vmovdqa %ymm4, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,3,u,u,5,0,2,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,3,u,u,5,0,2,7]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm11, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm14, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2],ymm2[3,4,5,6,7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [2,4,7,0,2,4,7,0]
@@ -2914,7 +2914,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm1 = mem[0],ymm1[1],mem[2,3],ymm1[4],mem[5],ymm1[6],mem[7,8],ymm1[9],mem[10,11],ymm1[12],mem[13],ymm1[14],mem[15]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2],xmm1[3,4],xmm2[5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <1,3,u,u,6,0,3,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [1,3,u,u,6,0,3,5]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm4 = [16,17,22,23,24,25,30,31,0,0,0,0,0,1,6,7,16,17,22,23,24,25,30,31,0,0,0,0,0,1,6,7]
; AVX2-FAST-NEXT: # ymm4 = mem[0,1,0,1]
@@ -3059,9 +3059,9 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm2 = ymm5[0,1],ymm0[2],ymm5[3],ymm0[4],ymm5[5,6],ymm0[7],ymm5[8,9],ymm0[10],ymm5[11],ymm0[12],ymm5[13,14],ymm0[15]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm11
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2],xmm11[3,4],xmm2[5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = <4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = [4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm12, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm2 = xmm8[0],xmm9[1],xmm8[2,3]
@@ -3094,9 +3094,9 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm2 = ymm0[0],ymm5[1,2],ymm0[3],ymm5[4],ymm0[5],ymm5[6,7],ymm0[8],ymm5[9,10],ymm0[11],ymm5[12],ymm0[13],ymm5[14,15]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm11
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm2 = xmm11[0],xmm2[1],xmm11[2],xmm2[3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm12, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm2 = xmm8[0,1],xmm9[2],xmm8[3]
@@ -3194,7 +3194,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm5 = ymm9[0],ymm3[1],ymm9[2,3],ymm3[4],ymm9[5],ymm3[6],ymm9[7,8],ymm3[9],ymm9[10,11],ymm3[12],ymm9[13],ymm3[14],ymm9[15]
; AVX512F-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm6
; AVX512F-SLOW-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1,2],ymm6[3],ymm5[4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,8,9,2,3,12,13,6,7,16,17,26,27,26,27,30,31,24,25,18,19,28,29,22,23>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,8,9,2,3,12,13,6,7,16,17,26,27,26,27,30,31,24,25,18,19,28,29,22,23]
; AVX512F-SLOW-NEXT: vpshufb %ymm7, %ymm5, %ymm5
; AVX512F-SLOW-NEXT: vmovdqa 176(%rdi), %xmm1
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm8 = xmm1[3,1,2,3]
@@ -3384,10 +3384,10 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa 192(%rdi), %ymm4
; AVX512F-FAST-NEXT: vmovdqa 224(%rdi), %ymm5
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm5[0],ymm4[1],ymm5[2,3],ymm4[4],ymm5[5],ymm4[6],ymm5[7,8],ymm4[9],ymm5[10,11],ymm4[12],ymm5[13],ymm4[14],ymm5[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <2,4,7,1,4,6,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [2,4,7,1,4,6,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm6, %ymm1
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,8,9,14,15,0,1,6,7,16,17,22,23,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <8,9,3,2,4,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [8,9,3,2,4,u,u,u]
; AVX512F-FAST-NEXT: vpermi2d %ymm0, %ymm1, %ymm6
; AVX512F-FAST-NEXT: vmovdqa 256(%rdi), %ymm0
; AVX512F-FAST-NEXT: vmovdqa 288(%rdi), %ymm1
@@ -3402,7 +3402,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa 64(%rdi), %ymm8
; AVX512F-FAST-NEXT: vmovdqa 96(%rdi), %ymm9
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm7 = ymm8[0],ymm9[1,2],ymm8[3],ymm9[4],ymm8[5],ymm9[6,7],ymm8[8],ymm9[9,10],ymm8[11],ymm9[12],ymm8[13],ymm9[14,15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <1,u,u,u,4,6,1,3>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [1,u,u,u,4,6,1,3]
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm12, %ymm7
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm7 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm7[2,3,16,17,22,23,24,25,30,31,20,21],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm12 = ymm10[0],ymm11[1],ymm10[2,3],ymm11[4],ymm10[5],ymm11[6],ymm10[7,8],ymm11[9],ymm10[10,11],ymm11[12],ymm10[13],ymm11[14],ymm10[15]
@@ -3410,7 +3410,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm12 = xmm12[0],xmm13[1,2,3],xmm12[4,5],xmm13[6,7]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[0,1,10,11,4,5,14,15,8,9,2,3,12,13],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm12[u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpor %ymm7, %ymm12, %ymm12
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <0,3,1,u,0,3,5,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [0,3,1,u,0,3,5,u]
; AVX512F-FAST-NEXT: vmovdqa 128(%rdi), %ymm7
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm13, %ymm13
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [4,5,2,3,4,5,6,7,8,9,2,3,4,5,10,11,20,21,18,19,20,21,22,23,24,25,18,19,20,21,26,27]
@@ -3421,14 +3421,14 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm12 = ymm11[0],ymm10[1],ymm11[2],ymm10[3],ymm11[4,5],ymm10[6],ymm11[7,8],ymm10[9],ymm11[10],ymm10[11],ymm11[12,13],ymm10[14],ymm11[15]
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm12, %xmm13
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm12 = xmm12[0,1],xmm13[2,3],xmm12[4,5,6],xmm13[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm6, %ymm12, %ymm12
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm13 = ymm9[0],ymm8[1],ymm9[2,3],ymm8[4],ymm9[5],ymm8[6],ymm9[7,8],ymm8[9],ymm9[10,11],ymm8[12],ymm9[13],ymm8[14],ymm9[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = <2,u,u,u,4,7,1,6>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = [2,u,u,u,4,7,1,6]
; AVX512F-FAST-NEXT: vpermd %ymm13, %ymm17, %ymm13
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm13 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm13[0,1,18,19,20,21,26,27,16,17,30,31],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpor %ymm13, %ymm12, %ymm12
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <1,3,2,u,1,3,6,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [1,3,2,u,1,3,6,u]
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm13, %ymm15
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [0,1,6,7,4,5,6,7,8,9,0,1,6,7,8,9,16,17,22,23,20,21,22,23,24,25,16,17,22,23,24,25]
; AVX512F-FAST-NEXT: vpshufb %ymm13, %ymm15, %ymm15
@@ -3437,7 +3437,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpsrlq $48, %xmm2, %xmm12
; AVX512F-FAST-NEXT: vpunpckldq {{.*#+}} xmm6 = xmm6[0],xmm12[0],xmm6[1],xmm12[1]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm12 = ymm4[0],ymm5[1],ymm4[2,3],ymm5[4],ymm4[5],ymm5[6],ymm4[7,8],ymm5[9],ymm4[10,11],ymm5[12],ymm4[13],ymm5[14],ymm4[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = <0,2,5,7,4,7,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = [0,2,5,7,4,7,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm12, %ymm17, %ymm12
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[2,3,4,5,4,5,0,1,6,7,8,9,14,15,4,5,18,19,20,21,20,21,16,17,22,23,24,25,30,31,20,21]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0,1,2],xmm12[3,4,5,6,7]
@@ -3457,7 +3457,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm15 = xmm3[u,u,u,u,u,u,u,u,4,5,14,15,u,u,u,u]
; AVX512F-FAST-NEXT: vpunpckhdq {{.*#+}} xmm15 = xmm15[2],xmm2[2],xmm15[3],xmm2[3]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm12 = ymm5[0],ymm4[1],ymm5[2],ymm4[3],ymm5[4,5],ymm4[6],ymm5[7,8],ymm4[9],ymm5[10],ymm4[11],ymm5[12,13],ymm4[14],ymm5[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm19 = <0,3,5,2,5,7,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm19 = [0,3,5,2,5,7,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm12, %ymm19, %ymm12
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[0,1,6,7,2,3,2,3,4,5,10,11,0,1,14,15,16,17,22,23,18,19,18,19,20,21,26,27,16,17,30,31]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm15 = xmm15[0,1,2],xmm12[3,4,5,6,7]
@@ -3468,7 +3468,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm12 = xmm12[0,1,2],xmm15[3,4],xmm12[5,6,7]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm12 = xmm12[4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm15 = ymm8[0],ymm9[1],ymm8[2,3],ymm9[4],ymm8[5],ymm9[6],ymm8[7,8],ymm9[9],ymm8[10,11],ymm9[12],ymm8[13],ymm9[14],ymm8[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm19 = <0,2,u,u,5,7,2,4>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm19 = [0,2,u,u,5,7,2,4]
; AVX512F-FAST-NEXT: vpermd %ymm15, %ymm19, %ymm15
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm15 = ymm15[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm12 = ymm12[0,1,2],ymm15[3,4,5,6,7]
@@ -3483,14 +3483,14 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm6 = xmm12[0],xmm6[1],xmm12[2],xmm6[3]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm12 = ymm9[0],ymm8[1],ymm9[2],ymm8[3],ymm9[4,5],ymm8[6],ymm9[7,8],ymm8[9],ymm9[10],ymm8[11],ymm9[12,13],ymm8[14],ymm9[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <0,3,u,u,5,0,2,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [0,3,u,u,5,0,2,7]
; AVX512F-FAST-NEXT: vpermd %ymm12, %ymm15, %ymm12
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2],ymm12[3,4,5,6,7]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm12 = xmm2[0],xmm3[1],xmm2[2,3]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm12 = xmm12[6,7,0,1,10,11,u,u,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm15 = ymm5[0,1],ymm4[2],ymm5[3],ymm4[4],ymm5[5,6],ymm4[7],ymm5[8,9],ymm4[10],ymm5[11],ymm4[12],ymm5[13,14],ymm4[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = <1,3,6,0,5,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = [1,3,6,0,5,u,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm15, %ymm18, %ymm15
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm15 = ymm15[2,3,2,3,4,5,0,1,6,7,8,9,14,15,4,5,18,19,18,19,20,21,16,17,22,23,24,25,30,31,20,21]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm12 = xmm12[0,1,2],xmm15[3,4,5,6,7]
@@ -3515,14 +3515,14 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm10 = xmm11[0,1,2],xmm10[3,4],xmm11[5,6,7]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[8,9,2,3,12,13,6,7,0,1,10,11,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0,1],ymm8[2],ymm9[3],ymm8[4],ymm9[5,6],ymm8[7],ymm9[8,9],ymm8[10],ymm9[11],ymm8[12],ymm9[13,14],ymm8[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <1,3,u,u,6,0,3,5>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [1,3,u,u,6,0,3,5]
; AVX512F-FAST-NEXT: vpermd %ymm8, %ymm9, %ymm8
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,16,17,22,23,16,17,22,23]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm8 = ymm10[0,1,2],ymm8[3,4,5,6,7]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm2 = xmm2[0,1],xmm3[2],xmm2[3]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[8,9,2,3,12,13,u,u,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm5[1,2],ymm4[3],ymm5[4],ymm4[5],ymm5[6,7],ymm4[8],ymm5[9,10],ymm4[11],ymm5[12],ymm4[13],ymm5[14,15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <1,4,6,3,6,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [1,4,6,3,6,u,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm3, %ymm4, %ymm3
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5,2,3,4,5,10,11,0,1,14,15,16,17,18,19,20,21,18,19,20,21,26,27,16,17,30,31]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2],xmm3[3,4,5,6,7]
@@ -3561,7 +3561,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [16,21,26,31,36,41,46,51,56,61,0,0,0,1,6,11,16,21,26,31,36,41,46,51,56,61,0,0,0,1,6,11]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm6
; AVX512BW-NEXT: movl $67100672, %eax # imm = 0x3FFE000
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -3571,7 +3571,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [49,54,59,0,5,10,15,20,25,30,0,0,0,34,39,44,49,54,59,0,5,10,15,20,25,30,0,0,0,34,39,44]
; AVX512BW-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm4, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu16 %zmm6, %zmm7 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,35,40,45,50,55,60]
@@ -3579,7 +3579,7 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm7 = [50,55,60,1,6,11,16,21,26,31,0,0,0,35,40,45,50,55,60,1,6,11,16,21,26,31,0,0,0,35,40,45]
; AVX512BW-NEXT: # zmm7 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm4, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm2, %zmm8
; AVX512BW-NEXT: vmovdqu16 %zmm7, %zmm8 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,36,41,46,51,56,61]
@@ -3587,14 +3587,14 @@ define void @load_i16_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [19,24,29,34,39,44,49,54,59,0,0,0,0,4,9,14,19,24,29,34,39,44,49,54,59,0,0,0,0,4,9,14]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm3, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm2, %zmm9
; AVX512BW-NEXT: movl $33546240, %eax # imm = 0x1FFE000
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm8, %zmm9 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,32,37,42,47,52,57,62]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm9, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm9
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm1 = [20,25,30,35,40,45,50,55,60,0,0,0,0,5,10,15,20,25,30,35,40,45,50,55,60,0,0,0,0,5,10,15]
; AVX512BW-NEXT: # zmm1 = mem[0,1,2,3,0,1,2,3]
@@ -4933,7 +4933,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX1-ONLY-NEXT: vpblendw $207, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0, %xmm1 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: # xmm1 = mem[0,1,2,3],xmm0[4,5],mem[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,8,9,2,3,12,13,12,13,12,13,12,13>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,8,9,2,3,12,13,12,13,12,13,12,13]
; AVX1-ONLY-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vpshufd $231, {{[-0-9]+}}(%r{{[sb]}}p), %xmm2 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: # xmm2 = mem[3,1,2,3]
@@ -5071,7 +5071,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX1-ONLY-NEXT: vpblendw $204, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0, %xmm3 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: # xmm3 = xmm0[0,1],mem[2,3],xmm0[4,5],mem[6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,0,1,10,11,4,5,14,15,14,15,14,15,14,15>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,0,1,10,11,4,5,14,15,14,15,14,15,14,15]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX1-ONLY-NEXT: vpsrlq $48, %xmm0, %xmm5
@@ -5514,9 +5514,9 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: # ymm1 = ymm15[0,1],mem[2],ymm15[3],mem[4],ymm15[5,6],mem[7],ymm15[8,9],mem[10],ymm15[11],mem[12],ymm15[13,14],mem[15]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2],xmm2[3,4],xmm1[5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm0, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0,1,2],ymm3[3,4,5,6,7]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm6 = xmm9[0],xmm14[1],xmm9[2,3]
@@ -5595,9 +5595,9 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: # ymm1 = mem[0],ymm15[1,2],mem[3],ymm15[4],mem[5],ymm15[6,7],mem[8],ymm15[9,10],mem[11],ymm15[12],mem[13],ymm15[14,15]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2],xmm1[3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm0, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0,1,2],ymm3[3,4,5,6,7]
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm15 # 16-byte Reload
@@ -5874,7 +5874,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendw $173, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm1 = mem[0],ymm1[1],mem[2,3],ymm1[4],mem[5],ymm1[6],mem[7,8],ymm1[9],mem[10,11],ymm1[12],mem[13],ymm1[14],mem[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,u,u,u,4,7,1,6>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,u,u,u,4,7,1,6]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm3
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [2,3,4,5,10,11,0,1,14,15,2,3,12,13,0,1,18,19,20,21,26,27,16,17,30,31,18,19,28,29,16,17]
; AVX2-FAST-NEXT: vpshufb %ymm1, %ymm3, %ymm3
@@ -5975,14 +5975,14 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm9[0,1],ymm4[2],ymm9[3],ymm4[4],ymm9[5,6],ymm4[7],ymm9[8,9],ymm4[10],ymm9[11],ymm4[12],ymm9[13,14],ymm4[15]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2],xmm1[3,4],xmm0[5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm6
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendw $82, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm3 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm3 = ymm1[0],mem[1],ymm1[2,3],mem[4],ymm1[5],mem[6],ymm1[7,8],mem[9],ymm1[10,11],mem[12],ymm1[13],mem[14],ymm1[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,u,u,5,7,2,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,u,u,5,7,2,4]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm1, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,6,7,16,17,22,23,24,25,30,31,20,21,22,23,16,17,22,23]
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm6[0,1,2],ymm7[3,4,5,6,7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [1,4,6,0,1,4,6,0]
@@ -6048,14 +6048,14 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm4[0],ymm9[1,2],ymm4[3],ymm9[4],ymm4[5],ymm9[6,7],ymm4[8],ymm9[9,10],ymm4[11],ymm9[12],ymm4[13],ymm9[14,15]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm0 = xmm1[0],xmm0[1],xmm1[2],xmm0[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm0, %xmm0
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendw $74, {{[-0-9]+}}(%r{{[sb]}}p), %ymm11, %ymm3 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm3 = ymm11[0],mem[1],ymm11[2],mem[3],ymm11[4,5],mem[6],ymm11[7,8],mem[9],ymm11[10],mem[11],ymm11[12,13],mem[14],ymm11[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,3,u,u,5,0,2,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,3,u,u,5,0,2,7]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm9, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,4,5,18,19,20,21,26,27,16,17,30,31,30,31,18,19,20,21]
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm6[3,4,5,6,7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [2,4,7,0,2,4,7,0]
@@ -6121,7 +6121,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm2 = ymm11[0,1],mem[2],ymm11[3],mem[4],ymm11[5,6],mem[7],ymm11[8,9],mem[10],ymm11[11],mem[12],ymm11[13,14],mem[15]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [8,9,2,3,12,13,6,7,0,1,10,11,0,1,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <1,3,u,u,6,0,3,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [1,3,u,u,6,0,3,5]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm5, %ymm2
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [16,17,22,23,24,25,30,31,0,0,0,0,0,1,6,7,16,17,22,23,24,25,30,31,0,0,0,0,0,1,6,7]
; AVX2-FAST-NEXT: # ymm7 = mem[0,1,0,1]
@@ -6432,10 +6432,10 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: # ymm1 = ymm5[0,1],mem[2],ymm5[3],mem[4],ymm5[5,6],mem[7],ymm5[8,9],mem[10],ymm5[11],mem[12],ymm5[13,14],mem[15]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2],xmm2[3,4],xmm1[5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm0, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm6, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [4,5,14,15,8,9,2,3,12,13,6,7,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm1[0,1,2],ymm2[3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm9 = xmm14[0],xmm4[1],xmm14[2,3]
@@ -6473,7 +6473,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0,1],ymm11[2],ymm10[3],ymm11[4],ymm10[5,6],ymm11[7],ymm10[8,9],ymm11[10],ymm10[11],ymm11[12],ymm10[13,14],ymm11[15]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm9, %xmm12
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0,1,2],xmm12[3,4],xmm9[5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,10,11,20,21,30,31,24,25,18,19,28,29,26,27,16,17,26,27]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm12, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm9, %xmm9
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm9[0,1,2],ymm2[3,4,5,6,7]
@@ -6513,9 +6513,9 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm8[0],ymm5[1,2],ymm8[3],ymm5[4],ymm8[5],ymm5[6,7],ymm8[8],ymm5[9,10],ymm8[11],ymm5[12],ymm8[13],ymm5[14,15]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2],xmm1[3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,u,2,3,12,13,22,23,16,17,26,27,20,21,30,31,30,31,18,19,28,29]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm3, %ymm0, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [6,7,0,1,10,11,4,5,14,15,8,9,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm1[0,1,2],ymm2[3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -6722,7 +6722,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm1[0],xmm3[1,2,3],xmm1[4,5],xmm3[6,7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,1,10,11,4,5,14,15,8,9,2,3,12,13,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,10,11,4,5,14,15,8,9,2,3,12,13,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm1, %ymm3, %ymm3
; AVX512F-SLOW-NEXT: vporq %ymm2, %ymm3, %ymm19
; AVX512F-SLOW-NEXT: vmovdqa 192(%rdi), %ymm15
@@ -6752,7 +6752,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm8[0],ymm7[1],ymm8[2],ymm7[3],ymm8[4,5],ymm7[6],ymm8[7,8],ymm7[9],ymm8[10],ymm7[11],ymm8[12,13],ymm7[14],ymm8[15]
; AVX512F-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm1[0,1],xmm3[2,3],xmm1[4,5,6],xmm3[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm1, %ymm3, %ymm3
; AVX512F-SLOW-NEXT: vpor %ymm2, %ymm3, %ymm2
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -6854,7 +6854,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-SLOW-NEXT: vmovdqa %xmm1, %xmm7
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm14 = xmm14[0,3,2,3,4,5,6,7]
; AVX512F-SLOW-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm14[0],xmm0[0],xmm14[1],xmm0[1]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,8,9,2,3,12,13,6,7,16,17,26,27,26,27,30,31,24,25,18,19,28,29,22,23>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,8,9,2,3,12,13,6,7,16,17,26,27,26,27,30,31,24,25,18,19,28,29,22,23]
; AVX512F-SLOW-NEXT: vpshufb %ymm14, %ymm2, %ymm2
; AVX512F-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm2[2,3,4,5,6,7]
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [12,13,14,15,4,5,14,15,8,9,2,3,12,13,6,7]
@@ -7203,7 +7203,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa64 %ymm4, %ymm30
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm1[0],xmm4[1,2,3],xmm1[4,5],xmm4[6,7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,1,10,11,4,5,14,15,8,9,2,3,12,13,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,10,11,4,5,14,15,8,9,2,3,12,13,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm1, %ymm4, %ymm4
; AVX512F-FAST-NEXT: vmovdqa 384(%rdi), %ymm6
; AVX512F-FAST-NEXT: vmovdqa 416(%rdi), %ymm9
@@ -7220,9 +7220,9 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm5, %xmm7
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm5[0,1,2],xmm7[3,4],xmm5[5,6,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm7 = ymm11[0],ymm10[1],ymm11[2,3],ymm10[4],ymm11[5],ymm10[6],ymm11[7,8],ymm10[9],ymm11[10,11],ymm10[12],ymm11[13],ymm10[14],ymm11[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,4,7,1,4,6,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,4,7,1,4,6,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm10, %ymm7
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,u,u,8,9,14,15,0,1,6,7,16,17,22,23,20,21,22,23,24,25,30,31,16,17,22,23>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,u,u,8,9,14,15,0,1,6,7,16,17,22,23,20,21,22,23,24,25,30,31,16,17,22,23]
; AVX512F-FAST-NEXT: vpshufb %ymm9, %ymm7, %ymm7
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = [8,9,3,2,4,5,7,6]
; AVX512F-FAST-NEXT: vpermt2d %ymm2, %ymm17, %ymm7
@@ -7290,20 +7290,20 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3],xmm0[4,5,6],xmm1[7]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = <2,u,u,u,4,7,1,6>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = [2,u,u,u,4,7,1,6]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm7[0],ymm11[1],ymm7[2,3],ymm11[4],ymm7[5],ymm11[6],ymm7[7,8],ymm11[9],ymm7[10,11],ymm11[12],ymm7[13],ymm11[14],ymm7[15]
; AVX512F-FAST-NEXT: vmovdqa64 %ymm11, %ymm19
; AVX512F-FAST-NEXT: vmovdqa64 %ymm7, %ymm22
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm17, %ymm1
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,0,1,18,19,20,21,26,27,16,17,30,31,128,128,128,128,128,128]
; AVX512F-FAST-NEXT: vpshufb %ymm6, %ymm1, %ymm1
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,3,12,13,6,7,0,1,10,11,4,5,14,15,128,128,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm2, %ymm0, %ymm0
; AVX512F-FAST-NEXT: vpor %ymm1, %ymm0, %ymm10
; AVX512F-FAST-NEXT: vpsrlq $48, %xmm31, %xmm0
; AVX512F-FAST-NEXT: vpshufb %xmm2, %xmm15, %xmm1
; AVX512F-FAST-NEXT: vpunpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = <0,2,5,7,4,7,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = [0,2,5,7,4,7,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm8[0],ymm13[1],ymm8[2,3],ymm13[4],ymm8[5],ymm13[6],ymm8[7,8],ymm13[9],ymm8[10,11],ymm13[12],ymm8[13],ymm13[14],ymm8[15]
; AVX512F-FAST-NEXT: vmovdqa %ymm8, %ymm5
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm20, %ymm4
@@ -7374,7 +7374,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm15[u,u,u,u,u,u,u,u,4,5,14,15,u,u,u,u]
; AVX512F-FAST-NEXT: vmovdqa64 %xmm15, %xmm20
; AVX512F-FAST-NEXT: vpunpckhdq {{.*#+}} xmm0 = xmm0[2],xmm31[2],xmm0[3],xmm31[3]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm24 = <0,3,5,2,5,7,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm24 = [0,3,5,2,5,7,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm13[0],ymm5[1],ymm13[2],ymm5[3],ymm13[4,5],ymm5[6],ymm13[7,8],ymm5[9],ymm13[10],ymm5[11],ymm13[12,13],ymm5[14],ymm13[15]
; AVX512F-FAST-NEXT: vmovdqa64 %ymm5, %ymm30
; AVX512F-FAST-NEXT: vmovdqa64 %ymm13, %ymm26
@@ -7392,7 +7392,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm0[2],ymm1[3],ymm0[4],ymm1[5,6],ymm0[7],ymm1[8,9],ymm0[10],ymm1[11],ymm0[12],ymm1[13,14],ymm0[15]
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm1[0,1,2],xmm3[3,4],xmm1[5,6,7]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm25 = <0,2,u,u,5,7,2,4>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm25 = [0,2,u,u,5,7,2,4]
; AVX512F-FAST-NEXT: vmovdqa64 %ymm19, %ymm0
; AVX512F-FAST-NEXT: vmovdqa64 %ymm22, %ymm1
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5],ymm1[6],ymm0[7,8],ymm1[9],ymm0[10,11],ymm1[12],ymm0[13],ymm1[14],ymm0[15]
@@ -7451,7 +7451,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm1 = xmm15[0],xmm7[1],xmm15[2,3]
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} xmm0 = [6,7,0,1,10,11,0,0,6,7,0,1,10,11,0,0]
; AVX512F-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm2
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = <1,3,6,0,5,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = [1,3,6,0,5,u,u,u]
; AVX512F-FAST-NEXT: vmovdqa64 %ymm30, %ymm9
; AVX512F-FAST-NEXT: vmovdqa64 %ymm26, %ymm10
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm10[0,1],ymm9[2],ymm10[3],ymm9[4],ymm10[5,6],ymm9[7],ymm10[8,9],ymm9[10],ymm10[11],ymm9[12],ymm10[13,14],ymm9[15]
@@ -7484,7 +7484,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm9[0],ymm10[1,2],ymm9[3],ymm10[4],ymm9[5],ymm10[6,7],ymm9[8],ymm10[9,10],ymm9[11],ymm10[12],ymm9[13],ymm10[14,15]
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} xmm2 = [8,9,2,3,12,13,0,0,8,9,2,3,12,13,0,0]
; AVX512F-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm3
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = <1,4,6,3,6,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = [1,4,6,3,6,u,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm27, %ymm1
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,2,3,4,5,2,3,4,5,10,11,0,1,14,15,16,17,18,19,20,21,18,19,20,21,26,27,16,17,30,31]
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm1, %ymm1
@@ -7502,7 +7502,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vmovdqa64 %ymm4, %ymm18
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm3, %xmm4
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} xmm3 = xmm4[0],xmm3[1],xmm4[2],xmm3[3]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <0,3,u,u,5,0,2,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,3,u,u,5,0,2,7]
; AVX512F-FAST-NEXT: vmovdqa64 %ymm21, %ymm15
; AVX512F-FAST-NEXT: vmovdqa64 %ymm29, %ymm10
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm5 = ymm10[0],ymm15[1],ymm10[2],ymm15[3],ymm10[4,5],ymm15[6],ymm10[7,8],ymm15[9],ymm10[10],ymm15[11],ymm10[12,13],ymm15[14],ymm10[15]
@@ -7566,7 +7566,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2,3],ymm2[4,5,6,7]
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm2, %zmm24, %zmm2
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm10[0,1],ymm15[2],ymm10[3],ymm15[4],ymm10[5,6],ymm15[7],ymm10[8,9],ymm15[10],ymm10[11],ymm15[12],ymm10[13,14],ymm15[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <1,3,u,u,6,0,3,5>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [1,3,u,u,6,0,3,5]
; AVX512F-FAST-NEXT: vpermd %ymm3, %ymm4, %ymm3
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,6,7,8,9,14,15,0,1,6,7,0,1,6,7,16,17,22,23,24,25,30,31,16,17,22,23,16,17,22,23]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm3, %ymm3
@@ -7647,7 +7647,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm12 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512BW-NEXT: vpermt2w %zmm10, %zmm12, %zmm13
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,35,40,45,50,55,60,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm8
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm6, %zmm8
; AVX512BW-NEXT: movl $67100672, %eax # imm = 0x3FFE000
@@ -7663,7 +7663,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm15
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm14, %zmm15
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = <1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = [1,6,11,16,21,26,31,36,41,46,51,56,61,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm13
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm12, %zmm13
; AVX512BW-NEXT: vmovdqu16 %zmm15, %zmm13 {%k1}
@@ -7677,7 +7677,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [34,39,44,49,54,59,0,5,10,15,20,25,30,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm15
; AVX512BW-NEXT: vpermt2w %zmm9, %zmm14, %zmm15
; AVX512BW-NEXT: vmovdqu16 %zmm17, %zmm15 {%k1}
@@ -7691,7 +7691,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm10, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = <35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = [35,40,45,50,55,60,1,6,11,16,21,26,31,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512BW-NEXT: vpermt2w %zmm9, %zmm18, %zmm19
; AVX512BW-NEXT: movl $33546240, %eax # imm = 0x1FFE000
@@ -7703,7 +7703,7 @@ define void @load_i16_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm4, %zmm18
; AVX512BW-NEXT: vmovdqu16 %zmm16, %zmm18 {%k1}
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm17, %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = <4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = [4,9,14,19,24,29,34,39,44,49,54,59,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm16, %zmm9
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [20,25,30,35,40,45,50,55,60,0,0,0,0,5,10,15,20,25,30,35,40,45,50,55,60,0,0,0,0,5,10,15]
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-6.ll
index 77207838c2d047c..fffe46d3cbed1d6 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-6.ll
@@ -2126,41 +2126,41 @@ define void @load_i16_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa 160(%rdi), %ymm4
; AVX512BW-NEXT: vmovdqa 128(%rdi), %ymm5
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm5, %ymm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm1
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7],ymm1[8,9,10],ymm0[11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm1 = [0,0,0,3,9,15,21,27,0,0,0,3,9,15,21,27]
; AVX512BW-NEXT: # ymm1 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm5, %ymm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm1 = ymm6[0,1,2],ymm1[3,4,5,6,7],ymm6[8,9,10],ymm1[11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm1 = ymm6[0,1,2,3],ymm1[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [0,0,0,20,26,0,6,12,0,0,0,20,26,0,6,12]
; AVX512BW-NEXT: # ymm6 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %ymm5, %ymm4, %ymm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm7
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0,1,2],ymm6[3,4,5,6,7],ymm7[8,9,10],ymm6[11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0,1,2,3],ymm6[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [0,0,0,21,27,1,7,13,0,0,0,21,27,1,7,13]
; AVX512BW-NEXT: # ymm7 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %ymm5, %ymm4, %ymm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm8
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm7 = ymm8[0,1,2],ymm7[3,4,5,6,7],ymm8[8,9,10],ymm7[11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm7 = ymm8[0,1,2,3],ymm7[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [0,0,0,6,12,18,24,30,0,0,0,6,12,18,24,30]
; AVX512BW-NEXT: # ymm8 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm5, %ymm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm9
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm8 = ymm9[0,1,2,3,4],ymm8[5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm9 = [0,0,1,7,13,19,25,31,0,0,1,7,13,19,25,31]
; AVX512BW-NEXT: # ymm9 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %ymm4, %ymm5, %ymm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0,1,2,3,4],ymm9[5,6,7]
; AVX512BW-NEXT: vmovdqa %ymm0, (%rsi)
@@ -3258,7 +3258,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0],ymm0[1],ymm2[2,3,4,5],ymm0[6],ymm2[7]
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm1, %ymm6
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm4[2],ymm5[3,4],ymm4[5],ymm5[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm2, %xmm0
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm9 = xmm3[2,2,2,2,4,5,6,7]
@@ -3280,7 +3280,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm11, %ymm8
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm9, %ymm8, %ymm5
; AVX2-SLOW-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm3[1,1,2,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,5,5,5,5]
@@ -3295,7 +3295,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm6 = xmm7[1,1,2,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,5,5,5,5]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0],xmm6[1],xmm3[2,3],xmm6[4],xmm3[5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm3, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm5
@@ -3303,7 +3303,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm2, %xmm3
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm3[0,1],xmm0[2],xmm3[3],xmm0[4,5],xmm3[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm7 = ymm14[0],ymm13[1],ymm14[2,3,4,5],ymm13[6],ymm14[7]
; AVX2-SLOW-NEXT: vpshufb %ymm12, %ymm7, %ymm3
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm8 = ymm0[0,1,2],ymm3[3,4,5,6,7]
@@ -3314,7 +3314,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm3 = xmm0[2,2,2,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm9 = xmm3[0,1,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm4
; AVX2-SLOW-NEXT: vpshufb %xmm15, %xmm4, %xmm11
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm9 = xmm11[0,1,2],xmm9[3],xmm11[4,5],xmm9[6],xmm11[7]
@@ -3351,13 +3351,13 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm6 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm5, %xmm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm10, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1],xmm5[2],xmm2[3],xmm5[4,5],xmm2[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm7, %ymm7
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2],ymm7[3,4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm4, %xmm4
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,5]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm4[0,1,2],xmm0[3],xmm4[4,5],xmm0[6],xmm4[7]
@@ -3400,7 +3400,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm10[2,1,2,0,4,5,6,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1,2],xmm2[3],xmm1[4,5,6,7]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm11, %ymm8, %ymm2
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7],ymm2[8,9,10],ymm0[11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,5,4]
@@ -3444,7 +3444,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm5 = xmm5[0,1,3,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,7,7,7,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm5 = xmm7[0],xmm5[1,2],xmm7[3],xmm5[4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-SLOW-NEXT: vpshufb %ymm7, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm6 = ymm8[0,1,2],ymm6[3,4,5,6,7],ymm8[8,9,10],ymm6[11,12,13,14,15]
@@ -3548,7 +3548,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0],ymm0[1],ymm2[2,3,4,5],ymm0[6],ymm2[7]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm1, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm4[2],ymm5[3,4],ymm4[5],ymm5[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm8, %xmm2, %xmm0
; AVX2-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm10 = xmm3[2,2,2,2,4,5,6,7]
@@ -3572,7 +3572,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpbroadcastd {{.*#+}} xmm7 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm3, %xmm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0],xmm3[1],xmm2[2,3],xmm3[4],xmm2[5,6,7]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [10,11,6,7,4,5,6,7,6,7,6,7,2,3,14,15,26,27,22,23,20,21,22,23,22,23,22,23,18,19,30,31]
@@ -3584,7 +3584,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm6, %xmm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm11[0],ymm9[1],ymm11[2,3],ymm9[4],ymm11[5,6],ymm9[7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm6[0],xmm3[1],xmm6[2,3],xmm3[4],xmm6[5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm3, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb %xmm15, %xmm5, %xmm0
@@ -3593,7 +3593,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm12 = [8,9,12,13,0,1,0,0,8,9,12,13,0,1,0,0]
; AVX2-FAST-NEXT: vpshufb %xmm12, %xmm11, %xmm1
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3],xmm1[4,5],xmm0[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm4[0],ymm13[1],ymm4[2,3,4,5],ymm13[6],ymm4[7]
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm7, %ymm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3,4,5,6,7]
@@ -3604,7 +3604,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm2[2],ymm1[3,4],ymm2[5],ymm1[6,7]
; AVX2-FAST-NEXT: vpbroadcastd {{.*#+}} xmm4 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm1, %xmm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm10
; AVX2-FAST-NEXT: vpshufb %xmm9, %xmm10, %xmm8
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm8[0,1,2],xmm6[3],xmm8[4,5],xmm6[6],xmm8[7]
@@ -3637,15 +3637,15 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm12[0,1,2],ymm4[3,4,5,6,7],ymm12[8,9,10],ymm4[11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm12[0,1,2,3],ymm4[4,5,6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm14 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX2-FAST-NEXT: vpshufb %xmm14, %xmm11, %xmm11
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm5[0,1],xmm11[2],xmm5[3],xmm11[4,5],xmm5[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-FAST-NEXT: vpshufb %ymm11, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2],ymm7[3,4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm10, %xmm10
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm10[0,1,2],xmm1[3],xmm10[4,5],xmm1[6],xmm10[7]
@@ -3674,7 +3674,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm5 = xmm0[2,1,0,3]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm0[0,1,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm5, %xmm1
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm2 = xmm4[0,1,2,3,6,5,6,4]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5,6],xmm2[7]
@@ -3689,7 +3689,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm13 = xmm3[2,1,2,0,4,5,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm13[0],xmm7[1,2],xmm13[3],xmm7[4,5,6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm6, %ymm14
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm14[0,1,2],ymm1[3,4,5,6,7],ymm14[8,9,10],ymm1[11,12,13,14,15]
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm14 = xmm14[0,1,2,3,4,6,5,4]
@@ -3722,7 +3722,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm10 = xmm10[0,1,2,3,4,6,5,4]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0,1,2,3,4],xmm10[5,6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,5,6,5]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm5[0,1,2,3],xmm4[4],xmm5[5,6],xmm4[7]
@@ -3730,7 +3730,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm11, %xmm10
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm3 = xmm3[3,1,2,1,4,5,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0],xmm10[1,2],xmm3[3],xmm10[4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-FAST-NEXT: vpshufb %ymm10, %ymm6, %ymm6
; AVX2-FAST-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm6[0,1,2],ymm4[3,4,5,6,7],ymm6[8,9,10],ymm4[11,12,13,14,15]
@@ -3754,7 +3754,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,3,2,1]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm3 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm2, %xmm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm1, %xmm8
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm8[0,1,2,3],xmm5[4],xmm8[5],xmm5[6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm0, %ymm5
@@ -3773,7 +3773,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm3 = mem[0,1,2,3,4],ymm3[5,6,7]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm6 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm2, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5],xmm2[6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -3830,7 +3830,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0],ymm0[1],ymm2[2,3,4,5],ymm0[6],ymm2[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm1, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm4[2],ymm5[3,4],ymm4[5],ymm5[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm2, %xmm0
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm10 = xmm3[2,2,2,2,4,5,6,7]
@@ -3854,7 +3854,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd {{.*#+}} xmm7 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm3, %xmm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0],xmm3[1],xmm2[2,3],xmm3[4],xmm2[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [10,11,6,7,4,5,6,7,6,7,6,7,2,3,14,15,26,27,22,23,20,21,22,23,22,23,22,23,18,19,30,31]
@@ -3866,7 +3866,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm6, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm5 = ymm11[0],ymm9[1],ymm11[2,3],ymm9[4],ymm11[5,6],ymm9[7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm3 = xmm6[0],xmm3[1],xmm6[2,3],xmm3[4],xmm6[5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm3, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm15, %xmm5, %xmm0
@@ -3875,7 +3875,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm12 = [8,9,12,13,0,1,0,0,8,9,12,13,0,1,0,0]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm12, %xmm11, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3],xmm1[4,5],xmm0[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm7 = ymm4[0],ymm13[1],ymm4[2,3,4,5],ymm13[6],ymm4[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm7, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3,4,5,6,7]
@@ -3886,7 +3886,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm2[2],ymm1[3,4],ymm2[5],ymm1[6,7]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd {{.*#+}} xmm4 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm1, %xmm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm10
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm9, %xmm10, %xmm8
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm6 = xmm8[0,1,2],xmm6[3],xmm8[4,5],xmm6[6],xmm8[7]
@@ -3919,15 +3919,15 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm12[0,1,2],ymm4[3,4,5,6,7],ymm12[8,9,10],ymm4[11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm4 = ymm12[0,1,2,3],ymm4[4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm14 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm14, %xmm11, %xmm11
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm5 = xmm5[0,1],xmm11[2],xmm5[3],xmm11[4,5],xmm5[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm7, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2],ymm7[3,4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm10, %xmm10
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm10[0,1,2],xmm1[3],xmm10[4,5],xmm1[6],xmm10[7]
@@ -3956,7 +3956,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm5 = xmm0[2,1,0,3]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm4 = xmm0[0,1,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm5, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm2 = xmm4[0,1,2,3,6,5,6,4]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5,6],xmm2[7]
@@ -3971,7 +3971,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm13 = xmm3[2,1,2,0,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm7 = xmm13[0],xmm7[1,2],xmm13[3],xmm7[4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm13, %ymm6, %ymm14
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm14[0,1,2],ymm1[3,4,5,6,7],ymm14[8,9,10],ymm1[11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm14 = xmm14[0,1,2,3,4,6,5,4]
@@ -4004,7 +4004,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm10 = xmm10[0,1,2,3,4,6,5,4]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0,1,2,3,4],xmm10[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,5,6,5]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm4 = xmm5[0,1,2,3],xmm4[4],xmm5[5,6],xmm4[7]
@@ -4012,7 +4012,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm11, %xmm10
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm3 = xmm3[3,1,2,1,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0],xmm10[1,2],xmm3[3],xmm10[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm10, %ymm6, %ymm6
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm6[0,1,2],ymm4[3,4,5,6,7],ymm6[8,9,10],ymm4[11,12,13,14,15]
@@ -4036,7 +4036,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,3,2,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm3 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm2, %xmm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm1, %xmm8
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm5 = xmm8[0,1,2,3],xmm5[4],xmm8[5],xmm5[6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm0, %ymm5
@@ -4055,7 +4055,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: # ymm3 = mem[0,1,2,3,4],ymm3[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm6 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm2, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5],xmm2[6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -4092,7 +4092,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-LABEL: load_i16_stride6_vf32:
; AVX512F-ONLY-SLOW: # %bb.0:
; AVX512F-ONLY-SLOW-NEXT: subq $136, %rsp
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 224(%rdi), %ymm12
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 192(%rdi), %ymm1
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm1[0],ymm12[1],ymm1[2,3],ymm12[4],ymm1[5,6],ymm12[7]
@@ -4111,7 +4111,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm7, %ymm23
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm5, %ymm25
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm5
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm8, %xmm5, %xmm7
; AVX512F-ONLY-SLOW-NEXT: vpshuflw {{.*#+}} xmm9 = xmm1[2,2,2,2,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpshufd {{.*#+}} xmm9 = xmm9[0,1,2,2]
@@ -4158,10 +4158,10 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm9, %ymm29
; AVX512F-ONLY-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm9 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm9, %xmm0, %xmm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm11, %xmm2, %xmm2
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1],xmm0[2],xmm2[3],xmm0[4,5],xmm2[6,7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm2, %xmm5, %xmm5
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm5[0,1,2],xmm1[3],xmm5[4,5],xmm1[6],xmm5[7]
@@ -4276,7 +4276,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpternlogq $226, %zmm2, %zmm0, %zmm5
; AVX512F-ONLY-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm20
; AVX512F-ONLY-SLOW-NEXT: vpternlogq $184, %zmm5, %zmm17, %zmm20
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm30, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm31, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
@@ -4325,7 +4325,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} ymm15 = ymm10[8,9,4,5,4,5,6,7,0,1,4,5,0,1,12,13,24,25,20,21,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm14 = ymm15[0,1,2,3,4],ymm14[5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vinserti64x4 $1, %ymm14, %zmm0, %zmm14
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[1,1,2,3]
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
@@ -4377,7 +4377,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-LABEL: load_i16_stride6_vf32:
; AVX512F-ONLY-FAST: # %bb.0:
; AVX512F-ONLY-FAST-NEXT: subq $136, %rsp
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 224(%rdi), %ymm15
; AVX512F-ONLY-FAST-NEXT: vmovdqa 192(%rdi), %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0],ymm15[1],ymm1[2,3],ymm15[4],ymm1[5,6],ymm15[7]
@@ -4396,7 +4396,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpbroadcastd {{.*#+}} xmm7 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm1, %xmm8
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm1, %xmm4
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm10, %xmm4, %xmm9
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm8 = xmm9[0,1,2],xmm8[3],xmm9[4,5],xmm8[6],xmm9[7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%rdi), %ymm12
@@ -4440,12 +4440,12 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,4,6]
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm10[4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm0, %ymm29
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm3, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm10 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm10, %xmm11, %xmm11
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0,1],xmm11[2],xmm3[3],xmm11[4,5],xmm3[6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm11, %xmm4, %xmm4
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm4[0,1,2],xmm1[3],xmm4[4,5],xmm1[6],xmm4[7]
@@ -4482,7 +4482,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm13, %ymm21
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm0, %xmm2
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm8 = xmm0[2,1,0,3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm8, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm7 = xmm2[0,1,2,1]
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm2 = xmm7[0,1,2,3,6,5,6,4]
@@ -4530,7 +4530,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} xmm10 = xmm10[3,1,2,1,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm10[0],xmm9[1,2],xmm10[3],xmm9[4,5,6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm10, %xmm8, %xmm8
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,7,5,6,5]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm8[0,1,2,3],xmm7[4],xmm8[5,6],xmm7[7]
@@ -4555,7 +4555,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpternlogq $226, %zmm7, %zmm0, %zmm4
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm20
; AVX512F-ONLY-FAST-NEXT: vpternlogq $184, %zmm4, %zmm17, %zmm20
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm30, %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm31, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
@@ -4570,7 +4570,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[0,3,2,1]
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm8 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm8, %xmm6, %xmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm9, %xmm2, %xmm7
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm7[0,1,2,3],xmm3[4],xmm7[5],xmm3[6,7]
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm3, %ymm0, %ymm3
@@ -4605,7 +4605,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm13, %zmm0, %zmm13
; AVX512F-ONLY-FAST-NEXT: vpbroadcastd {{.*#+}} xmm15 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm15, %xmm10, %xmm10
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm11, %xmm7, %xmm7
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm7[0],xmm10[1],xmm7[2,3],xmm10[4],xmm7[5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm0[6,7,2,3,14,15,26,27,22,23],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
@@ -4615,7 +4615,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm5[1],xmm1[2,3],xmm5[4],xmm1[5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm5 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm5, %xmm6, %xmm6
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3],xmm6[4],xmm2[5],xmm6[6,7]
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -4653,7 +4653,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-LABEL: load_i16_stride6_vf32:
; AVX512DQ-SLOW: # %bb.0:
; AVX512DQ-SLOW-NEXT: pushq %rax
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vmovdqa 224(%rdi), %ymm0
; AVX512DQ-SLOW-NEXT: vmovdqa 192(%rdi), %ymm13
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm8 = ymm13[0],ymm0[1],ymm13[2,3],ymm0[4],ymm13[5,6],ymm0[7]
@@ -4672,7 +4672,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm6, %ymm22
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm4, %ymm23
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm15
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm7, %xmm15, %xmm4
; AVX512DQ-SLOW-NEXT: vpshuflw {{.*#+}} xmm6 = xmm5[2,2,2,2,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpshufd {{.*#+}} xmm6 = xmm6[0,1,2,2]
@@ -4722,10 +4722,10 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm0 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm0, %xmm14, %xmm9
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm14, %xmm8, %xmm8
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm8 = xmm8[0,1],xmm9[2],xmm8[3],xmm9[4,5],xmm8[6,7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm9, %xmm15, %xmm10
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,5,5,5,5]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm5 = xmm10[0,1,2],xmm5[3],xmm10[4,5],xmm5[6],xmm10[7]
@@ -4837,7 +4837,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpternlogq $226, %zmm7, %zmm0, %zmm3
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm16
; AVX512DQ-SLOW-NEXT: vpternlogq $184, %zmm3, %zmm17, %zmm16
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm20, %ymm0
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm30, %ymm1
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
@@ -4886,7 +4886,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} ymm14 = ymm11[8,9,4,5,4,5,6,7,0,1,4,5,0,1,12,13,24,25,20,21,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm9 = ymm14[0,1,2,3,4],ymm9[5,6,7]
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm9, %zmm0, %zmm9
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm14, %xmm0, %xmm0
; AVX512DQ-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[1,1,2,3]
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,5]
@@ -4930,7 +4930,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
;
; AVX512DQ-FAST-LABEL: load_i16_stride6_vf32:
; AVX512DQ-FAST: # %bb.0:
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512DQ-FAST-NEXT: vmovdqa 224(%rdi), %ymm12
; AVX512DQ-FAST-NEXT: vmovdqa 192(%rdi), %ymm2
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm2[0],ymm12[1],ymm2[2,3],ymm12[4],ymm2[5,6],ymm12[7]
@@ -4950,7 +4950,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpbroadcastd {{.*#+}} xmm8 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX512DQ-FAST-NEXT: vpshufb %xmm8, %xmm6, %xmm4
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm6, %xmm15
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm15, %xmm5
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm5[0,1,2],xmm4[3],xmm5[4,5],xmm4[6],xmm5[7]
; AVX512DQ-FAST-NEXT: vmovdqa 32(%rdi), %ymm9
@@ -4998,12 +4998,12 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: kmovw %eax, %k1
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm0, %zmm0, %zmm11 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm7, %xmm7
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm8 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX512DQ-FAST-NEXT: vpshufb %xmm8, %xmm14, %xmm14
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm7[0,1],xmm14[2],xmm7[3],xmm14[4,5],xmm7[6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512DQ-FAST-NEXT: vpshufb %xmm14, %xmm15, %xmm15
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,5,5,5,5]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm15[0,1,2],xmm6[3],xmm15[4,5],xmm6[6],xmm15[7]
@@ -5039,7 +5039,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1],ymm0[2],ymm2[3,4],ymm0[5],ymm2[6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm2
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm14 = xmm0[2,1,0,3]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm14, %xmm3
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm6 = xmm2[0,1,2,1]
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm2 = xmm6[0,1,2,3,6,5,6,4]
@@ -5087,7 +5087,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm9 = xmm15[3,1,2,1,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm8 = xmm9[0],xmm8[1,2],xmm9[3],xmm8[4,5,6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm9, %xmm14, %xmm10
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,5,6,5]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm10[0,1,2,3],xmm6[4],xmm10[5,6],xmm6[7]
@@ -5110,7 +5110,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpternlogq $226, %zmm6, %zmm0, %zmm2
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm16
; AVX512DQ-FAST-NEXT: vpternlogq $184, %zmm2, %zmm17, %zmm16
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm29, %ymm0
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm30, %ymm1
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
@@ -5125,7 +5125,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,3,2,1]
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm8 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX512DQ-FAST-NEXT: vpshufb %xmm8, %xmm5, %xmm4
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm9, %xmm1, %xmm7
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm7[0,1,2,3],xmm4[4],xmm7[5],xmm4[6,7]
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
@@ -5161,7 +5161,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm8, %zmm0, %zmm8
; AVX512DQ-FAST-NEXT: vpbroadcastd {{.*#+}} xmm15 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX512DQ-FAST-NEXT: vpshufb %xmm15, %xmm10, %xmm10
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm14, %xmm7, %xmm7
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm7[0],xmm10[1],xmm7[2,3],xmm10[4],xmm7[5,6,7]
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} ymm4 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[6,7,2,3,14,15,26,27,22,23],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
@@ -5171,7 +5171,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm3[1],xmm0[2,3],xmm3[4],xmm0[5,6,7]
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm3 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm3, %xmm5, %xmm5
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm5[4],xmm1[5],xmm5[6,7]
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -5212,7 +5212,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [0,6,12,18,24,30,0,0,0,0,0,34,40,46,52,58,0,6,12,18,24,30,0,0,0,0,0,34,40,46,52,58]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm4, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm3, %zmm6
; AVX512BW-NEXT: movl $4192256, %edi # imm = 0x3FF800
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -5226,7 +5226,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [1,7,13,19,25,31,0,0,0,0,0,35,41,47,53,59,1,7,13,19,25,31,0,0,0,0,0,35,41,47,53,59]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm4, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm3, %zmm7
; AVX512BW-NEXT: vmovdqu16 %zmm9, %zmm7 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm8, %zmm7 {%k2}
@@ -5236,7 +5236,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [34,40,46,52,58,0,0,0,0,0,0,4,10,16,22,28,34,40,46,52,58,0,0,0,0,0,0,4,10,16,22,28]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = <34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = [34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm5, %zmm10
; AVX512BW-NEXT: movl $2095104, %edi # imm = 0x1FF800
; AVX512BW-NEXT: kmovd %edi, %k2
@@ -5250,11 +5250,11 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [35,41,47,53,59,0,0,0,0,0,0,5,11,17,23,29,35,41,47,53,59,0,0,0,0,0,0,5,11,17,23,29]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = <35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = [35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm5, %zmm11
; AVX512BW-NEXT: vmovdqu16 %zmm9, %zmm11 {%k2}
; AVX512BW-NEXT: vmovdqu16 %zmm8, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm3, %zmm8
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [36,42,48,54,60,0,0,0,0,0,0,6,12,18,24,30,36,42,48,54,60,0,0,0,0,0,0,6,12,18,24,30]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
@@ -5266,7 +5266,7 @@ define void @load_i16_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm8
; AVX512BW-NEXT: vmovdqu16 %zmm8, %zmm9 {%k1}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm3, %zmm8
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [37,43,49,55,61,0,0,0,0,0,1,7,13,19,25,31,37,43,49,55,61,0,0,0,0,0,1,7,13,19,25,31]
; AVX512BW-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
@@ -7501,7 +7501,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm12, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [8,9,4,5,4,5,6,7,0,1,4,5,0,1,12,13,24,25,20,21,20,21,22,23,16,17,20,21,16,17,28,29]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm8[0,1],ymm9[2],ymm8[3,4],ymm9[5],ymm8[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm1, %xmm0
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm5
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm3 = xmm5[2,2,2,2,4,5,6,7]
@@ -7553,7 +7553,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm11, %ymm2
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm14, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm9, %xmm9
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm13 = xmm13[1,1,2,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm13 = xmm13[0,1,2,3,5,5,5,5]
@@ -7589,10 +7589,10 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm13, %xmm14
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm14[0,2,0,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm9, %xmm13, %xmm1
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0,1],xmm0[2],xmm1[3],xmm0[4,5],xmm1[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-SLOW-NEXT: vpblendd $189, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm1 = mem[0],ymm1[1],mem[2,3,4,5],ymm1[6],mem[7]
@@ -7607,7 +7607,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm11 = ymm2[0,1],ymm0[2],ymm2[3,4],ymm0[5],ymm2[6,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm0 = xmm11[2,2,2,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[0,1,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm11, %xmm3
; AVX2-SLOW-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm3, %xmm3
@@ -7705,14 +7705,14 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm9 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX2-SLOW-NEXT: vpshufb %xmm9, %xmm14, %xmm15
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm13, %xmm13
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm15 = xmm13[0,1],xmm15[2],xmm13[3],xmm15[4,5],xmm13[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm13, %ymm0, %ymm14
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm14 = ymm15[0,1,2],ymm14[3,4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm11 = xmm11[0,1,2,3,5,5,5,5]
@@ -7790,7 +7790,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm7[2,1,2,0,4,5,6,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1,2],xmm2[3],xmm1[4,5,6,7]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7],ymm2[8,9,10],ymm0[11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,5,4]
@@ -7902,7 +7902,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm5 = xmm5[0,1,3,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,7,7,7,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm7 = xmm7[0],xmm5[1,2],xmm7[3],xmm5[4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -8112,7 +8112,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [8,9,4,5,4,5,6,7,0,1,4,5,0,1,12,13,24,25,20,21,20,21,22,23,16,17,20,21,16,17,28,29]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm8[0,1],ymm9[2],ymm8[3,4],ymm9[5],ymm8[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm1, %xmm0
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm5
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm4 = xmm5[2,2,2,2,4,5,6,7]
@@ -8166,7 +8166,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpbroadcastd {{.*#+}} xmm2 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm11, %xmm14
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm7, %xmm7
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm14 = xmm7[0],xmm14[1],xmm7[2,3],xmm14[4],xmm7[5,6,7]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [10,11,6,7,4,5,6,7,6,7,6,7,2,3,14,15,26,27,22,23,20,21,22,23,22,23,22,23,18,19,30,31]
@@ -8196,13 +8196,13 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm10 = mem[0],ymm0[1],mem[2,3],ymm0[4],mem[5,6],ymm0[7]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm10, %xmm0
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm14 = xmm0[2,1,0,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm10, %xmm0
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm2 = [8,9,12,13,0,1,0,0,8,9,12,13,0,1,0,0]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm14, %xmm1
; AVX2-FAST-NEXT: vmovdqa %xmm2, %xmm4
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3],xmm1[4,5],xmm0[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendd $189, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm1 = mem[0],ymm1[1],mem[2,3,4,5],ymm1[6],mem[7]
@@ -8217,7 +8217,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm2[0,1],ymm0[2],ymm2[3,4],ymm0[5],ymm2[6,7]
; AVX2-FAST-NEXT: vpbroadcastd {{.*#+}} xmm0 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm9, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm9, %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vpshufb %xmm13, %xmm3, %xmm3
@@ -8311,16 +8311,16 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm7 = ymm11[0,1,2],ymm7[3,4,5,6,7],ymm11[8,9,10],ymm7[11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm11[0,1,2,3],ymm7[4,5,6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm10, %xmm7
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm10 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm14, %xmm15
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm15 = xmm7[0,1],xmm15[2],xmm7[3],xmm15[4,5],xmm7[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm0, %ymm14
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm14 = ymm15[0,1,2],ymm14[3,4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm9 = xmm9[0,1,2,3,5,5,5,5]
@@ -8383,7 +8383,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[0,1,2,1]
; AVX2-FAST-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm12, %xmm1, %xmm0
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm2[0,1,2,3,6,5,6,4]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4],xmm0[5,6],xmm1[7]
@@ -8398,7 +8398,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm2 = xmm7[2,1,2,0,4,5,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1,2],xmm2[3],xmm1[4,5,6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm11, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7],ymm2[8,9,10],ymm0[11,12,13,14,15]
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,5,4]
@@ -8494,7 +8494,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4],xmm3[5,6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm2[4,5,6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpshufhw $103, {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Folded Reload
@@ -8504,7 +8504,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm9, %xmm9
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm7 = xmm7[3,1,2,1,4,5,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm7[0],xmm9[1,2],xmm7[3],xmm9[4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm11, %ymm11
; AVX2-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -8562,7 +8562,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm8 = xmm1[0,3,2,1]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm2 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm7, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm8, %xmm4
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm4[0,1,2,3],xmm1[4],xmm4[5],xmm1[6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -8603,7 +8603,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm2 = mem[0,1,2,3,4],ymm2[5,6,7]
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm1 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm10, %xmm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm12, %xmm12
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm12[0,1,2,3],xmm6[4],xmm12[5],xmm6[6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
@@ -8702,7 +8702,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [8,9,4,5,4,5,6,7,0,1,4,5,0,1,12,13,24,25,20,21,20,21,22,23,16,17,20,21,16,17,28,29]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm8[0,1],ymm9[2],ymm8[3,4],ymm9[5],ymm8[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm1, %xmm0
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm4 = xmm5[2,2,2,2,4,5,6,7]
@@ -8756,7 +8756,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd {{.*#+}} xmm2 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm11, %xmm14
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm7, %xmm7
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm14 = xmm7[0],xmm14[1],xmm7[2,3],xmm14[4],xmm7[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [10,11,6,7,4,5,6,7,6,7,6,7,2,3,14,15,26,27,22,23,20,21,22,23,22,23,22,23,18,19,30,31]
@@ -8786,13 +8786,13 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: # ymm10 = mem[0],ymm0[1],mem[2,3],ymm0[4],mem[5,6],ymm0[7]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm10, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm14 = xmm0[2,1,0,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm10, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm2 = [8,9,12,13,0,1,0,0,8,9,12,13,0,1,0,0]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm14, %xmm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm2, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3],xmm1[4,5],xmm0[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,u,u,u,u,8,9,4,5,16,17,28,29,24,25,28,29,24,25,28,29,24,25,20,21]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpblendd $189, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm1 = mem[0],ymm1[1],mem[2,3,4,5],ymm1[6],mem[7]
@@ -8807,7 +8807,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm9 = ymm2[0,1],ymm0[2],ymm2[3,4],ymm0[5],ymm2[6,7]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd {{.*#+}} xmm0 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm9, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm13 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm9, %xmm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm13, %xmm3, %xmm3
@@ -8901,16 +8901,16 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm7 = ymm11[0,1,2],ymm7[3,4,5,6,7],ymm11[8,9,10],ymm7[11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm11[0,1,2,3],ymm7[4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm10, %xmm7
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm10 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm14, %xmm15
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm15 = xmm7[0,1],xmm15[2],xmm7[3],xmm15[4,5],xmm7[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,u,u,u,u,10,11,6,7,18,19,30,31,26,27,26,27,30,31,26,27,26,27,22,23]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm0, %ymm14
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm14 = ymm15[0,1,2],ymm14[3,4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm15 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm15 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm9 = xmm9[0,1,2,3,5,5,5,5]
@@ -8973,7 +8973,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[0,1,2,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm12, %xmm1, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm1 = xmm2[0,1,2,3,6,5,6,4]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4],xmm0[5,6],xmm1[7]
@@ -8988,7 +8988,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm2 = xmm7[2,1,2,0,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1,2],xmm2[3],xmm1[4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7],ymm2[8,9,10],ymm0[11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,5,4]
@@ -9084,7 +9084,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4],xmm3[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm2[4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufhw $103, {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Folded Reload
@@ -9094,7 +9094,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm9, %xmm9
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm7 = xmm7[3,1,2,1,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm9 = xmm7[0],xmm9[1,2],xmm7[3],xmm9[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm11, %ymm11
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -9152,7 +9152,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm8 = xmm1[0,3,2,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm2 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm7, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm8, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm4[0,1,2,3],xmm1[4],xmm4[5],xmm1[6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -9193,7 +9193,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: # ymm2 = mem[0,1,2,3,4],ymm2[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm1 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm10, %xmm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm12, %xmm12
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm6 = xmm12[0,1,2,3],xmm6[4],xmm12[5],xmm6[6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
@@ -9263,7 +9263,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-LABEL: load_i16_stride6_vf64:
; AVX512F-ONLY-SLOW: # %bb.0:
; AVX512F-ONLY-SLOW-NEXT: subq $1480, %rsp # imm = 0x5C8
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 608(%rdi), %ymm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 576(%rdi), %ymm1
@@ -9282,7 +9282,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm14 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7]
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm14, %xmm2
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm12, %xmm2, %xmm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %xmm2, %xmm21
; AVX512F-ONLY-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm14[2,2,2,2,4,5,6,7]
@@ -9331,7 +9331,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, 672(%rdi), %ymm1, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm1[0],ymm2[1],ymm1[2,3,4,5],ymm2[6],ymm1[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm3, %ymm2, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm3, %ymm17
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm2, %ymm29
@@ -9406,10 +9406,10 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm12 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm12, %xmm15, %xmm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm15, %xmm13, %xmm13
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm13[0,1],xmm0[2],xmm13[3],xmm0[4,5],xmm13[6,7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm13 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm13 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm13, %xmm6, %xmm6
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm4 = xmm6[0,1,2],xmm4[3],xmm6[4,5],xmm4[6],xmm6[7]
@@ -9426,7 +9426,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm13, %xmm5, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm9[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2],xmm2[3],xmm0[4,5],xmm2[6],xmm0[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm2, %ymm1, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7],ymm1[8,9,10],ymm0[11,12,13,14,15]
@@ -9524,7 +9524,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm31, %ymm4
; AVX512F-ONLY-SLOW-NEXT: vpblendd $148, {{[-0-9]+}}(%r{{[sb]}}p), %ymm4, %ymm13 # 32-byte Folded Reload
; AVX512F-ONLY-SLOW-NEXT: # ymm13 = ymm4[0,1],mem[2],ymm4[3],mem[4],ymm4[5,6],mem[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm5, %ymm13, %ymm4
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm5, %ymm24
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0,1,2],ymm3[3,4,5,6,7],ymm4[8,9,10],ymm3[11,12,13,14,15]
@@ -9627,7 +9627,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpshuflw {{.*#+}} xmm10 = xmm15[1,1,1,1,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm10 = xmm10[0,1,2,3,4,5,7,7]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm2 = xmm10[0,1,2,3],xmm2[4],xmm10[5,6],xmm2[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm10, %ymm13, %ymm13
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm13[0,1,2],ymm2[3,4,5,6,7],ymm13[8,9,10],ymm2[11,12,13,14,15]
@@ -9676,7 +9676,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512F-ONLY-SLOW-NEXT: vpblendd $36, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm2 # 32-byte Folded Reload
; AVX512F-ONLY-SLOW-NEXT: # ymm2 = ymm0[0,1],mem[2],ymm0[3,4],mem[5],ymm0[6,7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm1, %xmm2, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm4
; AVX512F-ONLY-SLOW-NEXT: vpshuflw {{.*#+}} xmm3 = xmm4[2,2,2,2,4,5,6,7]
@@ -9785,7 +9785,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm3[0,1,2,3,4],ymm1[5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm0, %ymm16
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm1, %xmm2, %xmm2
; AVX512F-ONLY-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm24[1,1,2,3]
; AVX512F-ONLY-SLOW-NEXT: vpshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,5,5,5,5]
@@ -9894,7 +9894,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-LABEL: load_i16_stride6_vf64:
; AVX512F-ONLY-FAST: # %bb.0:
; AVX512F-ONLY-FAST-NEXT: subq $1480, %rsp # imm = 0x5C8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 608(%rdi), %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa 576(%rdi), %ymm1
@@ -9916,7 +9916,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpbroadcastd {{.*#+}} xmm9 = [8,9,4,5,8,9,4,5,8,9,4,5,8,9,4,5]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm9, %xmm15, %xmm1
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm15, %xmm4
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm2, %xmm4, %xmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm4, %xmm22
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0,1,2],xmm1[3],xmm3[4,5],xmm1[6],xmm3[7]
@@ -9962,7 +9962,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, 672(%rdi), %ymm1, %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0],ymm3[1],ymm1[2,3,4,5],ymm3[6],ymm1[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm4, %ymm16
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm3, %ymm29
@@ -10033,12 +10033,12 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,4,6]
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm9[4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm9, %xmm11, %xmm0
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm11 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm11, %xmm7, %xmm7
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm0[0,1],xmm7[2],xmm0[3],xmm7[4,5],xmm0[6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm6, %xmm6
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm6[0,1,2],xmm5[3],xmm6[4,5],xmm5[6],xmm6[7]
@@ -10055,7 +10055,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm3, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm8[0,1,2,3,5,5,5,5]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0,1,2],xmm1[3],xmm3[4,5],xmm1[6],xmm3[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3,4,5,6,7],ymm2[8,9,10],ymm1[11,12,13,14,15]
@@ -10098,7 +10098,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: # ymm0 = ymm0[0,1],mem[2],ymm0[3,4],mem[5],ymm0[6,7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm3 = xmm0[2,1,2,3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,0,1,4,5,u,u,12,13,12,13,12,13,12,13>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,0,1,4,5,u,u,12,13,12,13,12,13,12,13]
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm2 = xmm1[0,3,2,1]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm2, %xmm1
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm2, %xmm24
@@ -10110,7 +10110,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: # ymm2 = mem[0,1],ymm2[2],mem[3,4],ymm2[5],mem[6,7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm2[2,1,0,3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm13, %xmm4, %xmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm4, %xmm21
; AVX512F-ONLY-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm3[0,1,2,1]
@@ -10153,7 +10153,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm31, %ymm6
; AVX512F-ONLY-FAST-NEXT: vpblendd $148, {{[-0-9]+}}(%r{{[sb]}}p), %ymm6, %ymm15 # 32-byte Folded Reload
; AVX512F-ONLY-FAST-NEXT: # ymm15 = ymm6[0,1],mem[2],ymm6[3],mem[4],ymm6[5,6],mem[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm7, %ymm15, %ymm6
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm7, %ymm25
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm5 = ymm6[0,1,2],ymm5[3,4,5,6,7],ymm6[8,9,10],ymm5[11,12,13,14,15]
@@ -10223,14 +10223,14 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpternlogq $226, %zmm11, %zmm29, %zmm1
; AVX512F-ONLY-FAST-NEXT: vpternlogq $184, %zmm1, %zmm22, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm24, %xmm0
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm1, %xmm25
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm23, %xmm1
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[3,1,2,1,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3],xmm0[4,5,6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm21, %xmm11
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm1, %xmm11, %xmm11
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm20, %xmm13
@@ -10254,7 +10254,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm1, %xmm13, %xmm13
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm14 = xmm14[0,1,2,3,7,5,6,5]
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm13 = xmm13[0,1,2,3],xmm14[4],xmm13[5,6],xmm14[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm14, %ymm15, %ymm15
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm13, %ymm0, %ymm13
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm13 = ymm15[0,1,2],ymm13[3,4,5,6,7],ymm15[8,9,10],ymm13[11,12,13,14,15]
@@ -10301,7 +10301,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512F-ONLY-FAST-NEXT: vpblendd $219, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm12 # 32-byte Folded Reload
; AVX512F-ONLY-FAST-NEXT: # ymm12 = mem[0,1],ymm0[2],mem[3,4],ymm0[5],mem[6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm1, %xmm12, %xmm0
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm12, %xmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -10315,7 +10315,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm7 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm3, %xmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm3, %xmm24
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm15, %xmm4, %xmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm4, %xmm25
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2,3],xmm2[4],xmm3[5],xmm2[6,7]
@@ -10410,7 +10410,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm0, %ymm16
; AVX512F-ONLY-FAST-NEXT: vpbroadcastd {{.*#+}} xmm3 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm3, %xmm9, %xmm9
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm11, %xmm11
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm11[0],xmm9[1],xmm11[2,3],xmm9[4],xmm11[5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,6,7,2,3,14,15,26,27,22,23,128,128,128,128,128,128,128,128,128,128,128,128]
@@ -10425,7 +10425,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm12 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm24, %xmm0
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm12, %xmm0, %xmm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm25, %xmm2
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm11, %xmm2, %xmm15
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm15[0,1,2,3],xmm0[4],xmm15[5],xmm0[6,7]
@@ -10516,7 +10516,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-LABEL: load_i16_stride6_vf64:
; AVX512DQ-SLOW: # %bb.0:
; AVX512DQ-SLOW-NEXT: subq $840, %rsp # imm = 0x348
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vmovdqa 608(%rdi), %ymm0
; AVX512DQ-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-SLOW-NEXT: vmovdqa 576(%rdi), %ymm1
@@ -10534,7 +10534,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm3 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7]
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm3, %xmm2
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm12, %xmm2, %xmm1
; AVX512DQ-SLOW-NEXT: vmovdqa64 %xmm2, %xmm23
; AVX512DQ-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm3[2,2,2,2,4,5,6,7]
@@ -10582,7 +10582,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vinserti128 $1, 672(%rdi), %ymm2, %ymm2
; AVX512DQ-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm4 = ymm2[0],ymm4[1],ymm2[2,3,4,5],ymm4[6],ymm2[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm5, %ymm4, %ymm2
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm5, %ymm27
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm4, %ymm18
@@ -10662,10 +10662,10 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm12 = [2,3,14,15,10,11,0,0,2,3,14,15,10,11,0,0]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm12, %xmm14, %xmm0
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm14, %xmm13, %xmm13
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm13[0,1],xmm0[2],xmm13[3],xmm0[4,5],xmm13[6,7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm13 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm13 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm13, %xmm8, %xmm6
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,5,5,5]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm4 = xmm6[0,1,2],xmm4[3],xmm6[4,5],xmm4[6],xmm6[7]
@@ -10680,7 +10680,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpshufb %xmm13, %xmm7, %xmm2
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm3 = xmm10[0,1,2,3,5,5,5,5]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2],xmm3[3],xmm2[4,5],xmm3[6],xmm2[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm3, %ymm1, %ymm1
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm1[0,1,2],ymm2[3,4,5,6,7],ymm1[8,9,10],ymm2[11,12,13,14,15]
@@ -10781,7 +10781,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm26, %ymm4
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm29, %ymm5
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm13 = ymm4[0,1],ymm5[2],ymm4[3],ymm5[4],ymm4[5,6],ymm5[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm5, %ymm13, %ymm4
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm5, %ymm27
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0,1,2],ymm3[3,4,5,6,7],ymm4[8,9,10],ymm3[11,12,13,14,15]
@@ -10883,7 +10883,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpshuflw {{.*#+}} xmm10 = xmm15[1,1,1,1,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm10 = xmm10[0,1,2,3,4,5,7,7]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm2 = xmm10[0,1,2,3],xmm2[4],xmm10[5,6],xmm2[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm10, %ymm13, %ymm13
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm13[0,1,2],ymm2[3,4,5,6,7],ymm13[8,9,10],ymm2[11,12,13,14,15]
@@ -10924,7 +10924,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512DQ-SLOW-NEXT: vpblendd $219, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm4 # 32-byte Folded Reload
; AVX512DQ-SLOW-NEXT: # ymm4 = mem[0,1],ymm0[2],mem[3,4],ymm0[5],mem[6,7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm0, %xmm4, %xmm1
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm4, %xmm14
; AVX512DQ-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm14[2,2,2,2,4,5,6,7]
@@ -11031,7 +11031,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm12[0,1,2,3,4],ymm0[5,6,7]
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm21
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm2, %xmm4, %xmm4
; AVX512DQ-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm14[1,1,2,3]
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm12 = xmm12[0,1,2,3,5,5,5,5]
@@ -11121,7 +11121,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-LABEL: load_i16_stride6_vf64:
; AVX512DQ-FAST: # %bb.0:
; AVX512DQ-FAST-NEXT: subq $904, %rsp # imm = 0x388
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,12,13,u,u,4,5,u,u,u,u,12,13,14,15]
; AVX512DQ-FAST-NEXT: vmovdqa 608(%rdi), %ymm0
; AVX512DQ-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-FAST-NEXT: vmovdqa 576(%rdi), %ymm1
@@ -11144,7 +11144,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm2, %xmm1
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm2, %xmm4
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm2, %ymm23
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,4,5,u,u,0,1,12,13,u,u,4,5]
; AVX512DQ-FAST-NEXT: vpshufb %xmm2, %xmm4, %xmm3
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm4, %xmm22
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0,1,2],xmm1[3],xmm3[4,5],xmm1[6],xmm3[7]
@@ -11188,7 +11188,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vinserti128 $1, 672(%rdi), %ymm3, %ymm3
; AVX512DQ-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm3[0],ymm5[1],ymm3[2,3,4,5],ymm5[6],ymm3[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,12,13,8,9,12,13,8,9,12,13,4,5,u,u,16,17,28,29,24,25,28,29,24,25,28,29,20,21,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm7, %ymm5, %ymm3
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm7, %ymm26
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm5, %ymm18
@@ -11264,12 +11264,12 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpternlogq $226, %zmm17, %zmm16, %zmm6
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm0, %zmm0, %zmm6 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [2,3,14,15,u,u,6,7,u,u,u,u,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm11, %xmm0
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm11 = [10,11,14,15,2,3,0,0,10,11,14,15,2,3,0,0]
; AVX512DQ-FAST-NEXT: vpshufb %xmm11, %xmm7, %xmm7
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm0[0,1],xmm7[2],xmm0[3],xmm7[4,5],xmm0[6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [6,7,2,3,4,5,u,u,2,3,14,15,u,u,6,7]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm13, %xmm6
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,5,5,5,5]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm6[0,1,2],xmm5[3],xmm6[4,5],xmm5[6],xmm6[7]
@@ -11284,7 +11284,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm3, %xmm3
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm8[0,1,2,3,5,5,5,5]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0,1,2],xmm1[3],xmm3[4,5],xmm1[6],xmm3[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,14,15,10,11,10,11,14,15,10,11,u,u,6,7,18,19,30,31,26,27,26,27,30,31,26,27,u,u,22,23]
; AVX512DQ-FAST-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3,4,5,6,7],ymm2[8,9,10],ymm1[11,12,13,14,15]
@@ -11330,7 +11330,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: # ymm0 = mem[0,1],ymm0[2],mem[3,4],ymm0[5],mem[6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm14 = xmm0[2,1,2,3]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,0,1,4,5,u,u,12,13,12,13,12,13,12,13>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,0,1,4,5,u,u,12,13,12,13,12,13,12,13]
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm15 = xmm1[0,3,2,1]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm15, %xmm1
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm2 = xmm14[2,1,2,0,4,5,6,7]
@@ -11340,7 +11340,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: # ymm2 = ymm2[0,1],mem[2],ymm2[3,4],mem[5],ymm2[6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm2[2,1,0,3]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,0,1,0,1,0,1,u,u,8,9,12,13,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm4, %xmm2
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm4, %xmm19
; AVX512DQ-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm3[0,1,2,1]
@@ -11384,7 +11384,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm30, %ymm6
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm31, %ymm7
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm6[0,1],ymm7[2],ymm6[3],ymm7[4],ymm6[5,6],ymm7[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [4,5,0,1,12,13,14,15,8,9,12,13,0,1,u,u,20,21,16,17,28,29,30,31,24,25,28,29,16,17,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm8, %ymm7, %ymm6
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm8, %ymm27
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm7, %ymm17
@@ -11455,12 +11455,12 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm29
; AVX512DQ-FAST-NEXT: vpternlogq $226, %zmm9, %zmm20, %zmm1
; AVX512DQ-FAST-NEXT: vpternlogq $184, %zmm1, %zmm26, %zmm29
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,2,3,6,7,u,u,14,15,14,15,14,15,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm1, %xmm15, %xmm0
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm1, %xmm28
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm1 = xmm14[3,1,2,1,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3],xmm0[4,5,6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [2,3,2,3,2,3,2,3,u,u,10,11,14,15,u,u]
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm19, %xmm1
; AVX512DQ-FAST-NEXT: vpshufb %xmm3, %xmm1, %xmm9
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm21, %xmm1
@@ -11484,7 +11484,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm18, %xmm0
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm14 = xmm0[0,1,2,3,7,5,6,5]
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm14 = xmm15[0,1,2,3],xmm14[4],xmm15[5,6],xmm14[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [6,7,2,3,14,15,14,15,14,15,10,11,u,u,2,3,22,23,18,19,30,31,30,31,30,31,26,27,u,u,18,19]
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm17, %ymm0
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm0, %ymm0
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm14, %ymm0, %ymm14
@@ -11524,7 +11524,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512DQ-FAST-NEXT: vpblendd $36, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm13 # 32-byte Folded Reload
; AVX512DQ-FAST-NEXT: # ymm13 = ymm0[0,1],mem[2],ymm0[3,4],mem[5],ymm0[6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [8,9,u,u,0,1,12,13,u,u,12,13,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm13, %xmm1
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm13, %xmm14
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm2 = xmm14[2,2,2,2,4,5,6,7]
@@ -11537,7 +11537,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm4 = [8,9,0,0,0,1,12,13,8,9,0,0,0,1,12,13]
; AVX512DQ-FAST-NEXT: vpshufb %xmm4, %xmm3, %xmm2
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm3, %xmm26
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,0,1,4,5,u,u,12,13,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm11, %xmm5, %xmm3
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm5, %xmm28
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2,3],xmm2[4],xmm3[5],xmm2[6,7]
@@ -11631,7 +11631,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm25
; AVX512DQ-FAST-NEXT: vpbroadcastd {{.*#+}} xmm1 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX512DQ-FAST-NEXT: vpshufb %xmm1, %xmm10, %xmm11
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [10,11,u,u,2,3,14,15,u,u,10,11,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm12, %xmm12
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm11 = xmm12[0],xmm11[1],xmm12[2,3],xmm11[4],xmm12[5,6,7]
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [128,128,128,128,128,128,128,128,128,128,6,7,2,3,14,15,26,27,22,23,128,128,128,128,128,128,128,128,128,128,128,128]
@@ -11645,7 +11645,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm12 = [10,11,0,0,2,3,14,15,10,11,0,0,2,3,14,15]
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm26, %xmm4
; AVX512DQ-FAST-NEXT: vpshufb %xmm12, %xmm4, %xmm4
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,2,3,2,3,6,7,u,u,14,15,u,u,u,u]
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm28, %xmm14
; AVX512DQ-FAST-NEXT: vpshufb %xmm13, %xmm14, %xmm14
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm14[0,1,2,3],xmm4[4],xmm14[5],xmm4[6,7]
@@ -11737,7 +11737,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,6,12,18,24,30,36,42,48,54,60,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm8
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm7, %zmm8
; AVX512BW-NEXT: movl $4192256, %edi # imm = 0x3FF800
@@ -11759,7 +11759,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm19
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm18, %zmm19
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [1,7,13,19,25,31,37,43,49,55,61,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm15
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm14, %zmm15
; AVX512BW-NEXT: vmovdqu16 %zmm19, %zmm15 {%k1}
@@ -11777,7 +11777,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm21
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm20, %zmm21
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = <34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = [34,40,46,52,58,0,6,12,18,24,30,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm10, %zmm16, %zmm17
; AVX512BW-NEXT: movl $2095104, %edi # imm = 0x1FF800
@@ -11799,7 +11799,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm22 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm23
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm22, %zmm23
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = <35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = [35,41,47,53,59,1,7,13,19,25,31,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm19
; AVX512BW-NEXT: vpermt2w %zmm10, %zmm18, %zmm19
; AVX512BW-NEXT: vmovdqu16 %zmm23, %zmm19 {%k2}
@@ -11809,7 +11809,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm18
; AVX512BW-NEXT: vmovdqu16 %zmm22, %zmm18 {%k2}
; AVX512BW-NEXT: vmovdqu16 %zmm20, %zmm18 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = <4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [4,10,16,22,28,34,40,46,52,58,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm21
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm20, %zmm21
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm22 = [36,42,48,54,60,0,0,0,0,0,0,6,12,18,24,30,36,42,48,54,60,0,0,0,0,0,0,6,12,18,24,30]
@@ -11829,7 +11829,7 @@ define void @load_i16_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm20
; AVX512BW-NEXT: vmovdqa32 %zmm20, %zmm22 {%k2}
; AVX512BW-NEXT: vmovdqu16 %zmm21, %zmm22 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = <5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [5,11,17,23,29,35,41,47,53,59,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm20, %zmm10
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm13 = [37,43,49,55,61,0,0,0,0,0,1,7,13,19,25,31,37,43,49,55,61,0,0,0,0,0,1,7,13,19,25,31]
; AVX512BW-NEXT: # zmm13 = mem[0,1,2,3,0,1,2,3]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-7.ll
index 08667aed4bb35dd..4b32647f6fb3e79 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i16-stride-7.ll
@@ -2145,7 +2145,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm10 = ymm3[0],ymm2[1],ymm3[2,3,4],ymm2[5],ymm3[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,2,2,3]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[0,1,14,15,12,13,14,15,2,3,6,7,12,13,2,3,16,17,30,31,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm8, %ymm10, %ymm8
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm4 = ymm8[0,1,2,3,4],ymm4[5,6,7]
; AVX2-SLOW-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2317,7 +2317,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm3[0],ymm2[1],ymm3[2,3,4],ymm2[5],ymm3[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,2,2,3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[0,1,14,15,12,13,14,15,2,3,6,7,12,13,2,3,16,17,30,31,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm10[0,1,2,3,4],ymm5[5,6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2326,7 +2326,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm10 = xmm11[0],xmm10[1],xmm11[2,3,4,5],xmm10[6],xmm11[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[2,3,0,1,14,15,12,13,10,11,10,11,10,11,10,11]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,5,1,u,4,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,5,1,u,4,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm13, %ymm11
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm11[2,3,2,3,2,3,2,3,8,9,0,1,6,7,8,9,18,19,18,19,18,19,18,19,24,25,16,17,22,23,24,25]
; AVX2-FAST-NEXT: vmovdqa %xmm12, %xmm11
@@ -2346,7 +2346,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm10 = xmm10[0],xmm12[1],xmm10[2,3,4,5],xmm12[6],xmm10[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[4,5,2,3,0,1,14,15,12,13,12,13,12,13,12,13]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm12 = ymm3[0,1],ymm2[2,3],ymm3[4,5],ymm2[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,6,1,u,5,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,6,1,u,5,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm12, %ymm13, %ymm12
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[0,1,2,3,4,5,2,3,8,9,2,3,4,5,10,11,16,17,18,19,20,21,18,19,24,25,18,19,20,21,26,27]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm10, %ymm12, %ymm10
@@ -2386,7 +2386,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm14 = xmm14[3,1,2,3,4,5,6,7]
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm14 = xmm15[0],xmm14[0],xmm15[1],xmm14[1],xmm15[2],xmm14[2],xmm15[3],xmm14[3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm15 = ymm2[0,1],ymm3[2],ymm2[3,4,5],ymm3[6],ymm2[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,3,7,2,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,3,7,2,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm15, %ymm9, %ymm9
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm15 = ymm6[0],ymm4[1],ymm6[2,3,4],ymm4[5],ymm6[6,7]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm15, %xmm5
@@ -2395,7 +2395,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[0,1,6,7,4,5,2,3,0,1,14,15,u,u,u,u]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm0, %ymm5
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3,4,5,6],ymm13[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm9, %ymm9
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm14 = xmm14[0,1],xmm9[2,3]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm5 = ymm9[0],ymm5[1,2,3,4,5,6,7],ymm9[8],ymm5[9,10,11,12,13,14,15]
@@ -2411,7 +2411,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm5 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,16,17,22,23,24,25,30,31,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1,2,3,4],ymm14[5,6,7],ymm5[8,9,10,11,12],ymm14[13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm14 = ymm2[0,1,2],ymm3[3],ymm2[4,5],ymm3[6],ymm2[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <0,4,7,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [0,4,7,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm14, %ymm15, %ymm14
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm15 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm15, %xmm8
@@ -2433,7 +2433,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm4 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,18,19,20,21,26,27,28,29,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm4[0,1,2,3,4],ymm6[5,6,7],ymm4[8,9,10,11,12],ymm6[13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0],ymm2[1],ymm3[2,3],ymm2[4],ymm3[5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,4,0,3,7,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,4,0,3,7,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpshufb %ymm13, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3],ymm1[4,5],ymm0[6],ymm1[7]
@@ -2484,7 +2484,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm11 = ymm3[0],ymm2[1],ymm3[2,3,4],ymm2[5],ymm3[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[0,1,14,15,12,13,14,15,2,3,6,7,12,13,2,3,16,17,30,31,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm8, %ymm11, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm4 = ymm8[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2573,7 +2573,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqa 208(%rdi), %xmm12
; AVX2-FAST-PERLANE-NEXT: vmovdqa 192(%rdi), %xmm13
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm14 = xmm13[0,1,2,3,4,5],xmm12[6],xmm13[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm15, %xmm14, %xmm14
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm14, %ymm0, %ymm14
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm4[0,1,2,3,4],ymm14[5,6,7],ymm4[8,9,10,11,12],ymm14[13,14,15]
@@ -2595,7 +2595,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm4 = xmm4[u,u,10,11,8,9,6,7,4,5,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm5 = xmm12[0],xmm13[1],xmm12[2,3,4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm0, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm4[0,1,2,3,4],ymm5[5,6,7],ymm4[8,9,10,11,12],ymm5[13,14,15]
@@ -2803,9 +2803,9 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vbroadcasti32x4 {{.*#+}} ymm16 = [2,5,9,12,2,5,9,12]
; AVX512F-FAST-NEXT: # ymm16 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [10,3,6,15,12,13,6,15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,6,9,u,13,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,6,9,u,13,u,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm2, %zmm8
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,5,9,u,12,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,5,9,u,12,u,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm2, %zmm6
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [8,1,12,5,12,5,14,15]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm2, %zmm3
@@ -2877,7 +2877,7 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm7 = ymm7[6,7,4,5,2,3,0,1,14,15],zero,zero,zero,zero,zero,zero,zero,zero,ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpor %ymm7, %ymm9, %ymm7
; AVX512F-FAST-NEXT: vpermd %zmm1, %zmm16, %zmm13
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,3,7,10,14,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,3,7,10,14,u,u,u]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm15 = ymm7[0],ymm15[1,2,3,4,5,6,7],ymm7[8],ymm15[9,10,11,12,13,14,15]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1,2,3],ymm15[4,5,6,7]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm9, %zmm9
@@ -2894,12 +2894,12 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm10, %xmm10
; AVX512F-FAST-NEXT: vpshuflw {{.*#+}} xmm10 = xmm10[3,1,2,3,4,5,6,7]
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} xmm14 = xmm14[0],xmm10[0],xmm14[1],xmm10[1],xmm14[2],xmm10[2],xmm14[3],xmm10[3]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512F-FAST-NEXT: vpshufb %ymm10, %ymm9, %ymm9
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm14[0,1],ymm9[2,3,4,5,6,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm11 = ymm9[0],ymm11[1,2,3,4,5,6,7],ymm9[8],ymm11[9,10,11,12,13,14,15]
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3],ymm11[4,5,6,7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,3,3,u,0,3,7,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,3,3,u,0,3,7,u]
; AVX512F-FAST-NEXT: vpermd %ymm2, %ymm11, %ymm11
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[u,u,u,u,u,u,u,u,u,u,0,1,6,7,8,9,u,u,u,u,u,u,u,u,u,u,16,17,22,23,24,25]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm13[u,u,0,1,6,7,8,9,14,15,u,u,u,u,u,u,u,u,16,17,22,23,24,25,30,31,u,u,u,u,u,u]
@@ -2907,10 +2907,10 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm13 = ymm5[0,1],ymm4[2,3],ymm5[4,5],ymm4[6,7]
; AVX512F-FAST-NEXT: vextracti128 $1, %ymm13, %xmm14
; AVX512F-FAST-NEXT: vpshufb %xmm12, %xmm14, %xmm12
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <1,4,8,11,15,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [1,4,8,11,15,u,u,u]
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm15 = [2,6,9,13,2,6,9,13]
; AVX512F-FAST-NEXT: # ymm15 = mem[0,1,0,1]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = <0,4,7,11,14,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = [0,4,7,11,14,u,u,u]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm13 = xmm13[10,11,6,7,4,5,6,7,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} xmm12 = xmm13[0],xmm12[0],xmm13[1],xmm12[1],xmm13[2],xmm12[2],xmm13[3],xmm12[3]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm16, %zmm13
@@ -2958,48 +2958,48 @@ define void @load_i16_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm4 = [0,0,6,13,20,27,34,41,0,0,6,13,20,27,34,41]
; AVX512BW-NEXT: # ymm4 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [0,0,7,14,21,28,35,42,0,0,7,14,21,28,35,42]
; AVX512BW-NEXT: # ymm5 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm5 = ymm6[0],ymm5[1,2,3,4,5,6,7],ymm6[8],ymm5[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0,1,2,3],ymm5[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [0,1,8,15,22,29,36,43,0,1,8,15,22,29,36,43]
; AVX512BW-NEXT: # ymm6 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0],ymm6[1,2,3,4,5,6,7],ymm7[8],ymm6[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0,1,2,3],ymm6[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [0,2,9,16,23,30,37,44,0,2,9,16,23,30,37,44]
; AVX512BW-NEXT: # ymm7 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm8
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm7 = ymm8[0],ymm7[1,2,3,4,5,6,7],ymm8[8],ymm7[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm7 = ymm8[0,1,2,3],ymm7[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [0,3,10,17,24,31,38,45,0,3,10,17,24,31,38,45]
; AVX512BW-NEXT: # ymm8 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm9
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1,2,3,4,5,6,7],ymm9[8],ymm8[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm8 = ymm9[0,1,2,3],ymm8[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm9 = [0,4,11,18,25,32,39,46,0,4,11,18,25,32,39,46]
; AVX512BW-NEXT: # ymm9 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = <37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = [37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm10
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0],ymm9[1,2,3,4,5,6,7],ymm10[8],ymm9[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm9 = ymm10[0,1,2,3],ymm9[4,5,6,7]
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm10 = [0,5,12,19,26,33,40,47,0,5,12,19,26,33,40,47]
; AVX512BW-NEXT: # ymm10 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm10
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0],ymm10[1,2,3,4,5,6,7],ymm2[8],ymm10[9,10,11,12,13,14,15]
; AVX512BW-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
@@ -4704,7 +4704,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Reload
; AVX2-SLOW-NEXT: vpblendd $187, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14, %ymm5 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm5 = mem[0,1],ymm14[2],mem[3,4,5],ymm14[6],mem[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX2-SLOW-NEXT: vpshufb %ymm7, %ymm5, %ymm8
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm8[0],ymm2[1,2,3,4,5,6,7],ymm8[8],ymm2[9,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,3,0,1]
@@ -4761,7 +4761,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[2,1,2,3]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm7 = xmm7[1,3,2,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm7 = xmm7[0],xmm8[0],xmm7[1],xmm8[1],xmm7[2],xmm8[2],xmm7[3],xmm8[3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29]
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm1[0],ymm2[1,2,3,4,5,6,7],ymm1[8],ymm2[9,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm1 = xmm7[0,1],xmm1[2,3]
@@ -4821,7 +4821,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[0,1,2,1]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,7,6,7]
; AVX2-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm4 = xmm6[4],xmm4[4],xmm6[5],xmm4[5],xmm6[6],xmm4[6],xmm6[7],xmm4[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31]
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm8[0],ymm0[1,2,3,4,5,6,7],ymm8[8],ymm0[9,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm4 = xmm4[0,1],xmm8[2,3]
@@ -4920,7 +4920,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm13[0,1],ymm14[2],ymm13[3,4],ymm14[5],ymm13[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <2,5,1,u,4,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [2,5,1,u,4,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,2,3,2,3,2,3,8,9,0,1,6,7,8,9,18,19,18,19,18,19,18,19,24,25,16,17,22,23,24,25]
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm1, %ymm1
@@ -4949,7 +4949,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu %ymm11, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0],xmm3[1],xmm2[2,3,4,5],xmm3[6],xmm2[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,6,1,u,5,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,6,1,u,5,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,1,2,3,4,5,2,3,8,9,2,3,4,5,10,11,16,17,18,19,20,21,18,19,24,25,18,19,20,21,26,27]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm1, %ymm1
@@ -5119,9 +5119,9 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm0[0,1],ymm3[2],ymm0[3,4,5],ymm3[6],ymm0[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,3,7,2,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,3,7,2,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm9, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm1 = xmm1[0,1],xmm6[2,3]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm6[0],ymm2[1,2,3,4,5,6,7],ymm6[8],ymm2[9,10,11,12,13,14,15]
@@ -5168,9 +5168,9 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm15 = ymm0[0,1,2],ymm3[3],ymm0[4,5],ymm3[6],ymm0[7]
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm9
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,4,7,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,4,7,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm15, %ymm2, %ymm15
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm15, %ymm15
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm15[0],ymm1[1,2,3,4,5,6,7],ymm15[8],ymm1[9,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm6 = xmm6[0,1],xmm15[2,3]
@@ -5223,9 +5223,9 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm9, %xmm9
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm8 = xmm8[0,1,2,3,6,4,6,7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm8 = xmm8[4],xmm9[4],xmm8[5],xmm9[5],xmm8[6],xmm9[6],xmm8[7],xmm9[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <1,4,0,3,7,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [1,4,0,3,7,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm9, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm3[0],ymm0[1,2,3,4,5,6,7],ymm3[8],ymm0[9,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm3 = xmm8[0,1],xmm3[2,3]
@@ -5505,7 +5505,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpblendd $187, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm3 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm3 = mem[0,1],ymm1[2],mem[3,4,5],ymm1[6],mem[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm5, %ymm3, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm8[0,1,2,3,4,5],ymm3[6],ymm8[7,8,9,10,11,12,13],ymm3[14],ymm8[15]
@@ -5570,7 +5570,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm14, %xmm14
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[10,11,6,7,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm10 = xmm10[0],xmm14[0],xmm10[1],xmm14[1],xmm10[2],xmm14[2],xmm10[3],xmm14[3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm2[0],ymm8[1,2,3,4,5,6,7],ymm2[8],ymm8[9,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm2 = xmm10[0,1],xmm2[2,3]
@@ -5627,7 +5627,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm9, %xmm10, %xmm10
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,6,4,6,7]
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm4 = xmm4[4],xmm10[4],xmm4[5],xmm10[5],xmm4[6],xmm10[6],xmm4[7],xmm10[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm10, %ymm5, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm5[0],ymm0[1,2,3,4,5,6,7],ymm5[8],ymm0[9,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm4 = xmm4[0,1],xmm5[2,3]
@@ -5762,7 +5762,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm9[0,1,2],ymm10[3],ymm9[4,5],ymm10[6],ymm9[7]
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2,3,4,5],xmm1[6],xmm2[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm2, %ymm28
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -6059,15 +6059,15 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vbroadcasti32x4 {{.*#+}} ymm19 = [3,6,10,13,3,6,10,13]
; AVX512F-ONLY-FAST-NEXT: # ymm19 = mem[0,1,2,3,0,1,2,3]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 384(%rdi), %zmm14
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <2,6,9,u,13,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [2,6,9,u,13,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm0, %zmm10
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm14, %zmm1, %zmm12
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <1,u,u,u,4,8,11,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [1,u,u,u,4,8,11,15]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm0, %zmm15
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <2,5,9,u,12,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [2,5,9,u,12,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm0, %zmm3
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm14, %zmm16, %zmm7
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,u,u,u,4,7,11,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,u,u,u,4,7,11,14]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm0, %zmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [8,1,12,5,12,5,14,15]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm0, %zmm0
@@ -6110,7 +6110,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm4[0,1,2],ymm5[3],ymm4[4,5],ymm5[6],ymm4[7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm0, %xmm7
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm7[0],xmm0[1],xmm7[2,3,4,5],xmm0[6],xmm7[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm1, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm3 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm3[0,1,6,7,8,9,18,19],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512F-ONLY-FAST-NEXT: vporq %ymm3, %ymm0, %ymm20
@@ -6146,7 +6146,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm9[0,1,2],ymm3[3],ymm9[4,5],ymm3[6],ymm9[7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1],xmm1[2,3,4,5],xmm0[6],xmm1[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} ymm10 = [2,5,2,5,2,5,2,5]
@@ -6160,7 +6160,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm0, %xmm11
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm11[4],xmm0[5],xmm11[6],xmm0[7]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm18, %zmm11
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <1,u,u,u,5,8,12,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [1,u,u,u,5,8,12,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm1, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm12, %zmm1
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm1[2,3,16,17,22,23,24,25,30,31,u,u,u,u,u,u,u,u]
@@ -6202,7 +6202,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[0,1,2,3,2,3,0,1,14,15,12,13,10,11],zero,zero
; AVX512F-ONLY-FAST-NEXT: vpor %ymm1, %ymm10, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm10 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,3,7,10,14,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,3,7,10,14,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0],ymm9[1],ymm3[2,3,4],ymm9[5],ymm3[6,7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0],xmm1[1],xmm3[2],xmm1[3],xmm3[4,5,6,7]
@@ -6222,13 +6222,13 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm9, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} xmm3 = xmm3[3,1,2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm9, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm25, %zmm16, %zmm1
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm16 = [65535,65535,65535,65535,65535,65535,65535,65535,65535,0,0,0,0,0,0,0,0,0,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535]
; AVX512F-ONLY-FAST-NEXT: vpternlogq $184, %zmm0, %zmm16, %zmm23
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,3,3,u,0,3,7,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,3,3,u,0,3,7,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %ymm28, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,0,1,6,7,8,9,u,u,u,u,u,u,u,u,u,u,16,17,22,23,24,25]
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,0,1,6,7,8,9,14,15,u,u,u,u,u,u,16,17,16,17,22,23,24,25,30,31,u,u,u,u,u,u]
@@ -6244,7 +6244,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm3[1],xmm1[2,3,4,5],xmm3[6],xmm1[7]
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,u,u,2,3,0,1,14,15,12,13,10,11]
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,u,u,u,6,9,13,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,u,u,u,6,9,13,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm3, %zmm3
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} ymm3 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm3[2,3,16,17,22,23,24,25],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm3[0,1,2],ymm0[3,4,5,6,7],ymm3[8,9,10],ymm0[11,12,13,14,15]
@@ -6258,7 +6258,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [2,6,9,13,2,6,9,13]
; AVX512F-ONLY-FAST-NEXT: # ymm3 = mem[0,1,0,1]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm25, %zmm3, %zmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,4,7,11,14,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,4,7,11,14,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm17, %zmm0, %zmm17
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm12, %zmm12
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[10,11,6,7,4,5,6,7,u,u,u,u,u,u,u,u]
@@ -6280,7 +6280,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm8[0],ymm6[1],ymm8[2,3,4],ymm6[5],ymm8[6,7]
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm3[0],xmm1[1],xmm3[2],xmm1[3],xmm3[4,5,6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <3,u,u,u,6,10,13,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [3,u,u,u,6,10,13,u]
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm10, %zmm0, %zmm10
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm3, %zmm3
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,u,u,4,5,2,3,0,1,14,15,12,13]
@@ -6290,7 +6290,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[0,1,2,3,6,7,4,5,2,3,0,1,14,15],zero,zero
; AVX512F-ONLY-FAST-NEXT: vpor %ymm3, %ymm1, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <1,4,8,11,15,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [1,4,8,11,15,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm11, %zmm0, %zmm3
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm1, %zmm1
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm9, %ymm1, %ymm1
@@ -6422,7 +6422,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm8[0,1,2],ymm9[3],ymm8[4,5],ymm9[6],ymm8[7]
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2,3,4,5],xmm1[6],xmm2[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm2, %ymm25
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
@@ -6720,13 +6720,13 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vbroadcasti32x4 {{.*#+}} ymm19 = [3,6,10,13,3,6,10,13]
; AVX512DQ-FAST-NEXT: # ymm19 = mem[0,1,2,3,0,1,2,3]
; AVX512DQ-FAST-NEXT: vmovdqa64 384(%rdi), %zmm20
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <2,6,9,u,13,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [2,6,9,u,13,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm0, %zmm7
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = <1,u,u,u,4,8,11,15>
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <2,5,9,u,12,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = [1,u,u,u,4,8,11,15]
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [2,5,9,u,12,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm0, %zmm10
; AVX512DQ-FAST-NEXT: vpermd %zmm20, %zmm18, %zmm0
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,u,u,u,4,7,11,14>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,u,u,u,4,7,11,14]
; AVX512DQ-FAST-NEXT: vpermd %zmm26, %zmm3, %zmm9
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [8,1,12,5,12,5,14,15]
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm3, %zmm3
@@ -6770,7 +6770,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm3[0,1,2],ymm4[3],ymm3[4,5],ymm4[6],ymm3[7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm11
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm11[0],xmm0[1],xmm11[2,3,4,5],xmm0[6],xmm11[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm11, %ymm0, %ymm0
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} ymm10 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm10[0,1,6,7,8,9,18,19],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512DQ-FAST-NEXT: vporq %ymm10, %ymm0, %ymm21
@@ -6795,7 +6795,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm10[0,1,2],ymm12[3],ymm10[4,5],ymm12[6],ymm10[7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm7
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm7[0],xmm0[1],xmm7[2,3,4,5],xmm0[6],xmm7[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm0, %xmm7, %xmm7
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} ymm11 = [2,5,2,5,2,5,2,5]
@@ -6823,7 +6823,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm2, %xmm14
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3],xmm14[4],xmm2[5],xmm14[6],xmm2[7]
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [10,3,6,15,12,13,6,15]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = <1,u,u,u,5,8,12,15>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm27 = [1,u,u,u,5,8,12,15]
; AVX512DQ-FAST-NEXT: vpshufb %ymm0, %ymm2, %ymm0
; AVX512DQ-FAST-NEXT: vpermd %zmm26, %zmm27, %zmm2
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm2[2,3,16,17,22,23,24,25,30,31,u,u,u,u,u,u,u,u]
@@ -6878,9 +6878,9 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm8, %xmm8
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm8 = xmm8[3,1,2,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm11[0],xmm8[0],xmm11[1],xmm8[1],xmm11[2],xmm8[2],xmm11[3],xmm8[3]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <0,3,7,10,14,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,3,7,10,14,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm8, %zmm14
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm8, %ymm14, %ymm14
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm11[0,1],ymm14[2,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm10 = ymm12[0],ymm10[1],ymm12[2,3,4],ymm10[5],ymm12[6,7]
@@ -6902,7 +6902,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm11 = ymm6[0],ymm5[1],ymm6[2,3],ymm5[4],ymm6[5,6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm11, %xmm12
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm11 = xmm11[0],xmm12[1],xmm11[2,3,4,5],xmm12[6],xmm11[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,u,u,u,6,9,13,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,u,u,u,6,9,13,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm26, %zmm12, %zmm12
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,u,u,2,3,0,1,14,15,12,13,10,11]
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -6914,7 +6914,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpermd %zmm23, %zmm17, %zmm0
; AVX512DQ-FAST-NEXT: vpermd %zmm23, %zmm18, %zmm12
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm11, %zmm0, %zmm10 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,3,3,u,0,3,7,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,3,3,u,0,3,7,u]
; AVX512DQ-FAST-NEXT: vpermd %ymm25, %ymm11, %ymm11
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[u,u,u,u,u,u,u,u,u,u,0,1,6,7,8,9,u,u,u,u,u,u,u,u,u,u,16,17,22,23,24,25]
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[0,1,0,1,6,7,8,9,14,15,u,u,u,u,u,u,16,17,16,17,22,23,24,25,30,31,u,u,u,u,u,u]
@@ -6925,7 +6925,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm12 = ymm4[0,1],ymm3[2,3],ymm4[4,5],ymm3[6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm12, %xmm14
; AVX512DQ-FAST-NEXT: vpshufb %xmm7, %xmm14, %xmm7
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,4,7,11,14,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,4,7,11,14,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm14, %zmm14
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm12 = xmm12[10,11,6,7,4,5,6,7,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} xmm7 = xmm12[0],xmm7[0],xmm12[1],xmm7[1],xmm12[2],xmm7[2],xmm12[3],xmm7[3]
@@ -6947,7 +6947,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm14
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm0 = xmm14[0],xmm0[1],xmm14[2],xmm0[3],xmm14[4,5,6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <3,u,u,u,6,10,13,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [3,u,u,u,6,10,13,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm26, %zmm14, %zmm14
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm7 = xmm7[u,u,u,u,u,u,4,5,2,3,0,1,14,15,12,13]
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
@@ -6958,7 +6958,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm7[4,5,6,7]
; AVX512DQ-FAST-NEXT: vpternlogq $226, %zmm11, %zmm19, %zmm12
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm0, %zmm0, %zmm12 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <1,4,8,11,15,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [1,4,8,11,15,u,u,u]
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm28, %zmm0, %zmm7
; AVX512DQ-FAST-NEXT: vpermd %zmm1, %zmm0, %zmm0
; AVX512DQ-FAST-NEXT: vpshufb %ymm8, %ymm0, %ymm0
@@ -7017,7 +7017,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [16,23,30,0,0,0,0,0,0,0,38,45,52,59,2,9,16,23,30,0,0,0,0,0,0,0,38,45,52,59,2,9]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm5, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -7037,7 +7037,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [0,0,0,19,20,21,22,23,24,25,26,27,37,44,51,58,0,0,0,19,20,21,22,23,24,25,26,27,37,44,51,58]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = <1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = [1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm10
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [49,56,63,0,0,0,0,0,0,0,7,14,21,28,35,42,49,56,63,0,0,0,0,0,0,0,7,14,21,28,35,42]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
@@ -7055,7 +7055,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [50,57,0,0,0,0,0,0,0,1,8,15,22,29,36,43,50,57,0,0,0,0,0,0,0,1,8,15,22,29,36,43]
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm4, %zmm11
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm9
; AVX512BW-NEXT: movl $261632, %edi # imm = 0x3FE00
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -7072,7 +7072,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [51,58,0,0,0,0,0,0,0,2,9,16,23,30,37,44,51,58,0,0,0,0,0,0,0,2,9,16,23,30,37,44]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm4, %zmm10
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = <3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = [3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm12
; AVX512BW-NEXT: vmovdqu16 %zmm10, %zmm12 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm12 {%k2}
@@ -7085,7 +7085,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [52,59,0,0,0,0,0,0,0,3,10,17,24,31,38,45,52,59,0,0,0,0,0,0,0,3,10,17,24,31,38,45]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm4, %zmm10
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = <36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = [36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm13
; AVX512BW-NEXT: vmovdqu16 %zmm10, %zmm13 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm13 {%k2}
@@ -7098,7 +7098,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [21,28,0,0,0,0,0,0,0,36,43,50,57,0,7,14,21,28,0,0,0,0,0,0,0,36,43,50,57,0,7,14]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm5, %zmm10
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm14
; AVX512BW-NEXT: vmovdqu16 %zmm10, %zmm14 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm14 {%k2}
@@ -7111,7 +7111,7 @@ define void @load_i16_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [22,29,0,0,0,0,0,0,0,37,44,51,58,1,8,15,22,29,0,0,0,0,0,0,0,37,44,51,58,1,8,15]
; AVX512BW-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm5, %zmm3
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm4
; AVX512BW-NEXT: vmovdqu16 %zmm3, %zmm4 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm6, %zmm4 {%k2}
@@ -10519,7 +10519,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Reload
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm4 = ymm14[0,1],ymm13[2],ymm14[3,4,5],ymm13[6],ymm14[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm4, %ymm5
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,3,0,1]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0,1,2,3,4,5],ymm4[6],ymm5[7,8,9,10,11,12,13],ymm4[14],ymm5[15]
@@ -10646,7 +10646,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[2,1,2,3]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[1,3,2,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm4 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0],ymm1[1,2,3,4,5,6,7],ymm0[8],ymm1[9,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm0 = xmm4[0,1],xmm0[2,3]
@@ -10772,7 +10772,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[0,1,2,1]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,5,7,6,7]
; AVX2-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm10 = xmm10[4],xmm2[4],xmm10[5],xmm2[5],xmm10[6],xmm2[6],xmm10[7],xmm2[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm6, %ymm6
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm7 = ymm6[0],ymm7[1,2,3,4,5,6,7],ymm6[8],ymm7[9,10,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} xmm6 = xmm10[0,1],xmm6[2,3]
@@ -11005,7 +11005,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqu %ymm11, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm9[0,1],ymm11[2],ymm9[3,4],ymm11[5],ymm9[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,5,1,u,4,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,5,1,u,4,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm4, %ymm6
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [2,3,2,3,2,3,2,3,8,9,0,1,6,7,8,9,18,19,18,19,18,19,18,19,24,25,16,17,22,23,24,25]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm6, %ymm8
@@ -11051,7 +11051,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vextracti128 $1, %ymm4, %xmm5
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm4[0],xmm5[1],xmm4[2,3,4,5],xmm5[6],xmm4[7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm9[0,1],ymm11[2,3],ymm9[4,5],ymm11[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,6,1,u,5,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,6,1,u,5,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm4, %ymm6
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,2,3,4,5,2,3,8,9,2,3,4,5,10,11,16,17,18,19,20,21,18,19,24,25,18,19,20,21,26,27]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm6, %ymm8
@@ -11393,9 +11393,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendd $68, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm5 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm5 = ymm0[0,1],mem[2],ymm0[3,4,5],mem[6],ymm0[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,3,7,2,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,3,7,2,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm12, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm5, %ymm5
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm3 = xmm3[0,1],xmm5[2,3]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0],ymm4[1,2,3,4,5,6,7],ymm5[8],ymm4[9,10,11,12,13,14,15]
@@ -11498,7 +11498,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[10,11,6,7,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm13 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm6[0,1,2],ymm5[3],ymm6[4,5],ymm5[6],ymm6[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,4,7,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,4,7,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm2, %ymm5
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29]
@@ -11530,7 +11530,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm13 = ymm7[0,1,2],ymm10[3],ymm7[4,5],ymm10[6],ymm7[7]
; AVX2-FAST-NEXT: vmovdqa %ymm5, %ymm4
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm5, %ymm13
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm13, %ymm13
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm13[0],ymm0[1,2,3,4,5,6,7],ymm13[8],ymm0[9,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm12 = xmm12[0,1],xmm13[2,3]
@@ -11605,9 +11605,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm10, %xmm10
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,6,4,6,7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm7 = xmm7[4],xmm10[4],xmm7[5],xmm10[5],xmm7[6],xmm10[6],xmm7[7],xmm10[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <1,4,0,3,7,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [1,4,0,3,7,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm10, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm6[0],ymm0[1,2,3,4,5,6,7],ymm6[8],ymm0[9,10,11,12,13,14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} xmm6 = xmm7[0,1],xmm6[2,3]
@@ -12203,7 +12203,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpblendd $187, {{[-0-9]+}}(%r{{[sb]}}p), %ymm13, %ymm6 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm6 = mem[0,1],ymm13[2],mem[3,4,5],ymm13[6],mem[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm6, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0,1,2,3,4,5],ymm6[6],ymm7[7,8,9,10,11,12,13],ymm6[14],ymm7[15]
@@ -12329,7 +12329,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd {{.*#+}} xmm4 = [10,11,6,7,10,11,6,7,10,11,6,7,10,11,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm8 = xmm2[0],xmm8[0],xmm2[1],xmm8[1],xmm2[2],xmm8[2],xmm2[3],xmm8[3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,2,3,0,1,14,15,12,13,26,27,26,27,26,27,26,27,18,19,16,17,30,31,28,29]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0],ymm1[1,2,3,4,5,6,7],ymm0[8],ymm1[9,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm0 = xmm8[0,1],xmm0[2,3]
@@ -12447,7 +12447,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm9, %xmm9
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,6,4,6,7]
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm9 = xmm7[4],xmm9[4],xmm7[5],xmm9[5],xmm7[6],xmm9[6],xmm7[7],xmm9[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,4,5,2,3,0,1,14,15,28,29,28,29,28,29,28,29,20,21,18,19,16,17,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm2[0],ymm8[1,2,3,4,5,6,7],ymm2[8],ymm8[9,10,11,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} xmm2 = xmm9[0,1],xmm2[2,3]
@@ -12599,7 +12599,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm1, %ymm18
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3],xmm1[4],xmm0[5],xmm1[6],xmm0[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm0, %ymm1, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 512(%rdi), %ymm3
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 544(%rdi), %ymm4
@@ -12655,7 +12655,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm6, %ymm19
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm6
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm5 = xmm6[0],xmm5[1],xmm6[2,3,4,5],xmm5[6],xmm6[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm6, %ymm5, %ymm5
; AVX512F-ONLY-SLOW-NEXT: vpor %ymm0, %ymm5, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -12725,7 +12725,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0],xmm7[1],xmm6[2,3,4,5],xmm7[6],xmm6[7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [128,128,128,128,128,128,128,128,128,128,10,11,8,9,6,7,20,21,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm7, %ymm5, %ymm8
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm5, %ymm6, %ymm6
; AVX512F-ONLY-SLOW-NEXT: vpor %ymm6, %ymm8, %ymm6
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -12788,7 +12788,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm5[1,3,2,3]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [128,128,128,128,128,128,128,128,128,128,4,5,10,11,0,1,22,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm5, %ymm6, %ymm8
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm6, %ymm7, %ymm7
; AVX512F-ONLY-SLOW-NEXT: vpor %ymm7, %ymm8, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -13041,7 +13041,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm14[0],ymm10[1],ymm14[2,3,4],ymm10[5],ymm14[6,7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm14, %ymm21
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <4,5,u,u,0,1,14,15,8,9,10,11,4,5,6,7,20,21,u,u,16,17,30,31,24,25,26,27,20,21,22,23>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [4,5,u,u,0,1,14,15,8,9,10,11,4,5,6,7,20,21,u,u,16,17,30,31,24,25,26,27,20,21,22,23]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm1, %ymm0, %ymm11
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm11[0],ymm0[1],ymm11[2,3,4,5,6,7,8],ymm0[9],ymm11[10,11,12,13,14,15]
@@ -13049,7 +13049,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm11 = ymm2[0,1],ymm14[2],ymm2[3,4,5],ymm14[6],ymm2[7]
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm11, %xmm12
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm11 = xmm11[0,1,2,3],xmm12[4],xmm11[5],xmm12[6],xmm11[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm12, %ymm11, %ymm11
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [65535,65535,65535,65535,65535,65535,65535,0,0,0,0,0,65535,65535,65535,65535]
; AVX512F-ONLY-SLOW-NEXT: vpternlogq $242, %ymm0, %ymm20, %ymm11
@@ -13106,7 +13106,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm13, %ymm26
; AVX512F-ONLY-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm11
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm11[4],xmm1[5],xmm11[6],xmm1[7]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = <0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm11, %xmm0, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa %ymm10, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm21, %ymm10
@@ -13164,7 +13164,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm16, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4,5],ymm1[6],ymm0[7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %ymm1, %ymm31
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <8,9,8,9,8,9,8,9,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [8,9,8,9,8,9,8,9,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm3, %ymm0, %ymm11
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm11[0,1,2,3,4,5],ymm0[6],ymm11[7,8,9,10,11,12,13],ymm0[14],ymm11[15]
@@ -13201,7 +13201,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,10,11,24,25,22,23,20,21,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm1, %ymm11, %ymm11
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm11[0,1,2],ymm0[3,4,5,6,7],ymm11[8,9,10],ymm0[11,12,13,14,15]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm2, %xmm13, %xmm13
; AVX512F-ONLY-SLOW-NEXT: vpor %ymm11, %ymm13, %ymm11
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm11[0,1,2,3],ymm0[4,5,6,7]
@@ -13273,7 +13273,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm11, %xmm1, %xmm1
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512F-ONLY-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm0[0,1,2],ymm1[3,4,5,6,7],ymm0[8,9,10],ymm1[11,12,13,14,15]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %xmm7, %xmm22
; AVX512F-ONLY-SLOW-NEXT: vpor %ymm0, %ymm2, %ymm0
@@ -13489,7 +13489,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: subq $1800, %rsp # imm = 0x708
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 512(%rdi), %zmm5
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 64(%rdi), %zmm4
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <2,5,9,u,12,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [2,5,9,u,12,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm4, %zmm1, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [8,1,12,5,12,5,14,15]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm4, %zmm3, %zmm2
@@ -13503,7 +13503,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa %ymm5, %ymm12
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm3, %xmm5
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm3[0,1,2,3],xmm5[4],xmm3[5],xmm5[6],xmm3[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm5, %ymm6
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [128,128,128,128,128,128,128,128,128,128,6,7,12,13,2,3,16,17,30,31,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm5, %ymm4, %ymm4
@@ -13539,7 +13539,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm7, %ymm19
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm3, %xmm4
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm3 = xmm4[0],xmm3[1],xmm4[2,3,4,5],xmm3[6],xmm4[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm8, %ymm3, %ymm3
; AVX512F-ONLY-FAST-NEXT: vpor %ymm0, %ymm3, %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -13590,7 +13590,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm27, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm8 = xmm4[0],xmm2[1],xmm4[2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm8, %xmm8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <2,6,9,u,13,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [2,6,9,u,13,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $1, %ymm8, %zmm6, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm18, %ymm1
@@ -13605,7 +13605,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm29, %zmm7, %zmm9
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [128,128,128,128,128,128,128,128,128,128,2,3,4,5,10,11,16,17,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm6, %ymm9, %ymm9
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm10, %ymm8, %ymm8
; AVX512F-ONLY-FAST-NEXT: vpor %ymm9, %ymm8, %ymm8
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -13662,7 +13662,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm31, %zmm1, %zmm10
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [128,128,128,128,128,128,128,128,128,128,4,5,10,11,0,1,22,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm6, %ymm10, %ymm10
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm12, %ymm9, %ymm9
; AVX512F-ONLY-FAST-NEXT: vpor %ymm10, %ymm9, %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -13723,7 +13723,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vbroadcasti32x4 {{.*#+}} ymm24 = [3,6,10,13,3,6,10,13]
; AVX512F-ONLY-FAST-NEXT: # ymm24 = mem[0,1,2,3,0,1,2,3]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm21, %zmm24, %zmm6
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,1,2,3,0,1,6,7,8,9,14,15,12,13,14,15,16,17,18,19,16,17,22,23,24,25,30,31,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,1,2,3,0,1,6,7,8,9,14,15,12,13,14,15,16,17,18,19,16,17,22,23,24,25,30,31,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm11, %ymm6, %ymm6
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm6[0,1,2,3,4,5,6],ymm4[7]
; AVX512F-ONLY-FAST-NEXT: movw $992, %ax # imm = 0x3E0
@@ -13768,7 +13768,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm3 = [8,9,6,7,4,5,10,11,8,9,6,7,4,5,10,11]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm3, %xmm6, %xmm8
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 768(%rdi), %zmm30
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <0,u,u,u,4,7,11,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,u,u,u,4,7,11,14]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm30, %zmm6, %zmm13
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,1,6,7,8,9,14,15,8,9,14,15,4,5,2,3,16,17,22,23,24,25,30,31,24,25,30,31,20,21,18,19]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm10, %ymm13, %ymm13
@@ -13839,7 +13839,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm6[0,1,2,3],xmm1[4],xmm6[5],xmm1[6],xmm6[7]
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq {{.*#+}} xmm6 = [10,11,8,9,6,7,12,13,10,11,8,9,6,7,12,13]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm6, %xmm1, %xmm1
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <1,u,u,u,4,8,11,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [1,u,u,u,4,8,11,15]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm16, %zmm11, %zmm13
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,4,5,10,11,12,13,0,1,0,1,0,1,0,1,18,19,20,21,26,27,28,29,16,17,16,17,16,17,16,17]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm13, %ymm13
@@ -13868,9 +13868,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa %ymm15, %ymm13
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5],xmm2[6],xmm1[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm1
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,u,u,u,5,8,12,15>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,u,u,u,5,8,12,15]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm16, %zmm3, %zmm6
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,2,3,16,17,22,23,24,25,30,31,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm11, %ymm6, %ymm6
@@ -13921,7 +13921,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,0,1,22,23,28,29,18,19,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm12, %ymm6, %ymm6
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm6[0,1,2],ymm0[3,4,5,6,7],ymm6[8,9,10],ymm0[11,12,13,14,15]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm13, %xmm1, %xmm1
; AVX512F-ONLY-FAST-NEXT: vpor %ymm6, %ymm1, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
@@ -13955,10 +13955,10 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vextracti128 $1, %ymm1, %xmm1
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[3,1,2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,3,7,10,14,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,3,7,10,14,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm24 # 64-byte Reload
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm24, %zmm5, %zmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm14, %ymm3, %ymm3
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm3[2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm29 = [65535,65535,65535,65535,65535,65535,65535,65535,65535,0,0,0,0,0,0,0,0,0,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535,65535]
@@ -13976,12 +13976,12 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [2,3,2,3,2,3,2,3,0,1,14,15,12,13,10,11]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm6, %xmm1, %xmm1
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = <2,u,u,u,6,9,13,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = [2,u,u,u,6,9,13,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm16, %zmm20, %zmm13
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,2,3,16,17,22,23,24,25,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm0, %ymm13, %ymm13
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm13[0,1,2],ymm1[3,4,5,6,7],ymm13[8,9,10],ymm1[11,12,13,14,15]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpor %ymm3, %ymm13, %ymm3
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2,3],ymm1[4,5,6,7]
@@ -14029,9 +14029,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm2, %xmm28
; AVX512F-ONLY-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm21 = <0,4,7,11,14,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm21 = [0,4,7,11,14,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm24, %zmm21, %zmm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm8, %ymm2, %ymm2
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm2[2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -14047,12 +14047,12 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [4,5,4,5,4,5,4,5,2,3,0,1,14,15,12,13]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm13, %xmm0, %xmm0
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = <3,u,u,u,6,10,13,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} ymm18 = [3,u,u,u,6,10,13,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm16, %zmm18, %zmm1
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,0,1,18,19,20,21,26,27,128,128,128,128,128,128,128,128,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm1, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7],ymm1[8,9,10],ymm0[11,12,13,14,15]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm10, %xmm6, %xmm6
; AVX512F-ONLY-FAST-NEXT: vpor %ymm1, %ymm6, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5,6,7]
@@ -14096,9 +14096,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm4, %xmm3, %xmm3
; AVX512F-ONLY-FAST-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,6,7]
; AVX512F-ONLY-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,4,8,11,15,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,4,8,11,15,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm24, %zmm3, %zmm8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm10, %ymm8, %ymm8
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm8[2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm19, %ymm5
@@ -14226,7 +14226,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm1, %ymm29
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2,3],xmm1[4],xmm0[5],xmm1[6],xmm0[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm0, %ymm1, %ymm2
; AVX512DQ-SLOW-NEXT: vmovdqa 512(%rdi), %ymm3
; AVX512DQ-SLOW-NEXT: vmovdqa 544(%rdi), %ymm4
@@ -14281,7 +14281,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm6, %ymm19
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm6
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm5 = xmm6[0],xmm5[1],xmm6[2,3,4,5],xmm5[6],xmm6[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm6, %ymm5, %ymm5
; AVX512DQ-SLOW-NEXT: vpor %ymm0, %ymm5, %ymm0
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -14350,7 +14350,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0],xmm7[1],xmm6[2,3,4,5],xmm7[6],xmm6[7]
; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [128,128,128,128,128,128,128,128,128,128,10,11,8,9,6,7,20,21,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm7, %ymm5, %ymm8
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm5, %ymm6, %ymm6
; AVX512DQ-SLOW-NEXT: vpor %ymm6, %ymm8, %ymm3
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -14416,7 +14416,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm5[1,3,2,3]
; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [128,128,128,128,128,128,128,128,128,128,4,5,10,11,0,1,22,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm5, %ymm6, %ymm8
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm6, %ymm7, %ymm7
; AVX512DQ-SLOW-NEXT: vpor %ymm7, %ymm8, %ymm0
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -14669,7 +14669,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm10[6,7]
; AVX512DQ-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm14[0],ymm3[1],ymm14[2,3,4],ymm3[5],ymm14[6,7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <4,5,u,u,0,1,14,15,8,9,10,11,4,5,6,7,20,21,u,u,16,17,30,31,24,25,26,27,20,21,22,23>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [4,5,u,u,0,1,14,15,8,9,10,11,4,5,6,7,20,21,u,u,16,17,30,31,24,25,26,27,20,21,22,23]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm11, %ymm0, %ymm10
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm10[0],ymm0[1],ymm10[2,3,4,5,6,7,8],ymm0[9],ymm10[10,11,12,13,14,15]
@@ -14678,7 +14678,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm10 = ymm13[0,1],ymm15[2],ymm13[3,4,5],ymm15[6],ymm13[7]
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm10, %xmm12
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm10 = xmm10[0,1,2,3],xmm12[4],xmm10[5],xmm12[6],xmm10[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm12, %ymm10, %ymm10
; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} ymm19 = [65535,65535,65535,65535,65535,65535,65535,0,0,0,0,0,65535,65535,65535,65535]
; AVX512DQ-SLOW-NEXT: vpternlogq $242, %ymm0, %ymm19, %ymm10
@@ -14743,7 +14743,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa64 %ymm23, %ymm26
; AVX512DQ-SLOW-NEXT: vextracti128 $1, %ymm11, %xmm12
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} xmm11 = xmm11[0,1,2,3],xmm12[4],xmm11[5],xmm12[6],xmm11[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm12, %xmm0, %xmm0
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm13 = ymm14[0,1],ymm3[2],ymm14[3,4],ymm3[5],ymm14[6,7]
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm13 = ymm13[1,1,2,0]
@@ -14799,7 +14799,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512DQ-SLOW-NEXT: vpblendd $68, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX512DQ-SLOW-NEXT: # ymm0 = ymm0[0,1],mem[2],ymm0[3,4,5],mem[6],ymm0[7]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <8,9,8,9,8,9,8,9,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [8,9,8,9,8,9,8,9,0,1,14,15,u,u,10,11,24,25,24,25,24,25,24,25,16,17,30,31,u,u,26,27]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm11, %ymm0, %ymm12
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm12[0,1,2,3,4,5],ymm0[6],ymm12[7,8,9,10,11,12,13],ymm0[14],ymm12[15]
@@ -14836,7 +14836,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,10,11,24,25,22,23,20,21,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm1, %ymm12, %ymm12
; AVX512DQ-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm12[0,1,2],ymm0[3,4,5,6,7],ymm12[8,9,10],ymm0[11,12,13,14,15]
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm2, %xmm10, %xmm10
; AVX512DQ-SLOW-NEXT: vpor %ymm12, %ymm10, %ymm10
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm22, %zmm0, %zmm28
@@ -15092,7 +15092,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: subq $1304, %rsp # imm = 0x518
; AVX512DQ-FAST-NEXT: vmovdqa64 512(%rdi), %zmm2
; AVX512DQ-FAST-NEXT: vmovdqa64 64(%rdi), %zmm4
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = <2,5,9,u,12,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm17 = [2,5,9,u,12,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm4, %zmm17, %zmm0
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [8,1,12,5,12,5,14,15]
; AVX512DQ-FAST-NEXT: vpermd %zmm4, %zmm3, %zmm1
@@ -15105,7 +15105,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm5, %ymm22
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm3, %xmm5
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm3[0,1,2,3],xmm5[4],xmm3[5],xmm5[6],xmm3[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm3, %ymm5, %ymm6
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [128,128,128,128,128,128,128,128,128,128,6,7,12,13,2,3,16,17,30,31,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm5, %ymm4, %ymm4
@@ -15142,7 +15142,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm7, %ymm16
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm4[0],xmm1[1],xmm4[2,3,4,5],xmm1[6],xmm4[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [2,3,0,1,14,15,12,13,10,11,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm8, %ymm1, %ymm1
; AVX512DQ-FAST-NEXT: vpor %ymm0, %ymm1, %ymm0
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -15194,7 +15194,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa %xmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm14[0],xmm11[1],xmm14[2,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpshufb %xmm7, %xmm4, %xmm7
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,6,9,u,13,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,6,9,u,13,u,u,u]
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm7, %zmm6, %zmm0
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm16, %ymm0
@@ -15208,7 +15208,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpermd %zmm27, %zmm4, %zmm8
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [128,128,128,128,128,128,128,128,128,128,2,3,4,5,10,11,16,17,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm7, %ymm8, %ymm8
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [4,5,2,3,0,1,14,15,12,13,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm9, %ymm6, %ymm6
; AVX512DQ-FAST-NEXT: vpor %ymm6, %ymm8, %ymm2
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -15269,7 +15269,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpermd %zmm27, %zmm16, %zmm9
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [128,128,128,128,128,128,128,128,128,128,4,5,10,11,0,1,22,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm8, %ymm9, %ymm9
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [6,7,4,5,2,3,0,1,14,15,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm13, %ymm4, %ymm4
; AVX512DQ-FAST-NEXT: vpor %ymm4, %ymm9, %ymm0
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -15327,7 +15327,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vbroadcasti32x4 {{.*#+}} ymm28 = [3,6,10,13,3,6,10,13]
; AVX512DQ-FAST-NEXT: # ymm28 = mem[0,1,2,3,0,1,2,3]
; AVX512DQ-FAST-NEXT: vpermd %zmm21, %zmm28, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,1,2,3,0,1,6,7,8,9,14,15,12,13,14,15,16,17,18,19,16,17,22,23,24,25,30,31,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,1,2,3,0,1,6,7,8,9,14,15,12,13,14,15,16,17,18,19,16,17,22,23,24,25,30,31,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm9, %ymm3, %ymm3
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2,3,4,5,6],ymm1[7]
; AVX512DQ-FAST-NEXT: movw $992, %ax # imm = 0x3E0
@@ -15372,7 +15372,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm10 = [8,9,6,7,4,5,10,11,8,9,6,7,4,5,10,11]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm3, %xmm3
; AVX512DQ-FAST-NEXT: vmovdqa64 768(%rdi), %zmm30
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = <0,u,u,u,4,7,11,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = [0,u,u,u,4,7,11,14]
; AVX512DQ-FAST-NEXT: vpermd %zmm30, %zmm20, %zmm14
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,6,7,8,9,14,15,8,9,14,15,4,5,2,3,16,17,22,23,24,25,30,31,24,25,30,31,20,21,18,19]
; AVX512DQ-FAST-NEXT: vpshufb %ymm6, %ymm14, %ymm14
@@ -15440,7 +15440,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm6[0,1,2,3],xmm1[4],xmm6[5],xmm1[6],xmm6[7]
; AVX512DQ-FAST-NEXT: vpbroadcastq {{.*#+}} xmm6 = [10,11,8,9,6,7,12,13,10,11,8,9,6,7,12,13]
; AVX512DQ-FAST-NEXT: vpshufb %xmm6, %xmm1, %xmm1
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = <1,u,u,u,4,8,11,15>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm20 = [1,u,u,u,4,8,11,15]
; AVX512DQ-FAST-NEXT: vpermd %zmm18, %zmm20, %zmm10
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,3,4,5,10,11,12,13,0,1,0,1,0,1,0,1,18,19,20,21,26,27,28,29,16,17,16,17,16,17,16,17]
; AVX512DQ-FAST-NEXT: vpshufb %ymm3, %ymm10, %ymm10
@@ -15469,9 +15469,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm11, %ymm16
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4],xmm1[5],xmm2[6],xmm1[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,0,1,14,15,12,13,10,11,8,9,128,128,128,128,128,128,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm1
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,u,u,u,5,8,12,15>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,u,u,u,5,8,12,15]
; AVX512DQ-FAST-NEXT: vpermd %zmm18, %zmm3, %zmm6
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,2,3,16,17,22,23,24,25,30,31,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm10, %ymm6, %ymm6
@@ -15531,7 +15531,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,0,1,22,23,28,29,18,19,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm10, %ymm6, %ymm6
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm6[0,1,2],ymm0[3,4,5,6,7],ymm6[8,9,10],ymm0[11,12,13,14,15]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,2,3,2,3,0,1,14,15,12,13,10,11,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %xmm12, %xmm2, %xmm2
; AVX512DQ-FAST-NEXT: vpor %ymm6, %ymm2, %ymm2
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
@@ -15570,10 +15570,10 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vextracti128 $1, %ymm0, %xmm0
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[3,1,2,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm21 = <0,3,7,10,14,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm21 = [0,3,7,10,14,u,u,u]
; AVX512DQ-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm28 # 64-byte Reload
; AVX512DQ-FAST-NEXT: vpermd %zmm28, %zmm21, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,0,1,6,7,8,9,14,15,16,17,22,23,20,21,22,23,16,17,22,23,24,25,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm1, %ymm3, %ymm3
; AVX512DQ-FAST-NEXT: vmovdqa %ymm1, %ymm5
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm3[2,3,4,5,6,7]
@@ -15590,12 +15590,12 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [2,3,2,3,2,3,2,3,0,1,14,15,12,13,10,11]
; AVX512DQ-FAST-NEXT: vpshufb %xmm7, %xmm0, %xmm0
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = <2,u,u,u,6,9,13,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = [2,u,u,u,6,9,13,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm18, %zmm16, %zmm2
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,2,3,16,17,22,23,24,25,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm1, %ymm2, %ymm2
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3,4,5,6,7],ymm2[8,9,10],ymm0[11,12,13,14,15]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [0,1,2,3,4,5,2,3,0,1,14,15,12,13,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %xmm10, %xmm3, %xmm3
; AVX512DQ-FAST-NEXT: vpor %ymm2, %ymm3, %ymm2
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
@@ -15642,9 +15642,9 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm2, %xmm24
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm22 = <0,4,7,11,14,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm22 = [0,4,7,11,14,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm28, %zmm22, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,u,u,2,3,4,5,10,11,12,13,18,19,18,19,18,19,18,19,18,19,20,21,26,27,28,29]
; AVX512DQ-FAST-NEXT: vpshufb %ymm9, %ymm3, %ymm3
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm3[2,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm19, %ymm6
@@ -15659,12 +15659,12 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [4,5,4,5,4,5,4,5,2,3,0,1,14,15,12,13]
; AVX512DQ-FAST-NEXT: vpshufb %xmm12, %xmm3, %xmm3
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm3, %ymm0, %ymm3
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = <3,u,u,u,6,10,13,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm16 = [3,u,u,u,6,10,13,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm18, %zmm16, %zmm1
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,0,1,18,19,20,21,26,27,128,128,128,128,128,128,128,128,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm1
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm1[0,1,2],ymm3[3,4,5,6,7],ymm1[8,9,10],ymm3[11,12,13,14,15]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,2,3,6,7,4,5,2,3,0,1,14,15,128,128]
; AVX512DQ-FAST-NEXT: vpshufb %xmm15, %xmm10, %xmm10
; AVX512DQ-FAST-NEXT: vpor %ymm1, %ymm10, %ymm1
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
@@ -15706,7 +15706,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512DQ-FAST-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX512DQ-FAST-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,6,4,6,7]
; AVX512DQ-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <1,4,8,11,15,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [1,4,8,11,15,u,u,u]
; AVX512DQ-FAST-NEXT: vpermd %zmm28, %zmm5, %zmm2
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm29, %ymm15
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm2, %ymm2
@@ -15821,7 +15821,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm14, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,7,14,21,28,35,42,49,56,63,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm8, %zmm9
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
@@ -15850,7 +15850,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm20 = [0,0,0,19,20,21,22,23,24,25,26,27,37,44,51,58,0,0,0,19,20,21,22,23,24,25,26,27,37,44,51,58]
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm20, %zmm19
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm21 = <1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm21 = [1,8,15,22,29,36,43,50,57,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm22
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm21, %zmm22
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm17 = [49,56,63,0,0,0,0,0,0,0,7,14,21,28,35,42,49,56,63,0,0,0,0,0,0,0,7,14,21,28,35,42]
@@ -15878,7 +15878,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm23 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm24
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm23, %zmm24
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = <2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = [2,9,16,23,30,37,44,51,58,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm18, %zmm19
; AVX512BW-NEXT: movl $261632, %edi # imm = 0x3FE00
@@ -15904,7 +15904,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm25 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm26
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm25, %zmm26
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = <3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [3,10,17,24,31,38,45,52,59,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm21
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm20, %zmm21
; AVX512BW-NEXT: vmovdqu16 %zmm26, %zmm21 {%k1}
@@ -15926,7 +15926,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm27 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm28
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm27, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = <36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = [36,43,50,57,0,7,14,21,28,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm23
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm22, %zmm23
; AVX512BW-NEXT: vmovdqu16 %zmm28, %zmm23 {%k1}
@@ -15948,7 +15948,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm27 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm28
; AVX512BW-NEXT: vpermt2w %zmm14, %zmm27, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm29 = <37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm29 = [37,44,51,58,1,8,15,22,29,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm30
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm29, %zmm30
; AVX512BW-NEXT: vmovdqu16 %zmm28, %zmm30 {%k1}
@@ -15968,7 +15968,7 @@ define void @load_i16_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm13 = [22,29,0,0,0,0,0,0,0,37,44,51,58,1,8,15,22,29,0,0,0,0,0,0,0,37,44,51,58,1,8,15]
; AVX512BW-NEXT: # zmm13 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2w %zmm14, %zmm13, %zmm12
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [38,45,52,59,2,9,16,23,30,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm14, %zmm1
; AVX512BW-NEXT: vmovdqu16 %zmm12, %zmm1 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm1 {%k2}
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-3.ll
index e398c68fa81ca1a..4ed9a99c58c3f1b 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-3.ll
@@ -260,18 +260,18 @@ define void @load_i32_stride3_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-SLOW-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-SLOW-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm1 = ymm2[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
@@ -290,21 +290,21 @@ define void @load_i32_stride3_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm5 = [0,1,4,7,0,1,4,7]
; AVX2-FAST-NEXT: # ymm5 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm5, %ymm2
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4],ymm2[5,6,7]
; AVX2-FAST-NEXT: vmovaps %ymm3, (%rsi)
@@ -321,18 +321,18 @@ define void @load_i32_stride3_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd {{.*#+}} ymm3 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm0[0],ymm1[1],ymm0[2,3],ymm1[4],ymm0[5,6],ymm1[7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm5 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm5 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm3 = ymm4[0,1,2,3,4,5],ymm3[6,7]
; AVX2-FAST-PERLANE-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-PERLANE-NEXT: # ymm4 = mem[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm5 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm6 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm6 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm5, %ymm6, %ymm5
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm1 = ymm2[0,1,0,3,4,5,4,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
@@ -534,7 +534,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vbroadcastsd {{.*#+}} ymm6 = [2,5,2,5,2,5,2,5]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm6, %ymm7
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm8 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5,6],ymm3[7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm9 = <0,3,6,1,4,7,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm9 = [0,3,6,1,4,7,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm7 = ymm8[0,1,2,3,4,5],ymm7[6,7]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm6, %ymm6
@@ -545,7 +545,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: # ymm8 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm8, %ymm9
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm10 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm11 = <1,4,7,2,5,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm11 = [1,4,7,2,5,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm10, %ymm11, %ymm10
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm9 = ymm10[0,1,2,3,4],ymm9[5,6,7]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm8, %ymm8
@@ -553,7 +553,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpermps %ymm10, %ymm11, %ymm10
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm8 = ymm10[0,1,2,3,4],ymm8[5,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm3 = <2,5,0,3,6,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm3 = [2,5,0,3,6,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm4 = ymm4[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm4 = ymm4[0,1,0,3]
@@ -583,7 +583,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vbroadcastsd {{.*#+}} ymm6 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm6, %ymm7
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm8 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5,6],ymm3[7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm9 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm9 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-NEXT: vpermps %ymm8, %ymm9, %ymm8
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm7 = ymm8[0,1,2,3,4,5],ymm7[6,7]
; AVX2-FAST-NEXT: vpermps %ymm1, %ymm6, %ymm6
@@ -594,7 +594,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm8 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm8, %ymm9
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm10 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm11 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm11 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm10, %ymm11, %ymm10
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm9 = ymm10[0,1,2,3,4],ymm9[5,6,7]
; AVX2-FAST-NEXT: vpermps %ymm1, %ymm8, %ymm8
@@ -604,7 +604,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm10 = [0,1,0,3,0,1,4,7]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm10, %ymm4
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm2 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm3 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm3 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1,2,3,4],ymm4[5,6,7]
; AVX2-FAST-NEXT: vpermps %ymm1, %ymm10, %ymm1
@@ -631,7 +631,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd {{.*#+}} ymm6 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm6, %ymm7
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm8 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5,6],ymm3[7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm9 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm9 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm8, %ymm9, %ymm8
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm7 = ymm8[0,1,2,3,4,5],ymm7[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm6, %ymm6
@@ -642,7 +642,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: # ymm8 = mem[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm8, %ymm9
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm10 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm11 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm11 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm9 = ymm10[0,1,2,3,4],ymm9[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm8, %ymm8
@@ -650,7 +650,7 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm8 = ymm10[0,1,2,3,4],ymm8[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm3 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm3 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm4 = ymm4[0,1,0,3,4,5,4,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm4 = ymm4[0,1,0,3]
@@ -674,15 +674,15 @@ define void @load_i32_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,5,6,7,8,9,10,17,20,23,26,29]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,5,6,7,8,9,10,18,21,24,27,30]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,16,19,22,25,28,31]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm0
@@ -1089,7 +1089,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vbroadcastsd {{.*#+}} ymm1 = [2,5,2,5,2,5,2,5]
; AVX2-SLOW-NEXT: vpermps %ymm11, %ymm1, %ymm8
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm9 = ymm13[0],ymm10[1],ymm13[2,3],ymm10[4],ymm13[5,6],ymm10[7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm14 = <0,3,6,1,4,7,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm14 = [0,3,6,1,4,7,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm9, %ymm14, %ymm9
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm9[0,1,2,3,4,5],ymm8[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -1112,7 +1112,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm8[0,1,2,3,4,5],ymm1[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm0, (%rsp) # 32-byte Spill
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm1 = ymm13[0,1],ymm10[2],ymm13[3,4],ymm10[5],ymm13[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = <1,4,7,2,5,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = [1,4,7,2,5,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-SLOW-NEXT: # ymm0 = mem[0,1,0,1]
@@ -1136,7 +1136,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpermps %ymm15, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm2[0,1,2,3,4],ymm0[5,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm10[0,1],ymm13[2],ymm10[3,4],ymm13[5],ymm10[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm10 = <2,5,0,3,6,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm10 = [2,5,0,3,6,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm10, %ymm2
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm11 = ymm11[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm11 = ymm11[0,1,0,3]
@@ -1195,7 +1195,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vbroadcastsd {{.*#+}} ymm8 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-NEXT: vpermps %ymm12, %ymm8, %ymm0
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm9 = ymm13[0],ymm10[1],ymm13[2,3],ymm10[4],ymm13[5,6],ymm10[7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm14 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm14 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-NEXT: vpermps %ymm9, %ymm14, %ymm9
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm9[0,1,2,3,4,5],ymm0[6,7]
; AVX2-FAST-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -1218,7 +1218,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm9[0,1,2,3,4,5],ymm8[6,7]
; AVX2-FAST-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm8 = ymm13[0,1],ymm10[2],ymm13[3,4],ymm10[5],ymm13[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm8, %ymm1, %ymm8
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-NEXT: # ymm0 = mem[0,1,0,1]
@@ -1243,7 +1243,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm1 = ymm10[0,1],ymm13[2],ymm10[3,4],ymm13[5],ymm10[6,7]
; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm10 = [0,1,0,3,0,1,4,7]
; AVX2-FAST-NEXT: vpermps %ymm12, %ymm10, %ymm12
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm13 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm13 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm1, %ymm13, %ymm1
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1,2,3,4],ymm12[5,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm3 = ymm3[0,1],ymm6[2],ymm3[3,4],ymm6[5],ymm3[6,7]
@@ -1298,7 +1298,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd {{.*#+}} ymm1 = [2,5,2,5,2,5,2,5]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm11, %ymm1, %ymm8
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm9 = ymm13[0],ymm10[1],ymm13[2,3],ymm10[4],ymm13[5,6],ymm10[7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm14 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm14 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm9, %ymm14, %ymm9
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm9[0,1,2,3,4,5],ymm8[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -1321,7 +1321,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm8[0,1,2,3,4,5],ymm1[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm0, (%rsp) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm1 = ymm13[0,1],ymm10[2],ymm13[3,4],ymm10[5],ymm13[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-PERLANE-NEXT: # ymm0 = mem[0,1,0,1]
@@ -1345,7 +1345,7 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm15, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm2[0,1,2,3,4],ymm0[5,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm10[0,1],ymm13[2],ymm10[3,4],ymm13[5],ymm10[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm10 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm10 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm10, %ymm2
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm11 = ymm11[0,1,0,3,4,5,4,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm11 = ymm11[0,1,0,3]
@@ -1395,21 +1395,21 @@ define void @load_i32_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm3
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm4
; AVX512-NEXT: vmovdqa64 192(%rdi), %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm5, %zmm7
; AVX512-NEXT: vpermt2d %zmm1, %zmm6, %zmm7
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,6,7,8,9,10,17,20,23,26,29]
; AVX512-NEXT: vpermt2d %zmm0, %zmm8, %zmm7
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
; AVX512-NEXT: vpermt2d %zmm4, %zmm8, %zmm6
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = <17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm1, %zmm9
; AVX512-NEXT: vpermt2d %zmm5, %zmm8, %zmm9
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,5,6,7,8,9,10,18,21,24,27,30]
; AVX512-NEXT: vpermt2d %zmm0, %zmm10, %zmm9
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm8
; AVX512-NEXT: vpermt2d %zmm4, %zmm10, %zmm8
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = <2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = [2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u]
; AVX512-NEXT: vpermt2d %zmm1, %zmm10, %zmm5
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,3,4,5,6,7,8,9,16,19,22,25,28,31]
; AVX512-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
@@ -2243,7 +2243,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm14[0],ymm13[1],ymm14[2,3],ymm13[4],ymm14[5,6],ymm13[7]
; AVX2-SLOW-NEXT: vmovups %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovups %ymm13, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm8 = <0,3,6,1,4,7,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm8 = [0,3,6,1,4,7,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm8, %ymm2
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm1 = ymm2[0,1,2,3,4,5],ymm1[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2310,7 +2310,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm14[0,1],ymm13[2],ymm14[3,4],ymm13[5],ymm14[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = <1,4,7,2,5,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm1 = [1,4,7,2,5,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm1, %ymm2
; AVX2-SLOW-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-SLOW-NEXT: # ymm0 = mem[0,1,0,1]
@@ -2370,7 +2370,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-SLOW-NEXT: vblendps $219, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm1 = mem[0,1],ymm1[2],mem[3,4],ymm1[5],mem[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm14 = <2,5,0,3,6,u,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm14 = [2,5,0,3,6,u,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm1, %ymm14, %ymm1
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm2 = ymm10[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm2 = ymm2[0,1,0,3]
@@ -2489,7 +2489,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm2 = ymm14[0],ymm11[1],ymm14[2,3],ymm11[4],ymm14[5,6],ymm11[7]
; AVX2-FAST-NEXT: vmovups %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovups %ymm11, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm6, %ymm2
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm1 = ymm2[0,1,2,3,4,5],ymm1[6,7]
; AVX2-FAST-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2553,7 +2553,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
; AVX2-FAST-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm14[0,1],ymm11[2],ymm14[3,4],ymm11[5],ymm14[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm1, %ymm2
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-NEXT: # ymm0 = mem[0,1,0,1]
@@ -2614,7 +2614,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: # ymm2 = mem[0,1],ymm0[2],mem[3,4],ymm0[5],mem[6,7]
; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm14 = [0,1,0,3,0,1,4,7]
; AVX2-FAST-NEXT: vpermps %ymm8, %ymm14, %ymm0
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm1, %ymm2
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm10 = ymm2[0,1,2,3,4],ymm0[5,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm5[0,1],ymm4[2],ymm5[3,4],ymm4[5],ymm5[6,7]
@@ -2723,7 +2723,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm14[0],ymm13[1],ymm14[2,3],ymm13[4],ymm14[5,6],ymm13[7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm13, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm8 = <0,3,6,1,4,7,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm8 = [0,3,6,1,4,7,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm8, %ymm2
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm1 = ymm2[0,1,2,3,4,5],ymm1[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -2790,7 +2790,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm14[0,1],ymm13[2],ymm14[3,4],ymm13[5],ymm14[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = <1,4,7,2,5,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm1 = [1,4,7,2,5,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm1, %ymm2
; AVX2-FAST-PERLANE-NEXT: vbroadcastf128 {{.*#+}} ymm0 = [0,0,3,6,0,0,3,6]
; AVX2-FAST-PERLANE-NEXT: # ymm0 = mem[0,1,0,1]
@@ -2850,7 +2850,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vblendps $219, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm1 = mem[0,1],ymm1[2],mem[3,4],ymm1[5],mem[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm14 = <2,5,0,3,6,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm14 = [2,5,0,3,6,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm1, %ymm14, %ymm1
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm2 = ymm10[0,1,0,3,4,5,4,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm2 = ymm2[0,1,0,3]
@@ -2957,7 +2957,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm10
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm11
; AVX512-NEXT: vmovdqa64 192(%rdi), %zmm3
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [0,3,6,9,12,15,18,21,24,27,30,u,u,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512-NEXT: vpermt2d %zmm9, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,2,3,4,5,6,7,8,9,10,17,20,23,26,29]
@@ -2970,7 +2970,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vpermt2d %zmm4, %zmm14, %zmm16
; AVX512-NEXT: vpermi2d %zmm10, %zmm2, %zmm12
; AVX512-NEXT: vpermt2d %zmm11, %zmm14, %zmm12
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = <17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [17,20,23,26,29,0,3,6,9,12,15,u,u,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm7, %zmm17
; AVX512-NEXT: vpermt2d %zmm1, %zmm14, %zmm17
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,1,2,3,4,5,6,7,8,9,10,18,21,24,27,30]
@@ -2983,7 +2983,7 @@ define void @load_i32_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vpermt2d %zmm4, %zmm18, %zmm20
; AVX512-NEXT: vpermi2d %zmm2, %zmm10, %zmm14
; AVX512-NEXT: vpermt2d %zmm11, %zmm18, %zmm14
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = <2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [2,5,8,11,14,17,20,23,26,29,u,u,u,u,u,u]
; AVX512-NEXT: vpermt2d %zmm9, %zmm18, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,3,4,5,6,7,8,9,16,19,22,25,28,31]
; AVX512-NEXT: vpermt2d %zmm8, %zmm9, %zmm3
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-5.ll
index 0998868da1f52bf..c4de574173b3962 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-5.ll
@@ -219,13 +219,13 @@ define void @load_i32_stride5_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [0,5,2,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm3 = ymm0[0,1],ymm1[2,3],ymm0[4,5],ymm1[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm3, %ymm2, %ymm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <1,6,3,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [1,6,3,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm4 = ymm0[0,1],ymm1[2,3],ymm0[4,5,6,7]
; AVX2-ONLY-NEXT: vpermd %ymm4, %ymm3, %ymm3
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %xmm4
; AVX2-ONLY-NEXT: vpbroadcastd %xmm4, %xmm5
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} xmm3 = xmm3[0,1,2],xmm5[3]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <2,7,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [2,7,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm6 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm6, %ymm5, %ymm5
; AVX2-ONLY-NEXT: vpbroadcastd 68(%rdi), %xmm6
@@ -234,7 +234,7 @@ define void @load_i32_stride5_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpalignr {{.*#+}} ymm7 = ymm0[12,13,14,15],ymm1[0,1,2,3,4,5,6,7,8,9,10,11],ymm0[28,29,30,31],ymm1[16,17,18,19,20,21,22,23,24,25,26,27]
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,3,2,3]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} xmm6 = xmm7[0,1,2],xmm6[3]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <4,1,6,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [4,1,6,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5],ymm1[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm0, %ymm7, %ymm0
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} xmm0 = xmm0[0,1,2],xmm4[3]
@@ -450,7 +450,7 @@ define void @load_i32_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0,1,2,3],ymm5[4,5,6,7]
; AVX2-ONLY-NEXT: vinserti128 $1, 128(%rdi), %ymm6, %ymm6
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2],ymm6[3],ymm5[4,5,6],ymm6[7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <1,6,3,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [1,6,3,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm7 = ymm1[0,1],ymm2[2,3],ymm1[4,5,6,7]
; AVX2-ONLY-NEXT: vpermd %ymm7, %ymm6, %ymm6
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm7 = ymm3[0,1],ymm4[2,3],ymm3[4,5],ymm4[6,7]
@@ -460,7 +460,7 @@ define void @load_i32_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2],ymm7[3,4,5,6,7]
; AVX2-ONLY-NEXT: vpbroadcastd 144(%rdi), %ymm7
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2,3,4,5,6],ymm7[7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <2,7,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [2,7,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm8 = ymm1[0,1,2,3],ymm2[4,5],ymm1[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm8, %ymm7, %ymm7
; AVX2-ONLY-NEXT: vinserti128 $1, 96(%rdi), %ymm0, %ymm8
@@ -480,7 +480,7 @@ define void @load_i32_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm8 = ymm9[0,1,2,3,4,5],ymm8[6,7]
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm3 = ymm3[0,1],ymm4[0,1]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm3 = ymm3[0,1,2,3,4],ymm4[5],ymm3[6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <4,1,6,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [4,1,6,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1,2,3],ymm1[4,5],ymm2[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm1, %ymm4, %ymm1
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2],ymm3[3,4,5,6,7]
@@ -499,24 +499,24 @@ define void @load_i32_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512: # %bb.0:
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm1
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm2
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm0 = <0,5,10,15,20,25,30,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm0 = [0,5,10,15,20,25,30,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm1, %zmm0
; AVX512-NEXT: vinserti128 $1, 128(%rdi), %ymm0, %ymm3
; AVX512-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm3[3],ymm0[4,5,6],ymm3[7]
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = <17,22,27,0,5,10,15,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = [17,22,27,0,5,10,15,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm2, %zmm3
; AVX512-NEXT: vpbroadcastd 144(%rdi), %ymm4
; AVX512-NEXT: vpblendd {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5,6],ymm4[7]
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = <2,7,12,17,22,27,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = [2,7,12,17,22,27,u,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm1, %zmm4
; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,2,3,4,5,8,13]
; AVX512-NEXT: vmovdqa 128(%rdi), %ymm6
; AVX512-NEXT: vpermi2d %ymm6, %ymm4, %ymm5
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = <3,8,13,18,23,28,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = [3,8,13,18,23,28,u,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm1, %zmm4
; AVX512-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,2,3,4,5,9,14]
; AVX512-NEXT: vpermi2d %ymm6, %ymm4, %ymm7
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = <4,9,14,19,24,29,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm4 = [4,9,14,19,24,29,u,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm1, %zmm4
; AVX512-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,4,5,10,15]
; AVX512-NEXT: vpermi2d %ymm6, %ymm4, %ymm1
@@ -947,7 +947,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vinserti128 $1, 128(%rdi), %ymm7, %ymm7
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm7 = ymm10[0,1,2],ymm7[3],ymm10[4,5,6],ymm7[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <1,6,3,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [1,6,3,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm11 = ymm4[0,1],ymm6[2,3],ymm4[4,5,6,7]
; AVX2-ONLY-NEXT: vpermd %ymm11, %ymm7, %ymm11
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm12 = ymm8[0,1],ymm9[2,3],ymm8[4,5],ymm9[6,7]
@@ -966,7 +966,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpbroadcastd 304(%rdi), %ymm12
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1,2,3,4,5,6],ymm12[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <2,7,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [2,7,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm13 = ymm4[0,1,2,3],ymm6[4,5],ymm4[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm13, %ymm7, %ymm13
; AVX2-ONLY-NEXT: vinserti128 $1, 96(%rdi), %ymm0, %ymm14
@@ -1003,7 +1003,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm8 = ymm8[0,1],ymm9[0,1]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm8 = ymm8[0,1,2,3,4],ymm9[5],ymm8[6,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm4 = ymm6[0,1,2,3],ymm4[4,5],ymm6[6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <4,1,6,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [4,1,6,u]
; AVX2-ONLY-NEXT: vpermd %ymm4, %ymm6, %ymm4
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm4 = ymm4[0,1,2],ymm8[3,4,5,6,7]
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm8 = [2,7,2,7,2,7,2,7]
@@ -1044,7 +1044,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [8,13,18,23,28,0,0,3,8,13,18,23,28,0,0,3]
; AVX512F-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm5
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm6
; AVX512F-NEXT: movw $8064, %ax # imm = 0x1F80
; AVX512F-NEXT: kmovw %eax, %k1
@@ -1054,12 +1054,12 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [9,14,19,24,29,0,0,4,9,14,19,24,29,0,0,4]
; AVX512F-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <17,22,27,0,5,10,15,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [17,22,27,0,5,10,15,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm2, %zmm7
; AVX512F-NEXT: vmovdqa32 %zmm6, %zmm7 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,5,6,7,8,9,10,11,12,18,23,28]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm7, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <2,7,12,17,22,27,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [2,7,12,17,22,27,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm7
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [10,15,20,25,30,0,0,5,10,15,20,25,30,0,0,5]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
@@ -1069,7 +1069,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,4,5,6,7,8,9,10,11,12,19,24,29]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm8, %zmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <3,8,13,18,23,28,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [3,8,13,18,23,28,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm8
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [27,0,5,10,15,0,17,22,27,0,5,10,15,0,17,22]
; AVX512F-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
@@ -1080,7 +1080,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [12,17,22,27,0,0,2,7,12,17,22,27,0,0,2,7]
; AVX512F-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm9
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <4,9,14,19,24,29,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [4,9,14,19,24,29,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm3
; AVX512F-NEXT: movb $56, %al
; AVX512F-NEXT: kmovw %eax, %k1
@@ -1105,7 +1105,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [8,13,18,23,28,0,0,3,8,13,18,23,28,0,0,3]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm6
; AVX512BW-NEXT: movw $8064, %ax # imm = 0x1F80
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -1115,12 +1115,12 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [9,14,19,24,29,0,0,4,9,14,19,24,29,0,0,4]
; AVX512BW-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <17,22,27,0,5,10,15,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [17,22,27,0,5,10,15,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm2, %zmm7
; AVX512BW-NEXT: vmovdqa32 %zmm6, %zmm7 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,5,6,7,8,9,10,11,12,18,23,28]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm7, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <2,7,12,17,22,27,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [2,7,12,17,22,27,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm7
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [10,15,20,25,30,0,0,5,10,15,20,25,30,0,0,5]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
@@ -1130,7 +1130,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,4,5,6,7,8,9,10,11,12,19,24,29]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm8, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <3,8,13,18,23,28,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [3,8,13,18,23,28,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm8
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [27,0,5,10,15,0,17,22,27,0,5,10,15,0,17,22]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
@@ -1141,7 +1141,7 @@ define void @load_i32_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [12,17,22,27,0,0,2,7,12,17,22,27,0,0,2,7]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <4,9,14,19,24,29,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [4,9,14,19,24,29,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm3
; AVX512BW-NEXT: movb $56, %al
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -2069,7 +2069,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vinserti128 $1, 128(%rdi), %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3],ymm1[4,5,6],ymm0[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <1,6,3,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [1,6,3,u]
; AVX2-ONLY-NEXT: vmovdqu (%rsp), %ymm8 # 32-byte Reload
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm4[0,1],ymm8[2,3],ymm4[4,5,6,7]
@@ -2113,7 +2113,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpbroadcastd 464(%rdi), %ymm1
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <2,7,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [2,7,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm4[0,1,2,3],ymm8[4,5],ymm4[6,7]
; AVX2-ONLY-NEXT: vmovdqa %ymm8, %ymm14
; AVX2-ONLY-NEXT: vpermd %ymm1, %ymm9, %ymm1
@@ -2217,7 +2217,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: # ymm0 = mem[0,1,2,3],ymm11[4,5],mem[6,7]
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm5 = ymm8[0,1],ymm1[0,1]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3,4],ymm1[5],ymm5[6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <4,1,6,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [4,1,6,u]
; AVX2-ONLY-NEXT: vpermd %ymm0, %ymm10, %ymm0
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2],ymm5[3,4,5,6,7]
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm5 = [2,7,2,7,2,7,2,7]
@@ -2307,7 +2307,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm12 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm13
; AVX512F-NEXT: vpermt2d %zmm11, %zmm12, %zmm13
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm8
; AVX512F-NEXT: vpermt2d %zmm10, %zmm6, %zmm8
; AVX512F-NEXT: movw $8064, %ax # imm = 0x1F80
@@ -2323,7 +2323,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm15
; AVX512F-NEXT: vpermt2d %zmm11, %zmm14, %zmm15
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = <17,22,27,0,5,10,15,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = [17,22,27,0,5,10,15,u]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm13
; AVX512F-NEXT: vpermt2d %zmm1, %zmm12, %zmm13
; AVX512F-NEXT: vmovdqa32 %zmm15, %zmm13 {%k1}
@@ -2333,7 +2333,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2d %zmm0, %zmm3, %zmm12
; AVX512F-NEXT: vmovdqa32 %zmm14, %zmm12 {%k1}
; AVX512F-NEXT: vpermt2d %zmm2, %zmm15, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = <2,7,12,17,22,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = [2,7,12,17,22,27,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm17
; AVX512F-NEXT: vpermt2d %zmm10, %zmm16, %zmm17
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm15 = [10,15,20,25,30,0,0,5,10,15,20,25,30,0,0,5]
@@ -2349,7 +2349,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2d %zmm3, %zmm0, %zmm16
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm15 {%k1}
; AVX512F-NEXT: vpermt2d %zmm2, %zmm17, %zmm15
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = <3,8,13,18,23,28,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = [3,8,13,18,23,28,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm17
; AVX512F-NEXT: vpermt2d %zmm10, %zmm16, %zmm17
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm18 = [27,0,5,10,15,0,17,22,27,0,5,10,15,0,17,22]
@@ -2366,7 +2366,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm16 = [12,17,22,27,0,0,2,7,12,17,22,27,0,0,2,7]
; AVX512F-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermt2d %zmm11, %zmm16, %zmm9
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <4,9,14,19,24,29,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [4,9,14,19,24,29,u,u]
; AVX512F-NEXT: vpermt2d %zmm10, %zmm11, %zmm1
; AVX512F-NEXT: movb $56, %al
; AVX512F-NEXT: kmovw %eax, %k1
@@ -2406,7 +2406,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm12 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm12, %zmm13
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,5,10,15,20,25,30,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,5,10,15,20,25,30,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm6, %zmm8
; AVX512BW-NEXT: movw $8064, %ax # imm = 0x1F80
@@ -2422,7 +2422,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm15
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm14, %zmm15
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = <17,22,27,0,5,10,15,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = [17,22,27,0,5,10,15,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm12, %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm15, %zmm13 {%k1}
@@ -2432,7 +2432,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm3, %zmm12
; AVX512BW-NEXT: vmovdqa32 %zmm14, %zmm12 {%k1}
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm15, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = <2,7,12,17,22,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = [2,7,12,17,22,27,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm16, %zmm17
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm15 = [10,15,20,25,30,0,0,5,10,15,20,25,30,0,0,5]
@@ -2448,7 +2448,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm0, %zmm16
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm15 {%k1}
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm17, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = <3,8,13,18,23,28,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = [3,8,13,18,23,28,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm16, %zmm17
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm18 = [27,0,5,10,15,0,17,22,27,0,5,10,15,0,17,22]
@@ -2465,7 +2465,7 @@ define void @load_i32_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm16 = [12,17,22,27,0,0,2,7,12,17,22,27,0,0,2,7]
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm16, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = <4,9,14,19,24,29,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = [4,9,14,19,24,29,u,u]
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm11, %zmm1
; AVX512BW-NEXT: movb $56, %al
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -4381,7 +4381,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vinserti128 $1, 128(%rdi), %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3],ymm1[4,5,6],ymm0[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <1,6,3,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [1,6,3,u]
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm6[0,1],ymm5[2,3],ymm6[4,5,6,7]
@@ -4476,7 +4476,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vpbroadcastd 1104(%rdi), %ymm1
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <2,7,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [2,7,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm6[0,1,2,3],ymm5[4,5],ymm6[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm0, %ymm3, %ymm0
; AVX2-ONLY-NEXT: vinserti128 $1, 256(%rdi), %ymm0, %ymm1
@@ -4672,7 +4672,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm7[0,1],ymm6[0,1]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4],ymm6[5],ymm0[6,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm1[0,1,2,3],ymm5[4,5],ymm1[6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = <4,1,6,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = [4,1,6,u]
; AVX2-ONLY-NEXT: vpermd %ymm5, %ymm8, %ymm5
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm0 = ymm5[0,1,2],ymm0[3,4,5,6,7]
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm7 = [2,7,2,7,2,7,2,7]
@@ -4843,7 +4843,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm4
; AVX512F-NEXT: vpermt2d %zmm30, %zmm19, %zmm4
; AVX512F-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = <0,5,10,15,20,25,30,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = [0,5,10,15,20,25,30,u]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm4
; AVX512F-NEXT: vpermt2d %zmm21, %zmm15, %zmm4
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm5
@@ -4858,7 +4858,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm5
; AVX512F-NEXT: vpermt2d %zmm30, %zmm16, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = <17,22,27,0,5,10,15,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = [17,22,27,0,5,10,15,u]
; AVX512F-NEXT: vmovdqa64 %zmm21, %zmm24
; AVX512F-NEXT: vpermt2d %zmm13, %zmm12, %zmm24
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm5
@@ -4868,7 +4868,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermt2d %zmm26, %zmm16, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermi2d %zmm20, %zmm1, %zmm16
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm25 = <2,7,12,17,22,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm25 = [2,7,12,17,22,27,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm5
; AVX512F-NEXT: vpermt2d %zmm21, %zmm25, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -4900,12 +4900,12 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2d %zmm1, %zmm20, %zmm28
; AVX512F-NEXT: vpermt2d %zmm20, %zmm0, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <3,8,13,18,23,28,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [3,8,13,18,23,28,u,u]
; AVX512F-NEXT: vpermt2d %zmm17, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm29
; AVX512F-NEXT: vpermt2d %zmm21, %zmm5, %zmm29
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = <4,9,14,19,24,29,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [4,9,14,19,24,29,u,u]
; AVX512F-NEXT: vpermt2d %zmm21, %zmm9, %zmm13
; AVX512F-NEXT: vmovdqa64 320(%rdi), %zmm17
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm31
@@ -5047,7 +5047,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm19, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm15 = <0,5,10,15,20,25,30,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm15 = [0,5,10,15,20,25,30,u]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm15, %zmm4
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm5
@@ -5062,7 +5062,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm16, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = <17,22,27,0,5,10,15,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = [17,22,27,0,5,10,15,u]
; AVX512BW-NEXT: vmovdqa64 %zmm21, %zmm24
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm12, %zmm24
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm5
@@ -5072,7 +5072,7 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm16, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermi2d %zmm20, %zmm1, %zmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm25 = <2,7,12,17,22,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm25 = [2,7,12,17,22,27,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm25, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -5104,12 +5104,12 @@ define void @load_i32_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm20, %zmm28
; AVX512BW-NEXT: vpermt2d %zmm20, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <3,8,13,18,23,28,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [3,8,13,18,23,28,u,u]
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm29
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm5, %zmm29
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <4,9,14,19,24,29,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [4,9,14,19,24,29,u,u]
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm9, %zmm13
; AVX512BW-NEXT: vmovdqa64 320(%rdi), %zmm17
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm31
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-6.ll
index 0c229457fb60243..d3f9b1c4e15a72a 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-6.ll
@@ -283,13 +283,13 @@ define void @load_i32_stride6_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm1
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <0,6,4,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [0,6,4,u]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0,1,2,3],ymm2[4,5],ymm1[6,7]
; AVX2-ONLY-NEXT: vpermd %ymm3, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %xmm4
; AVX2-ONLY-NEXT: vpshufd {{.*#+}} xmm5 = xmm4[2,2,2,2]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} xmm0 = xmm0[0,1,2],xmm5[3]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <1,7,5,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [1,7,5,u]
; AVX2-ONLY-NEXT: vpermd %ymm3, %ymm5, %ymm3
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} xmm3 = xmm3[0,1,2],xmm4[3]
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %xmm5
@@ -572,7 +572,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovaps (%rdi), %ymm2
; AVX2-SLOW-NEXT: vmovaps 32(%rdi), %ymm4
; AVX2-SLOW-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm3 = <0,6,4,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm3 = [0,6,4,u]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm7 = ymm2[0,1,2,3],ymm4[4,5],ymm2[6,7]
; AVX2-SLOW-NEXT: vpermps %ymm7, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vperm2f128 {{.*#+}} ymm8 = ymm6[0,1],ymm5[0,1]
@@ -583,7 +583,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vbroadcastsd {{.*#+}} ymm8 = [4,2,4,2,4,2,4,2]
; AVX2-SLOW-NEXT: vpermps %ymm10, %ymm8, %ymm11
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5],ymm11[6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm11 = <1,7,5,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm11 = [1,7,5,u]
; AVX2-SLOW-NEXT: vpermps %ymm7, %ymm11, %ymm7
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm9 = ymm9[1,3,2,3,5,7,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm7 = ymm7[0,1,2],ymm9[3,4,5,6,7]
@@ -655,7 +655,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovaps (%rdi), %ymm2
; AVX2-FAST-NEXT: vmovaps 32(%rdi), %ymm4
; AVX2-FAST-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm3 = <0,6,4,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm3 = [0,6,4,u]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm7 = ymm2[0,1,2,3],ymm4[4,5],ymm2[6,7]
; AVX2-FAST-NEXT: vpermps %ymm7, %ymm3, %ymm3
; AVX2-FAST-NEXT: vperm2f128 {{.*#+}} ymm8 = ymm6[0,1],ymm5[0,1]
@@ -666,7 +666,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vbroadcastsd {{.*#+}} ymm8 = [4,2,4,2,4,2,4,2]
; AVX2-FAST-NEXT: vpermps %ymm10, %ymm8, %ymm11
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5],ymm11[6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm11 = <1,7,5,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm11 = [1,7,5,u]
; AVX2-FAST-NEXT: vpermps %ymm7, %ymm11, %ymm7
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm9 = ymm9[1,3,2,3,5,7,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm7 = ymm7[0,1,2],ymm9[3,4,5,6,7]
@@ -738,7 +738,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovaps (%rdi), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovaps 32(%rdi), %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm3 = <0,6,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm3 = [0,6,4,u]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm7 = ymm2[0,1,2,3],ymm4[4,5],ymm2[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm7, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vperm2f128 {{.*#+}} ymm8 = ymm6[0,1],ymm5[0,1]
@@ -749,7 +749,7 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd {{.*#+}} ymm8 = [4,2,4,2,4,2,4,2]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm10, %ymm8, %ymm11
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5],ymm11[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm11 = <1,7,5,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm11 = [1,7,5,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm7, %ymm11, %ymm7
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm9 = ymm9[1,3,2,3,5,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm7 = ymm7[0,1,2],ymm9[3,4,5,6,7]
@@ -820,32 +820,32 @@ define void @load_i32_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-NEXT: vmovdqa 128(%rdi), %ymm0
; AVX512-NEXT: vmovdqa 160(%rdi), %ymm1
; AVX512-NEXT: vpblendd {{.*#+}} ymm4 = ymm1[0,1,2,3],ymm0[4,5,6,7]
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = <0,6,12,18,24,30,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = [0,6,12,18,24,30,u,u]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
; AVX512-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,2,3,4,5,12,10]
; AVX512-NEXT: vpermi2d %ymm4, %ymm5, %ymm6
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = <1,7,13,19,25,31,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = [1,7,13,19,25,31,u,u]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
; AVX512-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,2,3,4,5,13,11]
; AVX512-NEXT: vpermi2d %ymm4, %ymm5, %ymm7
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm4 = [0,0,6,12,0,0,6,12]
; AVX512-NEXT: # ymm4 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %ymm1, %ymm0, %ymm4
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = <2,8,14,20,26,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = [2,8,14,20,26,u,u,u]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
; AVX512-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [0,1,7,13,0,1,7,13]
; AVX512-NEXT: # ymm5 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %ymm1, %ymm0, %ymm5
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm8 = <3,9,15,21,27,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm8 = [3,9,15,21,27,u,u,u]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm8
; AVX512-NEXT: vpblendd {{.*#+}} ymm5 = ymm8[0,1,2,3,4],ymm5[5,6,7]
; AVX512-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5,6,7]
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm1 = <20,26,0,6,12,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm1 = [20,26,0,6,12,u,u,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm1
; AVX512-NEXT: vmovdqa {{.*#+}} ymm8 = [0,1,2,3,4,10,8,14]
; AVX512-NEXT: vpermi2d %ymm0, %ymm1, %ymm8
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm1 = <21,27,1,7,13,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm1 = [21,27,1,7,13,u,u,u]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm1
; AVX512-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,11,9,15]
; AVX512-NEXT: vpermi2d %ymm0, %ymm1, %ymm2
@@ -1370,7 +1370,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovups %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovaps 32(%rdi), %ymm13
; AVX2-SLOW-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm14 = <0,6,4,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm14 = [0,6,4,u]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm3[0,1,2,3],ymm13[4,5],ymm3[6,7]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm14, %ymm3
; AVX2-SLOW-NEXT: vperm2f128 {{.*#+}} ymm7 = ymm6[0,1],ymm15[0,1]
@@ -1397,7 +1397,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm5, %ymm11
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1,2,3,4,5],ymm11[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm2 = <1,7,5,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm2 = [1,7,5,u]
; AVX2-SLOW-NEXT: vpermps %ymm4, %ymm2, %ymm4
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm9 = ymm9[1,3,2,3,5,7,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm4 = ymm4[0,1,2],ymm9[3,4,5,6,7]
@@ -1556,7 +1556,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovups %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovaps 32(%rdi), %ymm13
; AVX2-FAST-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = <0,6,4,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = [0,6,4,u]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm4 = ymm3[0,1,2,3],ymm13[4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm0, %ymm7
; AVX2-FAST-NEXT: vperm2f128 {{.*#+}} ymm8 = ymm6[0,1],ymm15[0,1]
@@ -1583,7 +1583,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm5, %ymm11
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],ymm11[6,7]
; AVX2-FAST-NEXT: vmovups %ymm9, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm11 = <1,7,5,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm11 = [1,7,5,u]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm11, %ymm4
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm8 = ymm8[1,3,2,3,5,7,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm4 = ymm4[0,1,2],ymm8[3,4,5,6,7]
@@ -1740,7 +1740,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovaps 32(%rdi), %ymm13
; AVX2-FAST-PERLANE-NEXT: vmovaps 64(%rdi), %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm14 = <0,6,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm14 = [0,6,4,u]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm3[0,1,2,3],ymm13[4,5],ymm3[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm14, %ymm3
; AVX2-FAST-PERLANE-NEXT: vperm2f128 {{.*#+}} ymm7 = ymm6[0,1],ymm15[0,1]
@@ -1767,7 +1767,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm5, %ymm11
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1,2,3,4,5],ymm11[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm2 = <1,7,5,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm2 = [1,7,5,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm4, %ymm2, %ymm4
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm9 = ymm9[1,3,2,3,5,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm4 = ymm4[0,1,2],ymm9[3,4,5,6,7]
@@ -1921,7 +1921,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm7 = [0,6,12,0,0,0,20,26,0,6,12,0,0,0,20,26]
; AVX512F-NEXT: # zmm7 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm6, %zmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = <0,6,12,18,24,30,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = [0,6,12,18,24,30,u,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm2
; AVX512F-NEXT: movb $56, %dil
; AVX512F-NEXT: kmovw %edi, %k2
@@ -1935,14 +1935,14 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [1,7,13,0,0,0,21,27,1,7,13,0,0,0,21,27]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm6, %zmm8
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <1,7,13,19,25,31,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [1,7,13,19,25,31,u,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm7 {%k2}
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [0,0,0,3,9,15,21,27,0,0,0,3,9,15,21,27]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm1, %zmm8
; AVX512F-NEXT: vmovdqa32 %zmm8, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <2,8,14,20,26,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [2,8,14,20,26,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm8
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [18,24,30,0,0,0,6,12,18,24,30,0,0,0,6,12]
; AVX512F-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
@@ -1954,7 +1954,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm8
; AVX512F-NEXT: vmovdqa32 %zmm8, %zmm9 {%k1}
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <3,9,15,21,27,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [3,9,15,21,27,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm3, %zmm8
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [19,25,31,0,0,1,7,13,19,25,31,0,0,1,7,13]
; AVX512F-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
@@ -1967,7 +1967,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [20,26,0,0,0,2,8,14,20,26,0,0,0,2,8,14]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm8
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <20,26,0,6,12,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [20,26,0,6,12,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm4, %zmm11
; AVX512F-NEXT: movw $992, %di # imm = 0x3E0
; AVX512F-NEXT: kmovw %edi, %k1
@@ -1981,7 +1981,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [21,27,0,0,0,3,9,15,21,27,0,0,0,3,9,15]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm8
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <21,27,1,7,13,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [21,27,1,7,13,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm4, %zmm5
; AVX512F-NEXT: vmovdqa32 %zmm8, %zmm5 {%k1}
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [0,0,1,7,13,19,25,31,0,0,1,7,13,19,25,31]
@@ -2009,7 +2009,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm7 = [0,6,12,0,0,0,20,26,0,6,12,0,0,0,20,26]
; AVX512BW-NEXT: # zmm7 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm6, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,6,12,18,24,30,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,6,12,18,24,30,u,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm2
; AVX512BW-NEXT: movb $56, %dil
; AVX512BW-NEXT: kmovd %edi, %k2
@@ -2023,14 +2023,14 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [1,7,13,0,0,0,21,27,1,7,13,0,0,0,21,27]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm6, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <1,7,13,19,25,31,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [1,7,13,19,25,31,u,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm7 {%k2}
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [0,0,0,3,9,15,21,27,0,0,0,3,9,15,21,27]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm1, %zmm8
; AVX512BW-NEXT: vmovdqa32 %zmm8, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <2,8,14,20,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [2,8,14,20,26,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm8
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [18,24,30,0,0,0,6,12,18,24,30,0,0,0,6,12]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
@@ -2042,7 +2042,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm8
; AVX512BW-NEXT: vmovdqa32 %zmm8, %zmm9 {%k1}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <3,9,15,21,27,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [3,9,15,21,27,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm3, %zmm8
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [19,25,31,0,0,1,7,13,19,25,31,0,0,1,7,13]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
@@ -2055,7 +2055,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [20,26,0,0,0,2,8,14,20,26,0,0,0,2,8,14]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = <20,26,0,6,12,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = [20,26,0,6,12,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm4, %zmm11
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -2069,7 +2069,7 @@ define void @load_i32_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [21,27,0,0,0,3,9,15,21,27,0,0,0,3,9,15]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <21,27,1,7,13,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [21,27,1,7,13,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm4, %zmm5
; AVX512BW-NEXT: vmovdqa32 %zmm8, %zmm5 {%k1}
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [0,0,1,7,13,19,25,31,0,0,1,7,13,19,25,31]
@@ -3175,7 +3175,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovaps 64(%rdi), %ymm5
; AVX2-SLOW-NEXT: vmovups %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = <0,6,4,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = [0,6,4,u]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm13 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
; AVX2-SLOW-NEXT: vpermps %ymm13, %ymm8, %ymm0
; AVX2-SLOW-NEXT: vperm2f128 {{.*#+}} ymm1 = ymm5[0,1],ymm4[0,1]
@@ -3251,7 +3251,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpermps %ymm11, %ymm12, %ymm15
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm8 = ymm8[0,1,2,3,4,5],ymm15[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = <1,7,5,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = [1,7,5,u]
; AVX2-SLOW-NEXT: vpermps %ymm13, %ymm8, %ymm13
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm5 = ymm5[1,3,2,3,5,7,6,7]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm5 = ymm13[0,1,2],ymm5[3,4,5,6,7]
@@ -3591,7 +3591,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovaps 64(%rdi), %ymm5
; AVX2-FAST-NEXT: vmovups %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm8 = <0,6,4,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm8 = [0,6,4,u]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm13 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
; AVX2-FAST-NEXT: vpermps %ymm13, %ymm8, %ymm0
; AVX2-FAST-NEXT: vperm2f128 {{.*#+}} ymm1 = ymm5[0,1],ymm4[0,1]
@@ -3665,7 +3665,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpermps %ymm11, %ymm12, %ymm15
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm8 = ymm8[0,1,2,3,4,5],ymm15[6,7]
; AVX2-FAST-NEXT: vmovups %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm8 = <1,7,5,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm8 = [1,7,5,u]
; AVX2-FAST-NEXT: vpermps %ymm13, %ymm8, %ymm13
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm6 = ymm6[1,3,2,3,5,7,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm6 = ymm13[0,1,2],ymm6[3,4,5,6,7]
@@ -4001,7 +4001,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovaps 64(%rdi), %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = <0,6,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = [0,6,4,u]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm13 = ymm0[0,1,2,3],ymm1[4,5],ymm0[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm13, %ymm8, %ymm0
; AVX2-FAST-PERLANE-NEXT: vperm2f128 {{.*#+}} ymm1 = ymm5[0,1],ymm4[0,1]
@@ -4077,7 +4077,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm11, %ymm12, %ymm15
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm8 = ymm8[0,1,2,3,4,5],ymm15[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = <1,7,5,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = [1,7,5,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm13, %ymm8, %ymm13
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm5 = ymm5[1,3,2,3,5,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm5 = ymm13[0,1,2],ymm5[3,4,5,6,7]
@@ -4416,7 +4416,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm15
; AVX512F-NEXT: vpermt2d %zmm12, %zmm14, %zmm15
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <0,6,12,18,24,30,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [0,6,12,18,24,30,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm8
; AVX512F-NEXT: vpermt2d %zmm1, %zmm7, %zmm8
; AVX512F-NEXT: movb $56, %dil
@@ -4438,7 +4438,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm17
; AVX512F-NEXT: vpermt2d %zmm12, %zmm16, %zmm17
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <1,7,13,19,25,31,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [1,7,13,19,25,31,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm15
; AVX512F-NEXT: vpermt2d %zmm1, %zmm14, %zmm15
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm15 {%k2}
@@ -4452,7 +4452,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2d %zmm0, %zmm3, %zmm14
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm14 {%k2}
; AVX512F-NEXT: vmovdqa32 %zmm17, %zmm14 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm18 = <2,8,14,20,26,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm18 = [2,8,14,20,26,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512F-NEXT: vpermt2d %zmm1, %zmm18, %zmm19
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm17 = [18,24,30,0,0,0,6,12,18,24,30,0,0,0,6,12]
@@ -4472,7 +4472,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2d %zmm0, %zmm3, %zmm18
; AVX512F-NEXT: vmovdqa32 %zmm18, %zmm17 {%k2}
; AVX512F-NEXT: vmovdqa32 %zmm19, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm20 = <3,9,15,21,27,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm20 = [3,9,15,21,27,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm21
; AVX512F-NEXT: vpermt2d %zmm1, %zmm20, %zmm21
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm19 = [19,25,31,0,0,1,7,13,19,25,31,0,0,1,7,13]
@@ -4494,7 +4494,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm21
; AVX512F-NEXT: vpermt2d %zmm13, %zmm20, %zmm21
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm22 = <20,26,0,6,12,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm22 = [20,26,0,6,12,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm23
; AVX512F-NEXT: vpermt2d %zmm11, %zmm22, %zmm23
; AVX512F-NEXT: movw $992, %di # imm = 0x3E0
@@ -4515,7 +4515,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm20 = [21,27,0,0,0,3,9,15,21,27,0,0,0,3,9,15]
; AVX512F-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermt2d %zmm13, %zmm20, %zmm12
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = <21,27,1,7,13,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = [21,27,1,7,13,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm11, %zmm13, %zmm1
; AVX512F-NEXT: vmovdqa32 %zmm12, %zmm1 {%k1}
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [0,0,1,7,13,19,25,31,0,0,1,7,13,19,25,31]
@@ -4561,7 +4561,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm15
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm14, %zmm15
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <0,6,12,18,24,30,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,6,12,18,24,30,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm7, %zmm8
; AVX512BW-NEXT: movb $56, %dil
@@ -4583,7 +4583,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <1,7,13,19,25,31,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [1,7,13,19,25,31,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm15
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm14, %zmm15
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm15 {%k2}
@@ -4597,7 +4597,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm3, %zmm14
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm14 {%k2}
; AVX512BW-NEXT: vmovdqa32 %zmm17, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = <2,8,14,20,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = [2,8,14,20,26,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm18, %zmm19
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm17 = [18,24,30,0,0,0,6,12,18,24,30,0,0,0,6,12]
@@ -4617,7 +4617,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm3, %zmm18
; AVX512BW-NEXT: vmovdqa32 %zmm18, %zmm17 {%k2}
; AVX512BW-NEXT: vmovdqa32 %zmm19, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = <3,9,15,21,27,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [3,9,15,21,27,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm21
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm20, %zmm21
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm19 = [19,25,31,0,0,1,7,13,19,25,31,0,0,1,7,13]
@@ -4639,7 +4639,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm21
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm20, %zmm21
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = <20,26,0,6,12,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = [20,26,0,6,12,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm23
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm22, %zmm23
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
@@ -4660,7 +4660,7 @@ define void @load_i32_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm20 = [21,27,0,0,0,3,9,15,21,27,0,0,0,3,9,15]
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm20, %zmm12
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = <21,27,1,7,13,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = [21,27,1,7,13,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm13, %zmm1
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm1 {%k1}
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [0,0,1,7,13,19,25,31,0,0,1,7,13,19,25,31]
@@ -6963,7 +6963,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovaps 192(%rdi), %ymm1
; AVX2-SLOW-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = <0,6,4,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm8 = [0,6,4,u]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5],ymm1[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpermps %ymm0, %ymm8, %ymm0
@@ -7140,7 +7140,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vpermps %ymm15, %ymm14, %ymm9
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm9[6,7]
; AVX2-SLOW-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm0 = <1,7,5,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm0 = [1,7,5,u]
; AVX2-SLOW-NEXT: vpermps {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm9 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vpermilps $237, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm14 = mem[1,3,2,3,5,7,6,7]
@@ -7834,7 +7834,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovaps 192(%rdi), %ymm2
; AVX2-FAST-NEXT: vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = <0,6,4,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = [0,6,4,u]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm14 = ymm2[0,1,2,3],ymm1[4,5],ymm2[6,7]
; AVX2-FAST-NEXT: vpermps %ymm14, %ymm0, %ymm1
; AVX2-FAST-NEXT: vperm2f128 {{.*#+}} ymm2 = ymm9[0,1],ymm8[0,1]
@@ -8006,7 +8006,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpermps %ymm1, %ymm13, %ymm7
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm7[6,7]
; AVX2-FAST-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = <1,7,5,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm0 = [1,7,5,u]
; AVX2-FAST-NEXT: vpermps %ymm14, %ymm0, %ymm7
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm14 = ymm15[1,3,2,3,5,7,6,7]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm14 = ymm7[0,1,2],ymm14[3,4,5,6,7]
@@ -8695,7 +8695,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovaps 192(%rdi), %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = <0,6,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm8 = [0,6,4,u]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2,3],ymm0[4,5],ymm1[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm0, %ymm8, %ymm0
@@ -8872,7 +8872,7 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm15, %ymm14, %ymm9
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm9[6,7]
; AVX2-FAST-PERLANE-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm0 = <1,7,5,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm0 = [1,7,5,u]
; AVX2-FAST-PERLANE-NEXT: vpermps {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm9 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: vpermilps $237, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm14 = mem[1,3,2,3,5,7,6,7]
@@ -9711,24 +9711,24 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermt2d %zmm1, %zmm0, %zmm22
; AVX512F-NEXT: vmovdqa64 448(%rdi), %zmm2
; AVX512F-NEXT: vmovdqa64 384(%rdi), %zmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <0,6,12,18,24,30,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [0,6,12,18,24,30,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm20
; AVX512F-NEXT: vpermt2d %zmm2, %zmm3, %zmm20
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = <1,7,13,19,25,31,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = [1,7,13,19,25,31,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm21
; AVX512F-NEXT: vpermt2d %zmm2, %zmm4, %zmm21
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = <2,8,14,20,26,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = [2,8,14,20,26,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm12, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <3,9,15,21,27,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [3,9,15,21,27,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm14, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <20,26,0,6,12,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [20,26,0,6,12,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm23
; AVX512F-NEXT: vpermt2d %zmm0, %zmm5, %zmm23
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm24 = <21,27,1,7,13,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm24 = [21,27,1,7,13,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm24, %zmm2
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512F-NEXT: vmovdqa64 64(%rdi), %zmm6
@@ -10061,24 +10061,24 @@ define void @load_i32_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm0, %zmm22
; AVX512BW-NEXT: vmovdqa64 448(%rdi), %zmm2
; AVX512BW-NEXT: vmovdqa64 384(%rdi), %zmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <0,6,12,18,24,30,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,6,12,18,24,30,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm3, %zmm20
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <1,7,13,19,25,31,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [1,7,13,19,25,31,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm21
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm4, %zmm21
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = <2,8,14,20,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm12 = [2,8,14,20,26,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm12, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <3,9,15,21,27,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [3,9,15,21,27,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm14, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <20,26,0,6,12,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [20,26,0,6,12,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm23
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm5, %zmm23
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm24 = <21,27,1,7,13,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm24 = [21,27,1,7,13,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm24, %zmm2
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm6
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-7.ll
index 705fda41f71e17e..e4f94b368b7fd53 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i32-stride-7.ll
@@ -329,7 +329,7 @@ define void @load_i32_stride7_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: movq {{[0-9]+}}(%rsp), %r10
; AVX2-SLOW-NEXT: vmovaps (%rdi), %ymm0
; AVX2-SLOW-NEXT: vmovaps 32(%rdi), %ymm1
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm2 = <0,7,6,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm2 = [0,7,6,u]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm3 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-SLOW-NEXT: vpermps %ymm3, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vbroadcastss 84(%rdi), %xmm3
@@ -387,7 +387,7 @@ define void @load_i32_stride7_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: movq {{[0-9]+}}(%rsp), %r10
; AVX2-FAST-NEXT: vmovaps (%rdi), %ymm0
; AVX2-FAST-NEXT: vmovaps 32(%rdi), %ymm1
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm2 = <0,7,6,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm2 = [0,7,6,u]
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm3 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-FAST-NEXT: vpermps %ymm3, %ymm2, %ymm2
; AVX2-FAST-NEXT: vbroadcastss 84(%rdi), %xmm3
@@ -446,7 +446,7 @@ define void @load_i32_stride7_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: movq {{[0-9]+}}(%rsp), %r10
; AVX2-FAST-PERLANE-NEXT: vmovaps (%rdi), %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovaps 32(%rdi), %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm2 = <0,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm2 = [0,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm3 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm3, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vbroadcastss 84(%rdi), %xmm3
@@ -803,7 +803,7 @@ define void @load_i32_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqa 96(%rdi), %ymm10
; AVX2-SLOW-NEXT: vpbroadcastq 80(%rdi), %ymm2
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm10[4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <0,7,6,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [0,7,6,u]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm6 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-SLOW-NEXT: vpermd %ymm6, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3,4,5,6,7]
@@ -906,7 +906,7 @@ define void @load_i32_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm10
; AVX2-FAST-NEXT: vpbroadcastq 80(%rdi), %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm10[4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <0,7,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [0,7,6,u]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3,4,5,6,7]
@@ -1011,7 +1011,7 @@ define void @load_i32_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm10
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 80(%rdi), %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm10[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <0,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [0,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm6 = ymm0[0,1,2,3,4,5],ymm1[6],ymm0[7]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm6, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3,4,5,6,7]
@@ -1113,25 +1113,25 @@ define void @load_i32_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm4 = [0,3,10,17,0,3,10,17]
; AVX512-NEXT: # ymm4 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = <0,7,14,21,28,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm5 = [0,7,14,21,28,u,u,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm5
; AVX512-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0,1,2,3,4],ymm4[5,6,7]
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [0,4,11,18,0,4,11,18]
; AVX512-NEXT: # ymm5 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm6 = <1,8,15,22,29,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm6 = [1,8,15,22,29,u,u,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm6
; AVX512-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0,1,2,3,4],ymm5[5,6,7]
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [0,5,12,19,0,5,12,19]
; AVX512-NEXT: # ymm6 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm7 = <18,25,0,7,14,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm7 = [18,25,0,7,14,u,u,u]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm7
; AVX512-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0,1,2,3,4],ymm6[5,6,7]
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [0,6,13,20,0,6,13,20]
; AVX512-NEXT: # ymm7 = mem[0,1,0,1]
; AVX512-NEXT: vpermi2d %zmm3, %zmm2, %zmm7
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm8 = <19,26,1,8,15,u,u,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm8 = [19,26,1,8,15,u,u,u]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm8
; AVX512-NEXT: vpblendd {{.*#+}} ymm7 = ymm8[0,1,2,3,4],ymm7[5,6,7]
; AVX512-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [0,7,14,21,0,7,14,21]
@@ -1771,7 +1771,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqa 96(%rdi), %ymm12
; AVX2-SLOW-NEXT: vpbroadcastq 80(%rdi), %ymm1
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm12[4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,7,6,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,7,6,u]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm3 = ymm9[0,1,2,3,4,5],ymm4[6],ymm9[7]
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm7
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm2, %ymm3
@@ -1997,7 +1997,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm1
; AVX2-FAST-NEXT: vpbroadcastq 80(%rdi), %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <0,7,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [0,7,6,u]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm8 = ymm5[0,1,2,3,4,5],ymm6[6],ymm5[7]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm7, %ymm8
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm8[0,1,2],ymm2[3,4,5,6,7]
@@ -2223,7 +2223,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm12
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 80(%rdi), %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm12[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <0,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [0,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm3 = ymm9[0,1,2,3,4,5],ymm4[6],ymm9[7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm4, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm2, %ymm3
@@ -2449,7 +2449,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [8,15,0,0,0,19,26,1,8,15,0,0,0,19,26,1]
; AVX512F-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <0,7,14,21,28,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [0,7,14,21,28,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm0, %zmm3
; AVX512F-NEXT: movw $992, %di # imm = 0x3E0
; AVX512F-NEXT: kmovw %edi, %k1
@@ -2479,7 +2479,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm13 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm13
; AVX512F-NEXT: vpermt2d %zmm7, %zmm6, %zmm8
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <1,8,15,22,29,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [1,8,15,22,29,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm0, %zmm14
; AVX512F-NEXT: movw $480, %di # imm = 0x1E0
; AVX512F-NEXT: kmovw %edi, %k2
@@ -2496,7 +2496,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm8 = [26,5,12,19,26,5,12,19,26,5,12,19,26,5,12,19]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermt2d %zmm7, %zmm8, %zmm9
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <18,25,0,7,14,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [18,25,0,7,14,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm2, %zmm7
; AVX512F-NEXT: vmovdqa32 %zmm9, %zmm7 {%k2}
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [0,1,8,15,22,29,0,0,0,1,8,15,22,29,0,0]
@@ -2506,7 +2506,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm9, %zmm15
; AVX512F-NEXT: vmovdqa32 %zmm15, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = <19,26,1,8,15,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [19,26,1,8,15,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm2, %zmm9
; AVX512F-NEXT: vmovdqa32 %zmm10, %zmm9 {%k2}
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [0,18,25,0,7,14,0,0,0,18,25,0,7,14,0,0]
@@ -2566,7 +2566,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [8,15,0,0,0,19,26,1,8,15,0,0,0,19,26,1]
; AVX512BW-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <0,7,14,21,28,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,7,14,21,28,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm0, %zmm3
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -2596,7 +2596,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm13 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm7, %zmm6, %zmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <1,8,15,22,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [1,8,15,22,29,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm0, %zmm14
; AVX512BW-NEXT: movw $480, %di # imm = 0x1E0
; AVX512BW-NEXT: kmovd %edi, %k2
@@ -2613,7 +2613,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm8 = [26,5,12,19,26,5,12,19,26,5,12,19,26,5,12,19]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2d %zmm7, %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <18,25,0,7,14,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [18,25,0,7,14,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm2, %zmm7
; AVX512BW-NEXT: vmovdqa32 %zmm9, %zmm7 {%k2}
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [0,1,8,15,22,29,0,0,0,1,8,15,22,29,0,0]
@@ -2623,7 +2623,7 @@ define void @load_i32_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm9, %zmm15
; AVX512BW-NEXT: vmovdqa32 %zmm15, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <19,26,1,8,15,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [19,26,1,8,15,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm2, %zmm9
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm9 {%k2}
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [0,18,25,0,7,14,0,0,0,18,25,0,7,14,0,0]
@@ -3982,7 +3982,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpbroadcastq 80(%rdi), %ymm0
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm0[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2,3,4,5],ymm6[6],ymm3[7]
; AVX2-SLOW-NEXT: vmovdqa %ymm6, %ymm12
; AVX2-SLOW-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -4501,7 +4501,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vpbroadcastq 80(%rdi), %ymm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm0[0,1,2,3],ymm2[4,5,6,7]
; AVX2-FAST-NEXT: vmovdqa %ymm2, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2,3,4,5],ymm9[6],ymm3[7]
; AVX2-FAST-NEXT: vmovdqu %ymm9, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm6
@@ -5003,7 +5003,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 80(%rdi), %ymm0
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm0[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2,3,4,5],ymm6[6],ymm3[7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm6, %ymm12
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5526,7 +5526,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm17
; AVX512F-NEXT: vpermt2d %zmm14, %zmm16, %zmm17
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <0,7,14,21,28,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [0,7,14,21,28,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm8
; AVX512F-NEXT: vpermt2d %zmm11, %zmm7, %zmm8
; AVX512F-NEXT: movw $992, %di # imm = 0x3E0
@@ -5552,7 +5552,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm19
; AVX512F-NEXT: vpermt2d %zmm12, %zmm18, %zmm19
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = <1,8,15,22,29,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = [1,8,15,22,29,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm17
; AVX512F-NEXT: vpermt2d %zmm11, %zmm16, %zmm17
; AVX512F-NEXT: movw $480, %di # imm = 0x1E0
@@ -5579,7 +5579,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm21 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm22
; AVX512F-NEXT: vpermt2d %zmm12, %zmm21, %zmm22
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm19 = <18,25,0,7,14,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm19 = [18,25,0,7,14,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm20
; AVX512F-NEXT: vpermt2d %zmm10, %zmm19, %zmm20
; AVX512F-NEXT: vmovdqa32 %zmm22, %zmm20 {%k2}
@@ -5602,7 +5602,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm24 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm25
; AVX512F-NEXT: vpermt2d %zmm12, %zmm24, %zmm25
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm22 = <19,26,1,8,15,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm22 = [19,26,1,8,15,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm23
; AVX512F-NEXT: vpermt2d %zmm10, %zmm22, %zmm23
; AVX512F-NEXT: vmovdqa32 %zmm25, %zmm23 {%k2}
@@ -5718,7 +5718,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm16 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm14, %zmm16, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <0,7,14,21,28,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,7,14,21,28,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm7, %zmm8
; AVX512BW-NEXT: movw $992, %di # imm = 0x3E0
@@ -5744,7 +5744,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm19
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm18, %zmm19
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = <1,8,15,22,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm16 = [1,8,15,22,29,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm11, %zmm16, %zmm17
; AVX512BW-NEXT: movw $480, %di # imm = 0x1E0
@@ -5771,7 +5771,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm21 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm22
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm21, %zmm22
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm19 = <18,25,0,7,14,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm19 = [18,25,0,7,14,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm19, %zmm20
; AVX512BW-NEXT: vmovdqa32 %zmm22, %zmm20 {%k2}
@@ -5794,7 +5794,7 @@ define void @load_i32_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm24 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm25
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm24, %zmm25
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = <19,26,1,8,15,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm22 = [19,26,1,8,15,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm23
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm22, %zmm23
; AVX512BW-NEXT: vmovdqa32 %zmm25, %zmm23 {%k2}
@@ -8578,7 +8578,7 @@ define void @load_i32_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-SLOW-NEXT: vmovdqu %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa 256(%rdi), %ymm2
; AVX2-SLOW-NEXT: vmovdqa 224(%rdi), %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2,3,4,5],ymm2[6],ymm3[7]
; AVX2-SLOW-NEXT: vmovdqa %ymm3, %ymm11
; AVX2-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -9629,7 +9629,7 @@ define void @load_i32_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-NEXT: vmovdqa 320(%rdi), %ymm12
; AVX2-FAST-NEXT: vmovdqa 256(%rdi), %ymm2
; AVX2-FAST-NEXT: vmovdqa 224(%rdi), %ymm10
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm10[0,1,2,3,4,5],ymm2[6],ymm10[7]
; AVX2-FAST-NEXT: vmovdqu %ymm10, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa %ymm2, %ymm11
@@ -10669,7 +10669,7 @@ define void @load_i32_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 256(%rdi), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa 224(%rdi), %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <0,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [0,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2,3,4,5],ymm2[6],ymm3[7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm3, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -11947,16 +11947,16 @@ define void @load_i32_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 512(%rdi), %zmm0
; AVX512F-NEXT: vmovdqa64 448(%rdi), %zmm17
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = <0,7,14,21,28,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = [0,7,14,21,28,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm22
; AVX512F-NEXT: vpermt2d %zmm0, %zmm2, %zmm22
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <1,8,15,22,29,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [1,8,15,22,29,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm23
; AVX512F-NEXT: vpermt2d %zmm0, %zmm3, %zmm23
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = <18,25,0,7,14,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = [18,25,0,7,14,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm24
; AVX512F-NEXT: vpermt2d %zmm17, %zmm4, %zmm24
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <19,26,1,8,15,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [19,26,1,8,15,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm29
; AVX512F-NEXT: vpermt2d %zmm17, %zmm7, %zmm29
; AVX512F-NEXT: vmovdqa64 {{.*#+}} xmm18 = [4,11,18,25]
@@ -12399,16 +12399,16 @@ define void @load_i32_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 512(%rdi), %zmm0
; AVX512BW-NEXT: vmovdqa64 448(%rdi), %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,7,14,21,28,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,7,14,21,28,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm22
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm2, %zmm22
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <1,8,15,22,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [1,8,15,22,29,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm23
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm3, %zmm23
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <18,25,0,7,14,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [18,25,0,7,14,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm24
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm4, %zmm24
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <19,26,1,8,15,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [19,26,1,8,15,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm29
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm7, %zmm29
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm18 = [4,11,18,25]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-3.ll
index 5ec2453c3564f24..f08bee7d93f2334 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-3.ll
@@ -301,15 +301,15 @@ define void @load_i64_stride3_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,3,6,9,12,15,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,3,6,9,12,15,u,u]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,5,10,13]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <1,4,7,10,13,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [1,4,7,10,13,u,u,u]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,8,11,14]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <10,13,0,3,6,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [10,13,0,3,6,u,u,u]
; AVX512-NEXT: vpermi2q %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,9,12,15]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm0
@@ -571,21 +571,21 @@ define void @load_i64_stride3_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm3
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm4
; AVX512-NEXT: vmovdqa64 192(%rdi), %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,3,6,9,12,15,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,3,6,9,12,15,u,u]
; AVX512-NEXT: vmovdqa64 %zmm5, %zmm7
; AVX512-NEXT: vpermt2q %zmm1, %zmm6, %zmm7
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,10,13]
; AVX512-NEXT: vpermt2q %zmm0, %zmm8, %zmm7
; AVX512-NEXT: vpermi2q %zmm3, %zmm2, %zmm6
; AVX512-NEXT: vpermt2q %zmm4, %zmm8, %zmm6
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = <1,4,7,10,13,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [1,4,7,10,13,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm5, %zmm9
; AVX512-NEXT: vpermt2q %zmm1, %zmm8, %zmm9
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,8,11,14]
; AVX512-NEXT: vpermt2q %zmm0, %zmm10, %zmm9
; AVX512-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
; AVX512-NEXT: vpermt2q %zmm4, %zmm10, %zmm8
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = <10,13,0,3,6,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm10 = [10,13,0,3,6,u,u,u]
; AVX512-NEXT: vpermt2q %zmm5, %zmm10, %zmm1
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,9,12,15]
; AVX512-NEXT: vpermt2q %zmm0, %zmm5, %zmm1
@@ -1154,7 +1154,7 @@ define void @load_i64_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 64(%rdi), %zmm3
; AVX512-NEXT: vmovdqa64 128(%rdi), %zmm10
; AVX512-NEXT: vmovdqa64 192(%rdi), %zmm11
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <0,3,6,9,12,15,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [0,3,6,9,12,15,u,u]
; AVX512-NEXT: vmovdqa64 %zmm11, %zmm13
; AVX512-NEXT: vpermt2q %zmm1, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,2,3,4,5,10,13]
@@ -1167,7 +1167,7 @@ define void @load_i64_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vpermt2q %zmm4, %zmm14, %zmm16
; AVX512-NEXT: vpermi2q %zmm3, %zmm9, %zmm12
; AVX512-NEXT: vpermt2q %zmm10, %zmm14, %zmm12
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = <1,4,7,10,13,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [1,4,7,10,13,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm7, %zmm17
; AVX512-NEXT: vpermt2q %zmm2, %zmm14, %zmm17
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,1,2,3,4,8,11,14]
@@ -1180,7 +1180,7 @@ define void @load_i64_stride3_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vpermt2q %zmm4, %zmm18, %zmm20
; AVX512-NEXT: vpermi2q %zmm3, %zmm9, %zmm14
; AVX512-NEXT: vpermt2q %zmm10, %zmm18, %zmm14
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = <10,13,0,3,6,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [10,13,0,3,6,u,u,u]
; AVX512-NEXT: vpermt2q %zmm11, %zmm18, %zmm1
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,3,4,9,12,15]
; AVX512-NEXT: vpermt2q %zmm8, %zmm11, %zmm1
@@ -2429,7 +2429,7 @@ define void @load_i64_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vmovdqa64 256(%rdi), %zmm7
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm25
; AVX512-NEXT: vmovdqa64 192(%rdi), %zmm30
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,3,6,9,12,15,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,3,6,9,12,15,u,u]
; AVX512-NEXT: vmovdqa64 %zmm30, %zmm6
; AVX512-NEXT: vpermt2q %zmm7, %zmm11, %zmm6
; AVX512-NEXT: vmovdqa64 %zmm28, %zmm15
@@ -2444,10 +2444,10 @@ define void @load_i64_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512-NEXT: vpermt2q %zmm13, %zmm11, %zmm20
; AVX512-NEXT: vmovdqa64 %zmm22, %zmm23
; AVX512-NEXT: vpermt2q %zmm14, %zmm11, %zmm23
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm21 = <1,4,7,10,13,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm21 = [1,4,7,10,13,u,u,u]
; AVX512-NEXT: vmovdqa64 %zmm30, %zmm24
; AVX512-NEXT: vpermt2q %zmm7, %zmm21, %zmm24
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm31 = <10,13,0,3,6,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm31 = [10,13,0,3,6,u,u,u]
; AVX512-NEXT: vpermt2q %zmm30, %zmm31, %zmm7
; AVX512-NEXT: vmovdqa64 %zmm29, %zmm30
; AVX512-NEXT: vpermt2q %zmm9, %zmm21, %zmm30
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-5.ll
index ccc15a4119ba457..9b772bc109ef96b 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-5.ll
@@ -226,20 +226,20 @@ define void @load_i64_stride5_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,5,10,15]
; AVX512-SLOW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <1,6,11,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [1,6,11,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-SLOW-NEXT: vmovdqa 128(%rdi), %xmm4
; AVX512-SLOW-NEXT: vpbroadcastq %xmm4, %ymm5
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm3 = ymm3[0,1,2,3,4,5],ymm5[6,7]
; AVX512-SLOW-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <2,7,12,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [2,7,12,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm1, %zmm0, %zmm5
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0,1,2,3,4,5],ymm4[6,7]
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <11,0,5,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [11,0,5,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm0, %zmm1, %zmm5
; AVX512-SLOW-NEXT: vpbroadcastq 144(%rdi), %ymm6
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3,4,5],ymm6[6,7]
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <12,1,6,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [12,1,6,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm0, %zmm1, %zmm6
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm6[0,1,2,3,4,5],mem[6,7]
; AVX512-SLOW-NEXT: vmovdqa %ymm2, (%rsi)
@@ -256,20 +256,20 @@ define void @load_i64_stride5_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-FAST-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,5,10,15]
; AVX512-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,6,11,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,6,11,u]
; AVX512-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,1,2,4]
; AVX512-FAST-NEXT: vmovdqa 128(%rdi), %ymm5
; AVX512-FAST-NEXT: vpermi2q %ymm5, %ymm3, %ymm4
; AVX512-FAST-NEXT: vinserti128 $1, 128(%rdi), %ymm0, %ymm3
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <2,7,12,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [2,7,12,u]
; AVX512-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm6
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm6[0,1,2,3,4,5],ymm3[6,7]
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <11,0,5,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [11,0,5,u]
; AVX512-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm6
; AVX512-FAST-NEXT: vpbroadcastq 144(%rdi), %ymm7
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2,3,4,5],ymm7[6,7]
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <12,1,6,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [12,1,6,u]
; AVX512-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm7
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm7[0,1,2,3,4,5],ymm5[6,7]
; AVX512-FAST-NEXT: vmovdqa %ymm2, (%rsi)
@@ -544,21 +544,21 @@ define void @load_i64_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm4 = zmm4[0,1,2,3],zmm5[4,5,6,7]
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,5,6,11]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm4, %zmm5
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = <1,6,11,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = [1,6,11,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm2, %zmm4
; AVX512F-NEXT: movb $7, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,5,6,12]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm6, %zmm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <2,7,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [2,7,12,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm2, %zmm6
; AVX512F-NEXT: movb $56, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm6 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,4,5,8,13]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm6, %zmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <11,0,5,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [11,0,5,u]
; AVX512F-NEXT: vpermi2q %zmm2, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm6 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,9,14]
@@ -602,21 +602,21 @@ define void @load_i64_stride5_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm4 = zmm4[0,1,2,3],zmm5[4,5,6,7]
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,4,5,6,11]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm4, %zmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <1,6,11,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [1,6,11,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm2, %zmm4
; AVX512BW-NEXT: movb $7, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,5,6,12]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm6, %zmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <2,7,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [2,7,12,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm2, %zmm6
; AVX512BW-NEXT: movb $56, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm6 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,4,5,8,13]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm6, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <11,0,5,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [11,0,5,u]
; AVX512BW-NEXT: vpermi2q %zmm2, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm6 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,4,5,9,14]
@@ -1200,7 +1200,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermi2q %zmm0, %zmm3, %zmm12
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm12 = zmm12[0,1,2,3],zmm14[4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm2, %zmm13, %zmm12
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = <1,6,11,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = [1,6,11,u]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm16
; AVX512F-NEXT: vpermt2q %zmm1, %zmm15, %zmm16
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm14 = [5,10,15,0,5,10,15,0]
@@ -1220,7 +1220,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm16
; AVX512F-NEXT: vpermt2q %zmm10, %zmm15, %zmm16
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm17 = <2,7,12,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm17 = [2,7,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm18
; AVX512F-NEXT: vpermt2q %zmm1, %zmm17, %zmm18
; AVX512F-NEXT: movb $56, %al
@@ -1236,7 +1236,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm16
; AVX512F-NEXT: vpermt2q %zmm10, %zmm15, %zmm16
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm19 = <11,0,5,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm19 = [11,0,5,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm20
; AVX512F-NEXT: vpermt2q %zmm9, %zmm19, %zmm20
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm20 {%k1}
@@ -1297,7 +1297,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm3, %zmm12
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm12 = zmm12[0,1,2,3],zmm14[4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm13, %zmm12
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm15 = <1,6,11,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm15 = [1,6,11,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm16
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm15, %zmm16
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm14 = [5,10,15,0,5,10,15,0]
@@ -1317,7 +1317,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm16
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm15, %zmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm17 = <2,7,12,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm17 = [2,7,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm18
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm17, %zmm18
; AVX512BW-NEXT: movb $56, %al
@@ -1333,7 +1333,7 @@ define void @load_i64_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm16
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm15, %zmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm19 = <11,0,5,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm19 = [11,0,5,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm20
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm19, %zmm20
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm20 {%k1}
@@ -2587,7 +2587,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512F-NEXT: vpermt2q %zmm21, %zmm11, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm28 = <1,6,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm28 = [1,6,11,u]
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm15 = [5,10,15,0,5,10,15,0]
; AVX512F-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm30
@@ -2605,7 +2605,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm23, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = <2,7,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = [2,7,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512F-NEXT: vpermt2q %zmm2, %zmm23, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2635,7 +2635,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vpermt2q %zmm17, %zmm13, %zmm21
; AVX512F-NEXT: vpermt2q %zmm18, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm31 = <11,0,5,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm31 = [11,0,5,u]
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm25
; AVX512F-NEXT: vpermt2q %zmm19, %zmm31, %zmm25
; AVX512F-NEXT: vpermt2q %zmm19, %zmm11, %zmm17
@@ -2788,7 +2788,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm21, %zmm11, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm28 = <1,6,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm28 = [1,6,11,u]
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm15 = [5,10,15,0,5,10,15,0]
; AVX512BW-NEXT: # zmm15 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm30
@@ -2806,7 +2806,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm23, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = <2,7,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = [2,7,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm23, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2836,7 +2836,7 @@ define void @load_i64_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm13, %zmm21
; AVX512BW-NEXT: vpermt2q %zmm18, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm31 = <11,0,5,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm31 = [11,0,5,u]
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm25
; AVX512BW-NEXT: vpermt2q %zmm19, %zmm31, %zmm25
; AVX512BW-NEXT: vpermt2q %zmm19, %zmm11, %zmm17
@@ -5630,15 +5630,15 @@ define void @load_i64_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [0,5,10,15]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm1
; AVX512F-NEXT: vpermt2q %zmm31, %zmm3, %zmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm29 = <1,6,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm29 = [1,6,11,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512F-NEXT: vpermt2q %zmm31, %zmm29, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <2,7,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [2,7,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512F-NEXT: vpermt2q %zmm31, %zmm6, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <11,0,5,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [11,0,5,u]
; AVX512F-NEXT: vmovdqa64 %zmm31, %zmm0
; AVX512F-NEXT: vpermt2q %zmm2, %zmm7, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -6098,15 +6098,15 @@ define void @load_i64_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,5,10,15]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm1
; AVX512BW-NEXT: vpermt2q %zmm31, %zmm3, %zmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm29 = <1,6,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm29 = [1,6,11,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm31, %zmm29, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <2,7,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [2,7,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm31, %zmm6, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <11,0,5,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [11,0,5,u]
; AVX512BW-NEXT: vmovdqa64 %zmm31, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm7, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-6.ll
index 288b7219260c9a9..fb0f09c059d33bb 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-6.ll
@@ -220,21 +220,21 @@ define void @load_i64_stride6_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-SLOW-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512-SLOW-NEXT: vmovdqa64 (%rdi), %zmm2
; AVX512-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm3
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,6,12,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,6,12,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm3, %zmm2, %zmm0
; AVX512-SLOW-NEXT: vpbroadcastq 144(%rdi), %ymm1
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm1[6,7]
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <1,7,13,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [1,7,13,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm3, %zmm2, %zmm1
; AVX512-SLOW-NEXT: vmovdqa 128(%rdi), %ymm4
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5],ymm4[6,7]
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <10,0,6,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [10,0,6,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm2, %zmm3, %zmm5
; AVX512-SLOW-NEXT: vmovdqa 160(%rdi), %xmm6
; AVX512-SLOW-NEXT: vpbroadcastq %xmm6, %ymm7
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0,1,2,3,4,5],ymm7[6,7]
; AVX512-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <11,1,7,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [11,1,7,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm2, %zmm3, %zmm7
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0,1,2,3,4,5],ymm6[6,7]
; AVX512-SLOW-NEXT: vmovdqa 160(%rdi), %ymm7
@@ -262,21 +262,21 @@ define void @load_i64_stride6_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-FAST-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512-FAST-NEXT: vmovdqa64 (%rdi), %zmm2
; AVX512-FAST-NEXT: vmovdqa64 64(%rdi), %zmm3
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,6,12,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,6,12,u]
; AVX512-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm0
; AVX512-FAST-NEXT: vpbroadcastq 144(%rdi), %ymm1
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm1[6,7]
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <1,7,13,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [1,7,13,u]
; AVX512-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm1
; AVX512-FAST-NEXT: vmovdqa 128(%rdi), %ymm4
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5],ymm4[6,7]
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <10,0,6,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [10,0,6,u]
; AVX512-FAST-NEXT: vpermi2q %zmm2, %zmm3, %zmm5
; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,2,4]
; AVX512-FAST-NEXT: vmovdqa 160(%rdi), %ymm7
; AVX512-FAST-NEXT: vpermi2q %ymm7, %ymm5, %ymm6
; AVX512-FAST-NEXT: vinserti128 $1, 160(%rdi), %ymm0, %ymm5
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <11,1,7,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [11,1,7,u]
; AVX512-FAST-NEXT: vpermi2q %zmm2, %zmm3, %zmm8
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm8[0,1,2,3,4,5],ymm5[6,7]
; AVX512-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [0,6,0,6]
@@ -607,7 +607,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [0,6,0,10,0,6,0,10]
; AVX512F-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm4, %zmm5
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = <0,6,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = [0,6,12,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: movb $56, %dil
; AVX512F-NEXT: kmovw %edi, %k1
@@ -634,7 +634,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm7 = [1,7,0,11,1,7,0,11]
; AVX512F-NEXT: # zmm7 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm4, %zmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = <1,7,13,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = [1,7,13,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm13
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm13 {%k1}
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm7 = [5,11,5,11,5,11,5,11]
@@ -644,7 +644,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm6 = [10,4,10,4,10,4,10,4]
; AVX512F-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm3, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <10,0,6,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [10,0,6,u]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm1, %zmm8
; AVX512F-NEXT: movb $24, %dil
; AVX512F-NEXT: kmovw %edi, %k2
@@ -655,7 +655,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm6 = [11,5,11,5,11,5,11,5]
; AVX512F-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm3, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = <11,1,7,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [11,1,7,u]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm1, %zmm9
; AVX512F-NEXT: vmovdqa64 %zmm6, %zmm9 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm9 {%k1}
@@ -692,7 +692,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [0,6,0,10,0,6,0,10]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm4, %zmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,6,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,6,12,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: movb $56, %dil
; AVX512BW-NEXT: kmovd %edi, %k1
@@ -719,7 +719,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm7 = [1,7,0,11,1,7,0,11]
; AVX512BW-NEXT: # zmm7 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm4, %zmm7
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = <1,7,13,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm13 = [1,7,13,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm13
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm13 {%k1}
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm7 = [5,11,5,11,5,11,5,11]
@@ -729,7 +729,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm6 = [10,4,10,4,10,4,10,4]
; AVX512BW-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm3, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <10,0,6,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [10,0,6,u]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm1, %zmm8
; AVX512BW-NEXT: movb $24, %dil
; AVX512BW-NEXT: kmovd %edi, %k2
@@ -740,7 +740,7 @@ define void @load_i64_stride6_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm6 = [11,5,11,5,11,5,11,5]
; AVX512BW-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm3, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <11,1,7,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [11,1,7,u]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm1, %zmm9
; AVX512BW-NEXT: vmovdqa64 %zmm6, %zmm9 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm9 {%k1}
@@ -1459,7 +1459,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm15
; AVX512F-NEXT: vpermt2q %zmm10, %zmm14, %zmm15
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = <0,6,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm6 = [0,6,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm7
; AVX512F-NEXT: vpermt2q %zmm11, %zmm6, %zmm7
; AVX512F-NEXT: movb $56, %dil
@@ -1482,7 +1482,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm17
; AVX512F-NEXT: vpermt2q %zmm10, %zmm18, %zmm17
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <1,7,13,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [1,7,13,u]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm15
; AVX512F-NEXT: vpermt2q %zmm11, %zmm14, %zmm15
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm15 {%k1}
@@ -1501,7 +1501,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm21
; AVX512F-NEXT: vpermt2q %zmm13, %zmm20, %zmm21
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm18 = <10,0,6,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm18 = [10,0,6,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512F-NEXT: vpermt2q %zmm9, %zmm18, %zmm19
; AVX512F-NEXT: movb $24, %dil
@@ -1523,7 +1523,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: # zmm22 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm23
; AVX512F-NEXT: vpermt2q %zmm13, %zmm22, %zmm23
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm20 = <11,1,7,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm20 = [11,1,7,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm21
; AVX512F-NEXT: vpermt2q %zmm9, %zmm20, %zmm21
; AVX512F-NEXT: vmovdqa64 %zmm23, %zmm21 {%k2}
@@ -1602,7 +1602,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm14 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm15
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm14, %zmm15
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,6,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,6,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm7
; AVX512BW-NEXT: vpermt2q %zmm11, %zmm6, %zmm7
; AVX512BW-NEXT: movb $56, %dil
@@ -1625,7 +1625,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm17
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm18, %zmm17
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = <1,7,13,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm14 = [1,7,13,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm15
; AVX512BW-NEXT: vpermt2q %zmm11, %zmm14, %zmm15
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm15 {%k1}
@@ -1644,7 +1644,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm21
; AVX512BW-NEXT: vpermt2q %zmm13, %zmm20, %zmm21
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = <10,0,6,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm18 = [10,0,6,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm19
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm18, %zmm19
; AVX512BW-NEXT: movb $24, %dil
@@ -1666,7 +1666,7 @@ define void @load_i64_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: # zmm22 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm23
; AVX512BW-NEXT: vpermt2q %zmm13, %zmm22, %zmm23
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = <11,1,7,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm20 = [11,1,7,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm21
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm20, %zmm21
; AVX512BW-NEXT: vmovdqa64 %zmm23, %zmm21 {%k2}
@@ -3245,7 +3245,7 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm7
; AVX512F-NEXT: vpermt2q %zmm5, %zmm6, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <0,6,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [0,6,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512F-NEXT: vpermt2q %zmm4, %zmm6, %zmm8
; AVX512F-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -3264,7 +3264,7 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm6, %zmm8
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = <1,7,13,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = [1,7,13,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512F-NEXT: vpermt2q %zmm4, %zmm8, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -3325,11 +3325,11 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermt2q %zmm19, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <10,0,6,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [10,0,6,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm11, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm31 = <11,1,7,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm31 = [11,1,7,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm31, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -3597,7 +3597,7 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm7
; AVX512BW-NEXT: vpermt2q %zmm5, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = <0,6,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,6,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm6, %zmm8
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -3616,7 +3616,7 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm6, %zmm8
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = <1,7,13,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = [1,7,13,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm1
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm8, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -3677,11 +3677,11 @@ define void @load_i64_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermt2q %zmm19, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = <10,0,6,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = [10,0,6,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm11, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm31 = <11,1,7,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} ymm31 = [11,1,7,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm31, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -7279,19 +7279,19 @@ define void @load_i64_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512F-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 448(%rdi), %zmm1
; AVX512F-NEXT: vmovdqa64 384(%rdi), %zmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <0,6,12,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [0,6,12,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512F-NEXT: vpermt2q %zmm1, %zmm8, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = <1,7,13,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [1,7,13,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512F-NEXT: vpermt2q %zmm1, %zmm9, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = <10,0,6,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = [10,0,6,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512F-NEXT: vpermt2q %zmm2, %zmm10, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <11,1,7,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [11,1,7,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512F-NEXT: vpermt2q %zmm2, %zmm11, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -8035,19 +8035,19 @@ define void @load_i64_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, pt
; AVX512BW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 448(%rdi), %zmm1
; AVX512BW-NEXT: vmovdqa64 384(%rdi), %zmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = <0,6,12,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,6,12,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm8, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = <1,7,13,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm9 = [1,7,13,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm9, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = <10,0,6,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm10 = [10,0,6,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm10, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = <11,1,7,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm11 = [11,1,7,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm11, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-7.ll
index db8bca5bc16bf40..37f4d0d57076caf 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i64-stride-7.ll
@@ -302,10 +302,10 @@ define void @load_i64_stride7_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm4
; AVX512-SLOW-NEXT: vmovdqa64 (%rdi), %zmm5
; AVX512-SLOW-NEXT: vinserti128 $1, 160(%rdi), %ymm0, %ymm0
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,7,14,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,7,14,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm4, %zmm5, %zmm1
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
-; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <9,0,7,u>
+; AVX512-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [9,0,7,u]
; AVX512-SLOW-NEXT: vpermi2q %zmm5, %zmm4, %zmm1
; AVX512-SLOW-NEXT: vpbroadcastq 176(%rdi), %ymm2
; AVX512-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5],ymm2[6,7]
@@ -356,10 +356,10 @@ define void @load_i64_stride7_vf4(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512-FAST-NEXT: vmovdqa64 64(%rdi), %zmm4
; AVX512-FAST-NEXT: vmovdqa64 (%rdi), %zmm5
; AVX512-FAST-NEXT: vinserti128 $1, 160(%rdi), %ymm0, %ymm0
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,7,14,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,7,14,u]
; AVX512-FAST-NEXT: vpermi2q %zmm4, %zmm5, %zmm1
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1,2,3,4,5],ymm0[6,7]
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <9,0,7,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [9,0,7,u]
; AVX512-FAST-NEXT: vpermi2q %zmm5, %zmm4, %zmm1
; AVX512-FAST-NEXT: vpbroadcastq 176(%rdi), %ymm2
; AVX512-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5],ymm2[6,7]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-3.ll
index bac0fbe9df6f607..d84733b12ff7ddf 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-3.ll
@@ -1265,9 +1265,9 @@ define void @load_i8_stride3_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm10, %xmm8
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm11, %xmm9
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm7, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <1,4,7,10,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [1,4,7,10,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm6, %xmm6
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <128,128,128,128,128,0,3,6,9,12,15,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [128,128,128,128,128,0,3,6,9,12,15,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm0, %xmm12
; AVX1-ONLY-NEXT: vpor %xmm6, %xmm12, %xmm6
; AVX1-ONLY-NEXT: vmovdqa %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll
index e2311246e711fa4..40894a14a681c7a 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-4.ll
@@ -2007,9 +2007,9 @@ define void @load_i8_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512BW-NEXT: vmovdqa64 128(%rdi), %zmm2
; AVX512BW-NEXT: vmovdqa64 192(%rdi), %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <128,128,128,128,0,4,8,12,u,u,u,u,u,u,u,u,128,128,128,128,16,20,24,28,u,u,u,u,u,u,u,u,128,128,128,128,32,36,40,44,u,u,u,u,u,u,u,u,128,128,128,128,48,52,56,60,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [128,128,128,128,0,4,8,12,u,u,u,u,u,u,u,u,128,128,128,128,16,20,24,28,u,u,u,u,u,u,u,u,128,128,128,128,32,36,40,44,u,u,u,u,u,u,u,u,128,128,128,128,48,52,56,60,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm4, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,4,8,12,128,128,128,128,u,u,u,u,u,u,u,u,16,20,24,28,128,128,128,128,u,u,u,u,u,u,u,u,32,36,40,44,128,128,128,128,u,u,u,u,u,u,u,u,48,52,56,60,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,4,8,12,128,128,128,128,u,u,u,u,u,u,u,u,16,20,24,28,128,128,128,128,u,u,u,u,u,u,u,u,32,36,40,44,128,128,128,128,u,u,u,u,u,u,u,u,48,52,56,60,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm6, %zmm2, %zmm7
; AVX512BW-NEXT: vporq %zmm5, %zmm7, %zmm5
; AVX512BW-NEXT: vpshufb %zmm4, %zmm1, %zmm4
@@ -2017,27 +2017,27 @@ define void @load_i8_stride4_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vporq %zmm4, %zmm6, %zmm4
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,4,8,12,1,5,9,13,16,20,24,28,17,21,25,29]
; AVX512BW-NEXT: vpermt2d %zmm5, %zmm6, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <128,128,128,128,1,5,9,13,u,u,u,u,u,u,u,u,128,128,128,128,17,21,25,29,u,u,u,u,u,u,u,u,128,128,128,128,33,37,41,45,u,u,u,u,u,u,u,u,128,128,128,128,49,53,57,61,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [128,128,128,128,1,5,9,13,u,u,u,u,u,u,u,u,128,128,128,128,17,21,25,29,u,u,u,u,u,u,u,u,128,128,128,128,33,37,41,45,u,u,u,u,u,u,u,u,128,128,128,128,49,53,57,61,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm5, %zmm3, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <1,5,9,13,128,128,128,128,u,u,u,u,u,u,u,u,17,21,25,29,128,128,128,128,u,u,u,u,u,u,u,u,33,37,41,45,128,128,128,128,u,u,u,u,u,u,u,u,49,53,57,61,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [1,5,9,13,128,128,128,128,u,u,u,u,u,u,u,u,17,21,25,29,128,128,128,128,u,u,u,u,u,u,u,u,33,37,41,45,128,128,128,128,u,u,u,u,u,u,u,u,49,53,57,61,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm8, %zmm2, %zmm9
; AVX512BW-NEXT: vporq %zmm7, %zmm9, %zmm7
; AVX512BW-NEXT: vpshufb %zmm5, %zmm1, %zmm5
; AVX512BW-NEXT: vpshufb %zmm8, %zmm0, %zmm8
; AVX512BW-NEXT: vporq %zmm5, %zmm8, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm7, %zmm6, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <128,128,128,128,2,6,10,14,u,u,u,u,u,u,u,u,128,128,128,128,18,22,26,30,u,u,u,u,u,u,u,u,128,128,128,128,34,38,42,46,u,u,u,u,u,u,u,u,128,128,128,128,50,54,58,62,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [128,128,128,128,2,6,10,14,u,u,u,u,u,u,u,u,128,128,128,128,18,22,26,30,u,u,u,u,u,u,u,u,128,128,128,128,34,38,42,46,u,u,u,u,u,u,u,u,128,128,128,128,50,54,58,62,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm7, %zmm3, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,6,10,14,128,128,128,128,u,u,u,u,u,u,u,u,18,22,26,30,128,128,128,128,u,u,u,u,u,u,u,u,34,38,42,46,128,128,128,128,u,u,u,u,u,u,u,u,50,54,58,62,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,6,10,14,128,128,128,128,u,u,u,u,u,u,u,u,18,22,26,30,128,128,128,128,u,u,u,u,u,u,u,u,34,38,42,46,128,128,128,128,u,u,u,u,u,u,u,u,50,54,58,62,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm9, %zmm2, %zmm10
; AVX512BW-NEXT: vporq %zmm8, %zmm10, %zmm8
; AVX512BW-NEXT: vpshufb %zmm7, %zmm1, %zmm7
; AVX512BW-NEXT: vpshufb %zmm9, %zmm0, %zmm9
; AVX512BW-NEXT: vporq %zmm7, %zmm9, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm8, %zmm6, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <128,128,128,128,3,7,11,15,u,u,u,u,u,u,u,u,128,128,128,128,19,23,27,31,u,u,u,u,u,u,u,u,128,128,128,128,35,39,43,47,u,u,u,u,u,u,u,u,128,128,128,128,51,55,59,63,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [128,128,128,128,3,7,11,15,u,u,u,u,u,u,u,u,128,128,128,128,19,23,27,31,u,u,u,u,u,u,u,u,128,128,128,128,35,39,43,47,u,u,u,u,u,u,u,u,128,128,128,128,51,55,59,63,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm8, %zmm3, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <3,7,11,15,128,128,128,128,u,u,u,u,u,u,u,u,19,23,27,31,128,128,128,128,u,u,u,u,u,u,u,u,35,39,43,47,128,128,128,128,u,u,u,u,u,u,u,u,51,55,59,63,128,128,128,128,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [3,7,11,15,128,128,128,128,u,u,u,u,u,u,u,u,19,23,27,31,128,128,128,128,u,u,u,u,u,u,u,u,35,39,43,47,128,128,128,128,u,u,u,u,u,u,u,u,51,55,59,63,128,128,128,128,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %zmm9, %zmm2, %zmm2
; AVX512BW-NEXT: vporq %zmm3, %zmm2, %zmm2
; AVX512BW-NEXT: vpshufb %zmm8, %zmm1, %zmm1
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-5.ll
index 10ccd40e486556f..130207fd9c2ebcb 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-5.ll
@@ -707,7 +707,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm3[u,u,u,u,u,u,u],zero,zero,zero,xmm3[2,7,12,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = xmm1[u,u,u,u,u,u,u,3,8,13],zero,zero,zero,xmm1[u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm5, %xmm6, %xmm5
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm6, %xmm4, %xmm5, %xmm4
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,2,3,4,5,6,7,8,9,10,11,12,128,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm4, %xmm5
@@ -764,7 +764,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY: # %bb.0:
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm2
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = zero,zero,zero,zero,xmm3[4,9,14],zero,zero,zero,xmm3[2,7,12,u,u,u]
@@ -775,7 +775,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %xmm2
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[1,6,11]
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm3, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm5, %ymm0, %ymm1, %ymm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = xmm5[1,6,11],zero,zero,zero,zero,xmm5[4,9,14],zero,zero,zero,xmm5[u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm5, %xmm5
@@ -784,7 +784,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm5, %xmm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[2,7,12]
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm5, %xmm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm6, %ymm1, %ymm0, %ymm6
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm6, %xmm7
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,zero,zero,xmm7[1,6,11],zero,zero,zero,zero,xmm7[4,9,14,u,u,u]
@@ -793,7 +793,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm6, %xmm6
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[3,8,13]
; AVX2-ONLY-NEXT: vpor %xmm7, %xmm6, %xmm6
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm7
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm8 = xmm7[3,8,13],zero,zero,zero,xmm7[1,6,11],zero,zero,zero,zero,xmm7[u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm7, %xmm7
@@ -802,7 +802,7 @@ define void @load_i8_stride5_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm7, %xmm4
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm2[4,9,14]
; AVX2-ONLY-NEXT: vpor %xmm7, %xmm4, %xmm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,zero,xmm1[3,8,13],zero,zero,zero,xmm1[1,6,11,u,u,u,u]
@@ -1551,7 +1551,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm10[u,u,u,u,u,u,u],zero,zero,zero,xmm10[2,7,12,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm8[u,u,u,u,u,u,u,3,8,13],zero,zero,zero,xmm8[u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm11, %xmm12, %xmm11
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = <255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = [255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm13, %xmm5, %xmm11, %xmm5
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm12 = [255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255]
; AVX1-ONLY-NEXT: vandps %ymm5, %ymm12, %ymm11
@@ -1574,7 +1574,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm2[u,u,u],zero,zero,zero,zero,xmm2[4,9,14,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm14 = xmm3[u,u,u,0,5,10,15],zero,zero,zero,xmm3[u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm13, %xmm14, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <128,128,128,3,4,5,6,7,8,9,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [128,128,128,3,4,5,6,7,8,9,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm13, %xmm13
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm15 = xmm4[1,6,11],zero,zero,zero,zero,zero,zero,zero,xmm4[u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm15, %xmm13, %xmm13
@@ -1617,7 +1617,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm2[u,u,u],zero,zero,zero,xmm2[1,6,11,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm14 = xmm3[u,u,u,2,7,12],zero,zero,zero,xmm3[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm13, %xmm14, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <128,128,128,3,4,5,6,7,8,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [128,128,128,3,4,5,6,7,8,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm13, %xmm13
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm15 = xmm4[3,8,13],zero,zero,zero,zero,zero,zero,xmm4[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm15, %xmm13, %xmm13
@@ -1671,20 +1671,20 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm4
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm3, %ymm4, %ymm5
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm5, %xmm6
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,zero,zero,xmm6[4,9,14],zero,zero,zero,xmm6[2,7,12,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[0,5,10,15],zero,zero,zero,xmm5[3,8,13],zero,zero,zero,xmm5[u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm5, %xmm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm0, %ymm1, %ymm6
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm7 = ymm6[2,3,0,1]
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm8 = [255,255,0,255,0,255,255,0,255,0,255,255,0,255,0,255,255,255,0,255,0,255,255,0,255,0,255,255,0,255,0,255]
; AVX2-ONLY-NEXT: # ymm8 = mem[0,1,0,1]
; AVX2-ONLY-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm6
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm6 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,1,6,11,16,21,26,31,20,25,30,19,24,29,u,u,u,u,u,u]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vmovdqa %xmm8, %xmm7
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm6
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm3, %ymm4, %ymm5
@@ -1692,7 +1692,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm5, %xmm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,zero,xmm5[0,5,10,15],zero,zero,zero,xmm5[3,8,13,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm9, %xmm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm1, %ymm0, %ymm10
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm11 = ymm10[2,3,0,1]
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm12 = [0,255,255,0,255,0,255,255,0,255,0,255,255,0,255,0,0,255,255,0,255,0,255,255,0,255,0,255,255,0,255,0]
@@ -1705,7 +1705,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = zero,zero,zero,xmm10[1,6,11],zero,zero,zero,zero,xmm10[4,9,14,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[2,7,12],zero,zero,zero,xmm9[0,5,10,15],zero,zero,zero,xmm9[u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm10, %xmm9, %xmm9
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm10, %ymm1, %ymm0, %ymm11
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm12 = ymm11[2,3,0,1]
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm13 = [255,0,255,255,0,255,0,255,255,0,255,0,255,255,0,255,255,0,255,255,0,255,0,255,255,0,255,0,255,255,0,255]
@@ -1713,7 +1713,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[u,u,u,u,u,u,u,u,u,u,u,u,u,3,8,13,18,23,28,17,22,27,16,21,26,31,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm9, %ymm11, %ymm7
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm1, %ymm0, %ymm11
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm12 = ymm11[2,3,0,1]
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm13 = [0,255,0,255,255,0,255,0,255,255,0,255,0,255,255,0,0,255,0,255,255,0,255,0,255,255,0,255,0,255,255,0]
@@ -1849,7 +1849,7 @@ define void @load_i8_stride5_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpternlogq $202, %ymm0, %ymm1, %ymm11
; AVX512F-NEXT: vpermq {{.*#+}} ymm12 = ymm11[2,3,0,1]
; AVX512F-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm11, %ymm12
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <128,128,128,128,128,128,128,128,128,128,128,128,128,4,9,14,19,24,29,18,23,28,17,22,27,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,4,9,14,19,24,29,18,23,28,17,22,27,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm11, %ymm12, %ymm12
; AVX512F-NEXT: vmovdqa %ymm4, %ymm13
; AVX512F-NEXT: vpternlogq $202, %ymm3, %ymm5, %ymm13
@@ -3169,7 +3169,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa %xmm7, %xmm12
; AVX1-ONLY-NEXT: vmovdqa %xmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm6, %xmm6
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm6, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 176(%rdi), %xmm1
@@ -3238,7 +3238,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm12, %xmm3
; AVX1-ONLY-NEXT: vpxor %xmm4, %xmm4, %xmm4
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm12 = xmm4[0,1],xmm3[2,3,4,5,6,7]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <0,5,10,15,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [0,5,10,15,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vmovdqa 80(%rdi), %xmm14
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm14, %xmm15
; AVX1-ONLY-NEXT: vpor %xmm15, %xmm12, %xmm12
@@ -3319,7 +3319,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa %xmm9, %xmm8
; AVX1-ONLY-NEXT: vmovdqa %xmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm4, %xmm3
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <128,128,128,3,4,5,6,7,8,9,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [128,128,128,3,4,5,6,7,8,9,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm10 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = xmm10[1,6,11],zero,zero,zero,zero,zero,zero,zero,xmm10[u,u,u,u,u,u]
@@ -3340,7 +3340,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm7, %xmm5
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm5, %xmm3
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm3, %xmm3
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <2,7,12,128,128,128,128,128,128,128,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [2,7,12,128,128,128,128,128,128,128,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm11, %xmm6, %xmm14
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm14, %xmm3
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm2 = xmm3[0,1,2,3,4],xmm0[5,6,7]
@@ -3413,7 +3413,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,2,7,12],zero,zero,zero,xmm1[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm1, %xmm0
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,zero,xmm0[3,4,5,6,7,8,u,u,u,u,u,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <3,8,13,128,128,128,128,128,128,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [3,8,13,128,128,128,128,128,128,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm15 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm2, %xmm15, %xmm1
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm0, %xmm0
@@ -3446,7 +3446,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm11 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = xmm11[u,u,u,2,7,12],zero,zero,zero,xmm11[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm5, %xmm7, %xmm5
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,128,3,4,5,6,7,8,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,128,3,4,5,6,7,8,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm5, %xmm5
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm14 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm2, %xmm14, %xmm2
@@ -3563,7 +3563,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm4
; AVX2-ONLY-NEXT: vmovdqa 224(%rdi), %ymm10
; AVX2-ONLY-NEXT: vmovdqa 256(%rdi), %ymm9
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm10, %ymm9, %ymm0
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [255,255,0,255,0,255,255,0,255,0,255,255,0,255,0,255,255,255,0,255,0,255,255,0,255,0,255,255,0,255,0,255]
@@ -3572,7 +3572,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm2, %ymm4, %ymm0
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm7
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm9, %ymm10, %ymm0
; AVX2-ONLY-NEXT: vmovdqa %ymm1, %ymm5
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
@@ -3583,7 +3583,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm5, %ymm4, %ymm2, %ymm0
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm11
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm9, %ymm10, %ymm0
; AVX2-ONLY-NEXT: vmovdqa %ymm1, %ymm5
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
@@ -3597,7 +3597,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
; AVX2-ONLY-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm9, %ymm10, %ymm0
; AVX2-ONLY-NEXT: vmovdqa %ymm1, %ymm5
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm0[2,3,0,1]
@@ -3611,13 +3611,13 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vmovdqa 160(%rdi), %ymm13
; AVX2-ONLY-NEXT: vmovdqa 192(%rdi), %ymm14
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm13, %ymm14, %ymm0
; AVX2-ONLY-NEXT: vmovdqa %ymm1, %ymm6
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,4,9,14,128,128,128,2,7,12,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,4,9,14,128,128,128,2,7,12,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm5, %xmm1, %xmm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <0,5,10,15,128,128,128,3,8,13,128,128,128,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [0,5,10,15,128,128,128,3,8,13,128,128,128,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vpor %xmm1, %xmm0, %xmm1
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [0,5,10,15,4,9,14,3,8,13,0,0,0,1,6,11,0,5,10,15,4,9,14,3,8,13,0,0,0,1,6,11]
@@ -3636,10 +3636,10 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm7, %ymm15
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm13, %ymm14, %ymm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <1,6,11,128,128,128,128,4,9,14,128,128,128,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [1,6,11,128,128,128,128,4,9,14,128,128,128,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm0, %xmm5
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,0,5,10,15,128,128,128,3,8,13,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,0,5,10,15,128,128,128,3,8,13,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [1,6,11,0,5,10,15,4,9,14,0,0,0,2,7,12,1,6,11,0,5,10,15,4,9,14,0,0,0,2,7,12]
@@ -3654,12 +3654,12 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm5, %xmm4
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm4, %xmm4
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm4, %ymm0, %ymm11
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm14, %ymm13, %ymm0
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,1,6,11,128,128,128,128,4,9,14,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,1,6,11,128,128,128,128,4,9,14,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm5, %xmm4, %xmm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <2,7,12,128,128,128,0,5,10,15,128,128,128,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [2,7,12,128,128,128,0,5,10,15,128,128,128,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vpor %xmm4, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm4 = [2,7,12,1,6,11,0,5,10,15,0,0,0,3,8,13,2,7,12,1,6,11,0,5,10,15,0,0,0,3,8,13]
@@ -3677,12 +3677,12 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm4, %xmm4
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm4, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm14, %ymm13, %ymm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <3,8,13,128,128,128,1,6,11,128,128,128,128,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [3,8,13,128,128,128,1,6,11,128,128,128,128,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm0, %xmm5
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,2,7,12,128,128,128,0,5,10,15,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,2,7,12,128,128,128,0,5,10,15,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm0, %xmm0
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [3,8,13,2,7,12,1,6,11,0,0,0,0,4,9,14,3,8,13,2,7,12,1,6,11,0,0,0,0,4,9,14]
@@ -3699,7 +3699,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm5, %xmm5
; AVX2-ONLY-NEXT: vpor %xmm4, %xmm5, %xmm4
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm4, %ymm0, %ymm6
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm4, %ymm10, %ymm9, %ymm0
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm4, {{[-0-9]+}}(%r{{[sb]}}p), %ymm2, %ymm2 # 32-byte Folded Reload
@@ -3711,7 +3711,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm5, %ymm2, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} xmm4 = [0,0,128,128,128,1,6,11,0,0,128,128,128,1,6,11]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u,255,255,0,0,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm0, %ymm14, %ymm13, %ymm13
; AVX2-ONLY-NEXT: vmovdqa 304(%rdi), %xmm2
; AVX2-ONLY-NEXT: vpblendvb %ymm0, %ymm1, %ymm3, %ymm9
@@ -3763,9 +3763,9 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm12, %ymm0, %ymm12
; AVX2-ONLY-NEXT: vpblendvb %ymm14, %ymm6, %ymm12, %ymm12
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm13, %xmm15
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,3,8,13,128,128,128,1,6,11,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,3,8,13,128,128,128,1,6,11,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm15, %xmm15
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <4,9,14,128,128,128,2,7,12,128,128,128,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [4,9,14,128,128,128,2,7,12,128,128,128,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm7, %xmm13, %xmm13
; AVX2-ONLY-NEXT: vpor %xmm15, %xmm13, %xmm13
; AVX2-ONLY-NEXT: vbroadcasti128 {{.*#+}} ymm15 = [4,9,14,3,8,13,2,7,12,0,0,0,0,5,10,15,4,9,14,3,8,13,2,7,12,0,0,0,0,5,10,15]
@@ -3972,7 +3972,7 @@ define void @load_i8_stride5_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpternlogq $202, %ymm21, %ymm22, %ymm0
; AVX512F-NEXT: vpermq {{.*#+}} ymm2 = ymm0[2,3,0,1]
; AVX512F-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm0 = <128,128,128,128,128,128,128,128,128,128,128,128,128,4,9,14,19,24,29,18,23,28,17,22,27,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,128,128,128,128,128,128,128,128,128,4,9,14,19,24,29,18,23,28,17,22,27,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm0, %ymm2, %ymm2
; AVX512F-NEXT: vmovdqa %ymm4, %ymm3
; AVX512F-NEXT: vpternlogq $202, %ymm23, %ymm24, %ymm3
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-6.ll
index 16808dca4511d7f..899f38951342e05 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-6.ll
@@ -802,7 +802,7 @@ define void @load_i8_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = zero,zero,zero,xmm2[4,10,u,u,u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm1[2,8,14],zero,zero,xmm1[u,u,u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm10, %xmm11, %xmm10
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm11, %xmm8, %xmm10, %xmm8
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = xmm5[u,u,u,u,u,u,u,u,u,u,u,4,10],zero,zero,zero
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm4[u,u,u,u,u,u,u,u,u,u,u],zero,zero,xmm4[0,6,12]
@@ -825,7 +825,7 @@ define void @load_i8_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm3[u,u,u,u,u,u,u,u,4,10,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm0[u,u,u,u,u,u,u,u,u,u,u,u,u,2,8,14]
; AVX1-ONLY-NEXT: vpunpckhqdq {{.*#+}} xmm11 = xmm12[1],xmm11[1]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm12, %xmm10, %xmm11, %xmm10
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm4[u,u,u,u,u,u,u,u,u,u],zero,zero,zero,xmm4[2,8,14]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm5[u,u,u,u,u,u,u,u,u,u,0,6,12],zero,zero,zero
@@ -855,7 +855,7 @@ define void @load_i8_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm3
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm0, %ymm3, %ymm4, %ymm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = xmm5[0,6,12],zero,zero,zero,xmm5[4,10],zero,zero,zero,xmm5[u,u,u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm5, %xmm6
@@ -875,7 +875,7 @@ define void @load_i8_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = xmm1[u,u,u,u,u,u,u,u,u,u,u,3,9,15],zero,zero
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm7, %xmm6
; AVX2-ONLY-NEXT: vpblendvb %xmm8, %xmm5, %xmm6, %xmm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm6, %ymm4, %ymm3, %ymm6
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm6, %xmm7
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = zero,zero,zero,xmm7[4,10],zero,zero,zero,xmm7[2,8,14,u,u,u,u,u]
@@ -892,7 +892,7 @@ define void @load_i8_stride6_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = xmm0[u,u,u,u,u,u,u,u,u,u,u],zero,zero,xmm0[1,7,13]
; AVX2-ONLY-NEXT: vpor %xmm7, %xmm10, %xmm7
; AVX2-ONLY-NEXT: vpblendvb %xmm8, %xmm6, %xmm7, %xmm6
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm4, %ymm3, %ymm3
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm3, %xmm4
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,zero,xmm4[0,6,12],zero,zero,zero,xmm4[4,10,u,u,u,u,u,u]
@@ -1711,7 +1711,7 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,zero,xmm7[4,10,u,u,u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm9[2,8,14],zero,zero,xmm9[u,u,u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm8[3,9,15,u,u,u,u,u,u,u,u,u,u,u,u,u]
@@ -1728,7 +1728,7 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = xmm8[u,u,u,u,u,u,u,u,4,10,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm6[u,u,u,u,u,u,u,u,u,u,u,u,u,2,8,14]
; AVX1-ONLY-NEXT: vpunpckhqdq {{.*#+}} xmm2 = xmm2[1],xmm0[1]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm9[5,11],zero,zero,zero,xmm9[u,u,u,u,u,u,u,u,u,u,u]
@@ -1954,13 +1954,13 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm3
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm8, %ymm2, %ymm3, %ymm9
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm9[0,6,12],zero,zero,zero,xmm9[4,10],zero,zero,zero,xmm9[u,u,u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm9, %xmm10
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,zero,xmm10[2,8,14],zero,zero,xmm10[0,6,12,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm6, %xmm11
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0]
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm5 = ymm0[0,1],ymm1[0,1]
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm6 = ymm0[2,3],ymm1[2,3]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm1
@@ -1972,13 +1972,13 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpor %xmm9, %xmm10, %xmm9
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,3,9,15,5,11,17,23,29,19,25,31,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm9, %ymm1, %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm3, %ymm2, %ymm9
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm9, %xmm10
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = zero,zero,zero,xmm10[4,10],zero,zero,zero,xmm10[2,8,14,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm9[2,8,14],zero,zero,xmm9[0,6,12],zero,zero,zero,xmm9[u,u,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm12, %xmm13, %xmm12
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0]
; AVX2-ONLY-NEXT: vpblendvb %ymm13, %ymm6, %ymm5, %ymm13
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm14 = ymm13[u,u,u,u,u,u,u,u,u,u,u,4,10,0,6,12,18,24,30,20,26,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm7, %ymm12, %ymm14, %ymm12
@@ -1996,7 +1996,7 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0]
; AVX2-ONLY-NEXT: vpblendvb %ymm10, %ymm12, %ymm7, %ymm7
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[u,u,u,u,u,1,7,13],zero,zero,zero,xmm8[5,11],zero,zero,zero
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[u,u,u,u,u],zero,zero,zero,xmm9[3,9,15],zero,zero,xmm9[1,7,13]
; AVX2-ONLY-NEXT: vpor %xmm8, %xmm9, %xmm8
@@ -2010,7 +2010,7 @@ define void @load_i8_stride6_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpor %xmm13, %xmm14, %xmm13
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm3, %ymm2, %ymm2
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = <0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = [0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm6, %ymm5, %ymm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,xmm3[0,6,12],zero,zero,zero,xmm3[4,10,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm2[4,10],zero,zero,zero,xmm2[2,8,14],zero,zero,xmm2[u,u,u,u,u,u]
@@ -3614,7 +3614,7 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa %xmm2, %xmm4
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm2, %xmm2
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [0,0,0,0,0,255,255,255,255,255,255,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm5, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm13, %xmm9, %xmm1
@@ -3668,7 +3668,7 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm5, %xmm2
; AVX1-ONLY-NEXT: vpunpckhqdq {{.*#+}} xmm2 = xmm2[1],xmm0[1]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm11, %xmm1
@@ -4221,21 +4221,21 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 96(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0]
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm2 = ymm0[0,1],ymm1[0,1]
; AVX2-ONLY-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-ONLY-NEXT: vpblendvb %ymm13, %ymm2, %ymm0, %ymm4
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm0 = ymm4[u,u,u,u,u,u,u,u,u,u,u,2,8,14,4,10,16,22,28,18,24,30,u,u,u,u,u,u,u,u,u,u]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm3, %ymm5, %ymm2
; AVX2-ONLY-NEXT: vmovdqa %ymm3, %ymm5
; AVX2-ONLY-NEXT: vmovdqu %ymm3, (%rsp) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm10, %xmm2, %xmm9
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm11, %xmm3, %xmm12
; AVX2-ONLY-NEXT: vpor %xmm9, %xmm12, %xmm9
; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
@@ -4256,9 +4256,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} ymm0 = ymm13[u,u,u,u,u,u,u,u,u,u,u,2,8,14,4,10,16,22,28,18,24,30,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm0, %xmm2, %xmm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-ONLY-NEXT: vpor %xmm1, %xmm3, %xmm1
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm3 = [1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11]
@@ -4271,17 +4271,17 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %ymm3, %ymm13, %ymm1
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm0, %ymm1, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm10 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm13, %ymm10, %ymm5, %ymm1
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm1, %xmm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm2, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm1, %xmm5
; AVX2-ONLY-NEXT: vpor %xmm3, %xmm5, %xmm5
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm14 = [2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0]
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm0, {{[-0-9]+}}(%r{{[sb]}}p), %ymm3, %ymm3 # 32-byte Folded Reload
; AVX2-ONLY-NEXT: vpshufb %ymm14, %ymm3, %ymm15
@@ -4297,9 +4297,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb %ymm14, %ymm0, %ymm6
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm4, %ymm6, %ymm4
; AVX2-ONLY-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm6, %xmm1, %xmm1
; AVX2-ONLY-NEXT: vpor %xmm2, %xmm1, %xmm1
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm2 = [3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13]
@@ -4313,10 +4313,10 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm1, %ymm0, %ymm14
; AVX2-ONLY-NEXT: vmovdqa 160(%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 128(%rdi), %ymm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm3, %ymm1
; AVX2-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm1, %ymm3, %ymm0, %ymm5
; AVX2-ONLY-NEXT: vpblendvb %ymm13, %ymm0, %ymm3, %ymm15
; AVX2-ONLY-NEXT: vmovdqa 352(%rdi), %ymm4
@@ -4329,15 +4329,15 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm10, %ymm0, %ymm8
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm7, %ymm0, %ymm10
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u]
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm2, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm13 # 32-byte Folded Reload
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm11, %ymm9, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm2, %xmm5, %xmm6
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm5, %xmm11
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12]
; AVX2-ONLY-NEXT: vpshufb %xmm0, %xmm11, %xmm9
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm9, %xmm6
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
@@ -4351,9 +4351,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpblendvb %ymm9, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,u,1,7,13,128,128,128,5,11,128,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,u,1,7,13,128,128,128,5,11,128,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm0, %xmm5, %xmm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,u,u,u,128,128,128,3,9,15,128,128,1,7,13>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,u,u,u,128,128,128,3,9,15,128,128,1,7,13]
; AVX2-ONLY-NEXT: vpshufb %xmm5, %xmm11, %xmm11
; AVX2-ONLY-NEXT: vpor %xmm2, %xmm11, %xmm2
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -4366,15 +4366,15 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm14, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovdqu %ymm0, (%rsp) # 32-byte Spill
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm15, %xmm14
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14]
; AVX2-ONLY-NEXT: vpshufb %xmm7, %xmm14, %xmm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm2, %xmm15, %xmm1
; AVX2-ONLY-NEXT: vpor %xmm0, %xmm1, %xmm1
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm8, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,0,6,12,128,128,128,4,10,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,0,6,12,128,128,128,4,10,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm5, %xmm3, %xmm6
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm4, %xmm8, %xmm11
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm11, %xmm6
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm11 = [4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14]
@@ -4398,14 +4398,14 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm4 = ymm4[0,1,2,3],ymm5[4,5,6,7]
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm4, %ymm0, %ymm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15]
; AVX2-ONLY-NEXT: vpshufb %xmm0, %xmm14, %xmm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm7, %xmm15, %xmm11
; AVX2-ONLY-NEXT: vpor %xmm4, %xmm11, %xmm4
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm11, %xmm3, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %xmm14, %xmm8, %xmm8
; AVX2-ONLY-NEXT: vpor %xmm3, %xmm8, %xmm3
; AVX2-ONLY-NEXT: vpbroadcastq {{.*#+}} ymm8 = [5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15]
@@ -4427,9 +4427,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpblendvb %ymm9, %ymm1, %ymm0, %ymm3
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm9, %xmm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10]
; AVX2-ONLY-NEXT: vpshufb %xmm1, %xmm0, %xmm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm7, %xmm9, %xmm8
; AVX2-ONLY-NEXT: vmovdqa %ymm9, %ymm10
; AVX2-ONLY-NEXT: vpor %xmm2, %xmm8, %xmm2
@@ -4447,9 +4447,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm1 = ymm7[0,1,2],ymm1[3,4,5,6,7],ymm7[8,9,10],ymm1[11,12,13,14,15]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm1 = ymm7[0,1,2,3],ymm1[4,5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11]
; AVX2-ONLY-NEXT: vpshufb %xmm7, %xmm0, %xmm0
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128]
; AVX2-ONLY-NEXT: vpshufb %xmm9, %xmm10, %xmm10
; AVX2-ONLY-NEXT: vpor %xmm0, %xmm10, %xmm0
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -4488,14 +4488,14 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-LABEL: load_i8_stride6_vf64:
; AVX512F: # %bb.0:
; AVX512F-NEXT: subq $40, %rsp
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm12 = [65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535]
; AVX512F-NEXT: vmovdqa64 224(%rdi), %ymm25
; AVX512F-NEXT: vmovdqa64 192(%rdi), %ymm26
; AVX512F-NEXT: vmovdqa %ymm12, %ymm0
; AVX512F-NEXT: vpternlogq $202, %ymm25, %ymm26, %ymm0
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u]
; AVX512F-NEXT: vextracti128 $1, %ymm0, %xmm4
; AVX512F-NEXT: vpshufb %xmm5, %xmm4, %xmm6
; AVX512F-NEXT: vpor %xmm3, %xmm6, %xmm9
@@ -4506,9 +4506,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vmovdqa %ymm12, %ymm6
; AVX512F-NEXT: vpternlogq $202, %ymm24, %ymm18, %ymm6
; AVX512F-NEXT: vextracti128 $1, %ymm6, %xmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10]
; AVX512F-NEXT: vpshufb %xmm3, %xmm7, %xmm10
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128]
; AVX512F-NEXT: vpshufb %xmm8, %xmm6, %xmm13
; AVX512F-NEXT: vpor %xmm10, %xmm13, %xmm10
; AVX512F-NEXT: vinserti128 $1, %xmm10, %ymm0, %ymm10
@@ -4529,15 +4529,15 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpshufb %xmm8, %xmm1, %xmm8
; AVX512F-NEXT: vpor %xmm3, %xmm8, %xmm3
; AVX512F-NEXT: vinserti128 $1, %xmm3, %ymm0, %ymm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = <1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = [1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm8, %xmm0, %xmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = <128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = [128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm10, %xmm4, %xmm4
; AVX512F-NEXT: vpor %xmm0, %xmm4, %xmm0
; AVX512F-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11]
; AVX512F-NEXT: vpshufb %xmm0, %xmm7, %xmm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128]
; AVX512F-NEXT: vpshufb %xmm7, %xmm6, %xmm6
; AVX512F-NEXT: vporq %xmm4, %xmm6, %xmm28
; AVX512F-NEXT: vpshufb %xmm8, %xmm9, %xmm4
@@ -4546,22 +4546,22 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpshufb %xmm0, %xmm5, %xmm0
; AVX512F-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX512F-NEXT: vporq %xmm0, %xmm1, %xmm27
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = <128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = [128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535]
; AVX512F-NEXT: vmovdqa %ymm9, %ymm4
; AVX512F-NEXT: vpternlogq $202, %ymm26, %ymm25, %ymm4
; AVX512F-NEXT: vextracti128 $1, %ymm4, %xmm15
; AVX512F-NEXT: vpshufb %xmm0, %xmm15, %xmm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = <2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = [2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm6, %xmm4, %xmm5
; AVX512F-NEXT: vpor %xmm1, %xmm5, %xmm1
; AVX512F-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-NEXT: vmovdqa %ymm12, %ymm5
; AVX512F-NEXT: vpternlogq $202, %ymm18, %ymm24, %ymm5
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128]
; AVX512F-NEXT: vpshufb %xmm8, %xmm5, %xmm7
; AVX512F-NEXT: vextracti128 $1, %ymm5, %xmm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12]
; AVX512F-NEXT: vpshufb %xmm10, %xmm1, %xmm13
; AVX512F-NEXT: vpor %xmm7, %xmm13, %xmm2
; AVX512F-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4578,9 +4578,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpshufb %xmm10, %xmm7, %xmm10
; AVX512F-NEXT: vpor %xmm8, %xmm10, %xmm0
; AVX512F-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = <128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm10 = [128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm10, %xmm15, %xmm15
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = <3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm8 = [3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm8, %xmm4, %xmm4
; AVX512F-NEXT: vpor %xmm4, %xmm15, %xmm0
; AVX512F-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4631,16 +4631,16 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpternlogq $226, %ymm26, %ymm12, %ymm11
; AVX512F-NEXT: vextracti128 $1, %ymm11, %xmm0
; AVX512F-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,xmm0[0,6,12],zero,zero,zero,xmm0[4,10,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm3, %xmm11, %xmm2
; AVX512F-NEXT: vmovdqa64 %xmm3, %xmm25
; AVX512F-NEXT: vporq %xmm1, %xmm2, %xmm26
; AVX512F-NEXT: vmovdqa64 %ymm18, %ymm14
; AVX512F-NEXT: vpternlogq $226, %ymm24, %ymm9, %ymm14
; AVX512F-NEXT: vextracti128 $1, %ymm14, %xmm10
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14]
; AVX512F-NEXT: vpshufb %xmm1, %xmm10, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128]
; AVX512F-NEXT: vpshufb %xmm3, %xmm14, %xmm4
; AVX512F-NEXT: vporq %xmm2, %xmm4, %xmm27
; AVX512F-NEXT: vpternlogq $202, %ymm30, %ymm31, %ymm12
@@ -4651,15 +4651,15 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-NEXT: vpshufb %xmm1, %xmm8, %xmm1
; AVX512F-NEXT: vpshufb %xmm3, %xmm9, %xmm2
; AVX512F-NEXT: vpor %xmm1, %xmm2, %xmm7
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovdqa64 %xmm1, %xmm22
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm13 = <5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm13 = [5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm13, %xmm11, %xmm1
; AVX512F-NEXT: vpor %xmm0, %xmm1, %xmm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15]
; AVX512F-NEXT: vpshufb %xmm2, %xmm10, %xmm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128]
; AVX512F-NEXT: vpshufb %xmm0, %xmm14, %xmm10
; AVX512F-NEXT: vpor %xmm1, %xmm10, %xmm10
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,4,10,0,6,12,18,24,30,20,26,128,128,128,128,128,128,128,128,128,128,128]
@@ -4741,14 +4741,14 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-LABEL: load_i8_stride6_vf64:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: movq {{[0-9]+}}(%rsp), %rax
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,6,12,128,128,128,4,10,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa 224(%rdi), %ymm0
; AVX512BW-NEXT: vmovdqa64 192(%rdi), %ymm23
; AVX512BW-NEXT: movw $18724, %r10w # imm = 0x4924
; AVX512BW-NEXT: kmovd %r10d, %k1
; AVX512BW-NEXT: vpblendmw %ymm0, %ymm23, %ymm9 {%k1}
; AVX512BW-NEXT: vpshufb %xmm2, %xmm9, %xmm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,2,8,14,128,128,0,6,12,u,u,u,u,u]
; AVX512BW-NEXT: vextracti128 $1, %ymm9, %xmm12
; AVX512BW-NEXT: vpshufb %xmm4, %xmm12, %xmm3
; AVX512BW-NEXT: vpor %xmm1, %xmm3, %xmm5
@@ -4759,9 +4759,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vmovdqa 160(%rdi), %ymm1
; AVX512BW-NEXT: vpblendmw %ymm26, %ymm1, %ymm15 {%k1}
; AVX512BW-NEXT: vextracti32x4 $1, %ymm15, %xmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [u,u,u,u,u,u,128,128,0,6,12,128,128,128,4,10]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm16, %xmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm18 = <u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm18 = [u,u,u,u,u,u,4,10,128,128,128,2,8,14,128,128]
; AVX512BW-NEXT: vpshufb %xmm18, %xmm15, %xmm13
; AVX512BW-NEXT: vpor %xmm11, %xmm13, %xmm11
; AVX512BW-NEXT: vinserti128 $1, %xmm11, %ymm0, %ymm11
@@ -4800,14 +4800,14 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: movabsq $-8796093022208, %rdi # imm = 0xFFFFF80000000000
; AVX512BW-NEXT: kmovq %rdi, %k3
; AVX512BW-NEXT: vmovdqu8 %zmm7, %zmm2 {%k3}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm7 = <1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm7 = [1,7,13,128,128,128,5,11,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm7, %xmm9, %xmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm9 = <128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm9 = [128,128,128,3,9,15,128,128,1,7,13,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm9, %xmm12, %xmm12
; AVX512BW-NEXT: vpor %xmm8, %xmm12, %xmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,u,u,u,u,128,128,1,7,13,128,128,128,5,11]
; AVX512BW-NEXT: vpshufb %xmm12, %xmm16, %xmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [u,u,u,u,u,u,5,11,128,128,128,3,9,15,128,128]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm15, %xmm15
; AVX512BW-NEXT: vporq %xmm16, %xmm15, %xmm15
; AVX512BW-NEXT: vinserti128 $1, %xmm15, %ymm0, %ymm15
@@ -4828,13 +4828,13 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vinserti64x4 $1, %ymm7, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqu8 %zmm7, %zmm9 {%k3}
; AVX512BW-NEXT: vpblendmw %ymm13, %ymm5, %ymm15 {%k4}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,128,4,10,128,128,128,2,8,14,u,u,u,u,u]
; AVX512BW-NEXT: movw $9362, %di # imm = 0x2492
; AVX512BW-NEXT: kmovd %edi, %k2
; AVX512BW-NEXT: vpblendmw %ymm10, %ymm3, %ymm8 {%k2}
; AVX512BW-NEXT: vextracti32x4 $1, %ymm8, %xmm16
; AVX512BW-NEXT: vpshufb %xmm7, %xmm16, %xmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [2,8,14,128,128,0,6,12,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm8, %xmm18
; AVX512BW-NEXT: vporq %xmm12, %xmm18, %xmm18
; AVX512BW-NEXT: vpbroadcastq {{.*#+}} ymm19 = [2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12]
@@ -4847,10 +4847,10 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vpshufb %xmm17, %xmm20, %xmm12
; AVX512BW-NEXT: vpor %xmm7, %xmm12, %xmm7
; AVX512BW-NEXT: vpblendmw %ymm1, %ymm26, %ymm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm22 = <u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm22 = [u,u,u,u,u,0,6,12,128,128,128,4,10,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm22, %xmm17, %xmm12
; AVX512BW-NEXT: vextracti32x4 $1, %ymm17, %xmm24
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm25 = <u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm25 = [u,u,u,u,u,128,128,128,2,8,14,128,128,0,6,12]
; AVX512BW-NEXT: vpshufb %xmm25, %xmm24, %xmm27
; AVX512BW-NEXT: vporq %xmm12, %xmm27, %xmm12
; AVX512BW-NEXT: vinserti128 $1, %xmm12, %ymm0, %ymm12
@@ -4872,9 +4872,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: kmovd %edi, %k4
; AVX512BW-NEXT: vmovdqu16 %ymm14, %ymm4 {%k4}
; AVX512BW-NEXT: vmovdqu16 %ymm13, %ymm5 {%k4}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm13 = <128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm13 = [128,128,128,5,11,128,128,128,3,9,15,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm13, %xmm16, %xmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm16 = <3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm16 = [3,9,15,128,128,1,7,13,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm16, %xmm8, %xmm8
; AVX512BW-NEXT: vpor %xmm14, %xmm8, %xmm8
; AVX512BW-NEXT: vpbroadcastq {{.*#+}} ymm14 = [3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13]
@@ -4882,9 +4882,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vpshufb %xmm13, %xmm21, %xmm13
; AVX512BW-NEXT: vpshufb %xmm16, %xmm20, %xmm15
; AVX512BW-NEXT: vpor %xmm13, %xmm15, %xmm13
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm15 = <u,u,u,u,u,1,7,13,128,128,128,5,11,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm15 = [u,u,u,u,u,1,7,13,128,128,128,5,11,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm15, %xmm17, %xmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <u,u,u,u,u,128,128,128,3,9,15,128,128,1,7,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [u,u,u,u,u,128,128,128,3,9,15,128,128,1,7,13]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm24, %xmm19
; AVX512BW-NEXT: vporq %xmm16, %xmm19, %xmm16
; AVX512BW-NEXT: vinserti32x4 $1, %xmm16, %ymm0, %ymm16
@@ -4899,11 +4899,11 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vmovdqu8 %zmm7, %zmm13 {%k3}
; AVX512BW-NEXT: vpbroadcastq {{.*#+}} ymm7 = [4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14,4,10,0,6,12,2,8,14]
; AVX512BW-NEXT: vpshufb %ymm7, %ymm5, %ymm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = <128,128,0,6,12,128,128,128,4,10,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = [128,128,0,6,12,128,128,128,4,10,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqu16 %ymm10, %ymm3 {%k1}
; AVX512BW-NEXT: vextracti128 $1, %ymm3, %xmm15
; AVX512BW-NEXT: vpshufb %xmm14, %xmm15, %xmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm16 = <4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm16 = [4,10,128,128,128,2,8,14,128,128,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm16, %xmm3, %xmm17
; AVX512BW-NEXT: vporq %xmm10, %xmm17, %xmm10
; AVX512BW-NEXT: vpblendw {{.*#+}} xmm10 = xmm10[0,1,2,3,4],xmm8[5,6,7]
@@ -4915,9 +4915,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vporq %xmm14, %xmm16, %xmm14
; AVX512BW-NEXT: vmovdqu16 %ymm26, %ymm1 {%k2}
; AVX512BW-NEXT: vextracti32x4 $1, %ymm1, %xmm16
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [u,u,u,u,u,128,128,128,4,10,128,128,128,2,8,14]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm16, %xmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm19 = <u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm19 = [u,u,u,u,u,2,8,14,128,128,0,6,12,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm19, %xmm1, %xmm20
; AVX512BW-NEXT: vporq %xmm18, %xmm20, %xmm18
; AVX512BW-NEXT: vinserti32x4 $1, %xmm18, %ymm0, %ymm18
@@ -4939,9 +4939,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vmovdqu16 %zmm7, %zmm10 {%k2}
; AVX512BW-NEXT: vpbroadcastq {{.*#+}} ymm7 = [5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15,5,11,1,7,13,3,9,15]
; AVX512BW-NEXT: vpshufb %ymm7, %ymm5, %ymm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = <128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = [128,128,1,7,13,128,128,128,5,11,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm14, %xmm15, %xmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [5,11,128,128,128,3,9,15,128,128,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm17, %xmm3, %xmm3
; AVX512BW-NEXT: vpor %xmm3, %xmm15, %xmm3
; AVX512BW-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0,1,2,3,4],xmm5[5,6,7]
@@ -4949,9 +4949,9 @@ define void @load_i8_stride6_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-NEXT: vpshufb %xmm14, %xmm8, %xmm5
; AVX512BW-NEXT: vpshufb %xmm17, %xmm0, %xmm0
; AVX512BW-NEXT: vpor %xmm5, %xmm0, %xmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,u,u,u,128,128,128,5,11,128,128,128,3,9,15]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm16, %xmm8
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = <u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm14 = [u,u,u,u,u,3,9,15,128,128,1,7,13,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm14, %xmm1, %xmm1
; AVX512BW-NEXT: vpor %xmm1, %xmm8, %xmm1
; AVX512BW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-7.ll
index 2b2cb554d6ac684..718d4973f97e251 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-load-i8-stride-7.ll
@@ -688,7 +688,7 @@ define void @load_i8_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: movq {{[0-9]+}}(%rsp), %r10
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,u,u,u,u,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,u,u,u,u,255,255,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm2
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = zero,zero,zero,xmm3[5,12],zero,zero,xmm3[1,u,u,u,u,u,u,u,u]
@@ -699,7 +699,7 @@ define void @load_i8_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm4 = zero,zero,zero,xmm4[6,13],zero,zero,xmm4[2,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[1,8,15],zero,zero,xmm3[4,11],zero,xmm3[u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm4, %xmm3, %xmm3
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,u,u,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,u,u,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm4
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm4, %xmm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,xmm5[0,7,14],zero,zero,xmm5[3,u,u,u,u,u,u,u,u]
@@ -710,7 +710,7 @@ define void @load_i8_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,xmm6[1,8,15],zero,zero,xmm6[4,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[3,10],zero,zero,zero,xmm5[6,13],zero,xmm5[u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm6, %xmm5, %xmm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm6, %ymm1, %ymm0, %ymm6
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm6, %xmm7
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,zero,xmm7[2,9],zero,zero,zero,xmm7[5,u,u,u,u,u,u,u,u]
@@ -721,7 +721,7 @@ define void @load_i8_stride7_vf8(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm8 = zero,zero,xmm8[3,10],zero,zero,zero,xmm8[6,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = xmm7[5,12],zero,zero,xmm7[1,8,15],zero,xmm7[u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpor %xmm7, %xmm8, %xmm7
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm8, %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[4,11,0,7,u,u,u,u,u,u,u,u,u,u,u,u]
@@ -1403,7 +1403,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm3[u,u,u,u,u,3,10],zero,zero,zero,xmm3[u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm4[u,u,u,u,u],zero,zero,xmm4[1,8,15,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm5, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm5, %xmm0, %xmm1, %xmm0
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 80(%rdi), %xmm1
@@ -1415,7 +1415,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = xmm4[u,u,u,u,u],zero,zero,xmm4[2,9,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = xmm3[u,u,u,u,u,4,11],zero,zero,xmm3[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm9, %xmm10, %xmm9
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm11, %xmm8, %xmm9, %xmm8
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = xmm1[u,u,u,u,5,12,u,u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = xmm5[u,u,u,u,u,0,7,14,u,u,u,u,u,u,u,u]
@@ -1525,7 +1525,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: movq {{[0-9]+}}(%rsp), %r10
; AVX2-ONLY-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-ONLY-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-ONLY-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm2
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = zero,zero,zero,xmm3[5,12],zero,zero,xmm3[1,8,15,u,u,u,u,u,u]
@@ -1539,7 +1539,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm2[u,u,u,u,u,u,u,u,u,u],zero,zero,xmm2[4,11],zero,zero
; AVX2-ONLY-NEXT: vpor %xmm5, %xmm4, %xmm4
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0,1,2,3,4],xmm4[5,6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm4
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm4, %xmm5
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,zero,xmm5[6,13],zero,zero,xmm5[2,9,u,u,u,u,u,u,u]
@@ -1553,7 +1553,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpor %xmm7, %xmm8, %xmm8
; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0]
; AVX2-ONLY-NEXT: vpblendvb %xmm7, %xmm6, %xmm8, %xmm6
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm8, %ymm0, %ymm1, %ymm8
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm8[2,9],zero,zero,zero,xmm8[5,12],zero,zero,xmm8[u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm8, %xmm8
@@ -1564,7 +1564,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm2[u,u,u,u,u,u,u,u,u],zero,zero,zero,xmm2[6,13],zero,zero
; AVX2-ONLY-NEXT: vpor %xmm12, %xmm11, %xmm11
; AVX2-ONLY-NEXT: vpblendvb %xmm7, %xmm8, %xmm11, %xmm8
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm11
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm11[3,10],zero,zero,zero,xmm11[6,13],zero,zero,xmm11[u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm11, %xmm11
@@ -1575,7 +1575,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm2[u,u,u,u,u,u,u,u,u],zero,zero,xmm2[0,7,14],zero,zero
; AVX2-ONLY-NEXT: vpor %xmm12, %xmm9, %xmm9
; AVX2-ONLY-NEXT: vpblendvb %xmm7, %xmm11, %xmm9, %xmm9
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm1, %ymm0, %ymm11
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm11, %xmm12
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = zero,zero,xmm12[2,9],zero,zero,zero,xmm12[5,12,u,u,u,u,u,u,u]
@@ -1585,7 +1585,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm2[u,u,u,u,u,u,u,u,u],zero,zero,xmm2[1,8,15],zero,zero
; AVX2-ONLY-NEXT: vpor %xmm12, %xmm10, %xmm10
; AVX2-ONLY-NEXT: vpblendvb %xmm7, %xmm11, %xmm10, %xmm10
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-ONLY-NEXT: vpblendvb %ymm11, %ymm1, %ymm0, %ymm11
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm11, %xmm12
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = zero,zero,xmm12[3,10],zero,zero,zero,xmm12[6,13,u,u,u,u,u,u,u]
@@ -1596,7 +1596,7 @@ define void @load_i8_stride7_vf16(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm2[u,u,u,u,u,u,u,u,u],zero,zero,xmm2[2,9],zero,zero,zero
; AVX2-ONLY-NEXT: vpor %xmm13, %xmm12, %xmm12
; AVX2-ONLY-NEXT: vpblendvb %xmm7, %xmm11, %xmm12, %xmm11
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-ONLY-NEXT: vpblendvb %ymm12, %ymm1, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm0[6,13],zero,zero,xmm0[2,9],zero,zero,zero,xmm0[u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vextracti128 $1, %ymm0, %xmm0
@@ -2921,7 +2921,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa 128(%rdi), %xmm9
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm9[u,u,u,5,12],zero,zero,xmm9[u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,0,0,0,0,255,255,255,255,255,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,0,0,0,0,255,255,255,255,255,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa (%rdi), %xmm3
@@ -2935,7 +2935,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm15[u,u,u,u,u,3,10],zero,zero,zero,xmm15[u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = xmm4[u,u,u,u,u],zero,zero,xmm4[1,8,15,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm3, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm3, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm6[u,u,u,u,u,u,u,2,9],zero,zero,zero,xmm6[u,u,u,u]
@@ -2952,7 +2952,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = xmm4[u,u,u,u,u],zero,zero,xmm4[2,9,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm15[u,u,u,u,u,4,11],zero,zero,xmm15[u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqu %ymm1, (%rsp) # 32-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm8[u,u],zero,zero,zero,xmm8[5,12,u,u,u,u,u,u,u,u,u]
@@ -2961,7 +2961,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm6[u,u,u,u,u,u,u,3,10],zero,zero,zero,xmm6[u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = xmm7[u,u,u,u,u,u,u],zero,zero,xmm7[1,8,15,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm3, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,255,255,255,255,255,0,0,0,0,0,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,255,255,255,255,255,0,0,0,0,0,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm3, %xmm1, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm10[2,9],zero,zero,zero,xmm10[u,u,u,u,u,u,u,u,u,u,u]
@@ -2988,7 +2988,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm0 = xmm7[u,u,u,u,u,u,u],zero,zero,xmm7[2,9,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm6[u,u,u,u,u,u,u,4,11],zero,zero,xmm6[u,u,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,255,255,255,255,255,0,0,0,0,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,255,255,255,255,255,0,0,0,0,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm0, %xmm1, %xmm2, %xmm2
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm9[u,u,2,9],zero,zero,zero,xmm9[u,u,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = xmm8[u,u],zero,zero,xmm8[0,7,14,u,u,u,u,u,u,u,u,u]
@@ -3009,7 +3009,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm12 = xmm1[u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,xmm1[5,12]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm13 = xmm5[u,u,u,u,u,u,u,u,u,u,u,0,7,14],zero,zero
; AVX1-ONLY-NEXT: vpor %xmm12, %xmm13, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,255,255,255,255,255,255,255,255,255,0,0,0,0,0>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
; AVX1-ONLY-NEXT: vpblendvb %xmm12, %xmm2, %xmm13, %xmm0
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm1[u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,xmm1[6,13]
@@ -3228,27 +3228,27 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqa 32(%rdi), %ymm7
; AVX2-SLOW-NEXT: vmovdqa 64(%rdi), %ymm13
; AVX2-SLOW-NEXT: vmovdqa 96(%rdi), %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm6, %ymm7, %ymm0
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,zero,xmm1[5,12],zero,zero,xmm1[1,8,15,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,7,14],zero,zero,xmm0[3,10],zero,zero,zero,xmm0[u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm0, %xmm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm3, %ymm13, %ymm1
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm14
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm4[2],ymm1[3,4],ymm4[5],ymm1[6,7,8,9],ymm4[10],ymm1[11,12],ymm4[13],ymm1[14,15]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,6,13,4,11,2,9,16,23,30,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm10, %ymm11, %ymm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm4 = xmm4[u,u,u],zero,zero,xmm4[3,10],zero,zero,zero,xmm4[6,13,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,5,12],zero,zero,xmm1[1,8,15],zero,zero,xmm1[u,u,u,u]
; AVX2-SLOW-NEXT: vpor %xmm4, %xmm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm6, %ymm7, %ymm4
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm4, %xmm5
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,zero,xmm5[6,13],zero,zero,xmm5[2,9,u,u,u,u,u,u,u]
@@ -3258,7 +3258,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm8
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1],ymm8[2],ymm5[3,4,5],ymm8[6],ymm5[7,8,9],ymm8[10],ymm5[11,12,13],ymm8[14],ymm5[15]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm5 = ymm5[u,u,u,u,u,u,u,u,u,0,7,14,5,12,3,10,17,24,31,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm5, %ymm4, %ymm8
; AVX2-SLOW-NEXT: vmovdqa 192(%rdi), %xmm4
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm9 = xmm4[u,u,u,u,u,u,4,11,u,u,u,u,u,u,u,u]
@@ -3298,7 +3298,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: # ymm0 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm8, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm11, %ymm10, %ymm1
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm8 = xmm1[u,u,2,9],zero,zero,zero,xmm1[5,12],zero,zero,xmm1[u,u,u,u,u]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm1
@@ -3311,7 +3311,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm8, %ymm0, %ymm8
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm8, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm11, %ymm10, %ymm1
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm12 = xmm1[u,u,3,10],zero,zero,zero,xmm1[6,13],zero,zero,xmm1[u,u,u,u,u]
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm1
@@ -3325,7 +3325,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm12, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm11, %ymm10, %ymm14
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm10, %xmm11
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm11 = xmm11[u,u],zero,zero,xmm11[2,9],zero,zero,zero,xmm11[5,12,u,u,u,u,u]
@@ -3338,14 +3338,14 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm11, %ymm0, %ymm11
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm10, %ymm11, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm7, %ymm6, %ymm10
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm6, %ymm7, %ymm11
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm15
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm7, %ymm6, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm7, %ymm6, %ymm6
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm13, %ymm3, %ymm0
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm13, %ymm3, %ymm9
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm13, %ymm3, %ymm8
@@ -3441,21 +3441,21 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %ymm6
; AVX2-FAST-NEXT: vmovdqa 64(%rdi), %ymm3
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm5, %ymm6, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm13
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,zero,xmm1[5,12],zero,zero,xmm1[1,8,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,7,14],zero,zero,xmm0[3,10],zero,zero,zero,xmm0[u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpor %xmm1, %xmm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm4, %ymm3, %ymm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm2[2],ymm1[3,4],ymm2[5],ymm1[6,7,8,9],ymm2[10],ymm1[11,12],ymm2[13],ymm1[14,15]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,6,13,4,11,2,9,16,23,30,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm8, %ymm9, %ymm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u],zero,zero,xmm2[3,10],zero,zero,zero,xmm2[6,13,u,u,u,u]
@@ -3472,7 +3472,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, (%rsp) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm5, %ymm6, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm12
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
@@ -3483,7 +3483,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm7
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm7[2],ymm1[3,4,5],ymm7[6],ymm1[7,8,9],ymm7[10],ymm1[11,12,13],ymm7[14],ymm1[15]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,0,7,14,5,12,3,10,17,24,31,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm8, %ymm9, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm7 = xmm1[u,u,u,6,13],zero,zero,xmm1[2,9],zero,zero,zero,xmm1[u,u,u,u]
@@ -3514,7 +3514,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: # ymm1 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm10, %ymm7, %ymm7
; AVX2-FAST-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm9, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm10 = xmm7[u,u,2,9],zero,zero,zero,xmm7[5,12],zero,zero,xmm7[u,u,u,u,u]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm7, %xmm7
@@ -3527,7 +3527,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vinserti128 $1, %xmm10, %ymm0, %ymm10
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm7, %ymm10, %ymm7
; AVX2-FAST-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm9, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm15 = xmm7[u,u,3,10],zero,zero,zero,xmm7[6,13],zero,zero,xmm7[u,u,u,u,u]
; AVX2-FAST-NEXT: vextracti128 $1, %ymm7, %xmm7
@@ -3539,9 +3539,9 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
; AVX2-FAST-NEXT: vinserti128 $1, %xmm12, %ymm0, %ymm12
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm7, %ymm12, %ymm15
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm9, %ymm8, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm8, %ymm9, %ymm8
; AVX2-FAST-NEXT: vextracti128 $1, %ymm8, %xmm9
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[u,u],zero,zero,xmm9[2,9],zero,zero,zero,xmm9[5,12,u,u,u,u,u]
@@ -3557,9 +3557,9 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm5, %ymm6, %ymm2
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm5, %ymm6, %ymm8
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm6, %ymm5, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm6, %ymm5, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm3, %ymm4, %ymm0
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm3, %ymm4, %ymm13
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm3, %ymm4, %ymm10
@@ -3652,27 +3652,27 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa 64(%rdi), %ymm13
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm6, %ymm7, %ymm0
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,zero,xmm1[5,12],zero,zero,xmm1[1,8,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,7,14],zero,zero,xmm0[3,10],zero,zero,zero,xmm0[u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm0, %xmm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm3, %ymm13, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm2, %ymm14
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm4[2],ymm1[3,4],ymm4[5],ymm1[6,7,8,9],ymm4[10],ymm1[11,12],ymm4[13],ymm1[14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,6,13,4,11,2,9,16,23,30,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm1, %ymm10, %ymm11, %ymm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm4 = xmm4[u,u,u],zero,zero,xmm4[3,10],zero,zero,zero,xmm4[6,13,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,5,12],zero,zero,xmm1[1,8,15],zero,zero,xmm1[u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpor %xmm4, %xmm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm6, %ymm7, %ymm4
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm4, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm5 = zero,zero,zero,xmm5[6,13],zero,zero,xmm5[2,9,u,u,u,u,u,u,u]
@@ -3682,7 +3682,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm5, %xmm8
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1],ymm8[2],ymm5[3,4,5],ymm8[6],ymm5[7,8,9],ymm8[10],ymm5[11,12,13],ymm8[14],ymm5[15]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm5 = ymm5[u,u,u,u,u,u,u,u,u,0,7,14,5,12,3,10,17,24,31,u,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm5, %ymm4, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa 192(%rdi), %xmm4
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm9 = xmm4[u,u,u,u,u,u,4,11,u,u,u,u,u,u,u,u]
@@ -3722,7 +3722,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: # ymm0 = mem[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm8, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm11, %ymm10, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm8 = xmm1[u,u,2,9],zero,zero,zero,xmm1[5,12],zero,zero,xmm1[u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm1
@@ -3735,7 +3735,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm8, %ymm0, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm1, %ymm8, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm11, %ymm10, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm12 = xmm1[u,u,3,10],zero,zero,zero,xmm1[6,13],zero,zero,xmm1[u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm1
@@ -3749,7 +3749,7 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm1, %ymm12, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm11, %ymm10, %ymm14
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm10, %xmm11
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm11 = xmm11[u,u],zero,zero,xmm11[2,9],zero,zero,zero,xmm11[5,12,u,u,u,u,u]
@@ -3762,14 +3762,14 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm11, %ymm0, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm10, %ymm11, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm7, %ymm6, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm6, %ymm7, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm15
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm7, %ymm6, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm7, %ymm6, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm13, %ymm3, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm13, %ymm3, %ymm9
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm13, %ymm3, %ymm8
@@ -4269,20 +4269,20 @@ define void @load_i8_stride7_vf32(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-SLOW: # %bb.0:
; AVX512BW-SLOW-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512BW-SLOW-NEXT: movq {{[0-9]+}}(%rsp), %r10
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31]
; AVX512BW-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm0, %zmm0
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm4
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm5
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm11
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm12
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm10
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31]
; AVX512BW-SLOW-NEXT: vpermw %zmm1, %zmm2, %zmm6
; AVX512BW-SLOW-NEXT: vmovdqa (%rdi), %ymm3
; AVX512BW-SLOW-NEXT: vmovdqa 32(%rdi), %ymm2
@@ -6781,7 +6781,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: # xmm5 = mem[0,0]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm8, %xmm8
; AVX1-ONLY-NEXT: vpor %xmm6, %xmm8, %xmm8
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,0,0,0,0,0,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm6, %xmm7, %xmm8, %xmm7
; AVX1-ONLY-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 240(%rdi), %xmm7
@@ -6815,7 +6815,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm6, %xmm9
; AVX1-ONLY-NEXT: vmovdqa %xmm6, %xmm14
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm9, %xmm9
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <u,u,u,0,0,0,0,255,255,255,255,255,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [u,u,u,0,0,0,0,255,255,255,255,255,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm15, %xmm2, %xmm9, %xmm2
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 400(%rdi), %xmm2
@@ -6877,7 +6877,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: # xmm3 = mem[0,0]
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm10, %xmm13
; AVX1-ONLY-NEXT: vpor %xmm4, %xmm13, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,255,255,255,255,255,0,0,0,0,0,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,255,255,255,255,255,0,0,0,0,0,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm4, %xmm5, %xmm13, %xmm5
; AVX1-ONLY-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm0, %xmm8, %xmm0
@@ -6904,7 +6904,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm7 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm7, %xmm13
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm13, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [255,255,255,255,255,0,0,0,0,u,u,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm9, %xmm3, %xmm13, %xmm0
; AVX1-ONLY-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -6984,7 +6984,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm12 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm12, %xmm13
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm13, %xmm13
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,255,255,255,255,255,0,0,0,0,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,255,255,255,255,255,0,0,0,0,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpblendvb %xmm2, %xmm1, %xmm13, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm7 # 16-byte Reload
@@ -7059,7 +7059,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX1-ONLY-NEXT: vpshufb %xmm11, %xmm0, %xmm13
; AVX1-ONLY-NEXT: vmovdqa %xmm0, %xmm5
; AVX1-ONLY-NEXT: vpor %xmm4, %xmm13, %xmm4
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = <u,u,255,255,255,255,255,255,255,255,255,0,0,0,0,0>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = [u,u,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
; AVX1-ONLY-NEXT: vpblendvb %xmm13, {{[-0-9]+}}(%r{{[sb]}}p), %xmm4, %xmm0 # 16-byte Folded Reload
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 432(%rdi), %xmm1
@@ -7615,7 +7615,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqa 32(%rdi), %ymm2
; AVX2-SLOW-NEXT: vmovdqa 64(%rdi), %ymm4
; AVX2-SLOW-NEXT: vmovdqa 96(%rdi), %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vmovdqa %ymm3, %ymm13
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm10
@@ -7623,12 +7623,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqa %ymm1, %ymm12
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm2, %xmm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpor %xmm3, %xmm0, %xmm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm5, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vmovdqa %ymm9, %ymm14
; AVX2-SLOW-NEXT: vmovdqa %ymm5, %ymm9
@@ -7664,19 +7664,19 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm11, %ymm9, %ymm2
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm4 = ymm2[0,1],ymm3[2],ymm2[3,4,5],ymm3[6],ymm2[7,8,9],ymm3[10],ymm2[11,12,13],ymm3[14],ymm2[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm12, %ymm10, %ymm5
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm3, %xmm7
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm5, %xmm5
; AVX2-SLOW-NEXT: vpor %xmm7, %xmm5, %xmm5
; AVX2-SLOW-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10,1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10]
; AVX2-SLOW-NEXT: # ymm7 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpshufb %ymm7, %ymm4, %ymm4
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm15, %ymm6, %ymm4
@@ -7692,14 +7692,14 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa 160(%rdi), %ymm9
; AVX2-SLOW-NEXT: vmovdqa 128(%rdi), %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm9, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vmovdqa %ymm3, %ymm15
; AVX2-SLOW-NEXT: vmovdqu %ymm9, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm3, %xmm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpbroadcastw {{.*#+}} xmm12 = [2,9,2,9,2,9,2,9,2,9,2,9,2,9,2,9]
@@ -7735,12 +7735,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm6[7]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm9, %ymm15, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm10, %xmm0, %xmm6
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpor %xmm6, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpbroadcastw {{.*#+}} xmm6 = [3,10,3,10,3,10,3,10,3,10,3,10,3,10,3,10]
@@ -7768,13 +7768,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm3, %ymm4, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm1 = [0,0,0,128,128,128,5,12,0,0,0,128,128,128,5,12]
@@ -7805,12 +7805,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm6, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm11, %ymm4, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm1, %xmm7
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpor %xmm7, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm7 = [0,0,0,128,128,128,6,13,0,0,0,128,128,128,6,13]
@@ -7837,13 +7837,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm6, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm3, %ymm4, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm5, %xmm1, %xmm7
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpor %xmm7, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm7 = [0,0,0,2,9,128,128,128,0,0,0,2,9,128,128,128]
@@ -7871,15 +7871,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm6, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm3, %ymm4, %ymm2
; AVX2-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm4, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm6, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpor %xmm2, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} xmm2 = [0,0,0,3,10,128,128,128,0,0,0,3,10,128,128,128]
@@ -7909,15 +7909,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm0, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm0, %ymm2, %ymm12
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm2, %ymm0, %ymm13
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm2, %ymm0, %ymm3
; AVX2-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm2, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -7932,7 +7932,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm6, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm6, %ymm2, %ymm4
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm6, %ymm2, %ymm9
@@ -7947,12 +7947,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm7, %ymm8, %ymm10
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm8, %ymm7, %ymm5
; AVX2-SLOW-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm8, %ymm7, %ymm11
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm1, %xmm15
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm15, %xmm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm15
@@ -7972,10 +7972,10 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm12, %xmm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm12, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm2, %xmm1
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm4, %xmm2
@@ -7993,9 +7993,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpshufb %ymm4, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm14
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm13, %xmm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm13, %xmm3
; AVX2-SLOW-NEXT: vpor %xmm0, %xmm3, %xmm0
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm9, %xmm3
@@ -8015,9 +8015,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm2, %xmm1
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
@@ -8037,11 +8037,11 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6,7,8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14,15]
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm1, %ymm3, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm5, %xmm3
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm5, %xmm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm5, %xmm5
; AVX2-SLOW-NEXT: vpor %xmm3, %xmm5, %xmm3
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
@@ -8062,9 +8062,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm9, %xmm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm5, %xmm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm9, %xmm9
; AVX2-SLOW-NEXT: vpor %xmm5, %xmm9, %xmm5
; AVX2-SLOW-NEXT: vpbroadcastw {{.*#+}} xmm13 = [4,11,4,11,4,11,4,11,4,11,4,11,4,11,4,11]
@@ -8154,19 +8154,19 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %ymm12
; AVX2-FAST-NEXT: vmovdqa 64(%rdi), %ymm4
; AVX2-FAST-NEXT: vmovdqa 96(%rdi), %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm1, %ymm12, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm7
; AVX2-FAST-NEXT: vmovdqu %ymm12, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm13
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm2, %xmm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpor %xmm3, %xmm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm5, %ymm4, %ymm3
; AVX2-FAST-NEXT: vmovdqa %ymm5, %ymm9
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -8199,23 +8199,23 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm11, %ymm9, %ymm2
; AVX2-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm2[0,1],ymm3[2],ymm2[3,4,5],ymm3[6],ymm2[7,8,9],ymm3[10],ymm2[11,12,13],ymm3[14],ymm2[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm13, %ymm12, %ymm4
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm9
; AVX2-FAST-NEXT: vextracti128 $1, %ymm4, %xmm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm5, %xmm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm4, %xmm4
; AVX2-FAST-NEXT: vpor %xmm5, %xmm4, %xmm4
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm5 = [1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10,1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10]
; AVX2-FAST-NEXT: # ymm5 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm3, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm8, %ymm0, %ymm3
@@ -8230,7 +8230,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm3, %ymm2, %ymm5
; AVX2-FAST-NEXT: vmovdqa 160(%rdi), %ymm1
; AVX2-FAST-NEXT: vmovdqa 128(%rdi), %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm1, %ymm3, %ymm2
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm12
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -8238,9 +8238,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm9
; AVX2-FAST-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm3, %xmm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX2-FAST-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -8273,12 +8273,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-FAST-NEXT: vpblendvb %ymm6, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm10, %ymm12, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm4
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpor %xmm4, %xmm0, %xmm0
; AVX2-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -8300,13 +8300,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm5, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm14, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm14, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm1, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpor %xmm1, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm10 = [0,0,0,128,128,128,5,12,0,0,0,128,128,128,5,12]
@@ -8338,14 +8338,14 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vinserti128 $1, %xmm10, %ymm0, %ymm10
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm6, %ymm10, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm14, %ymm3, %ymm6
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm14
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm6, %xmm11
; AVX2-FAST-NEXT: vextracti128 $1, %ymm6, %xmm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm12, %xmm6, %xmm6
; AVX2-FAST-NEXT: vpor %xmm6, %xmm11, %xmm6
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm11 = [0,0,0,128,128,128,6,13,0,0,0,128,128,128,6,13]
@@ -8373,14 +8373,14 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm6, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vmovdqa %ymm14, %ymm3
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm2, %ymm14, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm14, %xmm1, %xmm10
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm12, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpor %xmm1, %xmm10, %xmm1
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm10 = [0,0,0,2,9,128,128,128,0,0,0,2,9,128,128,128]
@@ -8407,15 +8407,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm1, %ymm6, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm2, %ymm3, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm3, %ymm2, %ymm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpor %xmm2, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} xmm2 = [0,0,0,3,10,128,128,128,0,0,0,3,10,128,128,128]
@@ -8446,15 +8446,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm0, %ymm1, %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm9
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm1, %ymm0, %ymm10
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm1, %ymm0, %ymm2
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -8470,7 +8470,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm2, %ymm0, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm2, %ymm0, %ymm4
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm2, %ymm0, %ymm14
@@ -8488,10 +8488,10 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm5, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm11, %xmm0
; AVX2-FAST-NEXT: vextracti128 $1, %ymm11, %xmm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm11, %xmm11
; AVX2-FAST-NEXT: vpor %xmm0, %xmm11, %xmm0
; AVX2-FAST-NEXT: vextracti128 $1, %ymm1, %xmm11
@@ -8511,10 +8511,10 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm9, %xmm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm9, %xmm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpor %xmm1, %xmm2, %xmm1
; AVX2-FAST-NEXT: vextracti128 $1, %ymm4, %xmm2
@@ -8532,9 +8532,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm12
; AVX2-FAST-NEXT: vextracti128 $1, %ymm10, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm10, %xmm3
; AVX2-FAST-NEXT: vpor %xmm0, %xmm3, %xmm0
; AVX2-FAST-NEXT: vextracti128 $1, %ymm14, %xmm3
@@ -8554,9 +8554,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm15
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-FAST-NEXT: vextracti128 $1, %ymm3, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-NEXT: vpor %xmm0, %xmm3, %xmm0
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
@@ -8576,11 +8576,11 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm3[0],ymm2[1],ymm3[2,3],ymm2[4],ymm3[5,6,7,8],ymm2[9],ymm3[10,11],ymm2[12],ymm3[13,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm4, %xmm2
; AVX2-FAST-NEXT: vextracti128 $1, %ymm4, %xmm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm4, %xmm4
; AVX2-FAST-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
@@ -8602,9 +8602,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-FAST-NEXT: vextracti128 $1, %ymm7, %xmm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm4, %xmm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm7, %xmm7
; AVX2-FAST-NEXT: vpor %xmm4, %xmm7, %xmm4
; AVX2-FAST-NEXT: vinserti128 $1, %xmm4, %ymm0, %ymm4
@@ -8686,7 +8686,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa 64(%rdi), %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rdi), %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm3, %ymm13
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm2, %ymm10
@@ -8694,12 +8694,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm1, %ymm12
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,128,5,12,128,128,1,8,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm2, %xmm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [0,7,14,128,128,3,10,128,128,128,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpor %xmm3, %xmm0, %xmm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm5, %ymm4, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm9, %ymm14
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm5, %ymm9
@@ -8735,19 +8735,19 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm9, %ymm2
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm2[0,1],ymm3[2],ymm2[3,4,5],ymm3[6],ymm2[7,8,9],ymm3[10],ymm2[11,12,13],ymm3[14],ymm2[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm12, %ymm10, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm5, %xmm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,128,6,13,128,128,2,9,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm3, %xmm7
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [1,8,15,128,128,4,11,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpor %xmm7, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vbroadcasti128 {{.*#+}} ymm7 = [1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10,1,8,15,0,0,0,0,0,0,0,7,14,5,12,3,10]
; AVX2-FAST-PERLANE-NEXT: # ymm7 = mem[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm4, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm15, %ymm6, %ymm4
@@ -8763,14 +8763,14 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 160(%rdi), %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa 128(%rdi), %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm9, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm3, %ymm15
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm9, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,128,128,3,10,128,128,128,6,13,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm3, %xmm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,5,12,128,128,1,8,15,128,128,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpbroadcastw {{.*#+}} xmm12 = [2,9,2,9,2,9,2,9,2,9,2,9,2,9,2,9]
@@ -8806,12 +8806,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm6[7]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm9, %ymm15, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,6,13,128,128,2,9,128,128,128,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm0, %xmm6
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,128,128,4,11,128,128,0,7,14,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpor %xmm6, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastw {{.*#+}} xmm6 = [3,10,3,10,3,10,3,10,3,10,3,10,3,10,3,10]
@@ -8839,13 +8839,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5,6],ymm1[7]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm3, %ymm4, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,128,128,128,6,13,128,128,2,9,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,1,8,15,128,128,4,11,128,128,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm1 = [0,0,0,128,128,128,5,12,0,0,0,128,128,128,5,12]
@@ -8876,12 +8876,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm6, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm11, %ymm4, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,2,9,128,128,128,5,12,128,128,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm1, %xmm7
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,128,128,0,7,14,128,128,3,10,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpor %xmm7, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm7 = [0,0,0,128,128,128,6,13,0,0,0,128,128,128,6,13]
@@ -8908,13 +8908,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm1, %ymm6, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm3, %ymm4, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,3,10,128,128,128,6,13,128,128,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm1, %xmm7
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,128,128,1,8,15,128,128,4,11,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpor %xmm7, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm7 = [0,0,0,2,9,128,128,128,0,0,0,2,9,128,128,128]
@@ -8942,15 +8942,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm1, %ymm6, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm3, %ymm4, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm4, %ymm3, %ymm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,128,128,2,9,128,128,128,5,12,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,4,11,128,128,0,7,14,128,128,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpor %xmm2, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} xmm2 = [0,0,0,3,10,128,128,128,0,0,0,3,10,128,128,128]
@@ -8980,15 +8980,15 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm0, %ymm2, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm0, %ymm2, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm2, %ymm0, %ymm13
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm2, %ymm0, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm2, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -9003,7 +9003,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u,0,0,u,u,255,255,0,0,u,u,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm6, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm6, %ymm2, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm6, %ymm2, %ymm9
@@ -9018,12 +9018,12 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm7, %ymm8, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm8, %ymm7, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u,0,0,u,u,255,255,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm8, %ymm7, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [2,9,128,128,128,5,12,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm1, %xmm15
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,0,7,14,128,128,3,10,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm15, %xmm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm15
@@ -9043,10 +9043,10 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm5, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [3,10,128,128,128,6,13,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm12, %xmm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm12, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,1,8,15,128,128,4,11,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm2, %xmm1
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm4, %xmm2
@@ -9064,9 +9064,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm14
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm13, %xmm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,2,9,128,128,128,5,12,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm0, %xmm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [4,11,128,128,0,7,14,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm13, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpor %xmm0, %xmm3, %xmm0
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm9, %xmm3
@@ -9086,9 +9086,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm2, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,3,10,128,128,128,6,13,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm1, %xmm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = <5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = [5,12,128,128,1,8,15,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm2, %xmm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
@@ -9108,11 +9108,11 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6,7,8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm5, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm1, %ymm3, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = <6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = [6,13,128,128,2,9,128,128,128,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm5, %xmm3
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm5, %xmm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,4,11,128,128,0,7,14,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpor %xmm3, %xmm5, %xmm3
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
@@ -9133,9 +9133,9 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm9, %xmm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,128,128,128,5,12,128,128,1,8,15,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm5, %xmm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,0,7,14,128,128,3,10,128,128,128,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm9, %xmm9
; AVX2-FAST-PERLANE-NEXT: vpor %xmm5, %xmm9, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpbroadcastw {{.*#+}} xmm13 = [4,11,4,11,4,11,4,11,4,11,4,11,4,11,4,11]
@@ -9374,7 +9374,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[u,u,1,8,15],zero,zero,xmm0[4,11],zero,zero,xmm0[u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpor %xmm2, %xmm0, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm2, %xmm10, %xmm6
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm15 = xmm5[u,u,u,u,u,u,u,u,u,u,u,0,7,14],zero,zero
; AVX512F-ONLY-SLOW-NEXT: vmovdqa %xmm5, %xmm10
@@ -9497,7 +9497,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u],zero,zero,zero,xmm3[6,13],zero,zero,xmm3[2,9,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u,u,1,8,15],zero,zero,xmm2[4,11],zero,zero,xmm2[u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpor %xmm3, %xmm2, %xmm2
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm5, %ymm0, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vpternlogq $220, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa %ymm9, %ymm0
@@ -9551,7 +9551,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u],zero,zero,xmm2[0,7,14],zero,zero,xmm2[3,10,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm12 # 16-byte Reload
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %xmm1, %xmm12, %xmm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa %xmm10, %xmm13
; AVX512F-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm6 = xmm10[u,u,u,u,u,u,u,u,u,u,u,1,8,15],zero,zero
@@ -9787,7 +9787,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,1,8,15],zero,zero,xmm2[4,11],zero,zero,xmm2[u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpor %xmm7, %xmm2, %xmm2
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm7
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 208(%rdi), %xmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm8
@@ -9914,7 +9914,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[u,u,u,u],zero,zero,zero,xmm5[6,13],zero,zero,xmm5[2,9,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,1,8,15],zero,zero,xmm3[4,11],zero,zero,xmm3[u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpor %xmm5, %xmm3, %xmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX512F-ONLY-FAST-NEXT: vpternlogq $220, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa %ymm6, %ymm2
@@ -9967,7 +9967,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u],zero,zero,xmm3[0,7,14],zero,zero,xmm3[3,10,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpor %xmm2, %xmm3, %xmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm12 # 16-byte Reload
; AVX512F-ONLY-FAST-NEXT: vpshufb {{.*#+}} xmm8 = xmm12[u,u,u,u,u,u,u,u,u,u,u,1,8,15],zero,zero
@@ -10224,7 +10224,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[u,u,1,8,15],zero,zero,xmm0[4,11],zero,zero,xmm0[u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpor %xmm2, %xmm0, %xmm0
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm2, %xmm10, %xmm5
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm15 = xmm6[u,u,u,u,u,u,u,u,u,u,u,0,7,14],zero,zero
; AVX512DQ-SLOW-NEXT: vpor %xmm5, %xmm15, %xmm5
@@ -10346,7 +10346,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u],zero,zero,zero,xmm3[6,13],zero,zero,xmm3[2,9,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u,u,1,8,15],zero,zero,xmm2[4,11],zero,zero,xmm2[u,u,u]
; AVX512DQ-SLOW-NEXT: vpor %xmm3, %xmm2, %xmm2
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm8, %ymm0, %ymm0
; AVX512DQ-SLOW-NEXT: vpternlogq $220, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm2
; AVX512DQ-SLOW-NEXT: vmovdqa %ymm9, %ymm0
@@ -10402,7 +10402,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u],zero,zero,xmm2[0,7,14],zero,zero,xmm2[3,10,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512DQ-SLOW-NEXT: vmovdqa %xmm10, %xmm1
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u]
; AVX512DQ-SLOW-NEXT: vpshufb %xmm2, %xmm10, %xmm2
; AVX512DQ-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm4 # 16-byte Reload
; AVX512DQ-SLOW-NEXT: vpshufb {{.*#+}} xmm3 = xmm4[u,u,u,u,u,u,u,u,u,u,u,1,8,15],zero,zero
@@ -10638,7 +10638,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,1,8,15],zero,zero,xmm2[4,11],zero,zero,xmm2[u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpor %xmm7, %xmm2, %xmm2
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm7
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,5,12,19,26,17,24,31,22,29,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vmovdqa 208(%rdi), %xmm0
; AVX512DQ-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512DQ-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm8
@@ -10765,7 +10765,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[u,u,u,u],zero,zero,zero,xmm5[6,13],zero,zero,xmm5[2,9,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,1,8,15],zero,zero,xmm3[4,11],zero,zero,xmm3[u,u,u]
; AVX512DQ-FAST-NEXT: vpor %xmm5, %xmm3, %xmm3
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,128,128,128,128,128,128,128,128,128,0,7,14,21,28,19,26,17,24,31,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX512DQ-FAST-NEXT: vpternlogq $220, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm3
; AVX512DQ-FAST-NEXT: vmovdqa %ymm6, %ymm2
@@ -10819,7 +10819,7 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[u,u],zero,zero,xmm6[0,7,14],zero,zero,xmm6[3,10,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpor %xmm2, %xmm6, %xmm2
; AVX512DQ-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm4 # 16-byte Reload
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,6,13,20,27,18,25,16,23,30,u,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm6, %xmm4, %xmm6
; AVX512DQ-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm12 # 16-byte Reload
; AVX512DQ-FAST-NEXT: vpshufb {{.*#+}} xmm8 = xmm12[u,u,u,u,u,u,u,u,u,u,u,1,8,15],zero,zero
@@ -10910,13 +10910,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-ONLY-SLOW-LABEL: load_i8_stride7_vf64:
; AVX512BW-ONLY-SLOW: # %bb.0:
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm25
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm18
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm24
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm9
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm0
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa (%rdi), %ymm10
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa 32(%rdi), %ymm1
@@ -11213,11 +11213,11 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512BW-ONLY-SLOW-NEXT: vpor %xmm7, %xmm8, %xmm7
; AVX512BW-ONLY-SLOW-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
; AVX512BW-ONLY-SLOW-NEXT: vmovdqu8 %ymm7, %ymm6 {%k1}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm7, %zmm7
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm8, %zmm8
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermw %zmm25, %zmm10, %zmm10
; AVX512BW-ONLY-SLOW-NEXT: vextracti128 $1, %ymm12, %xmm11
; AVX512BW-ONLY-SLOW-NEXT: vpshufb {{.*#+}} xmm11 = zero,zero,xmm11[2,9],zero,zero,zero,xmm11[5,12,u,u,u,u,u,u,u]
@@ -11646,13 +11646,13 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQBW-SLOW-LABEL: load_i8_stride7_vf64:
; AVX512DQBW-SLOW: # %bb.0:
; AVX512DQBW-SLOW-NEXT: vmovdqa64 64(%rdi), %zmm25
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [8,1,18,11,4,5,22,15,u,25,10,u,12,29,14,u]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm18
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,1,18,11,4,21,14,7,8,25,10,u,28,13,u,15]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm24
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [0,17,10,3,4,21,14,7,24,9,u,11,28,13,u,31]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm10
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [16,17,10,3,20,13,6,23,24,25,u,27,28,u,30,31]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm0, %zmm0
; AVX512DQBW-SLOW-NEXT: vmovdqa (%rdi), %ymm9
; AVX512DQBW-SLOW-NEXT: vmovdqa 32(%rdi), %ymm1
@@ -11949,11 +11949,11 @@ define void @load_i8_stride7_vf64(ptr %in.vec, ptr %out.vec0, ptr %out.vec1, ptr
; AVX512DQBW-SLOW-NEXT: vpor %xmm7, %xmm8, %xmm7
; AVX512DQBW-SLOW-NEXT: vinserti128 $1, %xmm7, %ymm0, %ymm7
; AVX512DQBW-SLOW-NEXT: vmovdqu8 %ymm7, %ymm6 {%k1}
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [16,9,2,19,20,13,6,23,24,u,26,27,28,u,30,31]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm7, %zmm7
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [16,9,2,19,12,5,22,23,24,u,26,27,u,29,30,31]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm8, %zmm8
-; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [8,1,2,19,12,5,22,15,u,9,26,11,u,29,14,u]
; AVX512DQBW-SLOW-NEXT: vpermw %zmm25, %zmm11, %zmm11
; AVX512DQBW-SLOW-NEXT: vextracti128 $1, %ymm2, %xmm12
; AVX512DQBW-SLOW-NEXT: vpshufb {{.*#+}} xmm12 = zero,zero,xmm12[2,9],zero,zero,zero,xmm12[5,12,u,u,u,u,u,u,u]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-3.ll
index 16b5e8db76ef13c..afa114048077e74 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-3.ll
@@ -156,7 +156,7 @@ define void @store_i16_stride3_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vmovq {{.*#+}} xmm2 = mem[0],zero
; AVX512BW-NEXT: vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX512BW-NEXT: vinserti128 $1, %xmm0, %ymm1, %ymm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,4,8,1,5,9,2,6,10,3,7,11,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,4,8,1,5,9,2,6,10,3,7,11,u,u,u,u]
; AVX512BW-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512BW-NEXT: vmovq %xmm1, 16(%rcx)
@@ -249,7 +249,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,0,1,u,u,u,u,2,3,u,u,u,u,4,5,u,u,22,23,u,u,u,u,24,25,u,u,u,u,26,27]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6],ymm3[7],ymm4[8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14],ymm3[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-SLOW-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -268,7 +268,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm1
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm2
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,4,1,5,1,5,2,6]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm5, %ymm3
@@ -294,7 +294,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,0,1,u,u,u,u,2,3,u,u,u,u,4,5,u,u,22,23,u,u,u,u,24,25,u,u,u,u,26,27]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6],ymm3[7],ymm4[8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14],ymm3[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -317,7 +317,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,0,1]
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,0,1,u,u,u,u,2,3,u,u,u,u,4,5,u,u,22,23,u,u,u,u,24,25,u,u,u,u,26,27]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6],ymm3[7],ymm4[8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14],ymm3[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX512F-SLOW-NEXT: vpermd %ymm2, %ymm4, %ymm4
; AVX512F-SLOW-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm3, %ymm4
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
@@ -341,7 +341,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} xmm3 = xmm2[2,2,3,3]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm3[1],xmm1[2,3],xmm3[4],xmm1[5,6],xmm3[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,0,0,u,1,1,u,2>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,0,0,u,1,1,u,2]
; AVX512F-FAST-NEXT: vpermd %ymm2, %ymm3, %ymm2
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,4,1,5,1,5,2,6]
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm3, %ymm0
@@ -358,7 +358,7 @@ define void @store_i16_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vmovdqa (%rdi), %xmm0
; AVX512BW-NEXT: vinserti128 $1, (%rsi), %ymm0, %ymm0
; AVX512BW-NEXT: vinserti32x4 $2, (%rdx), %zmm0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,8,16,1,9,17,2,10,18,3,11,19,4,12,20,5,13,21,6,14,22,7,15,23,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,8,16,1,9,17,2,10,18,3,11,19,4,12,20,5,13,21,6,14,22,7,15,23,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vextracti32x4 $2, %zmm0, 32(%rcx)
; AVX512BW-NEXT: vmovdqa %ymm0, (%rcx)
@@ -513,7 +513,7 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm6[4],xmm3[4],xmm6[5],xmm3[5],xmm6[6],xmm3[6],xmm6[7],xmm3[7]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[4,5,10,11,10,11,8,9,8,9,14,15,12,13,14,15]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm3, %ymm4, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <5,5,u,6,6,u,7,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [5,5,u,6,6,u,7,7]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm3, %ymm4, %ymm3
@@ -524,16 +524,16 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm2[0],xmm5[0],xmm2[1],xmm5[1],xmm2[2],xmm5[2],xmm2[3],xmm5[3]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm4, %ymm2, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,u,1,1,u,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,u,1,1,u,2]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm2, %ymm4, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,3,3,u,4,4,u,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,3,3,u,4,4,u,5]
; AVX2-SLOW-NEXT: vpermd (%rdi), %ymm4, %ymm4
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm1, %ymm4, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <2,u,3,3,u,4,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [2,u,3,3,u,4,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm4, %ymm0
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
@@ -558,7 +558,7 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm6[0],xmm2[0],xmm6[1],xmm2[1],xmm6[2],xmm2[2],xmm6[3],xmm2[3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm5, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm2, %ymm5, %ymm2
@@ -568,16 +568,16 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm3[4],xmm7[4],xmm3[5],xmm7[5],xmm3[6],xmm7[6],xmm3[7],xmm7[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[4,5,10,11,10,11,8,9,8,9,14,15,12,13,14,15]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm3, %ymm4, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm4, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-NEXT: vpermd (%rdi), %ymm4, %ymm4
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm1, %ymm4, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm4, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
@@ -602,7 +602,7 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm6[0],xmm2[0],xmm6[1],xmm2[1],xmm6[2],xmm2[2],xmm6[3],xmm2[3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm2, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm5, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm2, %ymm5, %ymm2
@@ -612,16 +612,16 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm3[4],xmm7[4],xmm3[5],xmm7[5],xmm3[6],xmm7[6],xmm3[7],xmm7[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[4,5,10,11,10,11,8,9,8,9,14,15,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm3, %ymm4, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-PERLANE-NEXT: vpermd (%rdi), %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm1, %ymm4, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm4, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
@@ -650,7 +650,7 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[10,11,u,u],zero,zero,ymm2[12,13,u,u],zero,zero,ymm2[14,15,u,u],zero,zero,ymm2[16,17,u,u],zero,zero,ymm2[18,19,u,u],zero,zero,ymm2[20,21]
; AVX512F-NEXT: vpor %ymm1, %ymm2, %ymm1
; AVX512F-NEXT: vinserti64x4 $1, %ymm1, %zmm3, %zmm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = <u,0,0,u,1,1,u,2>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = [u,0,0,u,1,1,u,2]
; AVX512F-NEXT: vpermd %ymm0, %ymm2, %ymm2
; AVX512F-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm2
; AVX512F-NEXT: vpshufb {{.*#+}} ymm3 = zero,zero,ymm0[10,11],zero,zero,zero,zero,ymm0[12,13],zero,zero,zero,zero,ymm0[14,15],zero,zero,zero,zero,ymm0[16,17],zero,zero,zero,zero,ymm0[18,19],zero,zero,zero,zero
@@ -662,7 +662,7 @@ define void @store_i16_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm4[4],xmm7[4],xmm4[5],xmm7[5],xmm4[6],xmm7[6],xmm4[7],xmm7[7]
; AVX512F-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[4,5,u,u,10,11,8,9,u,u,14,15,12,13,u,u]
; AVX512F-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <5,5,u,6,6,u,7,7>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [5,5,u,6,6,u,7,7]
; AVX512F-NEXT: vpermd %ymm0, %ymm3, %ymm0
; AVX512F-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm0
; AVX512F-NEXT: vmovdqa %ymm0, 64(%rcx)
@@ -973,7 +973,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,4,4,4,4]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm3[0,1],xmm5[2],xmm3[3,4],xmm5[5],xmm3[6,7]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm3, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <5,5,u,6,6,u,7,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [5,5,u,6,6,u,7,7]
; AVX2-SLOW-NEXT: vpermd %ymm1, %ymm5, %ymm6
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm3, %ymm6, %ymm3
@@ -1000,7 +1000,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm6, %xmm6
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm7, %ymm6, %ymm6
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,0,0,u,1,1,u,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,0,0,u,1,1,u,2]
; AVX2-SLOW-NEXT: vpermd %ymm2, %ymm7, %ymm9
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm6, %ymm9, %ymm6
@@ -1015,11 +1015,11 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm8, %ymm7, %ymm7
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-SLOW-NEXT: vpshufb %ymm8, %ymm4, %ymm4
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,3,3,u,4,4,u,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,3,3,u,4,4,u,5]
; AVX2-SLOW-NEXT: vpermd (%rdi), %ymm9, %ymm10
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm4, %ymm10, %ymm4
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <2,u,3,3,u,4,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [2,u,3,3,u,4,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm2, %ymm10, %ymm2
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm4, %ymm2, %ymm2
@@ -1052,7 +1052,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm1, %xmm1
; AVX2-FAST-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm7, %ymm3
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm1, %ymm3, %ymm1
@@ -1074,7 +1074,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm4, %xmm4
; AVX2-FAST-NEXT: vmovdqa 16(%rdi), %xmm12
; AVX2-FAST-NEXT: vinserti128 $1, %xmm10, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm7, %ymm5
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm4, %ymm5, %ymm4
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm10, %ymm5
@@ -1091,15 +1091,15 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm2, %ymm8, %ymm2
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm6, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-NEXT: vpermd (%rdi), %ymm8, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm6, %ymm9, %ymm6
; AVX2-FAST-NEXT: vmovdqa 32(%rsi), %ymm9
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm9, %ymm7
; AVX2-FAST-NEXT: vpermd 32(%rdi), %ymm8, %ymm8
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm7, %ymm8, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm8, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm6, %ymm0, %ymm0
@@ -1129,7 +1129,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm7, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm1, %ymm3, %ymm1
@@ -1151,7 +1151,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm4, %xmm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa 16(%rdi), %xmm12
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm10, %ymm4, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm7, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm10, %ymm5
@@ -1168,15 +1168,15 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm2, %ymm8, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm6, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-PERLANE-NEXT: vpermd (%rdi), %ymm8, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm6, %ymm9, %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rsi), %ymm9
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm9, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermd 32(%rdi), %ymm8, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm7, %ymm8, %ymm7
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm8, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm6, %ymm0, %ymm0
@@ -1215,7 +1215,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpblendw {{.*#+}} xmm3 = xmm5[0,1],xmm3[2],xmm5[3,4],xmm3[5],xmm5[6,7]
; AVX512F-NEXT: vinserti128 $1, %xmm6, %ymm3, %ymm3
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm3 = zmm3[0,1,2,3],zmm4[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <5,5,u,6,6,u,7,7,u,8,8,u,9,9,u,10>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [5,5,u,6,6,u,7,7,u,8,8,u,9,9,u,10]
; AVX512F-NEXT: vpermd (%rdx), %zmm4, %zmm5
; AVX512F-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm3, %zmm5
; AVX512F-NEXT: vmovdqa 32(%rdi), %ymm3
@@ -1257,7 +1257,7 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vinserti128 $1, %xmm6, %ymm0, %ymm0
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm0[0,1,2,3],zmm3[4,5,6,7]
; AVX512F-NEXT: vpshufb %ymm10, %ymm7, %ymm1
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = <u,0,0,u,1,1,u,2>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm2 = [u,0,0,u,1,1,u,2]
; AVX512F-NEXT: vpermd %ymm7, %ymm2, %ymm2
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535]
; AVX512F-NEXT: vpandn %ymm2, %ymm3, %ymm2
@@ -1274,15 +1274,15 @@ define void @store_i16_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,32,u,1,33,u,2,34,u,3,35,u,4,36,u,5,37,u,6,38,u,7,39,u,8,40,u,9,41,u,10,42>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,32,u,1,33,u,2,34,u,3,35,u,4,36,u,5,37,u,6,38,u,7,39,u,8,40,u,9,41,u,10,42]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,32,3,4,33,6,7,34,9,10,35,12,13,36,15,16,37,18,19,38,21,22,39,24,25,40,27,28,41,30,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,11,43,u,12,44,u,13,45,u,14,46,u,15,47,u,16,48,u,17,49,u,18,50,u,19,51,u,20,52,u,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,11,43,u,12,44,u,13,45,u,14,46,u,15,47,u,16,48,u,17,49,u,18,50,u,19,51,u,20,52,u,21]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [42,1,2,43,4,5,44,7,8,45,10,11,46,13,14,47,16,17,48,19,20,49,22,23,50,25,26,51,28,29,52,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <21,u,54,22,u,55,23,u,56,24,u,57,25,u,58,26,u,59,27,u,60,28,u,61,29,u,62,30,u,63,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [21,u,54,22,u,55,23,u,56,24,u,57,25,u,58,26,u,59,27,u,60,28,u,61,29,u,62,30,u,63,31,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,53,2,3,54,5,6,55,8,9,56,11,12,57,14,15,58,17,18,59,20,21,60,23,24,61,26,27,62,29,30,63]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm0
@@ -1904,7 +1904,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm5[0,1,2,3,4,4,4,4]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1],xmm5[2],xmm2[3,4],xmm5[5],xmm2[6,7]
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm6, %ymm2, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <5,5,u,6,6,u,7,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [5,5,u,6,6,u,7,7]
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm8, %ymm5
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm2, %ymm5, %ymm2
@@ -1954,7 +1954,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-SLOW-NEXT: vpshufb %xmm11, %xmm9, %xmm9
; AVX2-SLOW-NEXT: vinserti128 $1, %xmm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,0,0,u,1,1,u,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,0,0,u,1,1,u,2]
; AVX2-SLOW-NEXT: vpermd %ymm4, %ymm12, %ymm9
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm8, %ymm9, %ymm8
@@ -1994,9 +1994,9 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm2, %ymm5, %ymm11
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm14, %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,3,3,u,4,4,u,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,3,3,u,4,4,u,5]
; AVX2-SLOW-NEXT: vpermd (%rdi), %ymm12, %ymm13
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm5, %ymm13, %ymm5
; AVX2-SLOW-NEXT: vmovdqa 64(%rsi), %ymm13
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm13, %ymm13
@@ -2010,7 +2010,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm15, %ymm2
; AVX2-SLOW-NEXT: vpermd 96(%rdi), %ymm12, %ymm12
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm2, %ymm12, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <2,u,3,3,u,4,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [2,u,3,3,u,4,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm4, %ymm12, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm5, %ymm4, %ymm4
@@ -2053,7 +2053,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-NEXT: vpshufb %xmm12, %xmm3, %xmm3
; AVX2-FAST-NEXT: vinserti128 $1, %xmm4, %ymm3, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm13, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm3, %ymm4, %ymm3
@@ -2067,7 +2067,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[1,1,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0,1],xmm5[2],xmm4[3,4],xmm5[5],xmm4[6,7]
; AVX2-FAST-NEXT: vinserti128 $1, %xmm7, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm10, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm4, %ymm5, %ymm3
@@ -2138,9 +2138,9 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm4, %ymm6
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm3, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-NEXT: vpermd (%rdi), %ymm4, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm3, %ymm9, %ymm3
; AVX2-FAST-NEXT: vmovdqa 64(%rsi), %ymm9
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm9, %ymm9
@@ -2154,7 +2154,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm14, %ymm0
; AVX2-FAST-NEXT: vpermd 96(%rdi), %ymm4, %ymm4
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm0, %ymm4, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm4, %ymm2
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm3, %ymm2, %ymm2
@@ -2197,7 +2197,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm12 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm12, %xmm3, %xmm3
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm4, %ymm3, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,0,0,u,1,1,u,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,0,0,u,1,1,u,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm2, %ymm13, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm3, %ymm4, %ymm3
@@ -2211,7 +2211,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0,1],xmm5[2],xmm4[3,4],xmm5[5],xmm4[6,7]
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm7, %ymm4, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <5,5,u,6,6,u,7,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [5,5,u,6,6,u,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm1, %ymm10, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm4, %ymm5, %ymm3
@@ -2282,9 +2282,9 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm0, %ymm4, %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [10,11,0,1,12,13,12,13,2,3,2,3,14,15,4,5,26,27,16,17,28,29,28,29,18,19,18,19,30,31,20,21]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm3, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,3,3,u,4,4,u,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,3,3,u,4,4,u,5]
; AVX2-FAST-PERLANE-NEXT: vpermd (%rdi), %ymm4, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm3, %ymm9, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa 64(%rsi), %ymm9
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm9, %ymm9
@@ -2298,7 +2298,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm14, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpermd 96(%rdi), %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm0, %ymm4, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <2,u,3,3,u,4,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [2,u,3,3,u,4,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm2, %ymm4, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm3, %ymm2, %ymm2
@@ -2357,7 +2357,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa 64(%rdx), %ymm14
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm9 = [128,128,10,11,128,128,128,128,12,13,128,128,128,128,14,15,128,128,128,128,16,17,128,128,128,128,18,19,128,128,128,128]
; AVX512F-NEXT: vpshufb %ymm9, %ymm3, %ymm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = <u,0,0,u,1,1,u,2>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} ymm16 = [u,0,0,u,1,1,u,2]
; AVX512F-NEXT: vpermd %ymm3, %ymm16, %ymm3
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535]
; AVX512F-NEXT: vpandn %ymm3, %ymm15, %ymm3
@@ -2382,7 +2382,7 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpblendw {{.*#+}} xmm12 = xmm12[0,1],xmm13[2],xmm12[3,4],xmm13[5],xmm12[6,7]
; AVX512F-NEXT: vinserti128 $1, %xmm0, %ymm12, %ymm0
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm0[0,1,2,3],zmm10[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = <5,5,u,6,6,u,7,7,u,8,8,u,9,9,u,10>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = [5,5,u,6,6,u,7,7,u,8,8,u,9,9,u,10]
; AVX512F-NEXT: vpermd 64(%rdx), %zmm18, %zmm10
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535,65535,0,65535]
; AVX512F-NEXT: vpternlogd $184, %zmm0, %zmm22, %zmm10
@@ -2486,17 +2486,17 @@ define void @store_i16_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm3
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm4
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,32,u,1,33,u,2,34,u,3,35,u,4,36,u,5,37,u,6,38,u,7,39,u,8,40,u,9,41,u,10,42>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,32,u,1,33,u,2,34,u,3,35,u,4,36,u,5,37,u,6,38,u,7,39,u,8,40,u,9,41,u,10,42]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,32,3,4,33,6,7,34,9,10,35,12,13,36,15,16,37,18,19,38,21,22,39,24,25,40,27,28,41,30,31]
; AVX512BW-NEXT: vpermt2w %zmm4, %zmm8, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <21,u,54,22,u,55,23,u,56,24,u,57,25,u,58,26,u,59,27,u,60,28,u,61,29,u,62,30,u,63,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [21,u,54,22,u,55,23,u,56,24,u,57,25,u,58,26,u,59,27,u,60,28,u,61,29,u,62,30,u,63,31,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm10
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm9, %zmm10
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,53,2,3,54,5,6,55,8,9,56,11,12,57,14,15,58,17,18,59,20,21,60,23,24,61,26,27,62,29,30,63]
; AVX512BW-NEXT: vpermt2w %zmm5, %zmm11, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,11,43,u,12,44,u,13,45,u,14,46,u,15,47,u,16,48,u,17,49,u,18,50,u,19,51,u,20,52,u,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,11,43,u,12,44,u,13,45,u,14,46,u,15,47,u,16,48,u,17,49,u,18,50,u,19,51,u,20,52,u,21]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512BW-NEXT: vpermt2w %zmm3, %zmm12, %zmm13
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [42,1,2,43,4,5,44,7,8,45,10,11,46,13,14,47,16,17,48,19,20,49,22,23,50,25,26,51,28,29,52,31]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll
index e8da0c4dcf47a41..5f0b08a024f8b1a 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-4.ll
@@ -272,10 +272,10 @@ define void @store_i16_stride4_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-ONLY-NEXT: vinserti128 $1, (%rsi), %ymm0, %ymm0
; AVX2-ONLY-NEXT: vinserti128 $1, (%rcx), %ymm1, %ymm1
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm2 = ymm1[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15]
; AVX2-ONLY-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm4 = ymm0[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm5, %ymm4, %ymm4
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4],ymm2[5],ymm4[6],ymm2[7]
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm1 = ymm1[1,3,1,3]
@@ -295,10 +295,10 @@ define void @store_i16_stride4_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vinserti128 $1, (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vinserti128 $1, (%rcx), %ymm1, %ymm1
; AVX512F-NEXT: vpermq {{.*#+}} ymm2 = ymm1[1,3,1,3]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15]
; AVX512F-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX512F-NEXT: vpermq {{.*#+}} ymm4 = ymm0[1,3,1,3]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm5, %ymm4, %ymm4
; AVX512F-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4],ymm2[5],ymm4[6],ymm2[7]
; AVX512F-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,0,2]
@@ -1409,26 +1409,26 @@ define void @store_i16_stride4_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: movw $-21846, %ax # imm = 0xAAAA
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm5 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm2, 192(%r8)
@@ -3014,32 +3014,32 @@ define void @store_i16_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm6
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm6, %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm11
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm10, %zmm11
; AVX512BW-NEXT: movw $-21846, %ax # imm = 0xAAAA
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm9, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512BW-NEXT: vpermt2w %zmm6, %zmm9, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm13, %zmm14
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm15
; AVX512BW-NEXT: vpermt2w %zmm6, %zmm12, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm16, %zmm17
; AVX512BW-NEXT: vmovdqa32 %zmm15, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55]
; AVX512BW-NEXT: vpermt2w %zmm6, %zmm15, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u]
; AVX512BW-NEXT: vpermt2w %zmm2, %zmm6, %zmm0
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm0 {%k1}
; AVX512BW-NEXT: vpermi2w %zmm7, %zmm5, %zmm8
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-5.ll
index 413cd7d63d8c67c..d174db5f66a8192 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-5.ll
@@ -71,7 +71,7 @@ define void @store_i16_stride5_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,3,0,3,4,7,4,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm0 = ymm0[3,3,3,3,4,5,6,7,11,11,11,11,12,13,14,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,0,0,255,255,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,0,0,255,255,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-SLOW-NEXT: vmovd %xmm1, 16(%r9)
@@ -141,7 +141,7 @@ define void @store_i16_stride5_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vpunpckldq {{.*#+}} xmm1 = xmm1[0],mem[0],xmm1[1],mem[1]
; AVX512BW-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX512BW-NEXT: vinserti128 $1, (%r8), %ymm0, %ymm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,6,8,1,3,5,7,9,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,6,8,1,3,5,7,9,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512BW-NEXT: vmovd %xmm1, 16(%r9)
@@ -386,7 +386,7 @@ define void @store_i16_stride5_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512BW-NEXT: vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX512BW-NEXT: vinserti32x4 $2, %xmm1, %zmm0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,4,8,12,16,1,5,9,13,17,2,6,10,14,18,3,7,11,15,19,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,4,8,12,16,1,5,9,13,17,2,6,10,14,18,3,7,11,15,19,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vextracti32x4 $2, %zmm0, %xmm1
; AVX512BW-NEXT: vmovq %xmm1, 32(%r9)
@@ -572,7 +572,7 @@ define void @store_i16_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm9 = ymm6[3,3,3,3,4,5,6,7,11,11,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm9 = ymm9[0,1,2,3,5,5,4,7,8,9,10,11,13,13,12,15]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0,1],ymm8[2,3],ymm9[4,5,6],ymm8[7],ymm9[8,9],ymm8[10,11],ymm9[12,13,14],ymm8[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,1,1]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -608,10 +608,10 @@ define void @store_i16_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm4
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm5
; AVX2-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <1,5,2,6,2,6,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [1,5,2,6,2,6,u,u]
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm7 = zero,zero,ymm7[2,3,6,7,2,3],zero,zero,zero,zero,ymm7[8,9,12,13,16,17],zero,zero,zero,zero,ymm7[18,19,22,23,28,29],zero,zero,zero,zero
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <5,2,6,u,2,6,3,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [5,2,6,u,2,6,3,7]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm8, %ymm8
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm8[2,3],zero,zero,zero,zero,zero,zero,ymm8[4,5,8,9],zero,zero,zero,zero,zero,zero,ymm8[18,19,22,23],zero,zero,zero,zero,zero,zero,ymm8[24,25,28,29]
; AVX2-FAST-NEXT: vpor %ymm7, %ymm8, %ymm7
@@ -656,7 +656,7 @@ define void @store_i16_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm6[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm9[u,u,u,u,6,7,10,11,u,u,u,u,u,u,8,9,u,u,u,u,22,23,26,27,u,u,u,u,u,u,24,25]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm8[0,1],ymm9[2,3],ymm8[4,5,6],ymm9[7],ymm8[8,9],ymm9[10,11],ymm8[12,13,14],ymm9[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,1,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -739,13 +739,13 @@ define void @store_i16_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm6
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm6[0,2,2,0]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm7 = zero,zero,zero,zero,ymm7[0,1,8,9,u,u],zero,zero,zero,zero,ymm7[2,3,18,19,u,u],zero,zero,zero,zero,ymm7[28,29,20,21,u,u],zero,zero
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <1,5,2,6,2,6,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [1,5,2,6,2,6,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm6, %ymm8, %ymm6
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm6 = zero,zero,ymm6[2,3,6,7,u,u],zero,zero,zero,zero,ymm6[8,9,12,13,u,u],zero,zero,zero,zero,ymm6[18,19,22,23,u,u],zero,zero,zero,zero
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm6, %zmm7, %zmm6
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm5[0,2,0,2]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm7 = ymm7[0,1,8,9],zero,zero,zero,zero,ymm7[u,u,2,3,10,11],zero,zero,zero,zero,ymm7[u,u,20,21,28,29],zero,zero,zero,zero,ymm7[u,u,22,23]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <5,2,6,u,2,6,3,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [5,2,6,u,2,6,3,7]
; AVX512F-FAST-NEXT: vpermd %ymm5, %ymm8, %ymm5
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm5 = ymm5[2,3],zero,zero,zero,zero,ymm5[u,u,4,5,8,9],zero,zero,zero,zero,ymm5[u,u,18,19,22,23],zero,zero,zero,zero,ymm5[u,u,24,25,28,29]
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm5, %zmm7, %zmm5
@@ -1136,7 +1136,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm6 = xmm6[0,2,1,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,4,4,5,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm6, %ymm5, %ymm5
; AVX2-SLOW-NEXT: vpbroadcastq (%r8), %ymm6
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255]
@@ -1150,7 +1150,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm11 = ymm11[0,1,2,3,4,4,4,4,8,9,10,11,12,12,12,12]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm10 = ymm11[0,1],ymm10[2],ymm11[3],ymm10[4],ymm11[5,6],ymm10[7],ymm11[8,9],ymm10[10],ymm11[11],ymm10[12],ymm11[13,14],ymm10[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm6, %ymm10, %ymm6
; AVX2-SLOW-NEXT: vpbroadcastq 16(%r8), %ymm10
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255]
@@ -1163,7 +1163,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm7 = xmm8[0],xmm7[1],xmm8[2],xmm7[3],xmm8[4,5],xmm7[6],xmm8[7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm9, %ymm7, %ymm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm1[0,1,1,1]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -1178,7 +1178,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm10 = ymm10[2,3,2,3,6,7,6,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0],ymm9[1],ymm10[2],ymm9[3,4],ymm10[5,6,7,8],ymm9[9],ymm10[10],ymm9[11,12],ymm10[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vpbroadcastq 24(%r8), %ymm9
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0]
@@ -1189,7 +1189,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[1,1,2,2]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2,3],ymm2[4],ymm0[5],ymm2[6],ymm0[7,8],ymm2[9],ymm0[10,11],ymm2[12],ymm0[13],ymm2[14],ymm0[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm3, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[1,1,2,2]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255]
@@ -1219,7 +1219,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm9 = xmm8[0],xmm7[0],xmm8[1],xmm7[1],xmm8[2],xmm7[2],xmm8[3],xmm7[3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm5, %ymm9, %ymm5
; AVX2-FAST-NEXT: vpbroadcastq (%r8), %ymm9
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255]
@@ -1232,7 +1232,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm7 = xmm8[0],xmm7[1],xmm8[2],xmm7[3],xmm8[4,5],xmm7[6],xmm8[7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,0]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm1[0,1,1,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -1245,7 +1245,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,18,19,u,u,20,21,u,u,24,25,24,25,u,u]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0,1],ymm8[2],ymm9[3],ymm8[4],ymm9[5,6],ymm8[7],ymm9[8,9],ymm8[10],ymm9[11],ymm8[12],ymm9[13,14],ymm8[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpbroadcastq 16(%r8), %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255]
@@ -1258,7 +1258,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,u,u,30,31,u,u,u,u,28,29,30,31,30,31]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0],ymm9[1],ymm10[2],ymm9[3,4],ymm10[5,6,7,8],ymm9[9],ymm10[10],ymm9[11,12],ymm10[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-NEXT: vpbroadcastq 24(%r8), %ymm9
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0]
@@ -1269,7 +1269,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[1,1,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2,3],ymm2[4],ymm0[5],ymm2[6],ymm0[7,8],ymm2[9],ymm0[10,11],ymm2[12],ymm0[13],ymm2[14],ymm0[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm3, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[1,1,2,2]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255]
@@ -1299,7 +1299,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm9 = xmm8[0],xmm7[0],xmm8[1],xmm7[1],xmm8[2],xmm7[2],xmm8[3],xmm7[3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm5, %ymm9, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq (%r8), %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255]
@@ -1312,7 +1312,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm7 = xmm8[0],xmm7[1],xmm8[2],xmm7[3],xmm8[4,5],xmm7[6],xmm8[7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,0]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm1[0,1,1,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -1325,7 +1325,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,18,19,u,u,20,21,u,u,24,25,24,25,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0,1],ymm8[2],ymm9[3],ymm8[4],ymm9[5,6],ymm8[7],ymm9[8,9],ymm8[10],ymm9[11],ymm8[12],ymm9[13,14],ymm8[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 16(%r8), %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255]
@@ -1338,7 +1338,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm10 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,u,u,30,31,u,u,u,u,28,29,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0],ymm9[1],ymm10[2],ymm9[3,4],ymm10[5,6,7,8],ymm9[9],ymm10[10],ymm9[11,12],ymm10[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 24(%r8), %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0]
@@ -1349,7 +1349,7 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2,3],ymm2[4],ymm0[5],ymm2[6],ymm0[7,8],ymm2[9],ymm0[10,11],ymm2[12],ymm0[13],ymm2[14],ymm0[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm3, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255]
@@ -1524,13 +1524,13 @@ define void @store_i16_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa (%r8), %ymm2
; AVX512BW-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,13,29,45,61,u,14,30,46,62,u,15,31,47,63,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,13,29,45,61,u,14,30,46,62,u,15,31,47,63,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,16,32,48,u,1,17,33,49,u,2,18,34,50,u,3,19,35,51,u,4,20,36,52,u,5,21,37,53,u,6,22>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,16,32,48,u,1,17,33,49,u,2,18,34,50,u,3,19,35,51,u,4,20,36,52,u,5,21,37,53,u,6,22]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm4
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,32,5,6,7,8,33,10,11,12,13,34,15,16,17,18,35,20,21,22,23,36,25,26,27,28,37,30,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm4, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <6,22,u,39,55,7,23,u,40,56,8,24,u,41,57,9,25,u,42,58,10,26,u,43,59,11,27,u,44,60,12,28>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [6,22,u,39,55,7,23,u,40,56,8,24,u,41,57,9,25,u,42,58,10,26,u,43,59,11,27,u,44,60,12,28]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm4
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,38,3,4,5,6,39,8,9,10,11,40,13,14,15,16,41,18,19,20,21,42,23,24,25,26,43,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm4, %zmm0
@@ -2274,7 +2274,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm11 = xmm11[0,2,1,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm11 = xmm11[0,1,2,3,4,4,5,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm11, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpbroadcastq (%r8), %ymm11
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255]
@@ -2291,17 +2291,17 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpbroadcastq 32(%r8), %ymm5
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm5, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, (%rsp) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-SLOW-NEXT: vpshufb %xmm11, %xmm12, %xmm0
; AVX2-SLOW-NEXT: vpbroadcastq 8(%rdi), %xmm12
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm12[1],xmm0[2,3],xmm12[4],xmm0[5],xmm12[6],xmm0[7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-SLOW-NEXT: vpshufb %xmm14, %xmm7, %xmm7
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm6 = xmm6[1,2,2,2]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm6 = xmm7[0],xmm6[1],xmm7[2],xmm6[3],xmm7[4,5],xmm6[6],xmm7[7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm0, %ymm6, %ymm0
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm1[0,1,1,1]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -2331,7 +2331,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm14 = ymm14[2,3,2,3,6,7,6,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm13 = ymm14[0],ymm13[1],ymm14[2],ymm13[3,4],ymm14[5,6,7,8],ymm13[9],ymm14[10],ymm13[11,12],ymm14[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,2,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm0, %ymm13, %ymm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm13 = ymm6[0,1,1,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm9, %ymm13, %ymm0
@@ -2367,7 +2367,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1],ymm6[2],ymm5[3],ymm6[4],ymm5[5,6],ymm6[7],ymm5[8,9],ymm6[10],ymm5[11],ymm6[12],ymm5[13,14],ymm6[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm1, %ymm5, %ymm1
; AVX2-SLOW-NEXT: vpshufb %ymm0, %ymm13, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm15[1,1,1,2,5,5,5,6]
@@ -2385,15 +2385,15 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpbroadcastq 16(%r8), %ymm1
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm4, %ymm1, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm4, %ymm11, %ymm5
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm6 = ymm10[3,0,3,0,7,4,7,4]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2],ymm5[3],ymm6[4,5],ymm5[6],ymm6[7,8],ymm5[9],ymm6[10],ymm5[11],ymm6[12,13],ymm5[14],ymm6[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm7[1,1,2,2]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2,3],ymm3[4],ymm2[5],ymm3[6],ymm2[7,8],ymm3[9],ymm2[10,11],ymm3[12],ymm2[13],ymm3[14],ymm2[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm5, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vpshufb %ymm4, %ymm13, %ymm4
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm15[3,0,3,0,7,4,7,4]
@@ -2439,20 +2439,20 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm0
; AVX2-FAST-NEXT: vmovdqa 32(%rsi), %xmm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm0, %xmm6
; AVX2-FAST-NEXT: vpbroadcastq 8(%rdi), %xmm9
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0],xmm9[1],xmm6[2,3],xmm9[4],xmm6[5],xmm9[6],xmm6[7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
; AVX2-FAST-NEXT: vmovdqa (%rcx), %xmm13
; AVX2-FAST-NEXT: vmovdqa 32(%rcx), %xmm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm13, %xmm9
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm14
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm12 = xmm14[1,2,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0],xmm12[1],xmm9[2],xmm12[3],xmm9[4,5],xmm12[6],xmm9[7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm12 = ymm9[0,1,0,0]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm6, %ymm12, %ymm12
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm1[0,1,1,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -2477,7 +2477,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm15, %xmm13, %xmm9
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm0, %ymm9, %ymm0
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm2[0,1,1,1]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm7, %ymm9, %ymm1
@@ -2509,7 +2509,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm13 = ymm9[3,2,3,3,7,6,7,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm11 = ymm11[0],ymm13[1],ymm11[2],ymm13[3,4],ymm11[5,6,7,8],ymm13[9],ymm11[10],ymm13[11,12],ymm11[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,2,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm1, %ymm11, %ymm1
; AVX2-FAST-NEXT: vmovdqa (%rsi), %ymm11
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm11, %ymm0
@@ -2540,7 +2540,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1],ymm7[2],ymm0[3],ymm7[4],ymm0[5,6],ymm7[7],ymm0[8,9],ymm7[10],ymm0[11],ymm7[12],ymm0[13,14],ymm7[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm14, %ymm1
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm15[1,1,1,2,5,5,5,6]
@@ -2556,15 +2556,15 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vpbroadcastq 16(%r8), %ymm2
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm10, %ymm6
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm7 = ymm9[3,0,3,0,7,4,7,4]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0],ymm6[1],ymm7[2],ymm6[3],ymm7[4,5],ymm6[6],ymm7[7,8],ymm6[9],ymm7[10],ymm6[11],ymm7[12,13],ymm6[14],ymm7[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm4, %ymm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[1,1,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0],ymm4[1],ymm3[2,3],ymm4[4],ymm3[5],ymm4[6],ymm3[7,8],ymm4[9],ymm3[10,11],ymm4[12],ymm3[13],ymm4[14],ymm3[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm6, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm14, %ymm2
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm6 = ymm15[3,0,3,0,7,4,7,4]
@@ -2609,20 +2609,20 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsi), %xmm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rsi), %xmm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm7, %xmm0, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 8(%rdi), %xmm9
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm6 = xmm6[0],xmm9[1],xmm6[2,3],xmm9[4],xmm6[5],xmm9[6],xmm6[7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rcx), %xmm13
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rcx), %xmm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm13, %xmm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdx), %xmm14
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm12 = xmm14[1,2,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm9 = xmm9[0],xmm12[1],xmm9[2],xmm12[3],xmm9[4,5],xmm12[6],xmm9[7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm9[0,1,0,0]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm6, %ymm12, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm1[0,1,1,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
@@ -2647,7 +2647,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm15 = [4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm15, %xmm13, %xmm9
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm0, %ymm9, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm2[0,1,1,1]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm7, %ymm9, %ymm1
@@ -2679,7 +2679,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm13 = ymm9[3,2,3,3,7,6,7,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm11 = ymm11[0],ymm13[1],ymm11[2],ymm13[3,4],ymm11[5,6,7,8],ymm13[9],ymm11[10],ymm13[11,12],ymm11[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,2,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm1, %ymm11, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsi), %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm11, %ymm0
@@ -2710,7 +2710,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1],ymm7[2],ymm0[3],ymm7[4],ymm0[5,6],ymm7[7],ymm0[8,9],ymm7[10],ymm0[11],ymm7[12],ymm0[13,14],ymm7[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm14, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm2 = ymm15[1,1,1,2,5,5,5,6]
@@ -2726,15 +2726,15 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm0, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 16(%r8), %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm1, %ymm2, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm10, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm7 = ymm9[3,0,3,0,7,4,7,4]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0],ymm6[1],ymm7[2],ymm6[3],ymm7[4,5],ymm6[6],ymm7[7,8],ymm6[9],ymm7[10],ymm6[11],ymm7[12,13],ymm6[14],ymm7[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm4, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0],ymm4[1],ymm3[2,3],ymm4[4],ymm3[5],ymm4[6],ymm3[7,8],ymm4[9],ymm3[10,11],ymm4[12],ymm3[13],ymm4[14],ymm3[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm6, %ymm3, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm14, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm6 = ymm15[3,0,3,0,7,4,7,4]
@@ -2806,25 +2806,25 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm2, %zmm0, %zmm19
; AVX512F-SLOW-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm3, %zmm19
; AVX512F-SLOW-NEXT: vmovdqa 32(%rsi), %ymm3
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm2, %ymm3, %ymm0
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm2, %ymm22
; AVX512F-SLOW-NEXT: vmovdqa64 32(%rdi), %ymm18
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm18[1,1,2,2]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm7[0],ymm0[1],ymm7[2,3],ymm0[4],ymm7[5],ymm0[6],ymm7[7,8],ymm0[9],ymm7[10,11],ymm0[12],ymm7[13],ymm0[14],ymm7[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX512F-SLOW-NEXT: vpshufb %xmm7, %xmm8, %xmm8
; AVX512F-SLOW-NEXT: vpbroadcastq 40(%rdi), %xmm11
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm8 = xmm8[0],xmm11[1],xmm8[2,3],xmm11[4],xmm8[5],xmm11[6],xmm8[7]
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,1,0,1]
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm0, %zmm8, %zmm11
; AVX512F-SLOW-NEXT: vmovdqa 32(%rcx), %ymm0
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm4, %ymm0, %ymm2
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm4, %ymm23
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm8 = ymm6[3,0,3,0,7,4,7,4]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm8[0],ymm2[1],ymm8[2],ymm2[3],ymm8[4,5],ymm2[6],ymm8[7,8],ymm2[9],ymm8[10],ymm2[11],ymm8[12,13],ymm2[14],ymm8[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX512F-SLOW-NEXT: vpshufb %xmm8, %xmm10, %xmm10
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm9 = xmm9[1,2,2,2]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm9 = xmm10[0],xmm9[1],xmm10[2],xmm9[3],xmm10[4,5],xmm9[6],xmm10[7]
@@ -2926,7 +2926,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-LABEL: store_i16_stride5_vf32:
; AVX512F-FAST: # %bb.0:
; AVX512F-FAST-NEXT: vmovdqa 32(%rsi), %ymm5
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm1, %ymm5, %ymm0
; AVX512F-FAST-NEXT: vmovdqa64 %ymm1, %ymm17
; AVX512F-FAST-NEXT: vmovdqa 32(%rdi), %ymm7
@@ -2934,7 +2934,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2,3],ymm0[4],ymm1[5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10,11],ymm0[12],ymm1[13],ymm0[14],ymm1[15]
; AVX512F-FAST-NEXT: vmovdqa (%rsi), %xmm10
; AVX512F-FAST-NEXT: vmovdqa 32(%rsi), %xmm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX512F-FAST-NEXT: vpshufb %xmm12, %xmm2, %xmm1
; AVX512F-FAST-NEXT: vmovdqa64 %xmm2, %xmm20
; AVX512F-FAST-NEXT: vpbroadcastq 40(%rdi), %xmm2
@@ -2943,7 +2943,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512F-FAST-NEXT: vmovdqa (%rcx), %ymm1
; AVX512F-FAST-NEXT: vmovdqa 32(%rcx), %ymm8
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm8, %ymm2
; AVX512F-FAST-NEXT: vmovdqa64 %ymm3, %ymm24
; AVX512F-FAST-NEXT: vmovdqa64 (%rdx), %ymm18
@@ -2951,7 +2951,7 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm4 = ymm9[3,0,3,0,7,4,7,4]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4,5],ymm2[6],ymm4[7,8],ymm2[9],ymm4[10],ymm2[11],ymm4[12,13],ymm2[14],ymm4[15]
; AVX512F-FAST-NEXT: vmovdqa 32(%rcx), %xmm11
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX512F-FAST-NEXT: vpshufb %xmm4, %xmm11, %xmm13
; AVX512F-FAST-NEXT: vmovdqa 32(%rdx), %xmm6
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} xmm15 = xmm6[1,2,2,2]
@@ -3090,43 +3090,43 @@ define void @store_i16_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm3
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u,6,38>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u,6,38]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: movl $415641996, %eax # imm = 0x18C6318C
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm5, %zmm6 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,32,5,6,7,8,33,10,11,12,13,34,15,16,17,18,35,20,21,22,23,36,25,26,27,28,37,30,31]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm6, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <6,38,u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [6,38,u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm7
; AVX512BW-NEXT: movl $831283992, %eax # imm = 0x318C6318
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqu16 %zmm6, %zmm7 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,38,3,4,5,6,39,8,9,10,11,40,13,14,15,16,41,18,19,20,21,42,23,24,25,26,43,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm7, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u,u,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u,u,19]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm8
; AVX512BW-NEXT: vmovdqu16 %zmm7, %zmm8 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [44,1,2,3,4,45,6,7,8,9,46,11,12,13,14,47,16,17,18,19,48,21,22,23,24,49,26,27,28,29,50,31]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm8, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,19,51,u,u,u,20,52,u,u,u,21,53,u,u,u,22,54,u,u,u,23,55,u,u,u,24,56,u,u,u,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,19,51,u,u,u,20,52,u,u,u,21,53,u,u,u,22,54,u,u,u,23,55,u,u,u,24,56,u,u,u,25]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <19,u,u,u,52,20,u,u,u,53,21,u,u,u,54,22,u,u,u,55,23,u,u,u,56,24,u,u,u,57,25,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [19,u,u,u,52,20,u,u,u,53,21,u,u,u,54,22,u,u,u,55,23,u,u,u,56,24,u,u,u,57,25,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm9
; AVX512BW-NEXT: movl $-1939662650, %eax # imm = 0x8C6318C6
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqu16 %zmm8, %zmm9 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,51,4,5,6,7,52,9,10,11,12,53,14,15,16,17,54,19,20,21,22,55,24,25,26,27,56,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm4, %zmm9, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,26,58,u,u,u,27,59,u,u,u,28,60,u,u,u,29,61,u,u,u,30,62,u,u,u,31,63,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,26,58,u,u,u,27,59,u,u,u,28,60,u,u,u,29,61,u,u,u,30,62,u,u,u,31,63,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <25,u,u,u,58,26,u,u,u,59,27,u,u,u,60,28,u,u,u,61,29,u,u,u,62,30,u,u,u,63,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [25,u,u,u,58,26,u,u,u,59,27,u,u,u,60,28,u,u,u,61,29,u,u,u,62,30,u,u,u,63,31,u]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm0
; AVX512BW-NEXT: vmovdqu16 %zmm9, %zmm0 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,57,2,3,4,5,58,7,8,9,10,59,12,13,14,15,60,17,18,19,20,61,22,23,24,25,62,27,28,29,30,63]
@@ -4602,7 +4602,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[0,2,1,3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,4,4,5,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm7, %ymm1, %ymm8
; AVX2-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm0[0],xmm12[0],xmm0[1],xmm12[1],xmm0[2],xmm12[2],xmm0[3],xmm12[3]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm1, %xmm1
@@ -4645,17 +4645,17 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpbroadcastq 96(%r8), %ymm5
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm8 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm11, %xmm4
; AVX2-SLOW-NEXT: vpbroadcastq 8(%rdi), %xmm5
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0],xmm5[1],xmm4[2,3],xmm5[4],xmm4[5],xmm5[6],xmm4[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-SLOW-NEXT: vpshufb %xmm14, %xmm9, %xmm5
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm3[1,2,2,2]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm5[0],xmm3[1],xmm5[2],xmm3[3],xmm5[4,5],xmm3[6],xmm5[7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,0,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,1,0,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm4, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm10, %xmm4
; AVX2-SLOW-NEXT: vpbroadcastq 40(%rdi), %xmm5
@@ -4726,7 +4726,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3],ymm3[4],ymm2[5,6],ymm3[7],ymm2[8,9],ymm3[10],ymm2[11],ymm3[12],ymm2[13,14],ymm3[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm1, %ymm2, %ymm10
; AVX2-SLOW-NEXT: vmovdqa 32(%rdx), %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -4807,7 +4807,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0],ymm5[1],ymm1[2],ymm5[3,4],ymm1[5,6,7,8],ymm5[9],ymm1[10],ymm5[11,12],ymm1[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,3,2,2]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm0 = ymm13[0,1,2,3,7,6,5,7,8,9,10,11,15,14,13,15]
@@ -4862,17 +4862,17 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpbroadcastq 120(%r8), %ymm0
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm5, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-SLOW-NEXT: vpshufb %ymm0, %ymm10, %ymm1
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm7[3,0,3,0,7,4,7,4]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm5[0],ymm1[1],ymm5[2],ymm1[3],ymm5[4,5],ymm1[6],ymm5[7,8],ymm1[9],ymm5[10],ymm1[11],ymm5[12,13],ymm1[14],ymm5[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm7, %ymm9
; AVX2-SLOW-NEXT: vpermq $165, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm14 = mem[1,1,2,2]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm9 = ymm14[0],ymm9[1],ymm14[2,3],ymm9[4],ymm14[5],ymm9[6],ymm14[7,8],ymm9[9],ymm14[10,11],ymm9[12],ymm14[13],ymm9[14],ymm14[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm1, %ymm9, %ymm1
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm0, %ymm7, %ymm9
@@ -4978,7 +4978,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm12, %xmm12
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm12[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm0, %ymm15, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm14[0],xmm2[0],xmm14[1],xmm2[1],xmm14[2],xmm2[2],xmm14[3],xmm2[3]
@@ -5019,17 +5019,17 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpbroadcastq 96(%r8), %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm5, %xmm0
; AVX2-FAST-NEXT: vpbroadcastq 8(%rdi), %xmm1
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4],xmm0[5],xmm1[6],xmm0[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm8, %xmm5
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[1,2,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm5 = xmm5[0],xmm7[1],xmm5[2],xmm7[3],xmm5[4,5],xmm7[6],xmm5[7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,0]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm1, %ymm5, %ymm5
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm2, %xmm1
; AVX2-FAST-NEXT: vpbroadcastq 40(%rdi), %xmm2
@@ -5101,7 +5101,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0,1],ymm4[2],ymm3[3],ymm4[4],ymm3[5,6],ymm4[7],ymm3[8,9],ymm4[10],ymm3[11],ymm4[12],ymm3[13,14],ymm4[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm2, %ymm3, %ymm8
; AVX2-FAST-NEXT: vmovdqa 32(%rdx), %ymm4
; AVX2-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5179,7 +5179,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm13 = ymm13[0],ymm14[1],ymm13[2],ymm14[3,4],ymm13[5,6,7,8],ymm14[9],ymm13[10],ymm14[11,12],ymm13[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,3,2,2]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,2,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm10, %ymm13, %ymm10
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm6, %ymm13
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
@@ -5228,19 +5228,19 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpbroadcastq 120(%r8), %ymm9
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm0, %ymm9, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm7, %ymm11
; AVX2-FAST-NEXT: vpshufd $51, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm14 = mem[3,0,3,0,7,4,7,4]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2],ymm11[3],ymm14[4,5],ymm11[6],ymm14[7,8],ymm11[9],ymm14[10],ymm11[11],ymm14[12,13],ymm11[14],ymm14[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm2, %ymm15
; AVX2-FAST-NEXT: vpermq $165, {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Folded Reload
; AVX2-FAST-NEXT: # ymm13 = mem[1,1,2,2]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm13 = ymm13[0],ymm15[1],ymm13[2,3],ymm15[4],ymm13[5],ymm15[6],ymm13[7,8],ymm15[9],ymm13[10,11],ymm15[12],ymm13[13],ymm15[14],ymm13[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm11, %ymm13, %ymm11
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm4, %ymm13
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm12 = ymm1[3,0,3,0,7,4,7,4]
@@ -5346,7 +5346,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [4,5,2,3,2,3,0,1,10,11,8,9,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm12, %xmm12
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm12[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm0, %ymm15, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm14[0],xmm2[0],xmm14[1],xmm2[1],xmm14[2],xmm2[2],xmm14[3],xmm2[3]
@@ -5387,17 +5387,17 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 96(%r8), %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm5, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 8(%rdi), %xmm1
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4],xmm0[5],xmm1[6],xmm0[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm8, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[1,2,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm5 = xmm5[0],xmm7[1],xmm5[2],xmm7[3],xmm5[4,5],xmm7[6],xmm5[7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,0]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm1, %ymm5, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm2, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 40(%rdi), %xmm2
@@ -5469,7 +5469,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0,1],ymm4[2],ymm3[3],ymm4[4],ymm3[5,6],ymm4[7],ymm3[8,9],ymm4[10],ymm3[11],ymm4[12],ymm3[13,14],ymm4[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm2, %ymm3, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdx), %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5547,7 +5547,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm13 = ymm13[0],ymm14[1],ymm13[2],ymm14[3,4],ymm13[5,6,7,8],ymm14[9],ymm13[10],ymm14[11,12],ymm13[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,3,2,2]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,2,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm10, %ymm13, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm6, %ymm13
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
@@ -5596,19 +5596,19 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq 120(%r8), %ymm9
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm0, %ymm9, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm7, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufd $51, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm14 = mem[3,0,3,0,7,4,7,4]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2],ymm11[3],ymm14[4,5],ymm11[6],ymm14[7,8],ymm11[9],ymm14[10],ymm11[11],ymm14[12,13],ymm11[14],ymm14[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm2, %ymm15
; AVX2-FAST-PERLANE-NEXT: vpermq $165, {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm13 = mem[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm13 = ymm13[0],ymm15[1],ymm13[2,3],ymm15[4],ymm13[5],ymm15[6],ymm13[7,8],ymm15[9],ymm13[10,11],ymm15[12],ymm13[13],ymm15[14],ymm13[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255,255,255,u,u,0,0,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm11, %ymm13, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm4, %ymm13
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm12 = ymm1[3,0,3,0,7,4,7,4]
@@ -5693,14 +5693,14 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW: # %bb.0:
; AVX512F-SLOW-NEXT: subq $488, %rsp # imm = 0x1E8
; AVX512F-SLOW-NEXT: vmovdqa 96(%rcx), %ymm11
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm1, %ymm11, %ymm0
; AVX512F-SLOW-NEXT: vmovdqa %ymm1, %ymm14
; AVX512F-SLOW-NEXT: vmovdqa64 96(%rdx), %ymm17
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm17[3,0,3,0,7,4,7,4]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10],ymm0[11],ymm1[12,13],ymm0[14],ymm1[15]
; AVX512F-SLOW-NEXT: vmovdqa 96(%rcx), %xmm0
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX512F-SLOW-NEXT: vpshufb %xmm7, %xmm0, %xmm3
; AVX512F-SLOW-NEXT: vmovdqa 96(%rdx), %xmm2
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm4 = xmm2[1,2,2,2]
@@ -5709,14 +5709,14 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm3, %zmm1
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 96(%rsi), %ymm9
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm3, %ymm9, %ymm1
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm3, %ymm20
; AVX512F-SLOW-NEXT: vmovdqa 96(%rdi), %ymm5
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm5[1,1,2,2]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0],ymm1[1],ymm3[2,3],ymm1[4],ymm3[5],ymm1[6],ymm3[7,8],ymm1[9],ymm3[10,11],ymm1[12],ymm3[13],ymm1[14],ymm3[15]
; AVX512F-SLOW-NEXT: vmovdqa 96(%rsi), %xmm1
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX512F-SLOW-NEXT: vpshufb %xmm12, %xmm1, %xmm4
; AVX512F-SLOW-NEXT: vpbroadcastq 104(%rdi), %xmm6
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0],xmm6[1],xmm4[2,3],xmm6[4],xmm4[5],xmm6[6],xmm4[7]
@@ -5889,7 +5889,7 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0,1],ymm3[2],ymm1[3],ymm3[4],ymm1[5,6],ymm3[7],ymm1[8,9],ymm3[10],ymm1[11],ymm3[12],ymm1[13,14],ymm3[15]
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm30
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm4, %ymm2, %ymm0
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm16[3,0,3,0,7,4,7,4]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10],ymm0[11],ymm1[12,13],ymm0[14],ymm1[15]
@@ -6019,14 +6019,14 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST: # %bb.0:
; AVX512F-FAST-NEXT: subq $360, %rsp # imm = 0x168
; AVX512F-FAST-NEXT: vmovdqa 96(%rcx), %ymm8
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,12,13,u,u,0,1,u,u,u,u,14,15,u,u,u,u,28,29,u,u,16,17,u,u,u,u,30,31,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm15, %ymm8, %ymm0
; AVX512F-FAST-NEXT: vmovdqa64 96(%rdx), %ymm19
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm1 = ymm19[3,0,3,0,7,4,7,4]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4,5],ymm0[6],ymm1[7,8],ymm0[9],ymm1[10],ymm0[11],ymm1[12,13],ymm0[14],ymm1[15]
; AVX512F-FAST-NEXT: vmovdqa 96(%rcx), %xmm2
; AVX512F-FAST-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [10,11,u,u,6,7,u,u,8,9,8,9,u,u,8,9]
; AVX512F-FAST-NEXT: vpshufb %xmm0, %xmm2, %xmm2
; AVX512F-FAST-NEXT: vmovdqa 96(%rdx), %xmm3
; AVX512F-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -6036,14 +6036,14 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm2, %zmm1
; AVX512F-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 96(%rsi), %ymm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,0,1,u,u,u,u,14,15,u,u,2,3,u,u,u,u,16,17,u,u,u,u,30,31,u,u,18,19,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm12, %ymm3, %ymm1
; AVX512F-FAST-NEXT: vmovdqa64 96(%rdi), %ymm24
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm24[1,1,2,2]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5],ymm1[6],ymm2[7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13],ymm1[14],ymm2[15]
; AVX512F-FAST-NEXT: vmovdqa 96(%rsi), %xmm4
; AVX512F-FAST-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [6,7,u,u,10,11,6,7,u,u,8,9,u,u,12,13]
; AVX512F-FAST-NEXT: vpshufb %xmm1, %xmm4, %xmm4
; AVX512F-FAST-NEXT: vpbroadcastq 104(%rdi), %xmm5
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0],xmm5[1],xmm4[2,3],xmm5[4],xmm4[5],xmm5[6],xmm4[7]
@@ -6331,10 +6331,10 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm15
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm5
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,u,0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,u,0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm7
; AVX512BW-NEXT: vpermt2w %zmm6, %zmm16, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm17 = <0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u,6,38>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm17 = [0,32,u,u,u,1,33,u,u,u,2,34,u,u,u,3,35,u,u,u,4,36,u,u,u,5,37,u,u,u,6,38]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm3
; AVX512BW-NEXT: vpermt2w %zmm4, %zmm17, %zmm3
; AVX512BW-NEXT: movl $415641996, %eax # imm = 0x18C6318C
@@ -6342,19 +6342,19 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu16 %zmm7, %zmm3 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,1,2,3,32,5,6,7,8,33,10,11,12,13,34,15,16,17,18,35,20,21,22,23,36,25,26,27,28,37,30,31]
; AVX512BW-NEXT: vpermt2w %zmm5, %zmm18, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <u,u,26,58,u,u,u,27,59,u,u,u,28,60,u,u,u,29,61,u,u,u,30,62,u,u,u,31,63,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [u,u,26,58,u,u,u,27,59,u,u,u,28,60,u,u,u,29,61,u,u,u,30,62,u,u,u,31,63,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm19, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <25,u,u,u,58,26,u,u,u,59,27,u,u,u,60,28,u,u,u,61,29,u,u,u,62,30,u,u,u,63,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [25,u,u,u,58,26,u,u,u,59,27,u,u,u,60,28,u,u,u,61,29,u,u,u,62,30,u,u,u,63,31,u]
; AVX512BW-NEXT: vmovdqa64 %zmm15, %zmm7
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm8, %zmm7
; AVX512BW-NEXT: vmovdqu16 %zmm9, %zmm7 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,57,2,3,4,5,58,7,8,9,10,59,12,13,14,15,60,17,18,19,20,61,22,23,24,25,62,27,28,29,30,63]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm20, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,19,51,u,u,u,20,52,u,u,u,21,53,u,u,u,22,54,u,u,u,23,55,u,u,u,24,56,u,u,u,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,19,51,u,u,u,20,52,u,u,u,21,53,u,u,u,22,54,u,u,u,23,55,u,u,u,24,56,u,u,u,25]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm15, %zmm21, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <19,u,u,u,52,20,u,u,u,53,21,u,u,u,54,22,u,u,u,55,23,u,u,u,56,24,u,u,u,57,25,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [19,u,u,u,52,20,u,u,u,53,21,u,u,u,54,22,u,u,u,55,23,u,u,u,56,24,u,u,u,57,25,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm10, %zmm9
; AVX512BW-NEXT: movl $-1939662650, %eax # imm = 0x8C6318C6
@@ -6362,10 +6362,10 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu16 %zmm14, %zmm9 {%k3}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,2,51,4,5,6,7,52,9,10,11,12,53,14,15,16,17,54,19,20,21,22,55,24,25,26,27,56,29,30,31]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm22, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm23 = <u,u,u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [u,u,u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm24
; AVX512BW-NEXT: vpermt2w %zmm15, %zmm23, %zmm24
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = <u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u,u,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [u,13,45,u,u,u,14,46,u,u,u,15,47,u,u,u,16,48,u,u,u,17,49,u,u,u,18,50,u,u,u,19]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm25, %zmm14
; AVX512BW-NEXT: movl $831283992, %eax # imm = 0x318C6318
@@ -6373,10 +6373,10 @@ define void @store_i16_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu16 %zmm24, %zmm14 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [44,1,2,3,4,45,6,7,8,9,46,11,12,13,14,47,16,17,18,19,48,21,22,23,24,49,26,27,28,29,50,31]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm24, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm27
; AVX512BW-NEXT: vpermt2w %zmm11, %zmm26, %zmm27
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <6,38,u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [6,38,u,u,u,7,39,u,u,u,8,40,u,u,u,9,41,u,u,u,10,42,u,u,u,11,43,u,u,u,12,44]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm29
; AVX512BW-NEXT: vpermt2w %zmm15, %zmm28, %zmm29
; AVX512BW-NEXT: vmovdqu16 %zmm27, %zmm29 {%k2}
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll
index b2c0e00825e63e0..962c7d942e0dcd1 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-6.ll
@@ -185,7 +185,7 @@ define void @store_i16_stride6_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX512BW-NEXT: vpunpckldq {{.*#+}} xmm1 = xmm2[0],mem[0],xmm2[1],mem[1]
; AVX512BW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,6,8,10,1,3,5,7,9,11,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,6,8,10,1,3,5,7,9,11,u,u,u,u]
; AVX512BW-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX512BW-NEXT: vmovq %xmm1, 16(%rax)
@@ -458,7 +458,7 @@ define void @store_i16_stride6_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vpunpcklqdq {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512BW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512BW-NEXT: vinserti32x4 $2, %xmm2, %zmm0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,4,8,12,16,20,1,5,9,13,17,21,2,6,10,14,18,22,3,7,11,15,19,23,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,4,8,12,16,20,1,5,9,13,17,21,2,6,10,14,18,22,3,7,11,15,19,23,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vextracti32x4 $2, %zmm0, 32(%rax)
; AVX512BW-NEXT: vmovdqa %ymm0, (%rax)
@@ -1334,7 +1334,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm5, %ymm12, %ymm5
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm1, %ymm14
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm13, %ymm12
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} ymm14 = ymm12[0],ymm14[0],ymm12[1],ymm14[1],ymm12[2],ymm14[2],ymm12[3],ymm14[3],ymm12[8],ymm14[8],ymm12[9],ymm14[9],ymm12[10],ymm14[10],ymm12[11],ymm14[11]
@@ -1515,7 +1515,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,2,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm6, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm11, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm9, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} ymm3 = ymm3[0],ymm5[0],ymm3[1],ymm5[1],ymm3[2],ymm5[2],ymm3[3],ymm5[3],ymm3[8],ymm5[8],ymm3[9],ymm5[9],ymm3[10],ymm5[10],ymm3[11],ymm5[11]
@@ -1562,7 +1562,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm8 = ymm0[2,1,2,3,6,5,6,7]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} ymm8 = ymm8[0,3,2,1,4,5,6,7,8,11,10,9,12,13,14,15]
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm7 = ymm8[0],ymm7[0],ymm8[1],ymm7[1],ymm8[2],ymm7[2],ymm8[3],ymm7[3],ymm8[8],ymm7[8],ymm8[9],ymm7[9],ymm8[10],ymm7[10],ymm8[11],ymm7[11]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,5,12,u,4,13,u,7>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,5,12,u,4,13,u,7]
; AVX512F-SLOW-NEXT: vpermi2d %ymm6, %ymm7, %ymm8
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm6 = ymm3[1,2,2,3,5,6,6,7]
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm7 = ymm1[1,2,2,3,5,6,6,7]
@@ -1571,7 +1571,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0],ymm8[1,2],ymm6[3],ymm8[4,5],ymm6[6],ymm8[7]
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} ymm7 = ymm0[4],ymm2[4],ymm0[5],ymm2[5],ymm0[6],ymm2[6],ymm0[7],ymm2[7],ymm0[12],ymm2[12],ymm0[13],ymm2[13],ymm0[14],ymm2[14],ymm0[15],ymm2[15]
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} ymm8 = ymm4[4],ymm5[4],ymm4[5],ymm5[5],ymm4[6],ymm5[6],ymm4[7],ymm5[7],ymm4[12],ymm5[12],ymm4[13],ymm5[13],ymm4[14],ymm5[14],ymm4[15],ymm5[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <5,u,14,6,u,15,7,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [5,u,14,6,u,15,7,u]
; AVX512F-SLOW-NEXT: vpermi2d %ymm7, %ymm8, %ymm9
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm9, %zmm0, %zmm7
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [8,21,10,11,22,13,14,23]
@@ -1584,7 +1584,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa (%rsi), %xmm8
; AVX512F-SLOW-NEXT: vmovdqa (%rdi), %xmm10
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm10[0],xmm8[0],xmm10[1],xmm8[1],xmm10[2],xmm8[2],xmm10[3],xmm8[3]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <0,8,u,1,9,u,2,10>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [0,8,u,1,9,u,2,10]
; AVX512F-SLOW-NEXT: vpermi2d %ymm9, %ymm11, %ymm13
; AVX512F-SLOW-NEXT: vmovdqa (%r9), %xmm9
; AVX512F-SLOW-NEXT: vmovdqa (%r8), %xmm11
@@ -1599,7 +1599,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm10[0,1,2,1]
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm12 = xmm12[0,1,2,3,4,7,6,5]
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm12 = xmm12[4],xmm15[4],xmm12[5],xmm15[5],xmm12[6],xmm15[6],xmm12[7],xmm15[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <u,1,8,u,0,9,u,3>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [u,1,8,u,0,9,u,3]
; AVX512F-SLOW-NEXT: vpermi2d %ymm14, %ymm12, %ymm15
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm9[1,2,2,3]
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm14 = xmm11[1,2,2,3]
@@ -1609,14 +1609,14 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm12, %zmm13, %zmm12
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm4 = ymm4[0],ymm5[0],ymm4[1],ymm5[1],ymm4[2],ymm5[2],ymm4[3],ymm5[3],ymm4[8],ymm5[8],ymm4[9],ymm5[9],ymm4[10],ymm5[10],ymm4[11],ymm5[11]
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm0[0],ymm2[0],ymm0[1],ymm2[1],ymm0[2],ymm2[2],ymm0[3],ymm2[3],ymm0[8],ymm2[8],ymm0[9],ymm2[9],ymm0[10],ymm2[10],ymm0[11],ymm2[11]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <4,12,u,5,13,u,6,14>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [4,12,u,5,13,u,6,14]
; AVX512F-SLOW-NEXT: vpermi2d %ymm4, %ymm0, %ymm2
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[8],ymm3[8],ymm1[9],ymm3[9],ymm1[10],ymm3[10],ymm1[11],ymm3[11]
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,2]
; AVX512F-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1],ymm0[2],ymm2[3,4],ymm0[5],ymm2[6,7]
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm10[4],xmm8[4],xmm10[5],xmm8[5],xmm10[6],xmm8[6],xmm10[7],xmm8[7]
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm7[4],xmm6[4],xmm7[5],xmm6[5],xmm7[6],xmm6[6],xmm7[7],xmm6[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <1,u,10,2,u,11,3,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [1,u,10,2,u,11,3,u]
; AVX512F-SLOW-NEXT: vpermi2d %ymm1, %ymm2, %ymm3
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm11[4],xmm9[4],xmm11[5],xmm9[5],xmm11[6],xmm9[6],xmm11[7],xmm9[7]
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,9,2,3,10,5,6,11]
@@ -1643,7 +1643,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa (%rsi), %xmm8
; AVX512F-FAST-NEXT: vmovdqa (%rdi), %xmm10
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm10[0],xmm8[0],xmm10[1],xmm8[1],xmm10[2],xmm8[2],xmm10[3],xmm8[3]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,8,u,1,9,u,2,10>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,8,u,1,9,u,2,10]
; AVX512F-FAST-NEXT: vpermi2d %ymm9, %ymm11, %ymm12
; AVX512F-FAST-NEXT: vmovdqa (%r9), %xmm9
; AVX512F-FAST-NEXT: vmovdqa (%r8), %xmm11
@@ -1657,7 +1657,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpsrldq {{.*#+}} xmm13 = xmm6[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpsrldq {{.*#+}} xmm15 = xmm7[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} xmm13 = xmm15[0],xmm13[0],xmm15[1],xmm13[1],xmm15[2],xmm13[2],xmm15[3],xmm13[3]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <u,1,8,u,0,9,u,3>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [u,1,8,u,0,9,u,3]
; AVX512F-FAST-NEXT: vpermi2d %ymm13, %ymm12, %ymm15
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm15, %zmm0, %zmm12
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [16,9,10,17,12,13,18,15]
@@ -1668,14 +1668,14 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm13, %zmm14, %zmm0
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} xmm8 = xmm10[4],xmm8[4],xmm10[5],xmm8[5],xmm10[6],xmm8[6],xmm10[7],xmm8[7]
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} xmm6 = xmm7[4],xmm6[4],xmm7[5],xmm6[5],xmm7[6],xmm6[6],xmm7[7],xmm6[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <1,u,10,2,u,11,3,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [1,u,10,2,u,11,3,u]
; AVX512F-FAST-NEXT: vpermi2d %ymm8, %ymm6, %ymm7
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} xmm6 = xmm11[4],xmm9[4],xmm11[5],xmm9[5],xmm11[6],xmm9[6],xmm11[7],xmm9[7]
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,9,2,3,10,5,6,11]
; AVX512F-FAST-NEXT: vpermi2d %ymm6, %ymm7, %ymm8
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm6 = ymm4[0],ymm5[0],ymm4[1],ymm5[1],ymm4[2],ymm5[2],ymm4[3],ymm5[3],ymm4[8],ymm5[8],ymm4[9],ymm5[9],ymm4[10],ymm5[10],ymm4[11],ymm5[11]
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm7 = ymm2[0],ymm3[0],ymm2[1],ymm3[1],ymm2[2],ymm3[2],ymm2[3],ymm3[3],ymm2[8],ymm3[8],ymm2[9],ymm3[9],ymm2[10],ymm3[10],ymm2[11],ymm3[11]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <4,12,u,5,13,u,6,14>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [4,12,u,5,13,u,6,14]
; AVX512F-FAST-NEXT: vpermi2d %ymm6, %ymm7, %ymm9
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm9, %zmm0, %zmm6
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [8,9,20,11,12,21,14,15]
@@ -1683,14 +1683,14 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm9 = ymm10[0],ymm1[0],ymm10[1],ymm1[1],ymm10[2],ymm1[2],ymm10[3],ymm1[3],ymm10[8],ymm1[8],ymm10[9],ymm1[9],ymm10[10],ymm1[10],ymm10[11],ymm1[11]
; AVX512F-FAST-NEXT: vpermi2d %zmm9, %zmm6, %zmm7
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm7, %zmm8, %zmm6
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm7, %ymm3, %ymm8
; AVX512F-FAST-NEXT: vpshufb %ymm7, %ymm2, %ymm7
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm7 = ymm7[0],ymm8[0],ymm7[1],ymm8[1],ymm7[2],ymm8[2],ymm7[3],ymm8[3],ymm7[8],ymm8[8],ymm7[9],ymm8[9],ymm7[10],ymm8[10],ymm7[11],ymm8[11]
; AVX512F-FAST-NEXT: vpsrldq {{.*#+}} ymm8 = ymm5[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,ymm5[22,23,24,25,26,27,28,29,30,31],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpsrldq {{.*#+}} ymm9 = ymm4[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,ymm4[22,23,24,25,26,27,28,29,30,31],zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm8 = ymm9[0],ymm8[0],ymm9[1],ymm8[1],ymm9[2],ymm8[2],ymm9[3],ymm8[3],ymm9[8],ymm8[8],ymm9[9],ymm8[9],ymm9[10],ymm8[10],ymm9[11],ymm8[11]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,5,12,u,4,13,u,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,5,12,u,4,13,u,7]
; AVX512F-FAST-NEXT: vpermi2d %ymm8, %ymm7, %ymm9
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm7 = ymm1[1,2,2,3,5,6,6,7]
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm8 = ymm16[1,2,2,3,5,6,6,7]
@@ -1699,7 +1699,7 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpermi2d %ymm7, %ymm9, %ymm8
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} ymm2 = ymm2[4],ymm3[4],ymm2[5],ymm3[5],ymm2[6],ymm3[6],ymm2[7],ymm3[7],ymm2[12],ymm3[12],ymm2[13],ymm3[13],ymm2[14],ymm3[14],ymm2[15],ymm3[15]
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm4[4],ymm5[4],ymm4[5],ymm5[5],ymm4[6],ymm5[6],ymm4[7],ymm5[7],ymm4[12],ymm5[12],ymm4[13],ymm5[13],ymm4[14],ymm5[14],ymm4[15],ymm5[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <5,u,14,6,u,15,7,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [5,u,14,6,u,15,7,u]
; AVX512F-FAST-NEXT: vpermi2d %ymm2, %ymm3, %ymm4
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm4, %zmm0, %zmm2
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [8,21,10,11,22,13,14,23]
@@ -1721,15 +1721,15 @@ define void @store_i16_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512BW-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,16,32,48,u,u,1,17,33,49,u,u,2,18,34,50,u,u,3,19,35,51,u,u,4,20,36,52,u,u,5,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,16,32,48,u,u,1,17,33,49,u,u,2,18,34,50,u,u,3,19,35,51,u,u,4,20,36,52,u,u,5,21]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,32,48,6,7,8,9,33,49,12,13,14,15,34,50,18,19,20,21,35,51,24,25,26,27,36,52,30,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <5,21,u,u,38,54,6,22,u,u,39,55,7,23,u,u,40,56,8,24,u,u,41,57,9,25,u,u,42,58,10,26>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [5,21,u,u,38,54,6,22,u,u,39,55,7,23,u,u,40,56,8,24,u,u,41,57,9,25,u,u,42,58,10,26]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,37,53,4,5,6,7,38,54,10,11,12,13,39,55,16,17,18,19,40,56,22,23,24,25,41,57,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,11,27,43,59,u,u,12,28,44,60,u,u,13,29,45,61,u,u,14,30,46,62,u,u,15,31,47,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,11,27,43,59,u,u,12,28,44,60,u,u,13,29,45,61,u,u,14,30,46,62,u,u,15,31,47,63,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [42,58,2,3,4,5,43,59,8,9,10,11,44,60,14,15,16,17,45,61,20,21,22,23,46,62,26,27,28,29,47,63]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm3, %zmm0
@@ -2606,7 +2606,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm6 = ymm6[1,2,3,3,5,6,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm6[0,1],ymm1[2],ymm6[3,4],ymm1[5],ymm6[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm2, %ymm14
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,1,2,3]
@@ -2792,7 +2792,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa 32(%rsi), %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} ymm1 = ymm2[0],ymm1[0],ymm2[1],ymm1[1],ymm2[2],ymm1[2],ymm2[3],ymm1[3],ymm2[8],ymm1[8],ymm2[9],ymm1[9],ymm2[10],ymm1[10],ymm2[11],ymm1[11]
@@ -2847,7 +2847,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm7[4],xmm6[4],xmm7[5],xmm6[5],xmm7[6],xmm6[6],xmm7[7],xmm6[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <1,2,1,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [1,2,1,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm6 = xmm12[4],xmm14[4],xmm12[5],xmm14[5],xmm12[6],xmm14[6],xmm12[7],xmm14[7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[1,1,1,1]
@@ -2889,12 +2889,12 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: # ymm3 = ymm3[4],mem[4],ymm3[5],mem[5],ymm3[6],mem[6],ymm3[7],mem[7],ymm3[12],mem[12],ymm3[13],mem[13],ymm3[14],mem[14],ymm3[15],mem[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[3,3,3,3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm3[2],ymm0[3,4],ymm3[5],ymm0[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm10, %ymm4, %ymm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,1,2,3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm6[1],ymm0[2,3],ymm6[4],ymm0[5,6],ymm6[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm4, %ymm13
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,1,2,3]
@@ -3068,7 +3068,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdi), %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rsi), %ymm13
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm13, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm11, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm11, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -3165,7 +3165,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm1[1,2,3,3,5,6,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm1, %ymm11, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,1,2,3]
@@ -3307,7 +3307,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vmovdqa32 %zmm0, %zmm2 {%k1}
; AVX512F-ONLY-SLOW-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 32(%r8), %ymm1
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm4, %ymm1, %ymm3
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,2,3]
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2,3],ymm3[4],ymm0[5,6],ymm3[7]
@@ -3367,7 +3367,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rdx), %xmm14
; AVX512F-ONLY-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm7 = xmm14[4],xmm13[4],xmm14[5],xmm13[5],xmm14[6],xmm13[6],xmm14[7],xmm13[7]
; AVX512F-ONLY-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm2 = ymm3[0],ymm2[0],ymm3[1],ymm2[1],ymm3[2],ymm2[2],ymm3[3],ymm2[3],ymm3[8],ymm2[8],ymm3[9],ymm2[9],ymm3[10],ymm2[10],ymm3[11],ymm2[11]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = <17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6]
; AVX512F-ONLY-SLOW-NEXT: vpermt2d %zmm7, %zmm24, %zmm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rsi), %xmm11
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rdi), %xmm15
@@ -3512,7 +3512,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST: # %bb.0:
; AVX512F-ONLY-FAST-NEXT: pushq %rax
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%rsi), %ymm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm0, %ymm2, %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%rdi), %ymm4
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm0, %ymm4, %ymm3
@@ -3537,7 +3537,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vmovdqa32 %zmm3, %zmm1 {%k1}
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [8,21,10,11,20,13,14,23]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%r8), %ymm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm7, %ymm2, %ymm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm1, %zmm4
; AVX512F-ONLY-FAST-NEXT: vpermt2d %zmm3, %zmm11, %zmm4
@@ -3547,13 +3547,13 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpermt2d %ymm3, %ymm17, %ymm1
; AVX512F-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm16 = zmm1[0,1,2,3],zmm4[0,1,2,3]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%r9), %ymm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm14, %ymm3, %ymm1
; AVX512F-ONLY-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm15 = [16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25,16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25]
; AVX512F-ONLY-FAST-NEXT: # ymm15 = mem[0,1,0,1]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm15, %ymm3, %ymm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm3, %ymm26
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = <2,2,u,3,10,u,10,11>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = [2,2,u,3,10,u,10,11]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm1, %zmm18, %zmm2
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa (%rsi), %ymm9
@@ -3632,7 +3632,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm2[4],xmm1[4],xmm2[5],xmm1[5],xmm2[6],xmm1[6],xmm2[7],xmm1[7]
; AVX512F-ONLY-FAST-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm0[0],xmm8[0],xmm0[1],xmm8[1],xmm0[2],xmm8[2],xmm0[3],xmm8[3]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm1, %zmm21, %zmm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,0,0,u,8,8,u,9>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,0,0,u,8,8,u,9]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 32(%r9), %xmm12
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [0,1,4,5,4,5,6,7,8,9,8,9,8,9,8,9]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm7, %xmm12, %xmm4
@@ -3661,7 +3661,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm8[4],xmm0[5],xmm8[5],xmm0[6],xmm8[6],xmm0[7],xmm8[7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [1,1,1,1,10,10,10,11]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm7, %zmm15, %zmm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14]
; AVX512F-ONLY-FAST-NEXT: vpermd %zmm18, %zmm7, %zmm18
; AVX512F-ONLY-FAST-NEXT: vmovdqa32 %zmm0, %zmm18 {%k1}
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [8,9,20,11,12,21,14,15]
@@ -3675,7 +3675,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} ymm4 = ymm11[0,0,2,1,4,5,6,7,8,8,10,9,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [8,9,12,13,12,13,10,11,8,9,10,11,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm11, %xmm5, %xmm8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,u,0,1,u,10,10,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,u,0,1,u,10,10,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm4, %zmm5, %zmm8
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm24, %ymm4
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm25, %ymm9
@@ -3824,7 +3824,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-SLOW-NEXT: vinserti32x8 $1, %ymm3, %zmm2, %zmm0 {%k1}
; AVX512DQ-SLOW-NEXT: vmovdqa 32(%r8), %ymm6
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm10, %ymm6, %ymm1
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,2,3]
; AVX512DQ-SLOW-NEXT: vextracti64x4 $1, %zmm0, %ymm2
@@ -3875,7 +3875,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vpshuflw {{.*#+}} ymm10 = ymm10[0,2,2,1,4,5,6,7,8,10,10,9,12,13,14,15]
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm25 = ymm10[2,1,2,3]
; AVX512DQ-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm3 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[8],ymm3[8],ymm1[9],ymm3[9],ymm1[10],ymm3[10],ymm1[11],ymm3[11]
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6]
; AVX512DQ-SLOW-NEXT: vpermt2d %zmm0, %zmm10, %zmm3
; AVX512DQ-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm15[4],xmm12[4],xmm15[5],xmm12[5],xmm15[6],xmm12[6],xmm15[7],xmm12[7]
; AVX512DQ-SLOW-NEXT: vmovdqa (%r9), %ymm1
@@ -3983,7 +3983,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa 32(%rdx), %xmm6
; AVX512DQ-FAST-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm2[4],xmm3[4],xmm2[5],xmm3[5],xmm2[6],xmm3[6],xmm2[7],xmm3[7]
; AVX512DQ-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm2
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = <1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = [1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14]
; AVX512DQ-FAST-NEXT: vpermd %zmm2, %zmm21, %zmm19
; AVX512DQ-FAST-NEXT: movw $18724, %ax # imm = 0x4924
; AVX512DQ-FAST-NEXT: kmovw %eax, %k1
@@ -4008,7 +4008,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,9,2,3,8,5,6,11]
; AVX512DQ-FAST-NEXT: vpermt2d %ymm1, %ymm6, %ymm19
; AVX512DQ-FAST-NEXT: vmovdqa (%r9), %ymm7
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = <0,u,0,1,u,10,10,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,u,0,1,u,10,10,u]
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} ymm1 = ymm7[0,0,2,1,4,5,6,7,8,8,10,9,12,13,14,15]
; AVX512DQ-FAST-NEXT: vmovdqa (%r9), %xmm0
; AVX512DQ-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4036,7 +4036,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa 32(%r9), %ymm5
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} ymm0 = ymm5[0,0,2,1,4,5,6,7,8,8,10,9,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm0, %zmm18, %zmm13
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm6, %ymm2, %ymm0
; AVX512DQ-FAST-NEXT: vpshufb %ymm6, %ymm1, %ymm15
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm15[0],ymm0[0],ymm15[1],ymm0[1],ymm15[2],ymm0[2],ymm15[3],ymm0[3],ymm15[8],ymm0[8],ymm15[9],ymm0[9],ymm15[10],ymm0[10],ymm15[11],ymm0[11]
@@ -4053,7 +4053,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm1, %zmm4, %zmm0 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [8,21,10,11,20,13,14,23]
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm0, %zmm9
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm1, %ymm3, %ymm15
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm1, %ymm16
; AVX512DQ-FAST-NEXT: vpermt2d %zmm15, %zmm4, %zmm9
@@ -4067,12 +4067,12 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} ymm0 = ymm3[0],ymm0[0],ymm3[1],ymm0[1],ymm3[2],ymm0[2],ymm3[3],ymm0[3],ymm3[8],ymm0[8],ymm3[9],ymm0[9],ymm3[10],ymm0[10],ymm3[11],ymm0[11]
; AVX512DQ-FAST-NEXT: vpunpckhwd {{.*#+}} ymm3 = ymm9[4],ymm14[4],ymm9[5],ymm14[5],ymm9[6],ymm14[6],ymm9[7],ymm14[7],ymm9[12],ymm14[12],ymm9[13],ymm14[13],ymm9[14],ymm14[14],ymm9[15],ymm14[15]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm3, %zmm2, %zmm0
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm3, %ymm5, %ymm6
; AVX512DQ-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm9 = [16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25,16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25]
; AVX512DQ-FAST-NEXT: # ymm9 = mem[0,1,0,1]
; AVX512DQ-FAST-NEXT: vpshufb %ymm9, %ymm5, %ymm2
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = <2,2,u,3,10,u,10,11>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = [2,2,u,3,10,u,10,11]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm6, %zmm5, %zmm2
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm27, %ymm14
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm26, %ymm1
@@ -4141,7 +4141,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm31, %xmm7
; AVX512DQ-FAST-NEXT: vpshufb %xmm8, %xmm7, %xmm4
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm7 = xmm7[0,0,2,1,4,5,6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,0,0,u,8,8,u,9>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,0,0,u,8,8,u,9]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm4, %zmm9, %zmm7
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm28, %xmm11
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm4 = xmm11[2,1,3,3,4,5,6,7]
@@ -4192,7 +4192,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movw $9362, %cx # imm = 0x2492
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm6, %zmm7 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,2,3,32,u,6,7,8,9,33,u,12,13,14,15,34,u,18,19,20,21,35,u,24,25,26,27,36,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,32,u,6,7,8,9,33,u,12,13,14,15,34,u,18,19,20,21,35,u,24,25,26,27,36,u,30,31]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm7, %zmm8
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,32,6,7,8,9,10,33,12,13,14,15,16,34,18,19,20,21,22,35,24,25,26,27,28,36,30,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm8, %zmm6
@@ -4205,7 +4205,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movw $18724, %cx # imm = 0x4924
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,37,u,4,5,6,7,38,u,10,11,12,13,39,u,16,17,18,19,40,u,22,23,24,25,41,u,28,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,37,u,4,5,6,7,38,u,10,11,12,13,39,u,16,17,18,19,40,u,22,23,24,25,41,u,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm8, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,37,4,5,6,7,8,38,10,11,12,13,14,39,16,17,18,19,20,40,22,23,24,25,26,41,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm7, %zmm8
@@ -4216,7 +4216,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm9
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm9 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <42,u,2,3,4,5,43,u,8,9,10,11,44,u,14,15,16,17,45,u,20,21,22,23,46,u,26,27,28,29,47,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [42,u,2,3,4,5,43,u,8,9,10,11,44,u,14,15,16,17,45,u,20,21,22,23,46,u,26,27,28,29,47,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm9, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,42,2,3,4,5,6,43,8,9,10,11,12,44,14,15,16,17,18,45,20,21,22,23,24,46,26,27,28,29,30,47]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm7, %zmm9
@@ -4227,7 +4227,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm10
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm10 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,48,u,6,7,8,9,49,u,12,13,14,15,50,u,18,19,20,21,51,u,24,25,26,27,52,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,48,u,6,7,8,9,49,u,12,13,14,15,50,u,18,19,20,21,51,u,24,25,26,27,52,u,30,31]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm10, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,48,6,7,8,9,10,49,12,13,14,15,16,50,18,19,20,21,22,51,24,25,26,27,28,52,30,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm7, %zmm10
@@ -4238,7 +4238,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm5, %zmm4, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,53,u,4,5,6,7,54,u,10,11,12,13,55,u,16,17,18,19,56,u,22,23,24,25,57,u,28,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,53,u,4,5,6,7,54,u,10,11,12,13,55,u,16,17,18,19,56,u,22,23,24,25,57,u,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm11, %zmm7
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,53,4,5,6,7,8,54,10,11,12,13,14,55,16,17,18,19,20,56,22,23,24,25,26,57,28,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm7, %zmm11
@@ -4249,7 +4249,7 @@ define void @store_i16_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm4 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm4 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <58,u,2,3,4,5,59,u,8,9,10,11,60,u,14,15,16,17,61,u,20,21,22,23,62,u,26,27,28,29,63,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [58,u,2,3,4,5,59,u,8,9,10,11,60,u,14,15,16,17,61,u,20,21,22,23,62,u,26,27,28,29,63,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm4, %zmm2
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,58,2,3,4,5,6,59,8,9,10,11,12,60,14,15,16,17,18,61,20,21,22,23,24,62,26,27,28,29,30,63]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm2, %zmm1
@@ -6185,7 +6185,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm1[1,2,3,3,5,6,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm0[2],ymm1[3,4],ymm0[5],ymm1[6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm0, %ymm3
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,2,3]
@@ -6413,7 +6413,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa (%rsi), %ymm2
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} ymm2 = ymm3[0],ymm2[0],ymm3[1],ymm2[1],ymm3[2],ymm2[2],ymm3[3],ymm2[3],ymm3[8],ymm2[8],ymm3[9],ymm2[9],ymm3[10],ymm2[10],ymm3[11],ymm2[11]
@@ -6667,7 +6667,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm1 # 16-byte Folded Reload
; AVX2-FAST-NEXT: # xmm1 = xmm1[4],mem[4],xmm1[5],mem[5],xmm1[6],mem[6],xmm1[7],mem[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,2,1,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,2,1,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm3, %ymm0
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[1,1,1,1]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6,7]
@@ -6745,12 +6745,12 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[3,3,3,3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm2[2],ymm1[3,4],ymm2[5],ymm1[6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm0, %ymm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,2,3]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm1[0],ymm4[1],ymm1[2,3],ymm4[4],ymm1[5,6],ymm4[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm1, %ymm0, %ymm14
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,1,2,3]
@@ -7139,7 +7139,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm14, %ymm6
@@ -7314,12 +7314,12 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm2 = ymm2[1,2,3,3,5,6,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm0, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm3 = ymm1[0],ymm3[1],ymm1[2,3],ymm3[4],ymm1[5,6],ymm3[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm0, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,2,3]
@@ -7447,7 +7447,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vmovdqa32 %zmm0, %zmm1 {%k1}
; AVX512F-ONLY-SLOW-NEXT: vextracti64x4 $1, %zmm1, %ymm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 96(%r8), %ymm5
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512F-ONLY-SLOW-NEXT: vpshufb %ymm7, %ymm5, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,1,2,3]
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm2[1],ymm0[2,3],ymm2[4],ymm0[5,6],ymm2[7]
@@ -7593,7 +7593,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm7 = zmm14[0,1,2,3],zmm7[4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm1 = ymm1[0],ymm2[0],ymm1[1],ymm2[1],ymm1[2],ymm2[2],ymm1[3],ymm2[3],ymm1[8],ymm2[8],ymm1[9],ymm2[9],ymm1[10],ymm2[10],ymm1[11],ymm2[11]
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rcx), %xmm7
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rdx), %xmm14
; AVX512F-ONLY-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm14[4],xmm7[4],xmm14[5],xmm7[5],xmm14[6],xmm7[6],xmm14[7],xmm7[7]
@@ -7945,7 +7945,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa 96(%rsi), %ymm12
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm12, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm6, %ymm12, %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa 96(%rdi), %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -7969,7 +7969,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm0, %zmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa 96(%r8), %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm13, %ymm1, %ymm4
; AVX512F-ONLY-FAST-NEXT: vmovdqa %ymm13, %ymm8
; AVX512F-ONLY-FAST-NEXT: vpermt2d %zmm4, %zmm25, %zmm3
@@ -7978,10 +7978,10 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpermt2d %ymm4, %ymm29, %ymm0
; AVX512F-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm0[0,1,2,3],zmm3[0,1,2,3]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm28 = <2,2,u,3,10,u,10,11>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm28 = [2,2,u,3,10,u,10,11]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 96(%r9), %ymm1
; AVX512F-ONLY-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm15, %ymm1, %ymm0
; AVX512F-ONLY-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm14 = [16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25,16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25]
; AVX512F-ONLY-FAST-NEXT: # ymm14 = mem[0,1,0,1]
@@ -8054,7 +8054,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpermt2d %ymm3, %ymm16, %ymm4
; AVX512F-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm4[0,1,2,3],zmm2[0,1,2,3]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = <u,0,0,u,8,8,u,9>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = [u,0,0,u,8,8,u,9]
; AVX512F-ONLY-FAST-NEXT: vmovdqa 96(%r9), %xmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,4,5,4,5,6,7,8,9,8,9,8,9,8,9]
@@ -8216,7 +8216,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm15[4],xmm0[5],xmm15[5],xmm0[6],xmm15[6],xmm0[7],xmm15[7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [1,1,1,1,10,10,10,11]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm2, %zmm9, %zmm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = <1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14]
; AVX512F-ONLY-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm15, %zmm2 # 64-byte Folded Reload
; AVX512F-ONLY-FAST-NEXT: vmovdqa32 %zmm0, %zmm2 {%k1}
; AVX512F-ONLY-FAST-NEXT: vpshuflw {{.*#+}} ymm0 = ymm3[0,1,1,3,4,5,6,7,8,9,9,11,12,13,14,15]
@@ -8231,7 +8231,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [8,9,12,13,12,13,10,11,8,9,10,11,12,13,14,15]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %xmm0, %xmm30
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm27 = <0,u,0,1,u,10,10,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,u,0,1,u,10,10,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm5, %zmm27, %zmm3
; AVX512F-ONLY-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX512F-ONLY-FAST-NEXT: vpunpcklwd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm5 # 32-byte Folded Reload
@@ -8529,7 +8529,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-SLOW-NEXT: vinserti32x8 $1, %ymm3, %zmm2, %zmm0 {%k1}
; AVX512DQ-SLOW-NEXT: vmovdqa 96(%r8), %ymm9
-; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512DQ-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512DQ-SLOW-NEXT: vpshufb %ymm13, %ymm9, %ymm1
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,2,3]
; AVX512DQ-SLOW-NEXT: vextracti64x4 $1, %zmm0, %ymm2
@@ -8650,7 +8650,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vpshufhw {{.*#+}} xmm13 = xmm13[0,1,2,3,4,4,4,4]
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm31 = ymm13[0,0,2,1]
; AVX512DQ-SLOW-NEXT: vpunpcklwd {{.*#+}} ymm2 = ymm2[0],ymm4[0],ymm2[1],ymm4[1],ymm2[2],ymm4[2],ymm2[3],ymm4[3],ymm2[8],ymm4[8],ymm2[9],ymm4[9],ymm2[10],ymm4[10],ymm2[11],ymm4[11]
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [17,18,17,18,u,u,19,19,5,4,2,2,5,4,6,6]
; AVX512DQ-SLOW-NEXT: vpermt2d %zmm0, %zmm4, %zmm2
; AVX512DQ-SLOW-NEXT: vmovdqa64 %xmm18, %xmm0
; AVX512DQ-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm15[4],xmm0[4],xmm15[5],xmm0[5],xmm15[6],xmm0[6],xmm15[7],xmm0[7]
@@ -8904,7 +8904,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512DQ-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm3[4],xmm2[4],xmm3[5],xmm2[5],xmm3[6],xmm2[6],xmm3[7],xmm2[7]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm1, %zmm18, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [1,2,1,2,u,u,3,3,13,12,10,10,13,12,14,14]
; AVX512DQ-FAST-NEXT: vpermd %zmm0, %zmm20, %zmm22
; AVX512DQ-FAST-NEXT: movw $18724, %ax # imm = 0x4924
; AVX512DQ-FAST-NEXT: kmovw %eax, %k1
@@ -8922,7 +8922,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [12,13,10,11,10,11,14,15,14,15,14,15,14,15,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX512DQ-FAST-NEXT: vpermt2d %ymm0, %ymm19, %ymm22
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm23 = <0,u,0,1,u,10,10,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,u,0,1,u,10,10,u]
; AVX512DQ-FAST-NEXT: vmovdqa (%r9), %ymm0
; AVX512DQ-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} ymm0 = ymm0[0,0,2,1,4,5,6,7,8,8,10,9,12,13,14,15]
@@ -9014,7 +9014,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} ymm0 = ymm0[0,0,2,1,4,5,6,7,8,8,10,9,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm0, %zmm23, %zmm1
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,4,5,10,11,u,u,u,u,u,u,u,u,24,25,22,23,20,21,26,27,u,u,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm9, %ymm0
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm5, %ymm1
; AVX512DQ-FAST-NEXT: vpunpcklwd {{.*#+}} ymm8 = ymm1[0],ymm0[0],ymm1[1],ymm0[1],ymm1[2],ymm0[2],ymm1[3],ymm0[3],ymm1[8],ymm0[8],ymm1[9],ymm0[9],ymm1[10],ymm0[10],ymm1[11],ymm0[11]
@@ -9033,7 +9033,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vinserti32x8 $1, %ymm0, %zmm1, %zmm8 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} ymm23 = [8,21,10,11,20,13,14,23]
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm8, %zmm0
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,u,u,14,15,14,15,14,15,14,15,28,29,26,27,26,27,30,31,30,31,30,31,30,31,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm2, %ymm10, %ymm1
; AVX512DQ-FAST-NEXT: vmovdqa %ymm2, %ymm9
; AVX512DQ-FAST-NEXT: vpermt2d %zmm1, %zmm23, %zmm0
@@ -9042,14 +9042,14 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpermt2d %ymm1, %ymm30, %ymm8
; AVX512DQ-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm8[0,1,2,3],zmm0[0,1,2,3]
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm1, %ymm11, %ymm0
; AVX512DQ-FAST-NEXT: vmovdqa %ymm1, %ymm8
; AVX512DQ-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25,16,17,20,21,20,21,22,23,24,25,24,25,24,25,24,25]
; AVX512DQ-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX512DQ-FAST-NEXT: vpshufb %ymm2, %ymm11, %ymm1
; AVX512DQ-FAST-NEXT: vmovdqa %ymm2, %ymm11
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <2,2,u,3,10,u,10,11>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [2,2,u,3,10,u,10,11]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm0, %zmm31, %zmm1
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm13, %ymm0
; AVX512DQ-FAST-NEXT: vpshufb %ymm15, %ymm12, %ymm2
@@ -9112,7 +9112,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpshufb %xmm4, %xmm5, %xmm3
; AVX512DQ-FAST-NEXT: vmovdqa %xmm4, %xmm8
; AVX512DQ-FAST-NEXT: vpshuflw {{.*#+}} xmm7 = xmm5[0,0,2,1,4,5,6,7]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,0,0,u,8,8,u,9>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,0,0,u,8,8,u,9]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm3, %zmm18, %zmm7
; AVX512DQ-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX512DQ-FAST-NEXT: vpshufb %xmm9, %xmm5, %xmm3
@@ -9174,7 +9174,7 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpermt2q %zmm8, %zmm29, %zmm5
; AVX512DQ-FAST-NEXT: vshufi64x2 {{.*#+}} zmm29 = zmm3[0,1,2,3],zmm6[0,1,2,3]
; AVX512DQ-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,8,9,10,11,12,13,14,15,24,25,28,29,28,29,26,27,24,25,26,27,28,29,30,31]
; AVX512DQ-FAST-NEXT: vpshufb %ymm13, %ymm3, %ymm8
; AVX512DQ-FAST-NEXT: vmovdqa %ymm11, %ymm15
; AVX512DQ-FAST-NEXT: vpshufb %ymm11, %ymm3, %ymm3
@@ -9374,22 +9374,22 @@ define void @store_i16_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa32 %zmm20, %zmm4 {%k1}
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm22, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,2,3,32,u,6,7,8,9,33,u,12,13,14,15,34,u,18,19,20,21,35,u,24,25,26,27,36,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,2,3,32,u,6,7,8,9,33,u,12,13,14,15,34,u,18,19,20,21,35,u,24,25,26,27,36,u,30,31]
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm15, %zmm2
; AVX512BW-NEXT: vmovdqa32 %zmm23, %zmm9 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <58,u,2,3,4,5,59,u,8,9,10,11,60,u,14,15,16,17,61,u,20,21,22,23,62,u,26,27,28,29,63,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [58,u,2,3,4,5,59,u,8,9,10,11,60,u,14,15,16,17,61,u,20,21,22,23,62,u,26,27,28,29,63,u]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm19, %zmm4
; AVX512BW-NEXT: vmovdqa32 %zmm26, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <0,1,53,u,4,5,6,7,54,u,10,11,12,13,55,u,16,17,18,19,56,u,22,23,24,25,57,u,28,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,1,53,u,4,5,6,7,54,u,10,11,12,13,55,u,16,17,18,19,56,u,22,23,24,25,57,u,28,29,30,31]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm20, %zmm6
; AVX512BW-NEXT: vmovdqa32 %zmm25, %zmm16 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <0,1,2,3,48,u,6,7,8,9,49,u,12,13,14,15,50,u,18,19,20,21,51,u,24,25,26,27,52,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,2,3,48,u,6,7,8,9,49,u,12,13,14,15,50,u,18,19,20,21,51,u,24,25,26,27,52,u,30,31]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm22, %zmm9
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm0 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <42,u,2,3,4,5,43,u,8,9,10,11,44,u,14,15,16,17,45,u,20,21,22,23,46,u,26,27,28,29,47,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [42,u,2,3,4,5,43,u,8,9,10,11,44,u,14,15,16,17,45,u,20,21,22,23,46,u,26,27,28,29,47,u]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm7, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm15, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,37,u,4,5,6,7,38,u,10,11,12,13,39,u,16,17,18,19,40,u,22,23,24,25,41,u,28,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,37,u,4,5,6,7,38,u,10,11,12,13,39,u,16,17,18,19,40,u,22,23,24,25,41,u,28,29,30,31]
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm15, %zmm16
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm5 {%k1}
; AVX512BW-NEXT: vpermt2w %zmm12, %zmm19, %zmm5
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-7.ll
index f3ec4420e4dda36..d3fb3bdc28da180 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-7.ll
@@ -139,7 +139,7 @@ define void @store_i16_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vinserti128 $1, (%r8), %ymm0, %ymm0
; AVX512BW-NEXT: vpunpckldq {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[4],ymm1[4],ymm0[5],ymm1[5]
; AVX512BW-NEXT: vinserti128 $1, (%r10), %ymm2, %ymm1
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,2,16,18,8,10,24,1,3,17,19,9,11,25,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,2,16,18,8,10,24,1,3,17,19,9,11,25,u,u]
; AVX512BW-NEXT: vpermi2w %ymm1, %ymm0, %ymm2
; AVX512BW-NEXT: vextracti128 $1, %ymm2, %xmm0
; AVX512BW-NEXT: vpextrd $2, %xmm0, 24(%rax)
@@ -329,9 +329,9 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm2[14,15,14,15,14,15,14,15,4,5,6,7,14,15,14,15,30,31,30,31,30,31,30,31,20,21,22,23,30,31,30,31]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm3[3,1,2,1]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[2,0,2,3,4,5,6,7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,0,0,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,0,0,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vmovdqa %ymm4, (%rax)
; AVX2-SLOW-NEXT: vextracti128 $1, %ymm0, %xmm1
@@ -357,7 +357,7 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpunpcklqdq {{.*#+}} ymm0 = ymm0[0],ymm3[0],ymm0[2],ymm3[2]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <5,7,1,3,7,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [5,7,1,3,7,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,4,5],zero,zero,zero,zero,zero,zero,ymm1[10,11,14,15,2,3,18,19],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [1,3,5,7,1,3,5,7]
@@ -414,9 +414,9 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm2[14,15,14,15,14,15,14,15,4,5,6,7,14,15,14,15,30,31,30,31,30,31,30,31,20,21,22,23,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm2 = xmm3[4,5,12,13,4,5,6,7,8,9,10,11,4,5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,0,0,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,0,0,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm4, (%rax)
; AVX2-FAST-PERLANE-NEXT: vextracti128 $1, %ymm0, %xmm1
@@ -481,7 +481,7 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2
; AVX512F-FAST-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u]
; AVX512F-FAST-NEXT: vpermi2q %ymm3, %ymm0, %ymm1
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm0 = [0,2,4,6,0,2,4,6]
; AVX512F-FAST-NEXT: # ymm0 = mem[0,1,0,1]
@@ -496,7 +496,7 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm2[2,3,0,1]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm3 = zero,zero,zero,zero,ymm3[0,1,8,9,u,u,u,u,u,u],zero,zero,ymm3[26,27],zero,zero,zero,zero,ymm3[u,u,u,u,u,u,20,21,28,29]
; AVX512F-FAST-NEXT: vpternlogq $168, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <5,7,1,3,7,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [5,7,1,3,7,u,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm2, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,4,5],zero,zero,zero,zero,zero,zero,ymm1[10,11,14,15,2,3,18,19],zero,zero,zero,zero,zero,zero,ymm1[u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm3, %zmm1
@@ -524,7 +524,7 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512BW-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512BW-SLOW-NEXT: vpunpcklqdq {{.*#+}} ymm0 = ymm0[0],ymm3[0],ymm0[2],ymm3[2]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,4,8,12,32,40,36,1,5,9,13,33,41,37,2,6,10,14,34,42,38,3,7,11,15,35,43,39,u,u,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,4,8,12,32,40,36,1,5,9,13,33,41,37,2,6,10,14,34,42,38,3,7,11,15,35,43,39,u,u,u,u]
; AVX512BW-SLOW-NEXT: vpermi2w %zmm0, %zmm2, %zmm1
; AVX512BW-SLOW-NEXT: vextracti32x4 $2, %zmm1, 32(%rax)
; AVX512BW-SLOW-NEXT: vextracti32x4 $3, %zmm1, %xmm0
@@ -548,10 +548,10 @@ define void @store_i16_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2
; AVX512BW-FAST-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512BW-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u]
; AVX512BW-FAST-NEXT: vpermi2q %ymm3, %ymm0, %ymm1
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm2, %zmm0
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,4,8,12,16,20,24,1,5,9,13,17,21,25,2,6,10,14,18,22,26,3,7,11,15,19,23,27,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,4,8,12,16,20,24,1,5,9,13,17,21,25,2,6,10,14,18,22,26,3,7,11,15,19,23,27,u,u,u,u]
; AVX512BW-FAST-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-FAST-NEXT: vextracti32x4 $2, %zmm0, 32(%rax)
; AVX512BW-FAST-NEXT: vextracti32x4 $3, %zmm0, %xmm1
@@ -840,12 +840,12 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm6 = ymm6[0],ymm10[1],ymm6[2,3,4,5],ymm10[6],ymm6[7,8],ymm10[9],ymm6[10,11,12,13],ymm10[14],ymm6[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm7[0,2,1,3]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,16,17,24,25,u,u,u,u]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm6, %ymm10, %ymm6
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm9[0,2,0,2]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm11 = ymm10[u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,22,23,30,31,u,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpbroadcastd 4(%r10), %ymm12
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm6, %ymm11, %ymm6
@@ -857,7 +857,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[1,3,3,1]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm9 = ymm9[0,1,8,9,u,u,u,u,u,u,u,u,u,u,2,3,18,19,u,u,u,u,u,u,u,u,u,u,28,29,20,21]
; AVX2-SLOW-NEXT: vpbroadcastd 8(%r10), %ymm12
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm9, %ymm12, %ymm9
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
@@ -868,7 +868,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpor %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm10[u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,18,19,26,27,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpbroadcastd (%r10), %ymm10
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm8, %ymm10, %ymm8
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm7, %ymm8, %ymm7
@@ -904,14 +904,14 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vinserti128 $1, %xmm3, %ymm1, %ymm9
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm7[0,2,1,3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm6 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm6[6,7,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm6[16,17,24,25],zero,zero,zero,zero
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <1,5,u,u,5,2,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [1,5,u,u,5,2,6,u]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm10, %ymm10
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[0,1,4,5],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm10[2,3,18,19],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm10[20,21,24,25]
; AVX2-FAST-NEXT: vpor %ymm6, %ymm10, %ymm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm9[0,2,0,2]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm11 = ymm10[u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,22,23,30,31,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpbroadcastd 4(%r10), %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm6, %ymm11, %ymm6
@@ -923,7 +923,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[1,3,3,1]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm9[0,1,8,9,u,u,u,u,u,u,u,u,u,u,2,3,18,19,u,u,u,u,u,u,u,u,u,u,28,29,20,21]
; AVX2-FAST-NEXT: vpbroadcastd 8(%r10), %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm9, %ymm12, %ymm9
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
@@ -934,7 +934,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpor %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm10[u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,18,19,26,27,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpbroadcastd (%r10), %ymm10
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm8, %ymm10, %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm7, %ymm8, %ymm7
@@ -974,12 +974,12 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm6 = ymm6[0],ymm10[1],ymm6[2,3,4,5],ymm10[6],ymm6[7,8],ymm10[9],ymm6[10,11,12,13],ymm10[14],ymm6[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm7[0,2,1,3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,16,17,24,25,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm6, %ymm10, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm9[0,2,0,2]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm11 = ymm10[u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,22,23,30,31,u,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 4(%r10), %ymm12
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm6, %ymm11, %ymm6
@@ -991,7 +991,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[1,3,3,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm9[0,1,8,9,u,u,u,u,u,u,u,u,u,u,2,3,18,19,u,u,u,u,u,u,u,u,u,u,28,29,20,21]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 8(%r10), %ymm12
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm9, %ymm12, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
@@ -1002,7 +1002,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm8 = ymm10[u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,18,19,26,27,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd (%r10), %ymm10
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm8, %ymm10, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm7, %ymm8, %ymm7
@@ -1128,7 +1128,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm3, %zmm2, %zmm2
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm6[0,2,2,0]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[0,1,8,9],zero,zero,zero,zero,ymm3[u,u,u,u,u,u,2,3,18,19],zero,zero,zero,zero,ymm3[u,u,u,u,u,u,28,29,20,21]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <1,5,u,u,5,2,6,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [1,5,u,u,5,2,6,u]
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm4, %ymm4
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm4 = ymm4[0,1,4,5,u,u,u,u,u,u],zero,zero,zero,zero,ymm4[2,3,18,19,u,u,u,u,u,u],zero,zero,zero,zero,ymm4[20,21,24,25]
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm4, %zmm3, %zmm3
@@ -1162,7 +1162,7 @@ define void @store_i16_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vinserti32x4 $2, (%r10), %zmm1, %zmm1
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,8,16,24,32,40,48,1,9,17,25,33,41,49,2,10,18,26,34,42,50,3,11,19,27,35,43,51,4,12,20,28]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <36,44,52,5,13,21,29,37,45,53,6,14,22,30,38,46,54,7,15,23,31,39,47,55,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [36,44,52,5,13,21,29,37,45,53,6,14,22,30,38,46,54,7,15,23,31,39,47,55,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm3
; AVX512BW-NEXT: vextracti32x4 $2, %zmm3, 96(%rax)
; AVX512BW-NEXT: vmovdqa64 %zmm2, (%rax)
@@ -1781,32 +1781,32 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqa (%r9), %ymm2
; AVX2-SLOW-NEXT: vmovdqa (%rax), %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm7, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm8 = ymm6[0,3,2,3,4,7,6,7]
; AVX2-SLOW-NEXT: vmovdqa %ymm6, %ymm7
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm8 = ymm8[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,u,u,u,4,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,u,u,u,4,u,u]
; AVX2-SLOW-NEXT: vpermd %ymm5, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vmovdqa %ymm5, %ymm6
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm9 = ymm13[0,0,0,0,4,5,6,7,8,8,8,8,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm9 = ymm9[0,1,2,3,7,7,7,7,8,9,10,11,15,15,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,3,u,u,u,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,3,u,u,u,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vmovdqa %ymm3, %ymm4
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm9 = ymm2[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm9 = ymm9[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <3,u,u,3,u,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [3,u,u,3,u,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm1, %ymm9, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
@@ -1822,7 +1822,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm8 = xmm8[2,1,2,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm8 = xmm8[0,1,2,3,4,5,5,4]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,0,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm8, %ymm0, %ymm1
; AVX2-SLOW-NEXT: vmovdqa (%r9), %xmm9
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm14
@@ -1830,7 +1830,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,0,1,1]
; AVX2-SLOW-NEXT: vpbroadcastd 8(%rax), %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm8, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm0, %ymm1, %ymm0
@@ -1844,14 +1844,14 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm15[0,1],xmm1[2],xmm15[3,4],xmm1[5],xmm15[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm14[0],xmm9[0],xmm14[1],xmm9[1],xmm14[2],xmm9[2],xmm14[3],xmm9[3]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm9 = xmm0[0,1,2,3,4,5,7,6]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm9 = xmm9[0,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,1,3]
; AVX2-SLOW-NEXT: vpbroadcastd 4(%rax), %ymm14
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm9, %ymm14, %ymm9
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm1, %ymm9, %ymm1
@@ -1869,7 +1869,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm14 = ymm15[0],ymm14[1],ymm15[2,3],ymm14[4],ymm15[5,6,7,8],ymm14[9],ymm15[10,11],ymm14[12],ymm15[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm14 = ymm14[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm1, %ymm14, %ymm1
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm7
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm14 = ymm2[0,1,2,3,5,6,7,7,8,9,10,11,13,14,15,15]
@@ -1878,7 +1878,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm15 = ymm4[3,3,3,3,7,7,7,7]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm14 = ymm14[0,1],ymm15[2],ymm14[3,4],ymm15[5],ymm14[6,7,8,9],ymm15[10],ymm14[11,12],ymm15[13],ymm14[14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm15, %ymm1, %ymm14, %ymm1
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm14 = ymm4[2,3,3,3,6,7,7,7]
@@ -1892,13 +1892,13 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm10 = xmm10[0,1,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm10 = xmm10[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,1,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm1, %ymm10, %ymm1
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,1,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,1,1,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,0,1]
; AVX2-SLOW-NEXT: vpbroadcastd (%rax), %ymm10
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm0, %ymm10, %ymm0
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm1, %ymm0, %ymm10
@@ -1913,7 +1913,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm11 = ymm11[2,2,2,2,6,6,6,6]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm1[0],ymm11[1],ymm1[2,3],ymm11[4],ymm1[5,6,7,8],ymm11[9],ymm1[10,11],ymm11[12],ymm1[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm12, %ymm1, %ymm3
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm15
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm2[1,2,2,3,5,6,6,7]
@@ -1925,7 +1925,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm13
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm11 = ymm4[0,1,2,2,4,5,6,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,1,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm1, %ymm11, %ymm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm3, %ymm1, %ymm7
@@ -1939,7 +1939,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm4 = ymm4[0,0,0,0,4,4,4,4]
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0,1],ymm4[2],ymm3[3,4],ymm4[5],ymm3[6,7,8,9],ymm4[10],ymm3[11,12],ymm4[13],ymm3[14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm3 = ymm15[0,0,2,1,4,4,6,5]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm2 = ymm2[1,2,2,3,4,5,6,7,9,10,10,11,12,13,14,15]
@@ -1948,7 +1948,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,3,3]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm3 = ymm13[0,1,1,3,4,5,5,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
@@ -1986,7 +1986,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm9 = ymm5[2,2,2,2,6,6,6,6]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1],ymm9[2,3],ymm8[4],ymm9[5,6,7,8],ymm8[9],ymm9[10,11],ymm8[12],ymm9[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [4,5,2,2,6,6,6,6]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm8, %ymm8
@@ -1994,7 +1994,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm10 = ymm15[1,2,2,3,5,6,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0,1],ymm9[2],ymm10[3,4],ymm9[5],ymm10[6,7,8,9],ymm9[10],ymm10[11,12],ymm9[13],ymm10[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm9, %ymm8, %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
@@ -2007,7 +2007,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm9 = ymm4[0,1,1,3,4,5,5,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0,1],ymm8[2],ymm9[3,4],ymm8[5],ymm9[6,7,8,9],ymm8[10],ymm9[11,12],ymm8[13],ymm9[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,1,4,5,4,5,5,7]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm8, %ymm8
@@ -2016,31 +2016,31 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,18,19,20,21,18,19,u,u,18,19,20,21,u,u,20,21]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm9 = ymm10[0,1,2],ymm9[3],ymm10[4,5],ymm9[6],ymm10[7,8,9,10],ymm9[11],ymm10[12,13],ymm9[14],ymm10[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm9, %ymm8, %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm7[0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm8, %ymm8
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm6[0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-NEXT: vpermd %ymm15, %ymm8, %ymm8
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm3[0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm9, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
@@ -2055,7 +2055,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm8 = xmm14[4],xmm13[4],xmm14[5],xmm13[5],xmm14[6],xmm13[6],xmm14[7],xmm13[7]
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm8 = xmm8[0,2,3,3,4,5,6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm8, %ymm11
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm8
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm2
@@ -2063,7 +2063,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm0[0,0,1,1]
; AVX2-FAST-NEXT: vpbroadcastd 8(%rax), %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm0, %ymm11, %ymm11
@@ -2075,13 +2075,13 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm9[0,1],xmm1[2],xmm9[3,4],xmm1[5],xmm9[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm0, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm2[0],xmm8[0],xmm2[1],xmm8[1],xmm2[2],xmm8[2],xmm2[3],xmm8[3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm0[0,1,2,3,8,9,10,11,14,15,12,13,14,15,12,13]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,1,3]
; AVX2-FAST-NEXT: vpbroadcastd 4(%rax), %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm2, %ymm8, %ymm2
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm1, %ymm2, %ymm8
@@ -2093,13 +2093,13 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2,3],ymm2[4],ymm4[5,6,7,8],ymm2[9],ymm4[10,11],ymm2[12],ymm4[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm15[3,3,3,3,7,7,7,7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,28,29,u,u,28,29,26,27,u,u,30,31,30,31]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7,8,9],ymm2[10],ymm3[11,12],ymm2[13],ymm3[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [6,7,3,3,7,7,6,7]
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm2, %ymm2 # 32-byte Folded Reload
@@ -2111,12 +2111,12 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm3 = xmm14[0],xmm13[0],xmm14[1],xmm13[1],xmm14[2],xmm13[2],xmm14[3],xmm13[3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,1,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,5,6,7,4,5,12,13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,0,1]
; AVX2-FAST-NEXT: vpbroadcastd (%rax), %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm0, %ymm3, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm2, %ymm0, %ymm0
@@ -2148,28 +2148,28 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rax), %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm7, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm8 = ymm6[0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm4, %ymm8, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm4, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm5[0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm5, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm8, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm2[0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm1, %ymm9, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
@@ -2184,7 +2184,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm8 = xmm5[4],xmm3[4],xmm5[5],xmm3[5],xmm5[6],xmm3[6],xmm5[7],xmm3[7]
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm8 = xmm8[0,2,3,3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %xmm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %xmm14
@@ -2192,7 +2192,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm8 = xmm8[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 8(%rax), %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm8, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255,255,255,0,0,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm0, %ymm1, %ymm0
@@ -2205,13 +2205,13 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm1 = xmm15[0,1],xmm1[2],xmm15[3,4],xmm1[5],xmm15[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm0, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm14[0],xmm9[0],xmm14[1],xmm9[1],xmm14[2],xmm9[2],xmm14[3],xmm9[3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm9 = xmm0[0,1,2,3,8,9,10,11,14,15,12,13,14,15,12,13]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,1,3]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 4(%rax), %ymm14
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm9, %ymm14, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm1, %ymm9, %ymm1
@@ -2227,7 +2227,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm14 = ymm15[0],ymm14[1],ymm15[2,3],ymm14[4],ymm15[5,6,7,8],ymm14[9],ymm15[10,11],ymm14[12],ymm15[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm14 = ymm14[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm1, %ymm14, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm14 = ymm6[3,3,3,3,7,7,7,7]
@@ -2235,7 +2235,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm15 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,28,29,u,u,28,29,26,27,u,u,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm14 = ymm15[0,1],ymm14[2],ymm15[3,4],ymm14[5],ymm15[6,7,8,9],ymm14[10],ymm15[11,12],ymm14[13],ymm15[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u,255,255,255,255,255,255,255,255,0,0,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm1, %ymm14, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm14 = ymm13[2,3,3,3,6,7,7,7]
@@ -2248,12 +2248,12 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpunpcklwd {{.*#+}} xmm10 = xmm5[0],xmm3[0],xmm5[1],xmm3[1],xmm5[2],xmm3[2],xmm5[3],xmm3[3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,1,1,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm1, %ymm10, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,5,6,7,4,5,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd (%rax), %ymm10
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm0, %ymm10, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm1, %ymm0, %ymm10
@@ -2265,7 +2265,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm11 = ymm8[2,2,2,2,6,6,6,6]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm11[0],ymm1[1],ymm11[2,3],ymm1[4],ymm11[5,6,7,8],ymm1[9],ymm11[10,11],ymm1[12],ymm11[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm2, %ymm1, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm12, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm12[u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u]
@@ -2275,7 +2275,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm11 = ymm13[0,1,2,2,4,5,6,6]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,1,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm1, %ymm11, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm3, %ymm1, %ymm6
@@ -2287,7 +2287,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm3 = ymm7[0,1,1,3,4,5,5,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm3[0,1],ymm4[2],ymm3[3,4],ymm4[5],ymm3[6,7,8,9],ymm4[10],ymm3[11,12],ymm4[13],ymm3[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm3 = ymm15[0,0,2,1,4,4,6,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,18,19,20,21,18,19,u,u,18,19,20,21,u,u,20,21]
@@ -2295,7 +2295,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm3 = ymm13[0,1,1,3,4,5,5,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
@@ -2387,7 +2387,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm15[0],xmm11[0],xmm15[1],xmm11[1],xmm15[2],xmm11[2],xmm15[3],xmm11[3]
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm11 = xmm0[0,1,2,3,4,5,7,6]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,1,3,2,4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u]
; AVX512F-SLOW-NEXT: vpermi2d %zmm11, %zmm0, %zmm15
; AVX512F-SLOW-NEXT: vprold $16, %ymm3, %ymm0
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} ymm11 = ymm2[1,2,2,3,5,6,6,7]
@@ -2479,7 +2479,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} ymm3 = ymm6[0,0,1,1,2,2,3,3,8,8,9,9,10,10,11,11]
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm9 = ymm4[0,1,1,3,4,5,5,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm9[0,1],ymm3[2],ymm9[3,4],ymm3[5],ymm9[6,7,8,9],ymm3[10],ymm9[11,12],ymm3[13],ymm9[14,15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,u,3,2,u,10,10,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,u,3,2,u,10,10,11]
; AVX512F-FAST-NEXT: vpermi2q %zmm0, %zmm3, %zmm9
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm0 = ymm4[2,2,2,2,6,6,6,6]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,u,u,24,25,26,27,u,u,26,27,26,27]
@@ -2538,7 +2538,7 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm6 = [0,13,4,0,0,14,5,0,0,13,4,0,0,14,5,0]
; AVX512F-FAST-NEXT: # zmm6 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vpermd %zmm8, %zmm6, %zmm6
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <6,u,u,u,7,u,u,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [6,u,u,u,7,u,u,7]
; AVX512F-FAST-NEXT: vpermd %ymm8, %ymm7, %ymm7
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[4,5,2,3,4,5,6,7,8,9,10,11,10,11,8,9]
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,3]
@@ -2590,27 +2590,27 @@ define void @store_i16_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512BW-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,14,30,46,62,u,u,u,15,31,47,63,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,14,30,46,62,u,u,u,15,31,47,63,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <29,45,u,u,u,u,14,30,46,u,u,u,u,15,31,47>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [29,45,u,u,u,u,14,30,46,u,u,u,u,15,31,47]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,16,32,u,u,u,u,1,17,33,u,u,u,u,2,18,34,u,u,u,u,3,19,35,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,16,32,u,u,u,u,1,17,33,u,u,u,u,2,18,34,u,u,u,u,3,19,35,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,16,32,48,u,u,u,1,17,33,49,u,u,u,2,18,34,50,u,u,u,3,19,35,51,u,u,u,4,20,36,52>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,16,32,48,u,u,u,1,17,33,49,u,u,u,2,18,34,50,u,u,u,3,19,35,51,u,u,u,4,20,36,52]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: movl $236730480, %ecx # imm = 0xE1C3870
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm6, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,20,36,u,u,u,u,5,21,37,u,u,u,u,6,22,38,u,u,u,u,7,23,39,u,u,u,u,8,24,40,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,20,36,u,u,u,u,5,21,37,u,u,u,u,6,22,38,u,u,u,u,7,23,39,u,u,u,u,8,24,40,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,5,21,37,53,u,u,u,6,22,38,54,u,u,u,7,23,39,55,u,u,u,8,24,40,56,u,u,u,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,5,21,37,53,u,u,u,6,22,38,54,u,u,u,7,23,39,55,u,u,u,8,24,40,56,u,u,u,9]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm8
; AVX512BW-NEXT: movl $1893843847, %ecx # imm = 0x70E1C387
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm6, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,9,25,41,u,u,u,u,10,26,42,u,u,u,u,11,27,43,u,u,u,u,12,28,44,u,u,u,u,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,9,25,41,u,u,u,u,10,26,42,u,u,u,u,11,27,43,u,u,u,u,12,28,44,u,u,u,u,13]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <57,9,25,u,u,u,42,58,10,26,u,u,u,43,59,11,27,u,u,u,44,60,12,28,u,u,u,45,61,13,29,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [57,9,25,u,u,u,42,58,10,26,u,u,u,43,59,11,27,u,u,u,44,60,12,28,u,u,u,45,61,13,29,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: movl $-2029118408, %ecx # imm = 0x870E1C38
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -3889,13 +3889,13 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqa 32(%rcx), %ymm9
; AVX2-SLOW-NEXT: vmovdqa 32(%r8), %ymm6
; AVX2-SLOW-NEXT: vmovdqa 32(%r9), %ymm7
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm8, %ymm0, %ymm1
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm2 = ymm10[0,3,2,3,4,7,6,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm2 = ymm2[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,3,u,u,u,4,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,3,u,u,u,4,u,u]
; AVX2-SLOW-NEXT: vpermd %ymm5, %ymm2, %ymm4
; AVX2-SLOW-NEXT: vmovdqa %ymm5, %ymm11
; AVX2-SLOW-NEXT: vpermd %ymm13, %ymm0, %ymm0
@@ -3904,22 +3904,22 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm0, %ymm5, %ymm0
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm3 = ymm9[0,0,0,0,4,5,6,7,8,8,8,8,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,7,7,7,7,8,9,10,11,15,15,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm4, %ymm3, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vpermd %ymm12, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm3 = ymm14[0,0,0,0,4,5,6,7,8,8,8,8,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,7,7,7,7,8,9,10,11,15,15,15,15]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm2, %ymm3, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,3,u,u,u,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,3,u,u,u,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm6, %ymm3, %ymm5
; AVX2-SLOW-NEXT: vmovdqa %ymm6, %ymm12
; AVX2-SLOW-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm0, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm2 = ymm7[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm2 = ymm2[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm5, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vmovdqa (%r8), %ymm5
; AVX2-SLOW-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -3931,10 +3931,10 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm3, %ymm5, %ymm3
; AVX2-SLOW-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-SLOW-NEXT: vmovdqa 32(%rax), %ymm6
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <3,u,u,3,u,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [3,u,u,3,u,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm6, %ymm4, %ymm5
; AVX2-SLOW-NEXT: vmovdqa %ymm6, %ymm13
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm2, %ymm5, %ymm2
; AVX2-SLOW-NEXT: vmovdqa (%rax), %ymm5
; AVX2-SLOW-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -3958,17 +3958,17 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm12[1,2,2,3,5,6,6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm7[2,1,2,3,6,5,6,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm1 = ymm1[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm13, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm13[0,1,2,2,4,5,6,6]
@@ -3986,14 +3986,14 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,1,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpbroadcastd 60(%r8), %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm1 = ymm7[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm13[2,3,3,3,6,7,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
@@ -4013,7 +4013,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,1,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm1, %ymm12
; AVX2-SLOW-NEXT: vmovdqa (%rdi), %xmm0
; AVX2-SLOW-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4037,7 +4037,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,1,1,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,0,1]
; AVX2-SLOW-NEXT: vpbroadcastd 32(%rax), %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm0, %ymm9, %ymm0
; AVX2-SLOW-NEXT: vmovdqa (%r9), %xmm5
; AVX2-SLOW-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4064,7 +4064,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm3 = xmm12[0,1],xmm3[2],xmm12[3,4],xmm3[5],xmm12[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,2,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm2, %ymm3, %ymm9
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm8, %xmm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm7[1,1,2,2]
@@ -4081,7 +4081,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm3[0,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,1,1,3]
; AVX2-SLOW-NEXT: vpbroadcastd 36(%rax), %ymm4
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,7,6]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,3,3]
@@ -4099,7 +4099,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[2,1,2,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,5,4]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm7[4],xmm8[4],xmm7[5],xmm8[5],xmm7[6],xmm8[6],xmm7[7],xmm8[7]
; AVX2-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm13[4],xmm2[4],xmm13[5],xmm2[5],xmm13[6],xmm2[6],xmm13[7],xmm2[7]
@@ -4116,7 +4116,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
; AVX2-SLOW-NEXT: vpbroadcastd 40(%rax), %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm3, %xmm3 # 16-byte Folded Reload
@@ -4142,7 +4142,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6,7,8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,1,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm12 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm12[0,1,2,3,5,6,7,7,8,9,10,11,13,14,15,15]
@@ -4154,7 +4154,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm4 = ymm13[2,3,3,3,6,7,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
@@ -4167,7 +4167,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0],ymm4[1],ymm5[2,3],ymm4[4],ymm5[5,6,7,8],ymm4[9],ymm5[10,11],ymm4[12],ymm5[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm4 = ymm12[3,3,3,3,4,5,6,7,11,11,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm4 = ymm4[0,1,2,3,4,4,4,4,8,9,10,11,12,12,12,12]
@@ -4176,7 +4176,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,3,2]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm13[0,1,2,2,4,5,6,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,1,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -4194,7 +4194,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm5 = ymm5[0,1],ymm6[2],ymm5[3,4],ymm6[5],ymm5[6,7,8,9],ymm6[10],ymm5[11,12],ymm6[13],ymm5[14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm7[1,1,1,1,5,5,5,5]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm7 = ymm10[1,2,3,3,4,5,6,7,9,10,11,11,12,13,14,15]
@@ -4217,7 +4217,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd $212, {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm7 = mem[0,1,1,3,4,5,5,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm6
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm7 = ymm11[0,0,2,1,4,4,6,5]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm9 = ymm12[1,2,2,3,4,5,6,7,9,10,10,11,12,13,14,15]
@@ -4278,7 +4278,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa %ymm6, %ymm8
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [6,7,3,3,7,7,6,7]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm1, %ymm1
@@ -4289,7 +4289,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm10[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,28,29,u,u,28,29,26,27,u,u,30,31,30,31]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm3[0,1],ymm2[2],ymm3[3,4],ymm2[5],ymm3[6,7,8,9],ymm2[10],ymm3[11,12],ymm2[13],ymm3[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
@@ -4301,7 +4301,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm4[1,2,2,3,5,6,6,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7,8,9],ymm1[10],ymm2[11,12],ymm1[13],ymm2[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm7, %ymm4
; AVX2-FAST-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -4318,7 +4318,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm3[0],ymm2[1],ymm3[2,3],ymm2[4],ymm3[5,6,7,8],ymm2[9],ymm3[10,11],ymm2[12],ymm3[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
@@ -4337,7 +4337,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm7[0,1],ymm2[2],ymm7[3,4],ymm2[5],ymm7[6,7,8,9],ymm2[10],ymm7[11,12],ymm2[13],ymm7[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm7
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm6, %ymm1
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm9[1,1,1,1,5,5,5,5]
@@ -4358,7 +4358,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa 32(%rax), %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,4,5,4,5,5,7]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm13, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm10, %ymm3
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
@@ -4373,42 +4373,42 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm8, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm15, %ymm0, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm12, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm1, %ymm3, %ymm7
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm0 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-NEXT: vpermd %ymm14, %ymm1, %ymm2
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm11, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm2, %ymm8, %ymm2
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm1 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm8, %ymm3
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm1, %ymm3, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm7, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm1, %ymm3
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm6, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm3, %ymm8, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm4, %ymm7
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm1, %ymm7, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm7, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm3, %ymm8, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm1, %ymm7, %ymm1
@@ -4425,15 +4425,15 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [5,6,2,3,6,7,5,6]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [4,5,2,2,6,6,6,6]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm1, %ymm1
@@ -4448,14 +4448,14 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm14[3,3,3,3,7,7,7,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpbroadcastd 60(%r8), %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} ymm1 = ymm6[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [6,7,3,3,7,7,6,7]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm1, %ymm1
@@ -4479,7 +4479,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm0, %ymm1, %ymm12
; AVX2-FAST-NEXT: vmovdqa (%rcx), %xmm13
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm9
@@ -4501,7 +4501,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm5, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
; AVX2-FAST-NEXT: vpbroadcastd 32(%rax), %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm1
; AVX2-FAST-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4530,7 +4530,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm11 = xmm12[0,1],xmm11[2],xmm12[3,4],xmm11[5],xmm12[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm3, %ymm11, %ymm3
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm13, %xmm0
; AVX2-FAST-NEXT: vpshufd {{.*#+}} xmm11 = xmm9[1,1,2,2]
@@ -4545,7 +4545,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,1,3]
; AVX2-FAST-NEXT: vpbroadcastd 36(%rax), %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm5, %ymm11, %ymm5
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm6, %xmm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,1,3]
@@ -4561,7 +4561,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,3]
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm3 = xmm3[0,2,3,3,4,5,6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm3, %ymm0
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm15[4],xmm8[4],xmm15[5],xmm8[5],xmm15[6],xmm8[6],xmm15[7],xmm8[7]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm3, %xmm3
@@ -4577,7 +4577,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm9, %xmm4, %xmm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
; AVX2-FAST-NEXT: vpbroadcastd 40(%rax), %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm4, %ymm7, %ymm4
; AVX2-FAST-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm14, %xmm1 # 16-byte Folded Reload
; AVX2-FAST-NEXT: # xmm1 = xmm14[4],mem[4],xmm14[5],mem[5],xmm14[6],mem[6],xmm14[7],mem[7]
@@ -4633,13 +4633,13 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm15, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rcx), %ymm10
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%r8), %ymm7
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm8, %ymm0, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm9, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm6, %ymm3, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm6, %ymm12
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
@@ -4647,21 +4647,21 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm14, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm10, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm5, %ymm2, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%r9), %ymm11
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm13, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm15, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm7, %ymm3, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm0, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm11, %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -4672,10 +4672,10 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm3, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rax), %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm6, %ymm3, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm6, %ymm13
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rax), %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -4698,16 +4698,16 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm7[1,2,2,3,5,6,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm11, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm11[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm13, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm13[0,1,2,2,4,5,6,6]
@@ -4723,14 +4723,14 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,1,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 60(%r8), %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} ymm1 = ymm11[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm13[2,3,3,3,6,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
@@ -4752,7 +4752,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rcx), %xmm14
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdx), %xmm13
@@ -4774,7 +4774,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm5, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 32(%rax), %ymm10
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm10, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %xmm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -4801,7 +4801,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm4 = xmm15[0,1],xmm4[2],xmm15[3,4],xmm4[5],xmm15[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm2, %ymm4, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm14, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm4 = xmm13[1,1,2,2]
@@ -4816,7 +4816,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm5, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,1,3]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 36(%rax), %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,3]
@@ -4833,7 +4833,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,3]
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,2,3,3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm8[4],xmm7[4],xmm8[5],xmm7[5],xmm8[6],xmm7[6],xmm8[7],xmm7[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm1, %xmm1
@@ -4849,7 +4849,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 40(%rax), %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm10, %xmm3 # 16-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # xmm3 = xmm10[4],mem[4],xmm10[5],mem[5],xmm10[6],mem[6],xmm10[7],mem[7]
@@ -4872,7 +4872,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0],ymm3[1],ymm4[2,3],ymm3[4],ymm4[5,6,7,8],ymm3[9],ymm4[10,11],ymm3[12],ymm4[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,1,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm12 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm3 = ymm12[3,3,3,3,7,7,7,7]
@@ -4883,7 +4883,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm4 = ymm14[2,3,3,3,6,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
@@ -4895,7 +4895,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0],ymm4[1],ymm5[2,3],ymm4[4],ymm5[5,6,7,8],ymm4[9],ymm5[10,11],ymm4[12],ymm5[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm4 = ymm13[u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm5 = ymm12[1,2,2,3,5,6,6,7]
@@ -4903,7 +4903,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,3,2]
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm5 = ymm14[0,1,2,2,4,5,6,6]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,1,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
@@ -4921,7 +4921,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0,1],ymm6[2],ymm7[3,4],ymm6[5],ymm7[6,7,8,9],ymm6[10],ymm7[11,12],ymm6[13],ymm7[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm4, %ymm6, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm5, %ymm10, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm6 = ymm8[1,1,1,1,5,5,5,5]
@@ -4942,7 +4942,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd $212, {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm8 = mem[0,1,1,3,4,5,5,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm13, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm8 = ymm12[0,0,2,1,4,4,6,5]
@@ -4988,7 +4988,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufb %ymm9, %ymm1, %ymm0
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm1, %ymm16
; AVX512F-SLOW-NEXT: vmovdqa (%rdx), %ymm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm13, %ymm2, %ymm1
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm2, %ymm29
; AVX512F-SLOW-NEXT: vpor %ymm0, %ymm1, %ymm0
@@ -4998,7 +4998,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufb %ymm12, %ymm1, %ymm0
; AVX512F-SLOW-NEXT: vmovdqa %ymm1, %ymm15
; AVX512F-SLOW-NEXT: vmovdqa (%rdi), %ymm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19]
; AVX512F-SLOW-NEXT: vpshufb %ymm14, %ymm2, %ymm1
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm2, %ymm17
; AVX512F-SLOW-NEXT: vpor %ymm0, %ymm1, %ymm0
@@ -5007,7 +5007,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [128,128,128,128,128,128,128,128,128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128]
; AVX512F-SLOW-NEXT: vpshufb %ymm1, %ymm10, %ymm2
; AVX512F-SLOW-NEXT: vmovdqa (%r8), %ymm11
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm0, %ymm11, %ymm3
; AVX512F-SLOW-NEXT: vpor %ymm2, %ymm3, %ymm2
; AVX512F-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5026,7 +5026,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm6[4],xmm3[4],xmm6[5],xmm3[5],xmm6[6],xmm3[6],xmm6[7],xmm3[7]
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,7,6]
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <16,18,19,19,19,19,u,u,0,1,0,1,2,3,2,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [16,18,19,19,19,19,u,u,0,1,0,1,2,3,2,3]
; AVX512F-SLOW-NEXT: vpermi2d %zmm2, %zmm3, %zmm4
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdi), %ymm3
@@ -5055,7 +5055,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2,3],xmm1[4],xmm2[5,6],xmm1[7]
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm0[0],xmm8[0],xmm0[1],xmm8[1],xmm0[2],xmm8[2],xmm0[3],xmm8[3]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[0,1,3,2,4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,16,u,u,17,17,u,u,0,u,u,1,2,u,u,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,16,u,u,17,17,u,u,0,u,u,1,2,u,u,3]
; AVX512F-SLOW-NEXT: vpermi2d %zmm2, %zmm1, %zmm4
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa (%r9), %xmm1
@@ -5063,7 +5063,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm4 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm5 = xmm4[0,1,2,3,4,5,7,6]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm4 = xmm4[0,1,3,2,4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u]
; AVX512F-SLOW-NEXT: vpermi2d %zmm5, %zmm4, %zmm9
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vprold $16, %ymm10, %ymm4
@@ -5323,7 +5323,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa %ymm1, %ymm10
; AVX512F-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vmovdqa (%rdx), %ymm4
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm1, %ymm4, %ymm3
; AVX512F-FAST-NEXT: vmovdqa64 %ymm4, %ymm18
; AVX512F-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5336,7 +5336,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa %ymm4, %ymm9
; AVX512F-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vmovdqa (%rdi), %ymm6
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm6, %ymm4
; AVX512F-FAST-NEXT: vmovdqa64 %ymm6, %ymm16
; AVX512F-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5346,7 +5346,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [128,128,128,128,128,128,128,128,128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128]
; AVX512F-FAST-NEXT: vpshufb %ymm2, %ymm13, %ymm6
; AVX512F-FAST-NEXT: vmovdqa (%r8), %ymm14
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm4, %ymm14, %ymm7
; AVX512F-FAST-NEXT: vporq %ymm6, %ymm7, %ymm25
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm11, %ymm6
@@ -5366,7 +5366,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa 32(%r8), %ymm6
; AVX512F-FAST-NEXT: vpshufb %ymm4, %ymm6, %ymm1
; AVX512F-FAST-NEXT: vporq %ymm0, %ymm1, %ymm21
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29]
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm11, %ymm1
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm8[3,3,3,3,7,7,7,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3],ymm2[4,5],ymm1[6],ymm2[7,8,9,10],ymm1[11],ymm2[12,13],ymm1[14],ymm2[15]
@@ -5374,7 +5374,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 %ymm11, %ymm30
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm4 = ymm8[2,2,2,2,6,6,6,6]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2,3],ymm2[4],ymm4[5,6,7,8],ymm2[9],ymm4[10,11],ymm2[12],ymm4[13,14,15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = <u,2,2,3,10,u,11,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = [u,2,2,3,10,u,11,u]
; AVX512F-FAST-NEXT: vpermi2q %zmm1, %zmm2, %zmm24
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm10[14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm2 = ymm18[3,3,3,3,7,7,7,7]
@@ -5383,7 +5383,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa 32(%rdx), %xmm5
; AVX512F-FAST-NEXT: vpunpcklwd {{.*#+}} xmm2 = xmm5[0],xmm10[0],xmm5[1],xmm10[1],xmm5[2],xmm10[2],xmm5[3],xmm10[3]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm23 = <u,2,2,3,8,u,9,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm23 = [u,2,2,3,8,u,9,u]
; AVX512F-FAST-NEXT: vpermi2q %zmm2, %zmm1, %zmm23
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm9, %ymm0
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm1 = ymm16[3,3,3,3,7,7,7,7]
@@ -5414,7 +5414,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm2[2],xmm1[3,4],xmm2[5],xmm1[6,7]
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [4,5,2,3,4,5,6,7,8,9,10,11,10,11,8,9]
; AVX512F-FAST-NEXT: vpshufb %xmm7, %xmm4, %xmm2
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,0,u,1,8,8,9,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,0,u,1,8,8,9,u]
; AVX512F-FAST-NEXT: vpermi2q %zmm2, %zmm1, %zmm26
; AVX512F-FAST-NEXT: vmovdqa (%r9), %xmm2
; AVX512F-FAST-NEXT: vmovdqa (%r8), %xmm9
@@ -5440,7 +5440,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm9[4],xmm2[4],xmm9[5],xmm2[5],xmm9[6],xmm2[6],xmm9[7],xmm2[7]
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm25, %zmm0, %zmm4
; AVX512F-FAST-NEXT: vpshufb %xmm11, %xmm2, %xmm2
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = <0,0,1,1,12,13,u,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = [0,0,1,1,12,13,u,15]
; AVX512F-FAST-NEXT: vpermi2q %zmm4, %zmm2, %zmm25
; AVX512F-FAST-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512F-FAST-NEXT: vpbroadcastd 8(%rax), %ymm2
@@ -5462,7 +5462,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm15, %ymm2
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm3 = ymm6[0,0,2,1,4,4,6,5]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm2[0,1,2],ymm3[3],ymm2[4,5],ymm3[6],ymm2[7,8,9,10],ymm3[11],ymm2[12,13],ymm3[14],ymm2[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm29 = <0,1,u,3,10,10,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm29 = [0,1,u,3,10,10,11,11]
; AVX512F-FAST-NEXT: vpermi2q %zmm2, %zmm21, %zmm29
; AVX512F-FAST-NEXT: vmovdqa 32(%rax), %ymm8
; AVX512F-FAST-NEXT: vpshufb %ymm1, %ymm8, %ymm1
@@ -5565,7 +5565,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm14 = ymm14[0,1],ymm15[2],ymm14[3,4],ymm15[5],ymm14[6,7,8,9],ymm15[10],ymm14[11,12],ymm15[13],ymm14[14,15]
; AVX512F-FAST-NEXT: vpermt2q %zmm0, %zmm30, %zmm14
; AVX512F-FAST-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm4, %zmm14
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <6,u,u,u,7,u,u,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [6,u,u,u,7,u,u,7]
; AVX512F-FAST-NEXT: vpermd %ymm9, %ymm0, %ymm0
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm9, %zmm3, %zmm3
; AVX512F-FAST-NEXT: vpermd %zmm3, %zmm21, %zmm3
@@ -5640,7 +5640,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm9 = [0,0,2,34,0,32,0,0,0,3,35,1,33,0,0,0,0,0,2,34,0,32,0,0,0,3,35,1,33,0,0,0]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,5,32,u,u,u,u,11,12,33,u,u,u,u,18,19,34,u,u,u,u,25,26,35,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,5,32,u,u,u,u,11,12,33,u,u,u,u,18,19,34,u,u,u,u,25,26,35,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm9, %zmm10
; AVX512BW-NEXT: movl $236730480, %ecx # imm = 0xE1C3870
; AVX512BW-NEXT: kmovd %ecx, %k2
@@ -5657,7 +5657,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [4,36,0,0,0,7,39,5,37,0,0,0,8,40,6,38,4,36,0,0,0,7,39,5,37,0,0,0,8,40,6,38]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,36,u,u,u,u,7,8,37,u,u,u,u,14,15,38,u,u,u,u,21,22,39,u,u,u,u,28,29,40,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,36,u,u,u,u,7,8,37,u,u,u,u,14,15,38,u,u,u,u,21,22,39,u,u,u,u,28,29,40,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm10, %zmm11
; AVX512BW-NEXT: movl $1893843847, %ecx # imm = 0x70E1C387
; AVX512BW-NEXT: kmovd %ecx, %k3
@@ -5674,7 +5674,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [0,11,43,9,41,0,0,0,12,44,10,42,0,0,0,13,0,11,43,9,41,0,0,0,12,44,10,42,0,0,0,13]
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,3,4,41,u,u,u,u,10,11,42,u,u,u,u,17,18,43,u,u,u,u,24,25,44,u,u,u,u,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,3,4,41,u,u,u,u,10,11,42,u,u,u,u,17,18,43,u,u,u,u,24,25,44,u,u,u,u,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm11, %zmm12
; AVX512BW-NEXT: movl $-2029118408, %ecx # imm = 0x870E1C38
; AVX512BW-NEXT: kmovd %ecx, %k3
@@ -5689,7 +5689,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [13,0,0,0,48,16,46,14,0,0,0,49,17,47,15,0,13,0,0,0,48,16,46,14,0,0,0,49,17,47,15,0]
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm2, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,45,u,u,u,u,6,7,46,u,u,u,u,13,14,47,u,u,u,u,20,21,48,u,u,u,u,27,28,49,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,45,u,u,u,u,6,7,46,u,u,u,u,13,14,47,u,u,u,u,20,21,48,u,u,u,u,27,28,49,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm11, %zmm13
; AVX512BW-NEXT: movl $946921923, %ecx # imm = 0x3870E1C3
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -5704,7 +5704,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [20,52,18,50,0,0,0,21,53,19,51,0,0,0,22,54,20,52,18,50,0,0,0,21,53,19,51,0,0,0,22,54]
; AVX512BW-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,2,3,50,u,u,u,u,9,10,51,u,u,u,u,16,17,52,u,u,u,u,23,24,53,u,u,u,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,2,3,50,u,u,u,u,9,10,51,u,u,u,u,16,17,52,u,u,u,u,23,24,53,u,u,u,u,30,31]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm11, %zmm14
; AVX512BW-NEXT: movl $-1014559204, %ecx # imm = 0xC3870E1C
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -5721,7 +5721,7 @@ define void @store_i16_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [0,0,0,25,57,23,55,0,0,0,26,58,24,56,0,0,0,0,0,25,57,23,55,0,0,0,26,58,24,56,0,0]
; AVX512BW-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm2, %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <54,u,u,u,u,5,6,55,u,u,u,u,12,13,56,u,u,u,u,19,20,57,u,u,u,u,26,27,58,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [54,u,u,u,u,5,6,55,u,u,u,u,12,13,56,u,u,u,u,19,20,57,u,u,u,u,26,27,58,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm0, %zmm3, %zmm1
; AVX512BW-NEXT: movl $473460961, %ecx # imm = 0x1C3870E1
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -8342,14 +8342,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa 64(%rax), %ymm6
; AVX2-SLOW-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,3,u,u,u,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,3,u,u,u,4,u]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm11
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm1[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm2 = ymm1[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm0, %ymm2, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <3,u,u,3,u,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [3,u,u,3,u,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm11, %ymm2
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm3 = ymm8[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
@@ -8359,7 +8359,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm4 = ymm4[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vpermd %ymm9, %ymm10, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm5
; AVX2-SLOW-NEXT: vpermd %ymm7, %ymm10, %ymm0
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm0, %ymm1
@@ -8371,10 +8371,10 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm2 = ymm2[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vmovdqa (%rdi), %ymm3
; AVX2-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <3,u,u,u,4,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [3,u,u,u,4,u,u,4]
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm7
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm3, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm6
; AVX2-SLOW-NEXT: vmovdqa (%rcx), %ymm3
@@ -8383,12 +8383,12 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,7,7,7,7,8,9,10,11,15,15,15,15]
; AVX2-SLOW-NEXT: vmovdqa (%rdx), %ymm4
; AVX2-SLOW-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,3,u,u,u,4,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,3,u,u,u,4,u,u]
; AVX2-SLOW-NEXT: vpermd %ymm4, %ymm8, %ymm4
; AVX2-SLOW-NEXT: vmovdqa %ymm8, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm4, %ymm3, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vmovdqa %ymm4, %ymm10
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255]
@@ -8450,20 +8450,20 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7,8,9],ymm1[10],ymm2[11,12],ymm1[13],ymm2[14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqa 96(%r8), %ymm2
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm2[0,0,2,1,4,4,6,5]
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm7
; AVX2-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqa 96(%r9), %ymm2
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm1 = ymm2[1,1,2,2,4,5,6,7,9,9,10,10,12,13,14,15]
; AVX2-SLOW-NEXT: vmovdqa %ymm2, %ymm8
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqa 96(%rax), %ymm2
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm2[0,1,1,3,4,5,5,7]
@@ -8484,17 +8484,17 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm7[1,2,2,3,5,6,6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm8[2,1,2,3,6,5,6,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm1 = ymm1[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm9, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm9[0,1,2,2,4,5,6,6]
@@ -8512,14 +8512,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,1,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpbroadcastd 124(%r8), %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm1 = ymm8[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm1 = ymm9[2,3,3,3,6,7,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
@@ -8543,7 +8543,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,1,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm0[0,1,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm13
; AVX2-SLOW-NEXT: vmovdqa 32(%rsi), %xmm2
; AVX2-SLOW-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -8605,7 +8605,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,1,1,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
; AVX2-SLOW-NEXT: vpbroadcastd (%rax), %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vmovdqa 32(%r9), %xmm0
; AVX2-SLOW-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -8661,7 +8661,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm6[0,1],xmm1[2],xmm6[3,4],xmm1[5],xmm6[6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm0[0,0,1,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm6, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -8708,7 +8708,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[0,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,1,3]
; AVX2-SLOW-NEXT: vpbroadcastd 4(%rax), %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm4, %ymm8, %ymm4
; AVX2-SLOW-NEXT: vpshufhw $180, {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Folded Reload
; AVX2-SLOW-NEXT: # xmm5 = mem[0,1,2,3,4,5,7,6]
@@ -8745,7 +8745,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[2,1,2,3,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,5,4]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm1 # 16-byte Folded Reload
@@ -8784,7 +8784,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufb %xmm8, %xmm4, %xmm4
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
; AVX2-SLOW-NEXT: vpbroadcastd 8(%rax), %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm5, %xmm5 # 16-byte Folded Reload
@@ -8817,32 +8817,32 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm6, %ymm3, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,u,4,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,u,4,u,u,4]
; AVX2-SLOW-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm1 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vpshufd $236, {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm2 = mem[0,3,2,3,4,7,6,7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm2 = ymm2[0,0,3,3,4,5,6,7,8,8,11,11,12,13,14,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,3,u,u,u,4,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,3,u,u,u,4,u,u]
; AVX2-SLOW-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm2 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vpshuflw $0, {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm3 = mem[0,0,0,0,4,5,6,7,8,8,8,8,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,7,7,7,7,8,9,10,11,15,15,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm2, %ymm3, %ymm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,3,u,u,u,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,3,u,u,u,4,u]
; AVX2-SLOW-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm2 # 32-byte Folded Reload
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vpshufd $196, {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm2 = mem[0,1,0,3,4,5,4,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm2 = ymm2[0,1,2,3,4,4,7,7,8,9,10,11,12,12,15,15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,3,u,u,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,3,u,u,u,4]
; AVX2-SLOW-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm0, %ymm2 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm0
@@ -8861,7 +8861,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0,1],ymm3[2],ymm4[3,4],ymm3[5],ymm4[6,7,8,9],ymm3[10],ymm4[11,12],ymm3[13],ymm4[14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm2, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshuflw $249, {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Folded Reload
@@ -8906,7 +8906,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm10 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm6 = ymm10[0,1,1,3,4,5,5,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm6 = ymm11[1,2,2,3,4,5,6,7,9,10,10,11,12,13,14,15]
@@ -8952,7 +8952,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm6 = ymm7[0],ymm6[1],ymm7[2,3],ymm6[4],ymm7[5,6,7,8],ymm6[9],ymm7[10,11],ymm6[12],ymm7[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm5 # 32-byte Reload
@@ -8981,7 +8981,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1],ymm9[2,3],ymm8[4],ymm9[5,6,7,8],ymm8[9],ymm9[10,11],ymm8[12],ymm9[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshuflw $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm8 = mem[3,3,3,3,4,5,6,7,11,11,11,11,12,13,14,15]
@@ -8992,7 +8992,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,1,3,2]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm9 = ymm10[0,1,2,2,4,5,6,6]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,1,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm9 = ymm11[3,3,3,3,4,5,6,7,11,11,11,11,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm9 = ymm9[0,1,2,3,4,4,4,4,8,9,10,11,12,12,12,12]
@@ -9032,7 +9032,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpblendw {{.*#+}} ymm9 = ymm9[0],ymm10[1],ymm9[2,3],ymm10[4],ymm9[5,6,7,8],ymm10[9],ymm9[10,11],ymm10[12],ymm9[13,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,1,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,2,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vpshufd $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm9 = mem[3,3,3,3,7,7,7,7]
@@ -9071,7 +9071,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-SLOW-NEXT: vpshufd $254, {{[-0-9]+}}(%r{{[sb]}}p), %ymm12 # 32-byte Folded Reload
; AVX2-SLOW-NEXT: # ymm12 = mem[2,3,3,3,6,7,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm12 = ymm1[3,3,3,3,7,7,7,7]
; AVX2-SLOW-NEXT: vpshufhw $249, {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Folded Reload
@@ -9165,31 +9165,31 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa 96(%r8), %ymm3
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm5, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = ymm4[0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa 96(%r9), %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-FAST-NEXT: vmovdqa 96(%rax), %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
@@ -9208,7 +9208,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm2, %ymm3
; AVX2-FAST-NEXT: vmovdqa %ymm4, %ymm12
; AVX2-FAST-NEXT: vmovdqa %ymm2, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm3, %ymm3
; AVX2-FAST-NEXT: vmovdqa 32(%r8), %ymm6
; AVX2-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -9254,7 +9254,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm2 = ymm5[0,1],ymm2[2],ymm5[3,4],ymm2[5],ymm5[6,7,8,9],ymm2[10],ymm5[11,12],ymm2[13],ymm5[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255]
@@ -9298,7 +9298,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm4, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u]
; AVX2-FAST-NEXT: vmovdqa %ymm10, %ymm5
; AVX2-FAST-NEXT: vmovdqu %ymm10, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm10, %ymm1
@@ -9311,7 +9311,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa %ymm12, %ymm8
; AVX2-FAST-NEXT: vmovdqu %ymm12, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpermd %ymm12, %ymm9, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm2, %ymm12
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm7, %ymm2
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
@@ -9338,7 +9338,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm13
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm14 = ymm15[2,2,2,2,6,6,6,6]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm4[0,1],ymm14[2],ymm4[3,4],ymm14[5],ymm4[6,7,8,9],ymm14[10],ymm4[11,12],ymm14[13],ymm4[14,15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm1, %ymm0, %ymm14
; AVX2-FAST-NEXT: vmovdqa %ymm1, %ymm0
@@ -9347,7 +9347,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm14 = ymm15[0],ymm14[1],ymm15[2,3],ymm14[4],ymm15[5,6,7,8],ymm14[9],ymm15[10,11],ymm14[12],ymm15[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm4, %ymm14, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm4, %ymm12, %ymm1
@@ -9364,7 +9364,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm15[0],ymm4[1],ymm15[2,3],ymm4[4],ymm15[5,6,7,8],ymm4[9],ymm15[10,11],ymm4[12],ymm15[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm1, %ymm4, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -9392,7 +9392,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [6,7,3,3,7,7,6,7]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm5, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm0, %ymm0
@@ -9408,14 +9408,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm5, %ymm14
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm2, %ymm14, %ymm2
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm3 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm3, %ymm4
; AVX2-FAST-NEXT: vmovdqa %ymm5, %ymm6
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufd {{.*#+}} ymm14 = ymm8[3,3,3,3,7,7,7,7]
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm14[0,1,2],ymm4[3],ymm14[4,5],ymm4[6],ymm14[7,8,9,10],ymm4[11],ymm14[12,13],ymm4[14],ymm14[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm2, %ymm14
; AVX2-FAST-NEXT: vpshufd $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm15 # 32-byte Folded Reload
@@ -9423,7 +9423,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm14 = ymm15[0],ymm14[1],ymm15[2,3],ymm14[4],ymm15[5,6,7,8],ymm14[9],ymm15[10,11],ymm14[12],ymm15[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,1,3,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm4, %ymm14, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0,255,255,255,255,255,255,255,255,0,0,0,0,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm4, %ymm1, %ymm1
@@ -9457,11 +9457,11 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm3, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpermd %ymm14, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa %ymm2, %ymm4
@@ -9470,12 +9470,12 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm4, %ymm2
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm7, %ymm3
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm2, %ymm3, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm5, %ymm3 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm5, %ymm4
; AVX2-FAST-NEXT: vmovdqa %ymm5, %ymm6
@@ -9484,16 +9484,16 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm6, %ymm5
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm15, %ymm6
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm5, %ymm6, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm0
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm1, %ymm4, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm2, %ymm5, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm6, %ymm3 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm4, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm6, %ymm4 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vmovdqa %ymm6, %ymm7
@@ -9504,9 +9504,9 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm6, %ymm7, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm8, %ymm6 # 32-byte Folded Reload
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm3, %ymm6, %ymm3
; AVX2-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %ymm8, %ymm6 # 32-byte Folded Reload
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm4, %ymm6, %ymm4
@@ -9531,17 +9531,17 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0,1],ymm1[2],ymm2[3,4],ymm1[5],ymm2[6,7,8,9],ymm1[10],ymm2[11,12],ymm1[13],ymm2[14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [4,4,2,1,6,5,6,5]
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
; AVX2-FAST-NEXT: vpermd %ymm7, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Reload
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} ymm1 = ymm8[1,1,2,2,4,5,6,7,9,9,10,10,12,13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,4,5,4,5,5,7]
@@ -9557,15 +9557,15 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [5,6,2,3,6,7,5,6]
; AVX2-FAST-NEXT: vpermd %ymm7, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [4,5,2,2,6,6,6,6]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm1, %ymm1
@@ -9580,14 +9580,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,1,3,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpbroadcastd 124(%r8), %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} ymm1 = ymm8[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [6,7,3,3,7,7,6,7]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm1, %ymm1
@@ -9613,7 +9613,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm1[0,1,1,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm2, %ymm3, %ymm10
; AVX2-FAST-NEXT: vmovdqa 32(%rsi), %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -9673,7 +9673,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
; AVX2-FAST-NEXT: vpbroadcastd (%rax), %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm6
; AVX2-FAST-NEXT: vmovdqa 32(%r9), %xmm0
; AVX2-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -9723,7 +9723,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendw {{.*#+}} xmm2 = xmm4[0,1],xmm2[2],xmm4[3,4],xmm2[5],xmm4[6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm13, %xmm0
@@ -9768,7 +9768,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm7, %xmm7
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,1,3]
; AVX2-FAST-NEXT: vpbroadcastd 4(%rax), %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm7, %ymm11, %ymm7
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm12, %xmm11
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,1,1,3]
@@ -9798,7 +9798,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,3]
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[0,2,3,3,4,5,6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,2,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm13[4],xmm10[4],xmm13[5],xmm10[5],xmm13[6],xmm10[6],xmm13[7],xmm10[7]
; AVX2-FAST-NEXT: vmovdqa (%rsp), %xmm3 # 16-byte Reload
@@ -9835,7 +9835,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpshufb %xmm9, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,1,1]
; AVX2-FAST-NEXT: vpbroadcastd 8(%rax), %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm0, %xmm6 # 16-byte Folded Reload
@@ -9930,32 +9930,32 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rcx), %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%r8), %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <3,u,u,u,4,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [3,u,u,u,4,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm14, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [0,1,0,1,14,15,14,15,8,9,10,11,12,13,14,15,16,17,16,17,30,31,30,31,24,25,26,27,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm15, %ymm2, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm1, %ymm4, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,3,u,u,u,4,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,3,u,u,u,4,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm10, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [0,1,0,1,0,1,0,1,14,15,14,15,14,15,14,15,16,17,16,17,16,17,16,17,30,31,30,31,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm5, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm4, %ymm6, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm1, %ymm4, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm7, %ymm0, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm1, %ymm4, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%r9), %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [0,1,2,3,4,5,6,7,0,1,0,1,14,15,14,15,16,17,18,19,20,21,22,23,16,17,16,17,30,31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm4, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm1, %ymm6, %ymm1
; AVX2-FAST-PERLANE-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX2-FAST-PERLANE-NEXT: vmovdqa 96(%rax), %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm6, %ymm0, %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255,255,255,255,255,255,255,255,255,255,255,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm1
@@ -10009,10 +10009,10 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,3,u,u,u,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,3,u,u,u,4,u]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm1, %ymm12, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm2, %ymm10
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u,u,u,u,u,u,u,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm8, %ymm10, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%r8), %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -10031,9 +10031,9 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm12, %ymm9, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rax), %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <3,u,u,3,u,u,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [3,u,u,3,u,u,u,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm1, %ymm2, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm8, %ymm11, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rax), %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -10061,16 +10061,16 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm7[0,1],ymm1[2],ymm7[3,4],ymm1[5],ymm7[6,7,8,9],ymm1[10],ymm7[11,12],ymm1[13],ymm7[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm10 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm10[0,0,2,1,4,4,6,5]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} ymm1 = ymm4[1,1,2,2,4,5,6,7,9,9,10,10,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm6[0,1,1,3,4,5,5,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
@@ -10085,15 +10085,15 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm7[0],ymm1[1],ymm7[2,3],ymm1[4],ymm7[5,6,7,8],ymm1[9],ymm7[10,11],ymm1[12],ymm7[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm10[1,2,2,3,5,6,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm6[0,1,2,2,4,5,6,6]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,3]
@@ -10108,14 +10108,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm1 = ymm2[0],ymm1[1],ymm2[2,3],ymm1[4],ymm2[5,6,7,8],ymm1[9],ymm2[10,11],ymm1[12],ymm2[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,1,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 124(%r8), %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u,255,255,255,255,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufhw {{.*#+}} ymm1 = ymm4[0,1,2,3,5,5,7,6,8,9,10,11,13,13,15,14]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[3,3,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u,255,255,255,255,255,255,255,255,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm1 = ymm6[2,3,3,3,6,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,1,3,2]
@@ -10139,7 +10139,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,2,3,0,1,2,3,6,7,4,5,6,7,4,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm1[0,1,1,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm1, %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rsi), %xmm3
@@ -10198,7 +10198,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm9, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd (%rax), %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm1, %ymm2, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%r9), %xmm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -10249,7 +10249,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} xmm4 = xmm6[0,1],xmm4[2],xmm6[3,4],xmm4[5],xmm6[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm0, %ymm4, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm14, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm4 = xmm11[1,1,2,2]
@@ -10291,7 +10291,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm9, %xmm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,1,3]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 4(%rax), %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm7, %ymm9, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm8, %xmm8
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,1,1,3]
@@ -10325,7 +10325,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,3]
; AVX2-FAST-PERLANE-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,2,3,3,4,5,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,2,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm14[4],xmm1[5],xmm14[5],xmm1[6],xmm14[6],xmm1[7],xmm14[7]
@@ -10362,7 +10362,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm4, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vpbroadcastd 8(%rax), %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpckhwd {{[-0-9]+}}(%r{{[sb]}}p), %xmm5, %xmm5 # 16-byte Folded Reload
@@ -10409,7 +10409,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm5 = ymm6[0,1],ymm5[2],ymm6[3,4],ymm5[5],ymm6[6,7,8,9],ymm5[10],ymm6[11,12],ymm5[13],ymm6[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm4, %ymm5, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -10450,7 +10450,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm9 = ymm13[0,1,1,3,4,5,5,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm1, %ymm9
@@ -10486,7 +10486,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd $170, {{[-0-9]+}}(%r{{[sb]}}p), %ymm8 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm8 = mem[2,2,2,2,6,6,6,6]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm7 = ymm7[0,1],ymm8[2],ymm7[3,4],ymm8[5],ymm7[6,7,8,9],ymm8[10],ymm7[11,12],ymm8[13],ymm7[14,15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm1, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpshufd $170, {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Folded Reload
@@ -10494,7 +10494,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1],ymm9[2,3],ymm8[4],ymm9[5,6,7,8],ymm8[9],ymm9[10,11],ymm8[12],ymm9[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = <255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = [255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u,u,u,u,u,0,0,0,0,255,255,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm1, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm7, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm7 # 32-byte Reload
@@ -10522,14 +10522,14 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm1, %ymm9, %ymm10, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,u,u,u,u,u,u,8,9,u,u,u,u,u,u,u,u,22,23,u,u,u,u,24,25,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm10, %ymm6, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm12 = ymm14[1,2,2,3,5,6,6,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm11 = ymm12[0,1],ymm11[2],ymm12[3,4],ymm11[5],ymm12[6,7,8,9],ymm11[10],ymm12[11,12],ymm11[13],ymm12[14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,1,3,2]
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm12 = ymm13[0,1,2,2,4,5,6,6]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,1,3,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm4 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm10, %ymm4, %ymm12
@@ -10553,21 +10553,21 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm8, %ymm12, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm9, %ymm10, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm1, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpshufd $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm11 = mem[3,3,3,3,7,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm10 = ymm11[0,1,2],ymm10[3],ymm11[4,5],ymm10[6],ymm11[7,8,9,10],ymm10[11],ymm11[12,13],ymm10[14],ymm11[15]
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm13, %ymm1, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufd $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm12 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm12 = mem[3,3,3,3,7,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm11 = ymm12[0],ymm11[1],ymm12[2,3],ymm11[4],ymm12[5,6,7,8],ymm11[9],ymm12[10,11],ymm11[12],ymm12[13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,1,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,2,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u,255,255,255,255,0,0,0,0,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm1, %ymm11
@@ -10604,7 +10604,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-PERLANE-NEXT: vpshufd $254, {{[-0-9]+}}(%r{{[sb]}}p), %ymm15 # 32-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # ymm15 = mem[2,3,3,3,6,7,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm15[2,1,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0,u,u,u,u,u,u,u,u,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm14, %ymm15, %ymm14
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm13, %ymm4, %ymm15
; AVX2-FAST-PERLANE-NEXT: vpshufd $255, {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Folded Reload
@@ -10692,12 +10692,12 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128,128,128,128,128]
; AVX512F-SLOW-NEXT: vpshufb %ymm0, %ymm2, %ymm1
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm2, %ymm18
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm11, %ymm9, %ymm2
; AVX512F-SLOW-NEXT: vporq %ymm1, %ymm2, %ymm16
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-SLOW-NEXT: vpshufb %ymm12, %ymm8, %ymm1
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19]
; AVX512F-SLOW-NEXT: vpshufb %ymm13, %ymm7, %ymm2
; AVX512F-SLOW-NEXT: vporq %ymm1, %ymm2, %ymm17
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [128,128,128,128,128,128,128,128,128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128]
@@ -10706,7 +10706,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa %ymm3, %ymm10
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm2, %ymm21
; AVX512F-SLOW-NEXT: vmovdqa 64(%r8), %ymm3
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm15, %ymm3, %ymm2
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm3, %ymm20
; AVX512F-SLOW-NEXT: vpor %ymm1, %ymm2, %ymm1
@@ -10996,7 +10996,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,7,6]
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [0,1,2,3,4,5,4,5,6,7,10,11,8,9,10,11]
; AVX512F-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm1
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <16,18,19,19,19,19,u,u,0,1,0,1,2,3,2,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [16,18,19,19,19,19,u,u,0,1,0,1,2,3,2,3]
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm29, %zmm1
; AVX512F-SLOW-NEXT: vpbroadcastd 100(%rax), %ymm2
; AVX512F-SLOW-NEXT: vpbroadcastd 104(%rax), %ymm3
@@ -11013,7 +11013,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[1,1,2,2]
; AVX512F-SLOW-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0],xmm1[1],xmm2[2,3],xmm1[4],xmm2[5,6],xmm1[7]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm2 = xmm3[0,1,3,2,4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <16,16,17,17,17,17,u,u,0,1,0,1,2,3,2,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [16,16,17,17,17,17,u,u,0,1,0,1,2,3,2,3]
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm3, %zmm1
; AVX512F-SLOW-NEXT: vmovdqa 64(%rdi), %xmm2
; AVX512F-SLOW-NEXT: vmovdqa 64(%rsi), %xmm4
@@ -11037,7 +11037,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm1[0,1,2,3,4,5,7,6]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,1,3,2,4,5,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,0,1,0,1,1,3,16,18,19,19,19,19,u,u]
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm6, %zmm1
; AVX512F-SLOW-NEXT: vpbroadcastd 64(%rax), %ymm2
; AVX512F-SLOW-NEXT: vpbroadcastd 68(%rax), %ymm5
@@ -11395,7 +11395,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpshufb %ymm15, %ymm0, %ymm2
; AVX512F-FAST-NEXT: vmovdqa64 %ymm0, %ymm20
; AVX512F-FAST-NEXT: vmovdqa 96(%rdx), %ymm9
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm9, %ymm3
; AVX512F-FAST-NEXT: vmovdqa %ymm0, %ymm10
; AVX512F-FAST-NEXT: vporq %ymm2, %ymm3, %ymm16
@@ -11403,7 +11403,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128,128,128,128,128,128,128,128,128]
; AVX512F-FAST-NEXT: vpshufb %ymm14, %ymm6, %ymm2
; AVX512F-FAST-NEXT: vmovdqa 96(%rdi), %ymm7
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u,u,u,u,u,16,17,18,19]
; AVX512F-FAST-NEXT: vpshufb %ymm11, %ymm7, %ymm3
; AVX512F-FAST-NEXT: vporq %ymm2, %ymm3, %ymm17
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [128,128,128,128,128,128,128,128,128,128,128,128,14,15,128,128,128,128,128,128,128,128,128,128,128,128,16,17,128,128,128,128]
@@ -11412,7 +11412,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm0, %ymm2
; AVX512F-FAST-NEXT: vmovdqa 64(%r8), %ymm0
; AVX512F-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,12,13,14,15,128,128,u,u,u,u,u,u,u,u,u,u,16,17,128,128,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm0, %ymm4
; AVX512F-FAST-NEXT: vporq %ymm2, %ymm4, %ymm18
; AVX512F-FAST-NEXT: vmovdqa 64(%rcx), %ymm0
@@ -11495,7 +11495,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm12 = [65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535]
; AVX512F-FAST-NEXT: vpternlogq $248, %ymm12, %ymm0, %ymm1
; AVX512F-FAST-NEXT: vextracti64x4 $1, %zmm3, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,4,u,u,u,5,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,4,u,u,u,5,u,u]
; AVX512F-FAST-NEXT: vpermd %ymm11, %ymm3, %ymm3
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = [65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0,65535,65535,65535,65535,65535,65535,0]
; AVX512F-FAST-NEXT: vpternlogq $184, %ymm0, %ymm17, %ymm3
@@ -11517,19 +11517,19 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 %ymm5, %ymm20
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm3, %zmm1
; AVX512F-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,u,14,15,12,13,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29]
; AVX512F-FAST-NEXT: vpshufb %ymm3, %ymm6, %ymm1
; AVX512F-FAST-NEXT: vmovdqa64 %ymm3, %ymm26
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm3 = ymm7[3,3,3,3,7,7,7,7]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm3[0,1,2],ymm1[3],ymm3[4,5],ymm1[6],ymm3[7,8,9,10],ymm1[11],ymm3[12,13],ymm1[14],ymm3[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [10,11,8,9,u,u,u,u,u,u,u,u,u,u,u,u,26,27,24,25,u,u,u,u,26,27,24,25,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm6, %ymm3
; AVX512F-FAST-NEXT: vmovdqa64 %ymm5, %ymm29
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm6 = ymm7[2,2,2,2,6,6,6,6]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm6[0],ymm3[1],ymm6[2,3],ymm3[4],ymm6[5,6,7,8],ymm3[9],ymm6[10,11],ymm3[12],ymm6[13,14,15]
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,2,2,3,10,9,11,11]
; AVX512F-FAST-NEXT: vpermt2q %zmm1, %zmm6, %zmm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [14,15,12,13,u,u,u,u,u,u,u,u,u,u,u,u,30,31,28,29,u,u,u,u,30,31,28,29,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm7, %ymm2, %ymm1
; AVX512F-FAST-NEXT: vmovdqa64 %ymm7, %ymm27
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm7 = ymm9[3,3,3,3,7,7,7,7]
@@ -11546,7 +11546,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpternlogq $226, %zmm3, %zmm1, %zmm7
; AVX512F-FAST-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqa64 %zmm1, %zmm24
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <5,u,u,u,6,u,u,6>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [5,u,u,u,6,u,u,6]
; AVX512F-FAST-NEXT: vpermd %ymm11, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vmovdqa64 64(%rax), %zmm12
@@ -11594,7 +11594,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpshufb %ymm7, %ymm14, %ymm0
; AVX512F-FAST-NEXT: vpshufd {{.*#+}} ymm3 = ymm13[0,0,2,1,4,4,6,5]
; AVX512F-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0,1,2],ymm3[3],ymm0[4,5],ymm3[6],ymm0[7,8,9,10],ymm3[11],ymm0[12,13],ymm3[14],ymm0[15]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,u,3,10,10,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,u,3,10,10,11,11]
; AVX512F-FAST-NEXT: vpermi2q %zmm0, %zmm21, %zmm3
; AVX512F-FAST-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqu %ymm8, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -12057,7 +12057,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movl $101455920, %ecx # imm = 0x60C1830
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm0, %zmm3 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,59,u,u,4,5,6,7,60,u,u,11,12,13,14,61,u,u,18,19,20,21,62,u,u,25,26,27,28,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,59,u,u,4,5,6,7,60,u,u,11,12,13,14,61,u,u,18,19,20,21,62,u,u,25,26,27,28,63,u,u]
; AVX512BW-NEXT: vpermi2w %zmm30, %zmm3, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm24 = [0,25,57,23,55,0,0,0,26,58,24,56,0,0,0,27,0,25,57,23,55,0,0,0,26,58,24,56,0,0,0,27]
@@ -12071,7 +12071,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movl $-2096755688, %ecx # imm = 0x83060C18
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqu16 %zmm0, %zmm3 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,1,2,3,4,55,u,u,8,9,10,11,56,u,u,15,16,17,18,57,u,u,22,23,24,25,58,u,u,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,1,2,3,4,55,u,u,8,9,10,11,56,u,u,15,16,17,18,57,u,u,22,23,24,25,58,u,u,29,30,31]
; AVX512BW-NEXT: vpermi2w %zmm30, %zmm3, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, (%rsp) # 64-byte Spill
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm19 = [2,34,0,32,0,0,0,3,35,1,33,0,0,0,4,36,2,34,0,32,0,0,0,3,35,1,33,0,0,0,4,36]
@@ -12103,7 +12103,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2w %zmm13, %zmm19, %zmm25
; AVX512BW-NEXT: vmovdqa64 %zmm29, %zmm19
; AVX512BW-NEXT: vpermt2w %zmm9, %zmm28, %zmm29
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,4,5,32,u,u,u,u,11,12,33,u,u,u,u,18,19,34,u,u,u,u,25,26,35,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,4,5,32,u,u,u,u,11,12,33,u,u,u,u,18,19,34,u,u,u,u,25,26,35,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm20, %zmm2, %zmm0
; AVX512BW-NEXT: movl $236730480, %ecx # imm = 0xE1C3870
; AVX512BW-NEXT: vmovdqu16 %zmm25, %zmm29 {%k3}
@@ -12127,7 +12127,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm2 = [61,27,59,0,0,0,30,62,28,60,0,0,0,31,63,29,61,27,59,0,0,0,30,62,28,60,0,0,0,31,63,29]
; AVX512BW-NEXT: # zmm2 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm4, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,59,u,u,u,u,8,9,60,u,u,u,u,15,16,61,u,u,u,u,22,23,62,u,u,u,u,29,30,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,59,u,u,u,u,8,9,60,u,u,u,u,15,16,61,u,u,u,u,22,23,62,u,u,u,u,29,30,63]
; AVX512BW-NEXT: vpermi2w %zmm20, %zmm2, %zmm21
; AVX512BW-NEXT: movl $-507279602, %eax # imm = 0xE1C3870E
; AVX512BW-NEXT: kmovd %eax, %k3
@@ -12147,7 +12147,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [0,0,0,25,57,23,55,0,0,0,26,58,24,56,0,0,0,0,0,25,57,23,55,0,0,0,26,58,24,56,0,0]
; AVX512BW-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm4, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = <54,u,u,u,u,5,6,55,u,u,u,u,12,13,56,u,u,u,u,19,20,57,u,u,u,u,26,27,58,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [54,u,u,u,u,5,6,55,u,u,u,u,12,13,56,u,u,u,u,19,20,57,u,u,u,u,26,27,58,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm20, %zmm3, %zmm24
; AVX512BW-NEXT: movl $473460961, %eax # imm = 0x1C3870E1
; AVX512BW-NEXT: kmovd %eax, %k2
@@ -12155,7 +12155,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm3 = [20,52,18,50,0,0,0,21,53,19,51,0,0,0,22,54,20,52,18,50,0,0,0,21,53,19,51,0,0,0,22,54]
; AVX512BW-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2w %zmm26, %zmm3, %zmm22
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = <u,u,2,3,50,u,u,u,u,9,10,51,u,u,u,u,16,17,52,u,u,u,u,23,24,53,u,u,u,u,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [u,u,2,3,50,u,u,u,u,9,10,51,u,u,u,u,16,17,52,u,u,u,u,23,24,53,u,u,u,u,30,31]
; AVX512BW-NEXT: vpermt2w %zmm25, %zmm24, %zmm22
; AVX512BW-NEXT: movl $-1014559204, %eax # imm = 0xC3870E1C
; AVX512BW-NEXT: kmovd %eax, %k2
@@ -12176,7 +12176,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2w %zmm12, %zmm5, %zmm2
; AVX512BW-NEXT: vpermi2w %zmm15, %zmm14, %zmm3
; AVX512BW-NEXT: vmovdqu16 %zmm2, %zmm3 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,45,u,u,u,u,6,7,46,u,u,u,u,13,14,47,u,u,u,u,20,21,48,u,u,u,u,27,28,49,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,45,u,u,u,u,6,7,46,u,u,u,u,13,14,47,u,u,u,u,20,21,48,u,u,u,u,27,28,49,u,u]
; AVX512BW-NEXT: vpermt2w %zmm25, %zmm2, %zmm10
; AVX512BW-NEXT: movl $946921923, %eax # imm = 0x3870E1C3
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -12196,7 +12196,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2w %zmm14, %zmm15, %zmm23
; AVX512BW-NEXT: vpermi2w %zmm12, %zmm5, %zmm0
; AVX512BW-NEXT: vmovdqu16 %zmm23, %zmm0 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,3,4,41,u,u,u,u,10,11,42,u,u,u,u,17,18,43,u,u,u,u,24,25,44,u,u,u,u,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,3,4,41,u,u,u,u,10,11,42,u,u,u,u,17,18,43,u,u,u,u,24,25,44,u,u,u,u,31]
; AVX512BW-NEXT: vpermt2w %zmm25, %zmm10, %zmm1
; AVX512BW-NEXT: movl $-2029118408, %eax # imm = 0x870E1C38
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -12218,7 +12218,7 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm1 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermt2w %zmm8, %zmm1, %zmm4
; AVX512BW-NEXT: vpermt2w %zmm26, %zmm1, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,36,u,u,u,u,7,8,37,u,u,u,u,14,15,38,u,u,u,u,21,22,39,u,u,u,u,28,29,40,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,36,u,u,u,u,7,8,37,u,u,u,u,14,15,38,u,u,u,u,21,22,39,u,u,u,u,28,29,40,u]
; AVX512BW-NEXT: vpermt2w %zmm20, %zmm1, %zmm4
; AVX512BW-NEXT: vpermt2w %zmm25, %zmm1, %zmm28
; AVX512BW-NEXT: vmovdqu16 %zmm5, %zmm14 {%k3}
@@ -12226,10 +12226,10 @@ define void @store_i16_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqu16 %zmm28, %zmm19 {%k1}
; AVX512BW-NEXT: vmovdqu16 %zmm4, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,59,u,4,5,6,7,8,60,u,11,12,13,14,15,61,u,18,19,20,21,22,62,u,25,26,27,28,29,63,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,59,u,4,5,6,7,8,60,u,11,12,13,14,15,61,u,18,19,20,21,22,62,u,25,26,27,28,29,63,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512BW-NEXT: vpermi2w %zmm26, %zmm2, %zmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,5,55,u,8,9,10,11,12,56,u,15,16,17,18,19,57,u,22,23,24,25,26,58,u,29,30,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,5,55,u,8,9,10,11,12,56,u,15,16,17,18,19,57,u,22,23,24,25,26,58,u,29,30,31]
; AVX512BW-NEXT: vmovdqu64 (%rsp), %zmm4 # 64-byte Reload
; AVX512BW-NEXT: vpermi2w %zmm26, %zmm4, %zmm2
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,59,4,5,6,7,8,9,60,11,12,13,14,15,16,61,18,19,20,21,22,23,62,25,26,27,28,29,30,63]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll
index bc77e91a2bd4041..cd18176532c3bf3 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i16-stride-8.ll
@@ -336,10 +336,10 @@ define void @store_i16_stride8_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [0,2,4,6,0,2,4,6]
; AVX2-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1],ymm3[2,3],ymm2[4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [1,3,5,7,1,3,5,7]
@@ -464,10 +464,10 @@ define void @store_i16_stride8_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [1,3,5,7,1,3,5,7]
; AVX512F-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,u,u,u,u,0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,u,u,u,u,0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15]
; AVX512F-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,1,4,5,8,9,12,13,u,u,u,u,u,u,u,u,2,3,6,7,10,11,14,15,u,u,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1],ymm3[2,3],ymm2[4,5],ymm3[6,7]
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [0,2,4,6,0,2,4,6]
@@ -689,28 +689,28 @@ define void @store_i16_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-ONLY-NEXT: vinserti128 $1, (%r9), %ymm2, %ymm4
; AVX2-ONLY-NEXT: vinserti128 $1, (%r10), %ymm3, %ymm3
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm5 = ymm3[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11]
; AVX2-ONLY-NEXT: vpshufb %ymm6, %ymm5, %ymm2
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm7 = ymm4[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm8, %ymm7, %ymm9
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm2 = ymm9[0,1,2],ymm2[3],ymm9[4,5,6],ymm2[7]
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm9 = ymm1[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm10, %ymm9, %ymm11
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm12 = ymm0[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = <0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = [0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm13, %ymm12, %ymm14
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2,3,4],ymm11[5],ymm14[6,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm2 = ymm11[0,1],ymm2[2,3],ymm11[4,5],ymm2[6,7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15]
; AVX2-ONLY-NEXT: vpshufb %ymm11, %ymm5, %ymm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm14, %ymm7, %ymm7
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm7[0,1,2],ymm5[3],ymm7[4,5,6],ymm5[7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm7, %ymm9, %ymm9
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm15 = <4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm15 = [4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm15, %ymm12, %ymm12
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm9 = ymm12[0],ymm9[1],ymm12[2,3,4],ymm9[5],ymm12[6,7]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm9[0,1],ymm5[2,3],ymm9[4,5],ymm5[6,7]
@@ -753,28 +753,28 @@ define void @store_i16_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vinserti128 $1, (%r9), %ymm2, %ymm2
; AVX512F-NEXT: vinserti128 $1, (%r10), %ymm3, %ymm3
; AVX512F-NEXT: vpermq {{.*#+}} ymm4 = ymm3[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15]
; AVX512F-NEXT: vpshufb %ymm5, %ymm4, %ymm6
; AVX512F-NEXT: vpermq {{.*#+}} ymm7 = ymm2[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm8, %ymm7, %ymm9
; AVX512F-NEXT: vpblendd {{.*#+}} ymm6 = ymm9[0,1,2],ymm6[3],ymm9[4,5,6],ymm6[7]
; AVX512F-NEXT: vpermq {{.*#+}} ymm9 = ymm1[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,u,u,4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm10, %ymm9, %ymm11
; AVX512F-NEXT: vpermq {{.*#+}} ymm12 = ymm0[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = <4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = [4,5,12,13,u,u,u,u,u,u,u,u,u,u,u,u,6,7,14,15,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm13, %ymm12, %ymm14
; AVX512F-NEXT: vpblendd {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2,3,4],ymm11[5],ymm14[6,7]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm6 = ymm11[0,1],ymm6[2,3],ymm11[4,5],ymm6[6,7]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11]
; AVX512F-NEXT: vpshufb %ymm11, %ymm4, %ymm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm14, %ymm7, %ymm7
; AVX512F-NEXT: vpblendd {{.*#+}} ymm4 = ymm7[0,1,2],ymm4[3],ymm7[4,5,6],ymm4[7]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,u,0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm7, %ymm9, %ymm9
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = <0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = [0,1,8,9,u,u,u,u,u,u,u,u,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm15, %ymm12, %ymm12
; AVX512F-NEXT: vpblendd {{.*#+}} ymm9 = ymm12[0],ymm9[1],ymm12[2,3,4],ymm9[5],ymm12[6,7]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm4 = ymm9[0,1],ymm4[2,3],ymm9[4,5],ymm4[6,7]
@@ -1358,32 +1358,32 @@ define void @store_i16_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa (%rax), %xmm5
; AVX2-FAST-NEXT: vmovdqa (%r10), %xmm4
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm4[0],xmm5[0],xmm4[1],xmm5[1],xmm4[2],xmm5[2],xmm4[3],xmm5[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm6
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm9
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm9[0],xmm6[0],xmm9[1],xmm6[1],xmm9[2],xmm6[2],xmm9[3],xmm6[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm2, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm7[0,1,2],ymm0[3],ymm7[4,5,6],ymm0[7]
; AVX2-FAST-NEXT: vmovdqa (%rcx), %xmm10
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm12
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm13 = xmm12[0],xmm10[0],xmm12[1],xmm10[1],xmm12[2],xmm10[2],xmm12[3],xmm10[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm0, %ymm14
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm15
; AVX2-FAST-NEXT: vmovdqa (%rdi), %xmm2
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm2[0],xmm15[0],xmm2[1],xmm15[1],xmm2[2],xmm15[2],xmm2[3],xmm15[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0],ymm14[1],ymm7[2,3,4],ymm14[5],ymm7[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm7[0,1],ymm3[2,3],ymm7[4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm7, %ymm1
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm7, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm3[0,1,2],ymm1[3],ymm3[4,5,6],ymm1[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm11, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm11, %ymm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2,3,4],ymm3[5],ymm0[6,7]
@@ -1406,13 +1406,13 @@ define void @store_i16_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1],ymm5[2,3],ymm2[4,5],ymm5[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa (%r10), %ymm13
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%rax), %ymm15
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm1, %ymm4
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm4[0,1,2],ymm0[3],ymm4[4,5,6],ymm0[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm7, %ymm1, %ymm4
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm8, %ymm5
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0],ymm4[1],ymm5[2,3,4],ymm4[5],ymm5[6,7]
@@ -1685,9 +1685,9 @@ define void @store_i16_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm14[0],xmm13[0],xmm14[1],xmm13[1],xmm14[2],xmm13[2],xmm14[3],xmm13[3]
; AVX512F-NEXT: vinserti32x4 $2, %xmm11, %zmm11, %zmm11
; AVX512F-NEXT: movq {{[0-9]+}}(%rsp), %rax
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <0,u,0,16,u,u,1,17,10,10,10,26,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [0,u,0,16,u,u,1,17,10,10,10,26,u,u,11,27]
; AVX512F-NEXT: vpermt2d %zmm16, %zmm12, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,16,1,u,1,17,u,u,10,26,11,11,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,16,1,u,1,17,u,u,10,26,11,11,11,27,u,u]
; AVX512F-NEXT: vpermt2d %zmm17, %zmm13, %zmm0
; AVX512F-NEXT: movb $-86, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -1723,26 +1723,26 @@ define void @store_i16_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512BW-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
; AVX512BW-NEXT: vinserti64x4 $1, (%r10), %zmm3, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,0,16,32,48,u,u,u,u,1,17,33,49,u,u,u,u,2,18,34,50,u,u,u,u,3,19,35,51>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,0,16,32,48,u,u,u,u,1,17,33,49,u,u,u,u,2,18,34,50,u,u,u,u,3,19,35,51]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,16,32,48,u,u,u,u,1,17,33,49,u,u,u,u,2,18,34,50,u,u,u,u,3,19,35,51,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,16,32,48,u,u,u,u,1,17,33,49,u,u,u,u,2,18,34,50,u,u,u,u,3,19,35,51,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: movb $-86, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,4,20,36,52,u,u,u,u,5,21,37,53,u,u,u,u,6,22,38,54,u,u,u,u,7,23,39,55>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,4,20,36,52,u,u,u,u,5,21,37,53,u,u,u,u,6,22,38,54,u,u,u,u,7,23,39,55]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,20,36,52,u,u,u,u,5,21,37,53,u,u,u,u,6,22,38,54,u,u,u,u,7,23,39,55,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,20,36,52,u,u,u,u,5,21,37,53,u,u,u,u,6,22,38,54,u,u,u,u,7,23,39,55,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,8,24,40,56,u,u,u,u,9,25,41,57,u,u,u,u,10,26,42,58,u,u,u,u,11,27,43,59>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,8,24,40,56,u,u,u,u,9,25,41,57,u,u,u,u,10,26,42,58,u,u,u,u,11,27,43,59]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,24,40,56,u,u,u,u,9,25,41,57,u,u,u,u,10,26,42,58,u,u,u,u,11,27,43,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,24,40,56,u,u,u,u,9,25,41,57,u,u,u,u,10,26,42,58,u,u,u,u,11,27,43,59,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,12,28,44,60,u,u,u,u,13,29,45,61,u,u,u,u,14,30,46,62,u,u,u,u,15,31,47,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,12,28,44,60,u,u,u,u,13,29,45,61,u,u,u,u,14,30,46,62,u,u,u,u,15,31,47,63]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <12,28,44,60,u,u,u,u,13,29,45,61,u,u,u,u,14,30,46,62,u,u,u,u,15,31,47,63,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [12,28,44,60,u,u,u,u,13,29,45,61,u,u,u,u,14,30,46,62,u,u,u,u,15,31,47,63,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm2, 192(%rax)
@@ -2763,7 +2763,7 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa (%r10), %xmm1
; AVX2-FAST-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm4, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -2777,7 +2777,7 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm7, %ymm4
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm6
; AVX2-FAST-NEXT: vmovdqa %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -2788,17 +2788,17 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm6[0],ymm4[1],ymm6[2,3,4],ymm4[5],ymm6[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm4[0,1],ymm0[2,3],ymm4[4,5],ymm0[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm0, %ymm4
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm0, %ymm2
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm7
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm2[0,1,2],ymm4[3],ymm2[4,5,6],ymm4[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm0, %ymm6
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm5, %ymm14, %ymm5
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm5[0],ymm6[1],ymm5[2,3,4],ymm6[5],ymm5[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm5[0,1],ymm4[2,3],ymm5[4,5],ymm4[6,7]
@@ -2823,11 +2823,11 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm14[0],ymm3[1],ymm14[2,3,4],ymm3[5],ymm14[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm3[0,1],ymm12[2,3],ymm3[4,5],ymm12[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm12, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm12, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2],ymm3[3],ymm2[4,5,6],ymm3[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm15, %ymm1
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm15, %ymm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2,3,4],ymm1[5],ymm0[6,7]
@@ -2835,16 +2835,16 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm8[0],xmm9[0],xmm8[1],xmm9[1],xmm8[2],xmm9[2],xmm8[3],xmm9[3]
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm1 = xmm10[0],xmm13[0],xmm10[1],xmm13[1],xmm10[2],xmm13[2],xmm10[3],xmm13[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm8, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm10, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3],ymm3[4,5,6],ymm2[7]
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm3 = xmm5[0],xmm11[0],xmm5[1],xmm11[1],xmm5[2],xmm11[2],xmm5[3],xmm11[3]
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm4 = xmm6[0],xmm7[0],xmm6[1],xmm7[1],xmm6[2],xmm7[2],xmm6[3],xmm7[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm11, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm7, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm2[2,3],ymm5[4,5],ymm2[6,7]
@@ -3351,9 +3351,9 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa (%r8), %xmm5
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm5[4],xmm1[4],xmm5[5],xmm1[5],xmm5[6],xmm1[6],xmm5[7],xmm1[7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <0,u,0,u,u,u,1,u,2,2,2,2,u,u,3,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [0,u,0,u,u,u,1,u,2,2,2,2,u,u,3,3]
; AVX512F-SLOW-NEXT: vpermd %zmm1, %zmm26, %zmm30
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm27 = <0,0,0,0,u,u,1,1,2,2,2,2,u,u,3,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,0,0,0,u,u,1,1,2,2,2,2,u,u,3,3]
; AVX512F-SLOW-NEXT: movw $-30584, %r11w # imm = 0x8888
; AVX512F-SLOW-NEXT: kmovw %r11d, %k1
; AVX512F-SLOW-NEXT: vpermd %zmm0, %zmm27, %zmm30 {%k1}
@@ -3363,9 +3363,9 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa (%rsi), %xmm10
; AVX512F-SLOW-NEXT: vmovdqa (%rdi), %xmm11
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm11[4],xmm10[4],xmm11[5],xmm10[5],xmm11[6],xmm10[6],xmm11[7],xmm10[7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <0,u,1,u,1,u,u,u,2,2,3,3,3,3,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [0,u,1,u,1,u,u,u,2,2,3,3,3,3,u,u]
; AVX512F-SLOW-NEXT: vpermd %zmm1, %zmm28, %zmm3
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <0,0,1,1,1,1,u,u,2,2,3,3,3,3,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [0,0,1,1,1,1,u,u,2,2,3,3,3,3,u,u]
; AVX512F-SLOW-NEXT: movw $8738, %r11w # imm = 0x2222
; AVX512F-SLOW-NEXT: kmovw %r11d, %k2
; AVX512F-SLOW-NEXT: vpermd %zmm0, %zmm29, %zmm3 {%k2}
@@ -3636,21 +3636,21 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpermd %zmm6, %zmm13, %zmm3 {%k2}
; AVX512F-FAST-NEXT: vpermd %zmm4, %zmm0, %zmm0
; AVX512F-FAST-NEXT: vpermd %zmm5, %zmm13, %zmm0 {%k2}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,0,0,0,u,u,1,1,10,10,10,10,u,u,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,0,0,0,u,u,1,1,10,10,10,10,u,u,11,11]
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm4, %zmm5 # 64-byte Folded Reload
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm4, %zmm5 {%k1} # 64-byte Folded Reload
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,u,0,u,u,u,1,u,10,10,10,10,u,u,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,u,0,u,u,u,1,u,10,10,10,10,u,u,11,11]
; AVX512F-FAST-NEXT: vpermd %zmm12, %zmm6, %zmm6
; AVX512F-FAST-NEXT: vpermd %zmm28, %zmm4, %zmm6 {%k1}
; AVX512F-FAST-NEXT: vpermd %zmm25, %zmm4, %zmm12
; AVX512F-FAST-NEXT: vpermd %zmm16, %zmm4, %zmm12 {%k1}
; AVX512F-FAST-NEXT: vpermd %zmm1, %zmm4, %zmm1
; AVX512F-FAST-NEXT: vpermd %zmm2, %zmm4, %zmm1 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,0,1,1,1,1,u,u,10,u,11,u,11,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,0,1,1,1,1,u,u,10,u,11,u,11,u,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm20, %zmm2, %zmm2
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,0,1,1,1,1,u,u,10,10,11,11,11,11,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,0,1,1,1,1,u,u,10,10,11,11,11,11,u,u]
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm4, %zmm2 {%k2} # 64-byte Folded Reload
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,u,1,u,1,u,u,u,10,10,11,11,11,11,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,u,1,u,1,u,u,u,10,10,11,11,11,11,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm30, %zmm13, %zmm13
; AVX512F-FAST-NEXT: vpermd %zmm29, %zmm4, %zmm13 {%k2}
; AVX512F-FAST-NEXT: vpermd %zmm15, %zmm4, %zmm15
@@ -3693,16 +3693,16 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm6
; AVX512BW-NEXT: vmovdqa64 (%r11), %zmm7
; AVX512BW-NEXT: vmovdqa64 (%r10), %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm9
; AVX512BW-NEXT: movw $-30584, %cx # imm = 0x8888
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm9 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm4
; AVX512BW-NEXT: movw $8738, %cx # imm = 0x2222
; AVX512BW-NEXT: kmovd %ecx, %k2
@@ -3710,80 +3710,80 @@ define void @store_i16_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movb $-86, %cl
; AVX512BW-NEXT: kmovd %ecx, %k3
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm4 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm10
; AVX512BW-NEXT: vmovdqa32 %zmm9, %zmm10 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm9
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm9 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm9 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm10
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm10 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm10 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm12
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm12 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm13, %zmm11 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm11 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm14
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm14 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm14 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm15
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm15 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm15 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm16
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm16 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm16 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63]
; AVX512BW-NEXT: vpermi2w %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u]
; AVX512BW-NEXT: vpermi2w %zmm6, %zmm5, %zmm7
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm3, %zmm2, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa32 %zmm5, %zmm2 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm2 {%k3}
@@ -5745,51 +5745,51 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqa (%rax), %xmm4
; AVX2-FAST-NEXT: vmovdqa (%r10), %xmm5
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm3 = xmm5[0],xmm4[0],xmm5[1],xmm4[1],xmm5[2],xmm4[2],xmm5[3],xmm4[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm0, %ymm1
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm6
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm7
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm10 = xmm7[0],xmm6[0],xmm7[1],xmm6[1],xmm7[2],xmm6[2],xmm7[3],xmm6[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm0, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0,1,2],ymm1[3],ymm2[4,5,6],ymm1[7]
; AVX2-FAST-NEXT: vmovdqa (%rcx), %xmm8
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm9
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm11 = xmm9[0],xmm8[0],xmm9[1],xmm8[1],xmm9[2],xmm8[2],xmm9[3],xmm8[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm0, %ymm12
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm13
; AVX2-FAST-NEXT: vmovdqa (%rdi), %xmm1
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm1[0],xmm13[0],xmm1[1],xmm13[1],xmm1[2],xmm13[2],xmm1[3],xmm13[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm14, %ymm15
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm12 = ymm15[0],ymm12[1],ymm15[2,3,4],ymm12[5],ymm15[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm12[0,1],ymm2[2,3],ymm12[4,5],ymm2[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm12, %ymm2
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm12, %ymm3
; AVX2-FAST-NEXT: vmovdqa %ymm12, %ymm15
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3],ymm3[4,5,6],ymm2[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm10, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm10, %ymm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2,3,4],ymm3[5],ymm0[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm2[2,3],ymm0[4,5],ymm2[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm5[4],xmm4[4],xmm5[5],xmm4[5],xmm5[6],xmm4[6],xmm5[7],xmm4[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm11, %ymm2
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm7[4],xmm6[4],xmm7[5],xmm6[5],xmm7[6],xmm6[6],xmm7[7],xmm6[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm12, %ymm4
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0,1,2],ymm2[3],ymm4[4,5,6],ymm2[7]
; AVX2-FAST-NEXT: vmovdqa 32(%rax), %xmm10
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm8 = xmm9[4],xmm8[4],xmm9[5],xmm8[5],xmm9[6],xmm8[6],xmm9[7],xmm8[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm14, %ymm5
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm13[4],xmm1[5],xmm13[5],xmm1[6],xmm13[6],xmm1[7],xmm13[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm4, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
; AVX2-FAST-NEXT: vmovdqa 32(%r10), %xmm6
@@ -5800,7 +5800,7 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm15, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm2[0,1,2],ymm0[3],ymm2[4,5,6],ymm0[7]
; AVX2-FAST-NEXT: vmovdqa 32(%r8), %xmm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm8, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0],ymm2[1],ymm1[2,3,4],ymm2[5],ymm1[6,7]
@@ -5818,17 +5818,17 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm13 = xmm8[0],xmm3[0],xmm8[1],xmm3[1],xmm8[2],xmm3[2],xmm8[3],xmm3[3]
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm14, %ymm15
; AVX2-FAST-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm11[0],xmm9[0],xmm11[1],xmm9[1],xmm11[2],xmm9[2],xmm11[3],xmm9[3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm12, %ymm14
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm14 = ymm14[0],ymm15[1],ymm14[2,3,4],ymm15[5],ymm14[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm14[0,1],ymm1[2,3],ymm14[4,5],ymm1[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm14, %ymm1
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm14, %ymm2
; AVX2-FAST-NEXT: vmovdqa %ymm14, %ymm15
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0,1,2],ymm1[3],ymm2[4,5,6],ymm1[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm13, %ymm4, %ymm2
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm4, %ymm0
; AVX2-FAST-NEXT: vmovdqa %ymm4, %ymm13
@@ -5837,14 +5837,14 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm6[4],xmm10[4],xmm6[5],xmm10[5],xmm6[6],xmm10[6],xmm6[7],xmm10[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm7[4],xmm5[4],xmm7[5],xmm5[5],xmm7[6],xmm5[6],xmm7[7],xmm5[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm7, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm10, %ymm4
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm4[0,1,2],ymm2[3],ymm4[4,5,6],ymm2[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm8[4],xmm3[4],xmm8[5],xmm3[5],xmm8[6],xmm3[6],xmm8[7],xmm3[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm4 = xmm11[4],xmm9[4],xmm11[5],xmm9[5],xmm11[6],xmm9[6],xmm11[7],xmm9[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm14, %ymm5
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm12, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
@@ -5878,12 +5878,12 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm13 = ymm15[0],ymm13[1],ymm15[2,3,4],ymm13[5],ymm15[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm13[0,1],ymm5[2,3],ymm13[4,5],ymm5[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm13, %ymm4
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm13, %ymm5
; AVX2-FAST-NEXT: vmovdqa %ymm13, %ymm15
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm4 = ymm5[0,1,2],ymm4[3],ymm5[4,5,6],ymm4[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm12, %ymm5
; AVX2-FAST-NEXT: vpermd %ymm14, %ymm12, %ymm6
; AVX2-FAST-NEXT: vmovdqa %ymm12, %ymm14
@@ -5892,16 +5892,16 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm2[4],xmm0[5],xmm2[5],xmm0[6],xmm2[6],xmm0[7],xmm2[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm3[4],xmm1[4],xmm3[5],xmm1[5],xmm3[6],xmm1[6],xmm3[7],xmm1[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm11, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm13, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3],ymm3[4,5,6],ymm2[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm8[4],xmm7[4],xmm8[5],xmm7[5],xmm8[6],xmm7[6],xmm8[7],xmm7[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm4 = xmm10[4],xmm9[4],xmm10[5],xmm9[5],xmm10[6],xmm9[6],xmm10[7],xmm9[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm10, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm12, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm2[2,3],ymm5[4,5],ymm2[6,7]
@@ -5934,11 +5934,11 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm10 = ymm15[0],ymm10[1],ymm15[2,3,4],ymm10[5],ymm15[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm10[0,1],ymm3[2,3],ymm10[4,5],ymm3[6,7]
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <2,2,2,2,u,u,3,3>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [2,2,2,2,u,u,3,3]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm10, %ymm2
; AVX2-FAST-NEXT: vpermd %ymm6, %ymm10, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3],ymm3[4,5,6],ymm2[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <2,2,3,3,3,3,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [2,2,3,3,3,3,u,u]
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm12, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm14, %ymm12, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm3 = ymm6[0],ymm3[1],ymm6[2,3,4],ymm3[5],ymm6[6,7]
@@ -5946,16 +5946,16 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm1 = xmm5[4],xmm4[4],xmm5[5],xmm4[5],xmm5[6],xmm4[6],xmm5[7],xmm4[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,0,0,0,u,u,1,1>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,0,0,u,u,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <0,u,0,u,u,u,1,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,u,0,u,u,u,1,u]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm3[0,1,2],ymm2[3],ymm3[4,5,6],ymm2[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm3 = xmm8[4],xmm7[4],xmm8[5],xmm7[5],xmm8[6],xmm7[6],xmm8[7],xmm7[7]
; AVX2-FAST-NEXT: vpunpckhwd {{.*#+}} xmm4 = xmm13[4],xmm11[4],xmm13[5],xmm11[5],xmm13[6],xmm11[6],xmm13[7],xmm11[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,0,1,1,1,1,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,0,1,1,1,1,u,u]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm5, %ymm5
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <0,u,1,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,u,1,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm4, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm5 = ymm6[0],ymm5[1],ymm6[2,3,4],ymm5[5],ymm6[6,7]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm5[0,1],ymm2[2,3],ymm5[4,5],ymm2[6,7]
@@ -6857,9 +6857,9 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa (%r8), %xmm7
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm2 = xmm7[4],xmm0[4],xmm7[5],xmm0[5],xmm7[6],xmm0[6],xmm7[7],xmm0[7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm30 = <0,u,0,u,u,u,1,u,2,2,2,2,u,u,3,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm30 = [0,u,0,u,u,u,1,u,2,2,2,2,u,u,3,3]
; AVX512F-SLOW-NEXT: vpermd %zmm2, %zmm30, %zmm0
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <0,0,0,0,u,u,1,1,2,2,2,2,u,u,3,3>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [0,0,0,0,u,u,1,1,2,2,2,2,u,u,3,3]
; AVX512F-SLOW-NEXT: movw $-30584, %r11w # imm = 0x8888
; AVX512F-SLOW-NEXT: kmovw %r11d, %k2
; AVX512F-SLOW-NEXT: vpermd %zmm1, %zmm29, %zmm0 {%k2}
@@ -7022,9 +7022,9 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermd %zmm6, %zmm30, %zmm8
; AVX512F-SLOW-NEXT: vpunpckhwd {{.*#+}} xmm6 = xmm12[4],xmm9[4],xmm12[5],xmm9[5],xmm12[6],xmm9[6],xmm12[7],xmm9[7]
; AVX512F-SLOW-NEXT: vpermd %zmm4, %zmm29, %zmm8 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <0,u,1,u,1,u,u,u,2,2,3,3,3,3,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [0,u,1,u,1,u,u,u,2,2,3,3,3,3,u,u]
; AVX512F-SLOW-NEXT: vpermd %zmm6, %zmm29, %zmm7
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm30 = <0,0,1,1,1,1,u,u,2,2,3,3,3,3,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm30 = [0,0,1,1,1,1,u,u,2,2,3,3,3,3,u,u]
; AVX512F-SLOW-NEXT: vpermd %zmm3, %zmm30, %zmm7 {%k1}
; AVX512F-SLOW-NEXT: vmovdqa 96(%rsi), %xmm3
; AVX512F-SLOW-NEXT: vmovdqa 96(%rdi), %xmm2
@@ -7381,10 +7381,10 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpermd %zmm20, %zmm1, %zmm4 {%k1}
; AVX512F-FAST-NEXT: vpermd %zmm5, %zmm0, %zmm5
; AVX512F-FAST-NEXT: vpermd %zmm10, %zmm1, %zmm5 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,0,0,0,u,u,1,1,10,10,10,10,u,u,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,0,0,0,u,u,1,1,10,10,10,10,u,u,11,11]
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm0, %zmm1 # 64-byte Folded Reload
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm0, %zmm1 {%k2} # 64-byte Folded Reload
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,u,0,u,u,u,1,u,10,10,10,10,u,u,11,11>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,u,0,u,u,u,1,u,10,10,10,10,u,u,11,11]
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm10, %zmm10 # 64-byte Folded Reload
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm0, %zmm10 {%k2} # 64-byte Folded Reload
; AVX512F-FAST-NEXT: vpermd {{[-0-9]+}}(%r{{[sb]}}p), %zmm0, %zmm16 # 64-byte Folded Reload
@@ -7400,15 +7400,15 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpermd %zmm12, %zmm0, %zmm23
; AVX512F-FAST-NEXT: vpermd %zmm14, %zmm0, %zmm23 {%k2}
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm11, %zmm11, %zmm0
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,0,1,1,1,1,u,u,10,u,11,u,11,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,0,1,1,1,1,u,u,10,u,11,u,11,u,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm11, %zmm12
; AVX512F-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm0, %zmm0, %zmm0
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,0,1,1,1,1,u,u,10,10,11,11,11,11,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,0,1,1,1,1,u,u,10,10,11,11,11,11,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm11, %zmm12 {%k1}
; AVX512F-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm0, %zmm0, %zmm0
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,u,1,u,1,u,u,u,10,10,11,11,11,11,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,u,1,u,1,u,u,u,10,10,11,11,11,11,u,u]
; AVX512F-FAST-NEXT: vpermd %zmm0, %zmm14, %zmm0
; AVX512F-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm8 # 64-byte Reload
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm8, %zmm8, %zmm14
@@ -7502,57 +7502,57 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%r10), %zmm0
; AVX512BW-NEXT: vmovdqa64 (%rax), %zmm30
; AVX512BW-NEXT: vmovdqa64 64(%rax), %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm5
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm4
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm3, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm5
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm2, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm5, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm8
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm7, %zmm8
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm8, %zmm9
; AVX512BW-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm10
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm9, %zmm10
; AVX512BW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm11
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm10, %zmm11
; AVX512BW-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm11, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm12, %zmm13
; AVX512BW-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm13, %zmm14
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u>
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u]
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm16
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm15, %zmm16
; AVX512BW-NEXT: vmovdqu64 %zmm16, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = <u,u,u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = [u,u,u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59]
; AVX512BW-NEXT: vpermt2w %zmm30, %zmm31, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
@@ -7574,10 +7574,10 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermt2w %zmm28, %zmm31, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm31
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm28, %zmm31
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u]
; AVX512BW-NEXT: vpermt2w %zmm27, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm27
@@ -7598,33 +7598,33 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2w %zmm1, %zmm0, %zmm27
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm16
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm20
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm20
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm21
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm21
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm22
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm22
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm23
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm23
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm24
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm24
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [u,u,16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm25
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm29, %zmm25
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = <u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = [u,u,28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm26
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm30, %zmm26
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm16
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -7643,28 +7643,28 @@ define void @store_i16_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm1, %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm3
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [4,36,u,u,u,u,u,u,5,37,u,u,u,u,u,u,6,38,u,u,u,u,u,u,7,39,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm18
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm12, %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,32,u,u,u,u,u,u,1,33,u,u,u,u,u,u,2,34,u,u,u,u,u,u,3,35,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm17
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm10, %zmm17
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [12,44,u,u,u,u,u,u,13,45,u,u,u,u,u,u,14,46,u,u,u,u,u,u,15,47,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm15
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm8, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,40,u,u,u,u,u,u,9,41,u,u,u,u,u,u,10,42,u,u,u,u,u,u,11,43,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm14
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm7, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [20,52,u,u,u,u,u,u,21,53,u,u,u,u,u,u,22,54,u,u,u,u,u,u,23,55,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm6, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [16,48,u,u,u,u,u,u,17,49,u,u,u,u,u,u,18,50,u,u,u,u,u,u,19,51,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm11
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm4, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [28,60,u,u,u,u,u,u,29,61,u,u,u,u,u,u,30,62,u,u,u,u,u,u,31,63,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm9
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [24,56,u,u,u,u,u,u,25,57,u,u,u,u,u,u,26,58,u,u,u,u,u,u,27,59,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2w %zmm0, %zmm19, %zmm3
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm0
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-3.ll
index 1fc25ee32933320..3abc8b1fb8250dd 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-3.ll
@@ -55,7 +55,7 @@ define void @store_i32_stride3_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-NEXT: vmovsd {{.*#+}} xmm2 = mem[0],zero
; AVX2-NEXT: vmovlhps {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX2-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
-; AVX2-NEXT: vmovaps {{.*#+}} ymm1 = <0,2,4,1,3,5,u,u>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [0,2,4,1,3,5,u,u]
; AVX2-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX2-NEXT: vmovlps %xmm1, 16(%rcx)
@@ -120,7 +120,7 @@ define void @store_i32_stride3_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovaps (%rdi), %xmm0
; AVX2-SLOW-NEXT: vmovaps (%rsi), %xmm1
; AVX2-SLOW-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm2
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm3 = <0,4,u,1,5,u,2,6>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm3 = [0,4,u,1,5,u,2,6]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vbroadcastsd (%rdx), %ymm3
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
@@ -140,7 +140,7 @@ define void @store_i32_stride3_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovddup {{.*#+}} xmm2 = mem[0,0]
; AVX2-FAST-NEXT: vblendps {{.*#+}} xmm1 = xmm2[0],xmm1[1,2],xmm2[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm2 = <0,4,u,1,5,u,2,6>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm2 = [0,4,u,1,5,u,2,6]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vbroadcastsd (%rdx), %ymm2
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm2[2],ymm0[3,4],ymm2[5],ymm0[6,7]
@@ -154,7 +154,7 @@ define void @store_i32_stride3_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovaps (%rdi), %xmm0
; AVX2-FAST-PERLANE-NEXT: vmovaps (%rsi), %xmm1
; AVX2-FAST-PERLANE-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm3 = <0,4,u,1,5,u,2,6>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm3 = [0,4,u,1,5,u,2,6]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vbroadcastsd (%rdx), %ymm3
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1],ymm3[2],ymm2[3,4],ymm3[5],ymm2[6,7]
@@ -171,7 +171,7 @@ define void @store_i32_stride3_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovaps (%rdi), %xmm0
; AVX512-NEXT: vinsertf128 $1, (%rsi), %ymm0, %ymm0
; AVX512-NEXT: vinsertf32x4 $2, (%rdx), %zmm0, %zmm0
-; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = <0,4,8,1,5,9,2,6,10,3,7,11,u,u,u,u>
+; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = [0,4,8,1,5,9,2,6,10,3,7,11,u,u,u,u]
; AVX512-NEXT: vpermps %zmm0, %zmm1, %zmm0
; AVX512-NEXT: vextractf32x4 $2, %zmm0, 32(%rcx)
; AVX512-NEXT: vmovaps %ymm0, (%rcx)
@@ -669,15 +669,15 @@ define void @store_i32_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512-NEXT: vmovdqa64 (%rdx), %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,16,3,4,17,6,7,18,9,10,19,12,13,20,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,21,2,3,22,5,6,23,8,9,24,11,12,25,14,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [26,1,2,27,4,5,28,7,8,29,10,11,30,13,14,31]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm0
@@ -1328,17 +1328,17 @@ define void @store_i32_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 64(%rsi), %zmm3
; AVX512-NEXT: vmovdqa64 (%rdx), %zmm4
; AVX512-NEXT: vmovdqa64 64(%rdx), %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5]
; AVX512-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512-NEXT: vpermt2d %zmm2, %zmm6, %zmm7
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,16,3,4,17,6,7,18,9,10,19,12,13,20,15]
; AVX512-NEXT: vpermt2d %zmm4, %zmm8, %zmm7
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u]
; AVX512-NEXT: vmovdqa64 %zmm1, %zmm10
; AVX512-NEXT: vpermt2d %zmm3, %zmm9, %zmm10
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm11 = [26,1,2,27,4,5,28,7,8,29,10,11,30,13,14,31]
; AVX512-NEXT: vpermt2d %zmm5, %zmm11, %zmm10
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10]
; AVX512-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512-NEXT: vpermt2d %zmm1, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,21,2,3,22,5,6,23,8,9,24,11,12,25,14,15]
@@ -2734,17 +2734,17 @@ define void @store_i32_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 64(%rdx), %zmm9
; AVX512-NEXT: vmovdqa64 128(%rdx), %zmm10
; AVX512-NEXT: vmovdqa64 192(%rdx), %zmm11
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [0,16,u,1,17,u,2,18,u,3,19,u,4,20,u,5]
; AVX512-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512-NEXT: vpermt2d %zmm4, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,16,3,4,17,6,7,18,9,10,19,12,13,20,15]
; AVX512-NEXT: vpermt2d %zmm8, %zmm14, %zmm13
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,11,27,u,12,28,u,13,29,u,14,30,u,15,31,u]
; AVX512-NEXT: vmovdqa64 %zmm0, %zmm16
; AVX512-NEXT: vpermt2d %zmm7, %zmm15, %zmm16
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm17 = [26,1,2,27,4,5,28,7,8,29,10,11,30,13,14,31]
; AVX512-NEXT: vpermt2d %zmm11, %zmm17, %zmm16
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = <5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [5,u,22,6,u,23,7,u,24,8,u,25,9,u,26,10]
; AVX512-NEXT: vmovdqa64 %zmm7, %zmm19
; AVX512-NEXT: vpermt2d %zmm0, %zmm18, %zmm19
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,21,2,3,22,5,6,23,8,9,24,11,12,25,14,15]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-4.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-4.ll
index a4480d5d7e957c1..9d91423ac13d70a 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-4.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-4.ll
@@ -192,14 +192,14 @@ define void @store_i32_stride4_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-ONLY-NEXT: vmovaps (%rdx), %xmm1
; AVX2-ONLY-NEXT: vinsertf128 $1, (%rsi), %ymm0, %ymm0
; AVX2-ONLY-NEXT: vinsertf128 $1, (%rcx), %ymm1, %ymm1
-; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm2 = <u,u,0,4,u,u,1,5>
+; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm2 = [u,u,0,4,u,u,1,5]
; AVX2-ONLY-NEXT: vpermps %ymm1, %ymm2, %ymm2
-; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = <0,4,u,u,1,5,u,u>
+; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = [0,4,u,u,1,5,u,u]
; AVX2-ONLY-NEXT: vpermps %ymm0, %ymm3, %ymm3
; AVX2-ONLY-NEXT: vblendps {{.*#+}} ymm2 = ymm3[0,1],ymm2[2,3],ymm3[4,5],ymm2[6,7]
-; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = <u,u,2,6,u,u,3,7>
+; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = [u,u,2,6,u,u,3,7]
; AVX2-ONLY-NEXT: vpermps %ymm1, %ymm3, %ymm1
-; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = <2,6,u,u,3,7,u,u>
+; AVX2-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = [2,6,u,u,3,7,u,u]
; AVX2-ONLY-NEXT: vpermps %ymm0, %ymm3, %ymm0
; AVX2-ONLY-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5],ymm1[6,7]
; AVX2-ONLY-NEXT: vmovaps %ymm0, 32(%r8)
@@ -661,26 +661,26 @@ define void @store_i32_stride4_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm5
; AVX512F-NEXT: movb $-86, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm6
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm2, 192(%r8)
@@ -696,26 +696,26 @@ define void @store_i32_stride4_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: movb $-86, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm2, 192(%r8)
@@ -1337,32 +1337,32 @@ define void @store_i32_stride4_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm6
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm9
; AVX512F-NEXT: vpermt2d %zmm6, %zmm8, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm11
; AVX512F-NEXT: vpermt2d %zmm2, %zmm10, %zmm11
; AVX512F-NEXT: movb $-86, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm11 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512F-NEXT: vpermt2d %zmm6, %zmm9, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512F-NEXT: vpermt2d %zmm2, %zmm13, %zmm14
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm14 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm15
; AVX512F-NEXT: vpermt2d %zmm6, %zmm12, %zmm15
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm17
; AVX512F-NEXT: vpermt2d %zmm2, %zmm16, %zmm17
; AVX512F-NEXT: vmovdqa64 %zmm15, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512F-NEXT: vpermt2d %zmm6, %zmm15, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512F-NEXT: vpermt2d %zmm2, %zmm6, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm0 {%k1}
; AVX512F-NEXT: vpermi2d %zmm7, %zmm5, %zmm8
@@ -1398,32 +1398,32 @@ define void @store_i32_stride4_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm6
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm9
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm11
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm10, %zmm11
; AVX512BW-NEXT: movb $-86, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm9, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm13, %zmm14
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm15
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm12, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm16, %zmm17
; AVX512BW-NEXT: vmovdqa64 %zmm15, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm15, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm6, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm0 {%k1}
; AVX512BW-NEXT: vpermi2d %zmm7, %zmm5, %zmm8
@@ -2725,32 +2725,32 @@ define void @store_i32_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm26
; AVX512F-NEXT: vmovdqa64 128(%rcx), %zmm19
; AVX512F-NEXT: vmovdqa64 192(%rcx), %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm8
; AVX512F-NEXT: vpermt2d %zmm21, %zmm14, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm4
; AVX512F-NEXT: vpermt2d %zmm17, %zmm7, %zmm4
; AVX512F-NEXT: movb $-86, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm4 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm10
; AVX512F-NEXT: vpermt2d %zmm21, %zmm16, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512F-NEXT: vpermt2d %zmm17, %zmm11, %zmm8
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm8 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm20
; AVX512F-NEXT: vpermt2d %zmm21, %zmm18, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm10
; AVX512F-NEXT: vpermt2d %zmm17, %zmm15, %zmm10
; AVX512F-NEXT: vmovdqa64 %zmm20, %zmm10 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512F-NEXT: vpermt2d %zmm21, %zmm20, %zmm22
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512F-NEXT: vpermt2d %zmm17, %zmm21, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm0 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm22
@@ -2838,32 +2838,32 @@ define void @store_i32_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm26
; AVX512BW-NEXT: vmovdqa64 128(%rcx), %zmm19
; AVX512BW-NEXT: vmovdqa64 192(%rcx), %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,4,20,u,u,5,21,u,u,6,22,u,u,7,23]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm14, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm7, %zmm4
; AVX512BW-NEXT: movb $-86, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm4 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,u,0,16,u,u,1,17,u,u,2,18,u,u,3,19]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm10
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm16, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,16,u,u,1,17,u,u,2,18,u,u,3,19,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm11, %zmm8
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,u,12,28,u,u,13,29,u,u,14,30,u,u,15,31]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm18, %zmm20
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [12,28,u,u,13,29,u,u,14,30,u,u,15,31,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm10
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm15, %zmm10
; AVX512BW-NEXT: vmovdqa64 %zmm20, %zmm10 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,8,24,u,u,9,25,u,u,10,26,u,u,11,27]
; AVX512BW-NEXT: vpermt2d %zmm21, %zmm20, %zmm22
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [8,24,u,u,9,25,u,u,10,26,u,u,11,27,u,u]
; AVX512BW-NEXT: vpermt2d %zmm17, %zmm21, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm0 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm22
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll
index 1073c24b228ba02..2da05c85bf8f514 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-5.ll
@@ -73,7 +73,7 @@ define void @store_i32_stride5_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-ONLY-NEXT: vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm1[0]
; AVX2-ONLY-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm0
; AVX2-ONLY-NEXT: vmovq %rax, %xmm2
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm3 = <0,2,4,6,u,1,3,5>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm3 = [0,2,4,6,u,1,3,5]
; AVX2-ONLY-NEXT: vpermd %ymm0, %ymm3, %ymm0
; AVX2-ONLY-NEXT: vmovd %eax, %xmm3
; AVX2-ONLY-NEXT: vpbroadcastd %xmm3, %ymm3
@@ -96,7 +96,7 @@ define void @store_i32_stride5_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX512-NEXT: vmovsd {{.*#+}} xmm1 = mem[0],zero
; AVX512-NEXT: vinsertf32x4 $2, %xmm1, %zmm0, %zmm0
-; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = <0,2,4,6,8,1,3,5,7,9,u,u,u,u,u,u>
+; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = [0,2,4,6,8,1,3,5,7,9,u,u,u,u,u,u]
; AVX512-NEXT: vpermps %zmm0, %zmm1, %zmm0
; AVX512-NEXT: vextractf32x4 $2, %zmm0, %xmm1
; AVX512-NEXT: vmovlps %xmm1, 32(%r9)
@@ -199,16 +199,16 @@ define void @store_i32_stride5_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovaps (%r8), %xmm4
; AVX2-SLOW-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm5
; AVX2-SLOW-NEXT: vinsertf128 $1, %xmm3, %ymm2, %ymm6
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = <u,u,0,4,u,u,u,1>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = [u,u,0,4,u,u,u,1]
; AVX2-SLOW-NEXT: vpermps %ymm6, %ymm7, %ymm7
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm8 = <u,4,u,u,u,1,5,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm8 = [u,4,u,u,u,1,5,u]
; AVX2-SLOW-NEXT: vpermps %ymm5, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm7 = ymm8[0,1],ymm7[2,3],ymm8[4,5,6],ymm7[7]
; AVX2-SLOW-NEXT: vinsertf128 $1, %xmm4, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm7[1,2,3],ymm0[4],ymm7[5,6,7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = <u,u,2,6,u,u,u,3>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = [u,u,2,6,u,u,u,3]
; AVX2-SLOW-NEXT: vpermps %ymm5, %ymm7, %ymm5
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = <5,u,u,u,2,6,u,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm7 = [5,u,u,u,2,6,u,u]
; AVX2-SLOW-NEXT: vpermps %ymm6, %ymm7, %ymm6
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm5 = ymm6[0,1],ymm5[2,3],ymm6[4,5],ymm5[6,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm6 = ymm4[0,1,2,1]
@@ -230,16 +230,16 @@ define void @store_i32_stride5_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovaps (%r8), %xmm3
; AVX2-FAST-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm4
; AVX2-FAST-NEXT: vinsertf128 $1, (%rcx), %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = <u,u,0,4,u,u,u,1>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = [u,u,0,4,u,u,u,1]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm5, %ymm5
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = <u,4,u,u,u,1,5,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm6 = [u,4,u,u,u,1,5,u]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm6, %ymm6
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm5 = ymm6[0,1],ymm5[2,3],ymm6[4,5,6],ymm5[7]
; AVX2-FAST-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm0
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm5[1,2,3],ymm0[4],ymm5[5,6,7]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = <u,u,2,6,u,u,u,3>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = [u,u,2,6,u,u,u,3]
; AVX2-FAST-NEXT: vpermps %ymm4, %ymm5, %ymm4
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = <5,u,u,u,2,6,u,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm5 = [5,u,u,u,2,6,u,u]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm5, %ymm5
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm4 = ymm5[0,1],ymm4[2,3],ymm5[4,5],ymm4[6,7]
; AVX2-FAST-NEXT: vpermpd {{.*#+}} ymm5 = ymm3[0,1,2,1]
@@ -264,16 +264,16 @@ define void @store_i32_stride5_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovaps (%r8), %xmm4
; AVX2-FAST-PERLANE-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm5
; AVX2-FAST-PERLANE-NEXT: vinsertf128 $1, %xmm3, %ymm2, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = <u,u,0,4,u,u,u,1>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = [u,u,0,4,u,u,u,1]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm6, %ymm7, %ymm7
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm8 = <u,4,u,u,u,1,5,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm8 = [u,4,u,u,u,1,5,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm5, %ymm8, %ymm8
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm7 = ymm8[0,1],ymm7[2,3],ymm8[4,5,6],ymm7[7]
; AVX2-FAST-PERLANE-NEXT: vinsertf128 $1, %xmm4, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm7[1,2,3],ymm0[4],ymm7[5,6,7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = <u,u,2,6,u,u,u,3>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = [u,u,2,6,u,u,u,3]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm5, %ymm7, %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = <5,u,u,u,2,6,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm7 = [5,u,u,u,2,6,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm6, %ymm7, %ymm6
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm5 = ymm6[0,1],ymm5[2,3],ymm6[4,5],ymm5[6,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm6 = ymm4[0,1,2,1]
@@ -528,7 +528,7 @@ define void @store_i32_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovaps (%rsi), %xmm6
; AVX2-FAST-NEXT: vmovaps (%rdi), %xmm7
; AVX2-FAST-NEXT: vunpckhps {{.*#+}} xmm5 = xmm7[2],xmm6[2],xmm7[3],xmm6[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm8 = <0,1,0,1,u,u,2,2>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm8 = [0,1,0,1,u,u,2,2]
; AVX2-FAST-NEXT: vpermps %ymm5, %ymm8, %ymm5
; AVX2-FAST-NEXT: vmovaps (%rdx), %xmm9
; AVX2-FAST-NEXT: vmovaps (%rcx), %xmm10
@@ -541,7 +541,7 @@ define void @store_i32_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vunpcklps {{.*#+}} xmm9 = xmm9[0],xmm10[0],xmm9[1],xmm10[1]
; AVX2-FAST-NEXT: vpermps %ymm9, %ymm8, %ymm8
; AVX2-FAST-NEXT: vunpcklps {{.*#+}} xmm6 = xmm7[0],xmm6[0],xmm7[1],xmm6[1]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm7 = <0,1,u,u,3,2,3,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm7 = [0,1,u,u,3,2,3,u]
; AVX2-FAST-NEXT: vpermps %ymm6, %ymm7, %ymm7
; AVX2-FAST-NEXT: vblendps {{.*#+}} ymm7 = ymm7[0,1],ymm8[2,3],ymm7[4,5,6],ymm8[7]
; AVX2-FAST-NEXT: vinsertf128 $1, (%r8), %ymm6, %ymm6
@@ -648,13 +648,13 @@ define void @store_i32_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovdqa (%r8), %ymm2
; AVX512-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = <6,14,u,23,31,7,15,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = [6,14,u,23,31,7,15,u]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm3
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = <11,19,27,u,4,12,20,28,u,5,13,21,29,u,6,14>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [11,19,27,u,4,12,20,28,u,5,13,21,29,u,6,14]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm4
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,19,4,5,6,7,20,9,10,11,12,21,14,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm4, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,8,16,24,u,1,9,17,25,u,2,10,18,26,u,3>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,8,16,24,u,1,9,17,25,u,2,10,18,26,u,3]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm4
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,16,5,6,7,8,17,10,11,12,13,18,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm4, %zmm0
@@ -1125,7 +1125,7 @@ define void @store_i32_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovaps (%rdi), %xmm0
; AVX2-FAST-NEXT: vmovaps 32(%rdi), %xmm10
; AVX2-FAST-NEXT: vunpckhps {{.*#+}} xmm5 = xmm0[2],xmm6[2],xmm0[3],xmm6[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm12 = <0,1,0,1,u,u,2,2>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm12 = [0,1,0,1,u,u,2,2]
; AVX2-FAST-NEXT: vpermps %ymm5, %ymm12, %ymm5
; AVX2-FAST-NEXT: vmovaps (%rdx), %xmm8
; AVX2-FAST-NEXT: vmovaps 32(%rdx), %xmm14
@@ -2486,7 +2486,7 @@ define void @store_i32_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovaps 64(%rdi), %xmm13
; AVX2-FAST-NEXT: vmovaps %xmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vunpckhps {{.*#+}} xmm0 = xmm7[2],xmm6[2],xmm7[3],xmm6[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm12 = <0,1,0,1,u,u,2,2>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm12 = [0,1,0,1,u,u,2,2]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm12, %ymm0
; AVX2-FAST-NEXT: vmovaps (%rdx), %xmm8
; AVX2-FAST-NEXT: vmovaps 32(%rdx), %xmm2
@@ -5324,7 +5324,7 @@ define void @store_i32_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX2-FAST-NEXT: vmovaps 64(%rdi), %xmm6
; AVX2-FAST-NEXT: vmovaps %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vunpckhps {{.*#+}} xmm0 = xmm14[2],xmm11[2],xmm14[3],xmm11[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm9 = <0,1,0,1,u,u,2,2>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm9 = [0,1,0,1,u,u,2,2]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm9, %ymm2
; AVX2-FAST-NEXT: vmovaps (%rdx), %xmm15
; AVX2-FAST-NEXT: vmovaps 32(%rdx), %xmm10
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll
index a781ae5eb896758..e8d7eb34988b39e 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-6.ll
@@ -104,7 +104,7 @@ define void @store_i32_stride6_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovlhps {{.*#+}} xmm3 = xmm2[0],xmm1[0]
; AVX2-FAST-NEXT: vshufps {{.*#+}} ymm3 = ymm0[1,3],ymm3[1,3],ymm0[5,7],ymm3[5,7]
; AVX2-FAST-NEXT: vpermpd {{.*#+}} ymm3 = ymm3[2,1,2,3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm4 = <0,2,4,6,u,u,1,3>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm4 = [0,2,4,6,u,u,1,3]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm4, %ymm0
; AVX2-FAST-NEXT: vunpcklps {{.*#+}} xmm1 = xmm2[0],xmm1[0],xmm2[1],xmm1[1]
; AVX2-FAST-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm1
@@ -152,7 +152,7 @@ define void @store_i32_stride6_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovlhps {{.*#+}} xmm2 = xmm3[0],xmm2[0]
; AVX512-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX512-NEXT: vinsertf32x4 $2, %xmm2, %zmm0, %zmm0
-; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = <0,2,4,6,8,10,1,3,5,7,9,11,u,u,u,u>
+; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = [0,2,4,6,8,10,1,3,5,7,9,11,u,u,u,u]
; AVX512-NEXT: vpermps %zmm0, %zmm1, %zmm0
; AVX512-NEXT: vextractf32x4 $2, %zmm0, 32(%rax)
; AVX512-NEXT: vmovaps %ymm0, (%rax)
@@ -734,15 +734,15 @@ define void @store_i32_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,8,16,24,u,u,1,9,17,25,u,u,2,10,18,26>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,8,16,24,u,u,1,9,17,25,u,u,2,10,18,26]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,16,24,6,7,8,9,17,25,12,13,14,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,3,11,19,27,u,u,4,12,20,28,u,u,5,13>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,3,11,19,27,u,u,4,12,20,28,u,u,5,13]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [18,26,2,3,4,5,19,27,8,9,10,11,20,28,14,15]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <5,13,u,u,22,30,6,14,u,u,23,31,7,15,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [5,13,u,u,22,30,6,14,u,u,23,31,7,15,u,u]
; AVX512-NEXT: vpermi2d %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,21,29,4,5,6,7,22,30,10,11,12,13,23,31]
; AVX512-NEXT: vpermi2d %zmm2, %zmm3, %zmm0
@@ -1657,17 +1657,17 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: movb $36, %cl
; AVX512F-SLOW-NEXT: kmovw %ecx, %k1
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm6 {%k1} = zmm8[0,1,0,1,2,3,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm6, %zmm7
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm6
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28]
; AVX512F-SLOW-NEXT: vpermi2d %zmm5, %zmm4, %zmm7
; AVX512F-SLOW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [12,28,11,27,0,0,13,29,12,28,11,27,0,0,13,29]
; AVX512F-SLOW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512F-SLOW-NEXT: vpermi2d %zmm3, %zmm2, %zmm8
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm8, %zmm7
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm8
@@ -1680,7 +1680,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: movb $-110, %cl
; AVX512F-SLOW-NEXT: kmovw %ecx, %k2
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm7, %zmm9 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm9, %zmm7
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm9
@@ -1691,7 +1691,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512F-SLOW-NEXT: vpermi2d %zmm3, %zmm2, %zmm10
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm7, %zmm10 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm10, %zmm7
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm10
@@ -1701,7 +1701,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa (%rdi), %ymm11
; AVX512F-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm11 = ymm11[2],mem[2],ymm11[3],mem[3],ymm11[6],mem[6],ymm11[7],mem[7]
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm7 {%k1} = zmm11[2,3,2,3,2,3,2,3]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm7, %zmm11
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm11, %zmm7
@@ -1710,7 +1710,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermi2d %zmm5, %zmm4, %zmm11
; AVX512F-SLOW-NEXT: vpunpckhdq {{.*#+}} zmm2 = zmm2[2],zmm3[2],zmm2[3],zmm3[3],zmm2[6],zmm3[6],zmm2[7],zmm3[7],zmm2[10],zmm3[10],zmm2[11],zmm3[11],zmm2[14],zmm3[14],zmm2[15],zmm3[15]
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm11 {%k1} = zmm2[6,7,6,7,6,7,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-SLOW-NEXT: vpermi2d %zmm1, %zmm11, %zmm2
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm2, %zmm1
@@ -1741,11 +1741,11 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: movb $-110, %cl
; AVX512F-FAST-NEXT: kmovw %ecx, %k2
; AVX512F-FAST-NEXT: vmovdqa64 %zmm6, %zmm7 {%k2}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm7, %zmm8
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm8, %zmm6
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,3,19,u,u,7,23,4,20,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,3,19,u,u,7,23,4,20,u,u,u,u]
; AVX512F-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm7
; AVX512F-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [4,20,3,19,0,0,5,21,4,20,3,19,0,0,5,21]
; AVX512F-FAST-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
@@ -1753,7 +1753,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: movb $36, %cl
; AVX512F-FAST-NEXT: kmovw %ecx, %k1
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm8, %zmm7
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm8
@@ -1764,7 +1764,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm9
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm9 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm9, %zmm7
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm9
@@ -1775,17 +1775,17 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vpermi2d %zmm5, %zmm3, %zmm10
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm10 {%k2}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm10, %zmm7
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm10
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28]
; AVX512F-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm7
; AVX512F-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [12,28,11,27,0,0,13,29,12,28,11,27,0,0,13,29]
; AVX512F-FAST-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm11 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm11, %zmm7
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm11
@@ -1796,7 +1796,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm3
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm3 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-FAST-NEXT: vpermi2d %zmm1, %zmm3, %zmm2
; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512F-FAST-NEXT: vpermi2d %zmm0, %zmm2, %zmm1
@@ -1827,17 +1827,17 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: movb $36, %cl
; AVX512BW-SLOW-NEXT: kmovd %ecx, %k1
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm6 {%k1} = zmm8[0,1,0,1,2,3,6,7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm6, %zmm7
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm6
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm5, %zmm4, %zmm7
; AVX512BW-SLOW-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [12,28,11,27,0,0,13,29,12,28,11,27,0,0,13,29]
; AVX512BW-SLOW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm3, %zmm2, %zmm8
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm8, %zmm7
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm8
@@ -1850,7 +1850,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: movb $-110, %cl
; AVX512BW-SLOW-NEXT: kmovd %ecx, %k2
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm7, %zmm9 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm9, %zmm7
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm9
@@ -1861,7 +1861,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm3, %zmm2, %zmm10
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm7, %zmm10 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm10, %zmm7
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm7, %zmm10
@@ -1871,7 +1871,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vmovdqa (%rdi), %ymm11
; AVX512BW-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm11 = ymm11[2],mem[2],ymm11[3],mem[3],ymm11[6],mem[6],ymm11[7],mem[7]
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm7 {%k1} = zmm11[2,3,2,3,2,3,2,3]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm7, %zmm11
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm11, %zmm7
@@ -1880,7 +1880,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermi2d %zmm5, %zmm4, %zmm11
; AVX512BW-SLOW-NEXT: vpunpckhdq {{.*#+}} zmm2 = zmm2[2],zmm3[2],zmm2[3],zmm3[3],zmm2[6],zmm3[6],zmm2[7],zmm3[7],zmm2[10],zmm3[10],zmm2[11],zmm3[11],zmm2[14],zmm3[14],zmm2[15],zmm3[15]
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm11 {%k1} = zmm2[6,7,6,7,6,7,6,7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm1, %zmm11, %zmm2
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512BW-SLOW-NEXT: vpermi2d %zmm0, %zmm2, %zmm1
@@ -1911,11 +1911,11 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: movb $-110, %cl
; AVX512BW-FAST-NEXT: kmovd %ecx, %k2
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm6, %zmm7 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm7, %zmm8
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm8, %zmm6
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,3,19,u,u,7,23,4,20,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,3,19,u,u,7,23,4,20,u,u,u,u]
; AVX512BW-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm7
; AVX512BW-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm8 = [4,20,3,19,0,0,5,21,4,20,3,19,0,0,5,21]
; AVX512BW-FAST-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3]
@@ -1923,7 +1923,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: movb $36, %cl
; AVX512BW-FAST-NEXT: kmovd %ecx, %k1
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm8, %zmm7
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm8
@@ -1934,7 +1934,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm9
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm9 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm9, %zmm7
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm9
@@ -1945,17 +1945,17 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-FAST-NEXT: vpermi2d %zmm5, %zmm3, %zmm10
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm10 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm10, %zmm7
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm10
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,0,16,11,27,u,u,15,31,12,28,u,u,12,28]
; AVX512BW-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm7
; AVX512BW-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm11 = [12,28,11,27,0,0,13,29,12,28,11,27,0,0,13,29]
; AVX512BW-FAST-NEXT: # zmm11 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-FAST-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm11 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm11, %zmm7
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm7, %zmm11
@@ -1966,7 +1966,7 @@ define void @store_i32_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: # zmm3 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-FAST-NEXT: vpermi2d %zmm4, %zmm2, %zmm3
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm3 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-FAST-NEXT: vpermi2d %zmm1, %zmm3, %zmm2
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512BW-FAST-NEXT: vpermi2d %zmm0, %zmm2, %zmm1
@@ -3910,7 +3910,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: movb $36, %dl
; AVX512F-SLOW-NEXT: kmovw %edx, %k1
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm17 {%k1} = zmm18[0,1,0,1,2,3,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm18, %zmm17
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm22, %zmm17
@@ -3923,7 +3923,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm0, %zmm20
; AVX512F-SLOW-NEXT: vpermt2d %zmm10, %zmm18, %zmm20
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm20, %zmm11 {%k1}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm8, %zmm20, %zmm11
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm9, %zmm21, %zmm11
@@ -3934,7 +3934,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: movb $-110, %cl
; AVX512F-SLOW-NEXT: kmovw %ecx, %k2
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm23, %zmm13 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm8, %zmm23, %zmm13
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm9, %zmm24, %zmm13
@@ -3943,7 +3943,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm0, %zmm26
; AVX512F-SLOW-NEXT: vpermt2d %zmm10, %zmm25, %zmm26
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm26, %zmm15 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm8, %zmm26, %zmm15
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm9, %zmm27, %zmm15
@@ -3967,7 +3967,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa64 64(%rdi), %ymm22
; AVX512F-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm22 = ymm22[2],mem[2],ymm22[3],mem[3],ymm22[6],mem[6],ymm22[7],mem[7]
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm20 {%k1} = zmm22[2,3,2,3,2,3,2,3]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-SLOW-NEXT: vpermt2d %zmm8, %zmm22, %zmm20
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512F-SLOW-NEXT: vpermt2d %zmm9, %zmm23, %zmm20
@@ -3981,7 +3981,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermt2d %zmm10, %zmm21, %zmm0
; AVX512F-SLOW-NEXT: vpunpckhdq {{.*#+}} zmm5 = zmm5[2],zmm7[2],zmm5[3],zmm7[3],zmm5[6],zmm7[6],zmm5[7],zmm7[7],zmm5[10],zmm7[10],zmm5[11],zmm7[11],zmm5[14],zmm7[14],zmm5[15],zmm7[15]
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm0 {%k1} = zmm5[6,7,6,7,6,7,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-SLOW-NEXT: vpermt2d %zmm8, %zmm5, %zmm0
; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512F-SLOW-NEXT: vpermt2d %zmm9, %zmm7, %zmm0
@@ -4051,7 +4051,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vpermt2d %zmm24, %zmm16, %zmm15
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm25
; AVX512F-FAST-NEXT: vpermt2d %zmm24, %zmm26, %zmm7
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = <3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = [3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u]
; AVX512F-FAST-NEXT: vpermt2d %zmm24, %zmm26, %zmm25
; AVX512F-FAST-NEXT: vbroadcasti32x4 {{.*#+}} zmm24 = [6,22,7,23,6,22,7,23,6,22,7,23,6,22,7,23]
; AVX512F-FAST-NEXT: # zmm24 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
@@ -4081,22 +4081,22 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 %zmm20, %zmm3 {%k1}
; AVX512F-FAST-NEXT: vmovdqa64 64(%r8), %zmm13
; AVX512F-FAST-NEXT: vmovdqa64 %zmm22, %zmm6 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm11, %zmm14, %zmm2
; AVX512F-FAST-NEXT: vmovdqa64 %zmm23, %zmm9 {%k2}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm19 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm19, %zmm3
; AVX512F-FAST-NEXT: vmovdqa64 %zmm26, %zmm15 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm20, %zmm6
; AVX512F-FAST-NEXT: vmovdqa64 %zmm25, %zmm18 {%k1}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm22, %zmm9
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm0 {%k2}
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm7, %zmm15
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm14, %zmm0
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm13, %zmm14, %zmm18
; AVX512F-FAST-NEXT: vmovdqa64 %zmm12, %zmm5 {%k1}
; AVX512F-FAST-NEXT: vpermt2d %zmm11, %zmm19, %zmm5
@@ -4185,7 +4185,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: movb $36, %dl
; AVX512BW-SLOW-NEXT: kmovd %edx, %k1
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm17 {%k1} = zmm18[0,1,0,1,2,3,6,7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm18, %zmm17
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,18,2,3,4,5,6,19,8,9,10,11,12,20,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm22, %zmm17
@@ -4198,7 +4198,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm0, %zmm20
; AVX512BW-SLOW-NEXT: vpermt2d %zmm10, %zmm18, %zmm20
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm20, %zmm11 {%k1}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm8, %zmm20, %zmm11
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,26,2,3,4,5,6,27,8,9,10,11,12,28,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm9, %zmm21, %zmm11
@@ -4209,7 +4209,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: movb $-110, %cl
; AVX512BW-SLOW-NEXT: kmovd %ecx, %k2
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm23, %zmm13 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm8, %zmm23, %zmm13
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [0,1,2,3,4,24,6,7,8,9,10,25,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm9, %zmm24, %zmm13
@@ -4218,7 +4218,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm0, %zmm26
; AVX512BW-SLOW-NEXT: vpermt2d %zmm10, %zmm25, %zmm26
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm26, %zmm15 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm8, %zmm26, %zmm15
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,1,2,3,4,16,6,7,8,9,10,17,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm9, %zmm27, %zmm15
@@ -4242,7 +4242,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vmovdqa64 64(%rdi), %ymm22
; AVX512BW-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm22 = ymm22[2],mem[2],ymm22[3],mem[3],ymm22[6],mem[6],ymm22[7],mem[7]
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm20 {%k1} = zmm22[2,3,2,3,2,3,2,3]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm8, %zmm22, %zmm20
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,2,21,4,5,6,7,8,22,10,11,12,13,14,23]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm9, %zmm23, %zmm20
@@ -4256,7 +4256,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermt2d %zmm10, %zmm21, %zmm0
; AVX512BW-SLOW-NEXT: vpunpckhdq {{.*#+}} zmm5 = zmm5[2],zmm7[2],zmm5[3],zmm7[3],zmm5[6],zmm7[6],zmm5[7],zmm7[7],zmm5[10],zmm7[10],zmm5[11],zmm7[11],zmm5[14],zmm7[14],zmm5[15],zmm7[15]
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm0 {%k1} = zmm5[6,7,6,7,6,7,6,7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm8, %zmm5, %zmm0
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,29,4,5,6,7,8,30,10,11,12,13,14,31]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm9, %zmm7, %zmm0
@@ -4326,7 +4326,7 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: vpermt2d %zmm24, %zmm16, %zmm15
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm25
; AVX512BW-FAST-NEXT: vpermt2d %zmm24, %zmm26, %zmm7
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = <3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm26 = [3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u]
; AVX512BW-FAST-NEXT: vpermt2d %zmm24, %zmm26, %zmm25
; AVX512BW-FAST-NEXT: vbroadcasti32x4 {{.*#+}} zmm24 = [6,22,7,23,6,22,7,23,6,22,7,23,6,22,7,23]
; AVX512BW-FAST-NEXT: # zmm24 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
@@ -4356,22 +4356,22 @@ define void @store_i32_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm20, %zmm3 {%k1}
; AVX512BW-FAST-NEXT: vmovdqa64 64(%r8), %zmm13
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm22, %zmm6 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm11, %zmm14, %zmm2
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm23, %zmm9 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm19 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm19, %zmm3
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm26, %zmm15 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm20, %zmm6
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm25, %zmm18 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm22, %zmm9
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm0 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm7, %zmm15
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm14, %zmm0
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm14 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm13, %zmm14, %zmm18
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm12, %zmm5 {%k1}
; AVX512BW-FAST-NEXT: vpermt2d %zmm11, %zmm19, %zmm5
@@ -8476,7 +8476,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm29
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm29 {%k1} = zmm0[0,1,0,1,2,3,6,7]
; AVX512F-SLOW-NEXT: vmovdqa64 (%r8), %zmm4
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm0, %zmm3
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa64 64(%r8), %zmm3
@@ -8532,7 +8532,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm31
; AVX512F-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm1, %zmm31 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm16
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm10
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm8
@@ -8540,7 +8540,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm5
; AVX512F-SLOW-NEXT: vmovdqu64 (%rsp), %zmm1 # 64-byte Reload
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm1, %zmm5 {%k2}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm20
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm25
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm9
@@ -8548,7 +8548,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm11
; AVX512F-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11 {%k1}
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm23
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm26
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm27
@@ -8565,14 +8565,14 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-SLOW-NEXT: vmovdqa 192(%rdi), %ymm1
; AVX512F-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm1 = ymm1[2],mem[2],ymm1[3],mem[3],ymm1[6],mem[6],ymm1[7],mem[7]
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm21 {%k1} = zmm1[2,3,2,3,2,3,2,3]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm19
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm22
; AVX512F-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm18
; AVX512F-SLOW-NEXT: vpermt2d %zmm6, %zmm1, %zmm21
; AVX512F-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm24 {%k1} = zmm1[6,7,6,7,6,7,6,7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm24
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm17 {%k1} = zmm30[6,7,6,7,6,7,6,7]
; AVX512F-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm17
@@ -8668,7 +8668,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 %zmm8, %zmm1
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512F-FAST-NEXT: vmovdqa64 %zmm1, %zmm16
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u]
; AVX512F-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [4,20,3,19,0,0,5,21,4,20,3,19,0,0,5,21]
; AVX512F-FAST-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512F-FAST-NEXT: vmovdqa64 %zmm8, %zmm1
@@ -8803,22 +8803,22 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-FAST-NEXT: vmovdqa64 %zmm7, %zmm19 {%k1}
; AVX512F-FAST-NEXT: vmovdqa64 %zmm25, %zmm5 {%k1}
; AVX512F-FAST-NEXT: vmovdqa64 (%r8), %zmm0
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm2, %zmm16
; AVX512F-FAST-NEXT: vmovdqu64 %zmm16, (%rsp) # 64-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm7, %zmm17
; AVX512F-FAST-NEXT: vmovdqu64 %zmm17, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm18, %zmm23
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm25, %zmm27
; AVX512F-FAST-NEXT: vmovdqu64 %zmm27, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm16 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm16 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm16, %zmm19
; AVX512F-FAST-NEXT: vmovdqa64 %zmm16, %zmm17
; AVX512F-FAST-NEXT: vmovdqu64 %zmm19, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512F-FAST-NEXT: vpermt2d %zmm0, %zmm31, %zmm5
; AVX512F-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm27 # 64-byte Reload
; AVX512F-FAST-NEXT: vmovdqa64 %zmm22, %zmm27 {%k2}
@@ -9045,7 +9045,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm29
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm29 {%k1} = zmm0[0,1,0,1,2,3,6,7]
; AVX512BW-SLOW-NEXT: vmovdqa64 (%r8), %zmm4
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm0, %zmm3
; AVX512BW-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-SLOW-NEXT: vmovdqa64 64(%r8), %zmm3
@@ -9101,7 +9101,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm31
; AVX512BW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm1, %zmm31 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm16
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm10
; AVX512BW-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm8
@@ -9109,7 +9109,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm5
; AVX512BW-SLOW-NEXT: vmovdqu64 (%rsp), %zmm1 # 64-byte Reload
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm1, %zmm5 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm20
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm25
; AVX512BW-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm9
@@ -9117,7 +9117,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vpermi2d %zmm7, %zmm12, %zmm11
; AVX512BW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11 {%k1}
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm23
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm26
; AVX512BW-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm27
@@ -9134,14 +9134,14 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-SLOW-NEXT: vmovdqa 192(%rdi), %ymm1
; AVX512BW-SLOW-NEXT: vpunpckhdq {{.*#+}} ymm1 = ymm1[2],mem[2],ymm1[3],mem[3],ymm1[6],mem[6],ymm1[7],mem[7]
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm21 {%k1} = zmm1[2,3,2,3,2,3,2,3]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm19
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm22
; AVX512BW-SLOW-NEXT: vpermt2d %zmm2, %zmm1, %zmm18
; AVX512BW-SLOW-NEXT: vpermt2d %zmm6, %zmm1, %zmm21
; AVX512BW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm24 {%k1} = zmm1[6,7,6,7,6,7,6,7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm4, %zmm1, %zmm24
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm17 {%k1} = zmm30[6,7,6,7,6,7,6,7]
; AVX512BW-SLOW-NEXT: vpermt2d %zmm3, %zmm1, %zmm17
@@ -9237,7 +9237,7 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm8, %zmm1
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm1, %zmm16
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [3,19,0,16,3,19,0,16,7,23,4,20,u,u,u,u]
; AVX512BW-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm10 = [4,20,3,19,0,0,5,21,4,20,3,19,0,0,5,21]
; AVX512BW-FAST-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm8, %zmm1
@@ -9372,22 +9372,22 @@ define void @store_i32_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm7, %zmm19 {%k1}
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm25, %zmm5 {%k1}
; AVX512BW-FAST-NEXT: vmovdqa64 (%r8), %zmm0
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,16,u,6,7,8,9,17,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm2, %zmm16
; AVX512BW-FAST-NEXT: vmovdqu64 %zmm16, (%rsp) # 64-byte Spill
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [18,u,2,3,4,5,19,u,8,9,10,11,20,u,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm7, %zmm17
; AVX512BW-FAST-NEXT: vmovdqu64 %zmm17, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = <0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm18 = [0,1,21,u,4,5,6,7,22,u,10,11,12,13,23,u]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm18, %zmm23
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = <0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm25 = [0,1,2,3,24,u,6,7,8,9,25,u,12,13,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm25, %zmm27
; AVX512BW-FAST-NEXT: vmovdqu64 %zmm27, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm16 = <26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm16 = [26,u,2,3,4,5,27,u,8,9,10,11,28,u,14,15]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm16, %zmm19
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm16, %zmm17
; AVX512BW-FAST-NEXT: vmovdqu64 %zmm19, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [0,1,29,u,4,5,6,7,30,u,10,11,12,13,31,u]
; AVX512BW-FAST-NEXT: vpermt2d %zmm0, %zmm31, %zmm5
; AVX512BW-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm27 # 64-byte Reload
; AVX512BW-FAST-NEXT: vmovdqa64 %zmm22, %zmm27 {%k2}
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-7.ll
index e5992acfaa2a770..9ea961e6802dd8a 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-7.ll
@@ -101,7 +101,7 @@ define void @store_i32_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vshufps {{.*#+}} ymm3 = ymm0[0,2,2,3,4,6,6,7]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm3 = ymm3[0,1,0,2]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6],ymm1[7]
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm3 = <3,5,7,u>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} xmm3 = [3,5,7,u]
; AVX2-SLOW-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vbroadcastf128 {{.*#+}} ymm3 = [3,5,0,1,3,5,0,1]
; AVX2-SLOW-NEXT: # ymm3 = mem[0,1,0,1]
@@ -131,13 +131,13 @@ define void @store_i32_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm3[0],ymm0[2],ymm3[2]
; AVX2-FAST-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,2,1,3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm1 = <3,5,7,u>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} xmm1 = [3,5,7,u]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm3 = [3,5,0,1,3,5,0,1]
; AVX2-FAST-NEXT: # ymm3 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermps %ymm0, %ymm3, %ymm3
; AVX2-FAST-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm3[3]
-; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm4 = <0,2,4,6,u,u,u,1>
+; AVX2-FAST-NEXT: vmovaps {{.*#+}} ymm4 = [0,2,4,6,u,u,u,1]
; AVX2-FAST-NEXT: vpermps %ymm2, %ymm4, %ymm2
; AVX2-FAST-NEXT: vbroadcastf128 {{.*#+}} ymm4 = [0,2,4,0,0,2,4,0]
; AVX2-FAST-NEXT: # ymm4 = mem[0,1,0,1]
@@ -172,7 +172,7 @@ define void @store_i32_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} ymm3 = ymm0[0,2,2,3,4,6,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm3 = ymm3[0,1,0,2]
; AVX2-FAST-PERLANE-NEXT: vblendps {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6],ymm1[7]
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm3 = <3,5,7,u>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} xmm3 = [3,5,7,u]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vbroadcastf128 {{.*#+}} ymm3 = [3,5,0,1,3,5,0,1]
; AVX2-FAST-PERLANE-NEXT: # ymm3 = mem[0,1,0,1]
@@ -201,7 +201,7 @@ define void @store_i32_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-SLOW-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512-SLOW-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512-SLOW-NEXT: vpunpcklqdq {{.*#+}} ymm0 = ymm0[0],ymm3[0],ymm0[2],ymm3[2]
-; AVX512-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,2,4,6,16,20,18,1,3,5,7,17,21,19,u,u>
+; AVX512-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,2,4,6,16,20,18,1,3,5,7,17,21,19,u,u]
; AVX512-SLOW-NEXT: vpermi2d %zmm0, %zmm2, %zmm1
; AVX512-SLOW-NEXT: vextracti32x4 $2, %zmm1, 32(%rax)
; AVX512-SLOW-NEXT: vextracti32x4 $3, %zmm1, %xmm0
@@ -225,10 +225,10 @@ define void @store_i32_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2
; AVX512-FAST-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u>
+; AVX512-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u]
; AVX512-FAST-NEXT: vpermi2q %ymm3, %ymm0, %ymm1
; AVX512-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm2, %zmm0
-; AVX512-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,2,4,6,8,10,12,1,3,5,7,9,11,13,u,u>
+; AVX512-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,2,4,6,8,10,12,1,3,5,7,9,11,13,u,u]
; AVX512-FAST-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512-FAST-NEXT: vextracti32x4 $2, %zmm0, 32(%rax)
; AVX512-FAST-NEXT: vextracti32x4 $3, %zmm0, %xmm1
@@ -529,7 +529,7 @@ define void @store_i32_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vinserti32x4 $2, (%r10), %zmm1, %zmm1
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,4,8,12,16,20,24,1,5,9,13,17,21,25,2,6]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <10,14,18,22,26,3,7,11,15,19,23,27,u,u,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [10,14,18,22,26,3,7,11,15,19,23,27,u,u,u,u]
; AVX512-NEXT: vpermi2d %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vextracti32x4 $2, %zmm3, 96(%rax)
; AVX512-NEXT: vmovdqa64 %zmm2, (%rax)
@@ -1118,23 +1118,23 @@ define void @store_i32_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vbroadcasti128 {{.*#+}} ymm1 = [6,23,31,7,6,23,31,7]
; AVX512F-NEXT: # ymm1 = mem[0,1,0,1]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm3, %zmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,8,16,u,u,u,u,1,9,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,8,16,u,u,u,u,1,9,17,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,16,24,u,u,u,1,9,17,25,u,u,u,2,10>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,16,24,u,u,u,1,9,17,25,u,u,u,2,10]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm4, %zmm7
; AVX512F-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa32 %zmm6, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,2,10,18,u,u,u,u,3,11,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,2,10,18,u,u,u,u,3,11,19,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <2,10,u,u,u,19,27,3,11,u,u,u,20,28,4,12>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [2,10,u,u,u,19,27,3,11,u,u,u,20,28,4,12]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm5, %zmm8
; AVX512F-NEXT: movw $3612, %cx # imm = 0xE1C
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa32 %zmm6, %zmm8 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,5,13,21,29,u,u,u,6,14,22,30,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,5,13,21,29,u,u,u,6,14,22,30,u,u]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm4, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <4,12,20,u,u,u,u,5,13,21,u,u,u,u,6,14>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [4,12,20,u,u,u,u,5,13,21,u,u,u,u,6,14]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
; AVX512F-NEXT: movw $15480, %cx # imm = 0x3C78
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -1164,23 +1164,23 @@ define void @store_i32_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vbroadcasti128 {{.*#+}} ymm1 = [6,23,31,7,6,23,31,7]
; AVX512BW-NEXT: # ymm1 = mem[0,1,0,1]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm3, %zmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,8,16,u,u,u,u,1,9,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,8,16,u,u,u,u,1,9,17,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,16,24,u,u,u,1,9,17,25,u,u,u,2,10>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,16,24,u,u,u,1,9,17,25,u,u,u,2,10]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm4, %zmm7
; AVX512BW-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm6, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,2,10,18,u,u,u,u,3,11,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,2,10,18,u,u,u,u,3,11,19,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <2,10,u,u,u,19,27,3,11,u,u,u,20,28,4,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [2,10,u,u,u,19,27,3,11,u,u,u,20,28,4,12]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm5, %zmm8
; AVX512BW-NEXT: movw $3612, %cx # imm = 0xE1C
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm6, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,5,13,21,29,u,u,u,6,14,22,30,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,5,13,21,29,u,u,u,6,14,22,30,u,u]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm4, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <4,12,20,u,u,u,u,5,13,21,u,u,u,u,6,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [4,12,20,u,u,u,u,5,13,21,u,u,u,u,6,14]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
; AVX512BW-NEXT: movw $15480, %cx # imm = 0x3C78
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -2518,96 +2518,96 @@ define void @store_i32_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%r8), %zmm1
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm2
; AVX512F-NEXT: vmovdqa64 (%r10), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm8
; AVX512F-NEXT: movw $6192, %cx # imm = 0x1830
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa32 %zmm7, %zmm8 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm2, %zmm7
; AVX512F-NEXT: movw $24769, %cx # imm = 0x60C1
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm7, %zmm8 {%k2}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,29,2,3,4,5,6,7,30,9,10,11,12,13,14,31]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm8, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm8
; AVX512F-NEXT: movw $1548, %cx # imm = 0x60C
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm9, %zmm8 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm9, %zmm10
; AVX512F-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm10, %zmm8 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm9
; AVX512F-NEXT: movw $12384, %cx # imm = 0x3060
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm10, %zmm9 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm10, %zmm11
; AVX512F-NEXT: movw $3612, %cx # imm = 0xE1C
; AVX512F-NEXT: kmovw %ecx, %k3
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm9 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm11
; AVX512F-NEXT: vmovdqa32 %zmm10, %zmm11 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm12, %zmm10
; AVX512F-NEXT: movw $15480, %cx # imm = 0x3C78
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm10 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm12
; AVX512F-NEXT: movw $3096, %cx # imm = 0xC18
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm12 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm11, %zmm13
; AVX512F-NEXT: movw $28897, %cx # imm = 0x70E1
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm13, %zmm12 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512F-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm6, %zmm13
; AVX512F-NEXT: movw $-31994, %cx # imm = 0x8306
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm13 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm11, %zmm14
; AVX512F-NEXT: movw $7224, %cx # imm = 0x1C38
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm14, %zmm13 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm4, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm5, %zmm4
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm4 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512F-NEXT: vpermi2d %zmm0, %zmm3, %zmm1
; AVX512F-NEXT: movw $-30962, %cx # imm = 0x870E
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -2633,96 +2633,96 @@ define void @store_i32_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm1
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm2
; AVX512BW-NEXT: vmovdqa64 (%r10), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm8
; AVX512BW-NEXT: movw $6192, %cx # imm = 0x1830
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm2, %zmm7
; AVX512BW-NEXT: movw $24769, %cx # imm = 0x60C1
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm7, %zmm8 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,29,2,3,4,5,6,7,30,9,10,11,12,13,14,31]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm8, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm8
; AVX512BW-NEXT: movw $1548, %cx # imm = 0x60C
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm9, %zmm8 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm9, %zmm10
; AVX512BW-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm8 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm9
; AVX512BW-NEXT: movw $12384, %cx # imm = 0x3060
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm9 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm10, %zmm11
; AVX512BW-NEXT: movw $3612, %cx # imm = 0xE1C
; AVX512BW-NEXT: kmovd %ecx, %k3
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm9 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm10, %zmm11 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm12, %zmm10
; AVX512BW-NEXT: movw $15480, %cx # imm = 0x3C78
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm10 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm12
; AVX512BW-NEXT: movw $3096, %cx # imm = 0xC18
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm12 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm11, %zmm13
; AVX512BW-NEXT: movw $28897, %cx # imm = 0x70E1
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm13, %zmm12 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512BW-NEXT: vpermi2d %zmm5, %zmm3, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm6, %zmm13
; AVX512BW-NEXT: movw $-31994, %cx # imm = 0x8306
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm13 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm11, %zmm14
; AVX512BW-NEXT: movw $7224, %cx # imm = 0x1C38
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm14, %zmm13 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm4, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm5, %zmm4
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm4 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512BW-NEXT: vpermi2d %zmm0, %zmm3, %zmm1
; AVX512BW-NEXT: movw $-30962, %cx # imm = 0x870E
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -5498,42 +5498,42 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm10
; AVX512F-NEXT: vmovdqa64 64(%r9), %zmm26
; AVX512F-NEXT: vmovdqa64 (%rax), %zmm15
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm23, %zmm0
; AVX512F-NEXT: vpermt2d %zmm13, %zmm21, %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm3
; AVX512F-NEXT: vpermt2d %zmm12, %zmm6, %zmm3
; AVX512F-NEXT: movw $6192, %cx # imm = 0x1830
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm3 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u]
; AVX512F-NEXT: vpermi2d %zmm22, %zmm3, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512F-NEXT: vpermt2d %zmm12, %zmm20, %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm3
; AVX512F-NEXT: vpermt2d %zmm23, %zmm11, %zmm3
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm3 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29]
; AVX512F-NEXT: vpermi2d %zmm22, %zmm3, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512F-NEXT: vpermt2d %zmm9, %zmm28, %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm25 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm25 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm7
; AVX512F-NEXT: vpermt2d %zmm8, %zmm25, %zmm7
; AVX512F-NEXT: movw $1548, %cx # imm = 0x60C
; AVX512F-NEXT: kmovw %ecx, %k2
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm7 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm5
; AVX512F-NEXT: vpermt2d %zmm10, %zmm0, %zmm5
; AVX512F-NEXT: vmovdqa64 %zmm23, %zmm30
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm17 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm17 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm16
; AVX512F-NEXT: vpermt2d %zmm18, %zmm17, %zmm16
; AVX512F-NEXT: vmovdqa64 %zmm23, %zmm27
@@ -5544,7 +5544,7 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm13, %zmm28, %zmm23
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm29
; AVX512F-NEXT: vpermt2d %zmm12, %zmm25, %zmm18
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512F-NEXT: vpermt2d %zmm15, %zmm2, %zmm5
; AVX512F-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512F-NEXT: vmovdqa32 %zmm23, %zmm18 {%k2}
@@ -5554,7 +5554,7 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm23
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm28
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm25 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm25 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm26, %zmm25, %zmm22
; AVX512F-NEXT: vmovdqa64 64(%rax), %zmm25
; AVX512F-NEXT: vpermt2d %zmm25, %zmm2, %zmm22
@@ -5562,14 +5562,14 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermi2d %zmm9, %zmm1, %zmm21
; AVX512F-NEXT: vpermi2d %zmm8, %zmm14, %zmm6
; AVX512F-NEXT: vmovdqa32 %zmm21, %zmm6 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512F-NEXT: vpermi2d %zmm4, %zmm10, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31]
; AVX512F-NEXT: vpermi2d %zmm15, %zmm2, %zmm21
; AVX512F-NEXT: movw $-7741, %ax # imm = 0xE1C3
; AVX512F-NEXT: kmovw %eax, %k2
; AVX512F-NEXT: vmovdqa32 %zmm21, %zmm6 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512F-NEXT: vpermt2d %zmm13, %zmm2, %zmm30
; AVX512F-NEXT: movw $-31994, %ax # imm = 0x8306
; AVX512F-NEXT: kmovw %eax, %k2
@@ -5577,56 +5577,56 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermi2d %zmm8, %zmm14, %zmm20
; AVX512F-NEXT: vpermi2d %zmm1, %zmm9, %zmm11
; AVX512F-NEXT: vmovdqa32 %zmm20, %zmm11 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512F-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512F-NEXT: vpermi2d %zmm15, %zmm20, %zmm21
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm26, %zmm20, %zmm5
; AVX512F-NEXT: movw $-30962, %ax # imm = 0x870E
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa32 %zmm21, %zmm11 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm25, %zmm21, %zmm5
; AVX512F-NEXT: movw $7224, %ax # imm = 0x1C38
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa32 %zmm5, %zmm16 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm13, %zmm22, %zmm27
; AVX512F-NEXT: vpermi2d %zmm9, %zmm1, %zmm2
; AVX512F-NEXT: vpermi2d %zmm14, %zmm8, %zmm17
; AVX512F-NEXT: vmovdqa32 %zmm2, %zmm17 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512F-NEXT: vpermt2d %zmm12, %zmm5, %zmm3
; AVX512F-NEXT: movw $3096, %ax # imm = 0xC18
; AVX512F-NEXT: kmovw %eax, %k2
; AVX512F-NEXT: vmovdqa32 %zmm27, %zmm3 {%k2}
; AVX512F-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
; AVX512F-NEXT: vpermt2d %zmm15, %zmm21, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512F-NEXT: vpermt2d %zmm26, %zmm2, %zmm0
; AVX512F-NEXT: vmovdqa32 %zmm20, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512F-NEXT: vpermt2d %zmm25, %zmm20, %zmm0
; AVX512F-NEXT: movw $28897, %ax # imm = 0x70E1
; AVX512F-NEXT: kmovw %eax, %k3
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm3 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512F-NEXT: vpermt2d %zmm13, %zmm0, %zmm31
; AVX512F-NEXT: vpermi2d %zmm9, %zmm1, %zmm22
; AVX512F-NEXT: vpermi2d %zmm8, %zmm14, %zmm5
; AVX512F-NEXT: vmovdqa32 %zmm22, %zmm5 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm12, %zmm21, %zmm24
; AVX512F-NEXT: movw $12384, %ax # imm = 0x3060
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa32 %zmm31, %zmm24 {%k1}
; AVX512F-NEXT: vpermi2d %zmm10, %zmm4, %zmm2
; AVX512F-NEXT: vpermt2d %zmm15, %zmm20, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512F-NEXT: vpermt2d %zmm26, %zmm20, %zmm23
; AVX512F-NEXT: vmovdqa32 %zmm2, %zmm5 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512F-NEXT: vpermt2d %zmm25, %zmm2, %zmm23
; AVX512F-NEXT: movw $15480, %ax # imm = 0x3C78
; AVX512F-NEXT: kmovw %eax, %k2
@@ -5637,17 +5637,17 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
; AVX512F-NEXT: vpermt2d %zmm15, %zmm2, %zmm20
; AVX512F-NEXT: vmovdqa32 %zmm21, %zmm20 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512F-NEXT: vpermt2d %zmm12, %zmm0, %zmm29
; AVX512F-NEXT: vpermt2d %zmm8, %zmm0, %zmm14
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512F-NEXT: vpermt2d %zmm13, %zmm0, %zmm19
; AVX512F-NEXT: vpermt2d %zmm9, %zmm0, %zmm1
; AVX512F-NEXT: vmovdqa32 %zmm29, %zmm19 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm10, %zmm0, %zmm4
; AVX512F-NEXT: vpermt2d %zmm26, %zmm0, %zmm28
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm15, %zmm0, %zmm4
; AVX512F-NEXT: vpermt2d %zmm25, %zmm0, %zmm28
; AVX512F-NEXT: vmovdqa32 %zmm14, %zmm1 {%k1}
@@ -5655,10 +5655,10 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa32 %zmm28, %zmm19 {%k1}
; AVX512F-NEXT: vmovdqa32 %zmm4, %zmm1 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512F-NEXT: vpermi2d %zmm26, %zmm2, %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm4 # 64-byte Reload
; AVX512F-NEXT: vpermi2d %zmm26, %zmm4, %zmm2
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,29,2,3,4,5,6,7,30,9,10,11,12,13,14,31]
@@ -5701,42 +5701,42 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm10
; AVX512BW-NEXT: vmovdqa64 64(%r9), %zmm26
; AVX512BW-NEXT: vmovdqa64 (%rax), %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm23, %zmm0
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm21, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm3
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm6, %zmm3
; AVX512BW-NEXT: movw $6192, %cx # imm = 0x1830
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm3 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u]
; AVX512BW-NEXT: vpermi2d %zmm22, %zmm3, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm20, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm3
; AVX512BW-NEXT: vpermt2d %zmm23, %zmm11, %zmm3
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm3 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29]
; AVX512BW-NEXT: vpermi2d %zmm22, %zmm3, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm0
; AVX512BW-NEXT: vpermt2d %zmm9, %zmm28, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm8, %zmm25, %zmm7
; AVX512BW-NEXT: movw $1548, %cx # imm = 0x60C
; AVX512BW-NEXT: kmovd %ecx, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm7 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm0, %zmm5
; AVX512BW-NEXT: vmovdqa64 %zmm23, %zmm30
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm17 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm17 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm16
; AVX512BW-NEXT: vpermt2d %zmm18, %zmm17, %zmm16
; AVX512BW-NEXT: vmovdqa64 %zmm23, %zmm27
@@ -5747,7 +5747,7 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm28, %zmm23
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm29
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm25, %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512BW-NEXT: vpermt2d %zmm15, %zmm2, %zmm5
; AVX512BW-NEXT: movw $14448, %cx # imm = 0x3870
; AVX512BW-NEXT: vmovdqa32 %zmm23, %zmm18 {%k2}
@@ -5757,7 +5757,7 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm23
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm25, %zmm22
; AVX512BW-NEXT: vmovdqa64 64(%rax), %zmm25
; AVX512BW-NEXT: vpermt2d %zmm25, %zmm2, %zmm22
@@ -5765,14 +5765,14 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2d %zmm9, %zmm1, %zmm21
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm14, %zmm6
; AVX512BW-NEXT: vmovdqa32 %zmm21, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512BW-NEXT: vpermi2d %zmm4, %zmm10, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31]
; AVX512BW-NEXT: vpermi2d %zmm15, %zmm2, %zmm21
; AVX512BW-NEXT: movw $-7741, %ax # imm = 0xE1C3
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm21, %zmm6 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm2, %zmm30
; AVX512BW-NEXT: movw $-31994, %ax # imm = 0x8306
; AVX512BW-NEXT: kmovd %eax, %k2
@@ -5780,56 +5780,56 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm14, %zmm20
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm9, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm20, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512BW-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512BW-NEXT: vpermi2d %zmm15, %zmm20, %zmm21
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm20, %zmm5
; AVX512BW-NEXT: movw $-30962, %ax # imm = 0x870E
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm21, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm25, %zmm21, %zmm5
; AVX512BW-NEXT: movw $7224, %ax # imm = 0x1C38
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm5, %zmm16 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm22, %zmm27
; AVX512BW-NEXT: vpermi2d %zmm9, %zmm1, %zmm2
; AVX512BW-NEXT: vpermi2d %zmm14, %zmm8, %zmm17
; AVX512BW-NEXT: vmovdqa32 %zmm2, %zmm17 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm5, %zmm3
; AVX512BW-NEXT: movw $3096, %ax # imm = 0xC18
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqa32 %zmm27, %zmm3 {%k2}
; AVX512BW-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm15, %zmm21, %zmm20
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa32 %zmm20, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512BW-NEXT: vpermt2d %zmm25, %zmm20, %zmm0
; AVX512BW-NEXT: movw $28897, %ax # imm = 0x70E1
; AVX512BW-NEXT: kmovd %eax, %k3
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm3 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm0, %zmm31
; AVX512BW-NEXT: vpermi2d %zmm9, %zmm1, %zmm22
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm14, %zmm5
; AVX512BW-NEXT: vmovdqa32 %zmm22, %zmm5 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm21, %zmm24
; AVX512BW-NEXT: movw $12384, %ax # imm = 0x3060
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm31, %zmm24 {%k1}
; AVX512BW-NEXT: vpermi2d %zmm10, %zmm4, %zmm2
; AVX512BW-NEXT: vpermt2d %zmm15, %zmm20, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm20, %zmm23
; AVX512BW-NEXT: vmovdqa32 %zmm2, %zmm5 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512BW-NEXT: vpermt2d %zmm25, %zmm2, %zmm23
; AVX512BW-NEXT: movw $15480, %ax # imm = 0x3C78
; AVX512BW-NEXT: kmovd %eax, %k2
@@ -5840,17 +5840,17 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2d %zmm10, %zmm4, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm15, %zmm2, %zmm20
; AVX512BW-NEXT: vmovdqa32 %zmm21, %zmm20 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512BW-NEXT: vpermt2d %zmm12, %zmm0, %zmm29
; AVX512BW-NEXT: vpermt2d %zmm8, %zmm0, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512BW-NEXT: vpermt2d %zmm13, %zmm0, %zmm19
; AVX512BW-NEXT: vpermt2d %zmm9, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqa32 %zmm29, %zmm19 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm0, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm26, %zmm0, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm15, %zmm0, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm25, %zmm0, %zmm28
; AVX512BW-NEXT: vmovdqa32 %zmm14, %zmm1 {%k1}
@@ -5858,10 +5858,10 @@ define void @store_i32_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa32 %zmm28, %zmm19 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm4, %zmm1 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512BW-NEXT: vpermi2d %zmm26, %zmm2, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm4 # 64-byte Reload
; AVX512BW-NEXT: vpermi2d %zmm26, %zmm4, %zmm2
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,29,2,3,4,5,6,7,30,9,10,11,12,13,14,31]
@@ -11476,47 +11476,47 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm6
; AVX512F-NEXT: vmovdqa64 64(%r9), %zmm5
; AVX512F-NEXT: vmovdqa64 128(%r9), %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm7
; AVX512F-NEXT: vpermt2d %zmm2, %zmm1, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm7, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm12
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm12
; AVX512F-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm13
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm13
; AVX512F-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512F-NEXT: vpermt2d %zmm6, %zmm1, %zmm14
; AVX512F-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm14
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm0
@@ -11556,23 +11556,23 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermt2d %zmm22, %zmm1, %zmm4
; AVX512F-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm0, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm11, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm30 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm30 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512F-NEXT: vpermt2d %zmm2, %zmm30, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm1
; AVX512F-NEXT: vpermt2d %zmm3, %zmm4, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm2, %zmm1, %zmm3
; AVX512F-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm24
@@ -11632,29 +11632,29 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm2, %zmm0, %zmm29
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm22
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm9, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm20, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm27 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm27 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm27, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm3
; AVX512F-NEXT: vpermt2d %zmm22, %zmm6, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm18
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm18
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm8, %zmm22
; AVX512F-NEXT: vmovdqa64 64(%rdi), %zmm5
; AVX512F-NEXT: vmovdqa64 64(%rsi), %zmm0
@@ -11714,7 +11714,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm7 {%k1}
; AVX512F-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512F-NEXT: vmovdqa64 (%rax), %zmm28
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm8 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm8
; AVX512F-NEXT: movw $28897, %cx # imm = 0x70E1
@@ -11741,7 +11741,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm25 {%k2}
; AVX512F-NEXT: vmovdqa32 %zmm30, %zmm6 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm2
; AVX512F-NEXT: movw $7224, %cx # imm = 0x1C38
@@ -11756,7 +11756,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm27, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqa32 %zmm2, %zmm6 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm2
; AVX512F-NEXT: movw $-30962, %cx # imm = 0x870E
@@ -11783,7 +11783,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa32 %zmm0, %zmm22 {%k1}
; AVX512F-NEXT: vmovdqa32 %zmm24, %zmm5 {%k1}
; AVX512F-NEXT: vmovdqa32 %zmm21, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm1
; AVX512F-NEXT: movw $-7741, %cx # imm = 0xE1C3
@@ -11795,7 +11795,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm27, %zmm0, %zmm1
; AVX512F-NEXT: vmovdqa32 %zmm1, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm1
; AVX512F-NEXT: movw $14448, %cx # imm = 0x3870
@@ -11821,12 +11821,12 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa32 %zmm1, %zmm19 {%k3}
; AVX512F-NEXT: vmovdqa64 192(%r8), %zmm1
; AVX512F-NEXT: vmovdqa64 192(%r9), %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
; AVX512F-NEXT: vmovdqa64 192(%rax), %zmm10
; AVX512F-NEXT: vpermt2d %zmm10, %zmm0, %zmm11
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm9 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm11 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm0, %zmm11
; AVX512F-NEXT: movw $3612, %ax # imm = 0xE1C
@@ -11847,23 +11847,23 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa32 %zmm11, %zmm24 {%k2}
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm18 # 64-byte Reload
; AVX512F-NEXT: vmovdqa32 %zmm20, %zmm18 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm1, %zmm13
; AVX512F-NEXT: vpermt2d %zmm10, %zmm0, %zmm13
; AVX512F-NEXT: vmovdqa32 %zmm13, %zmm18 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,1,2,3,4,23,u,u,8,9,10,11,24,u,u,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,1,2,3,4,23,u,u,8,9,10,11,24,u,u,15]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm16, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,2,25,u,u,6,7,8,9,26,u,u,13,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,25,u,u,6,7,8,9,26,u,u,13,14,15]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm14, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm14 # 64-byte Reload
; AVX512F-NEXT: vpermi2d %zmm1, %zmm14, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm15, %zmm14
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm15 # 64-byte Reload
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512F-NEXT: vmovdqa32 %zmm15, %zmm0 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm16 # 64-byte Reload
; AVX512F-NEXT: vpermt2d %zmm28, %zmm15, %zmm16
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm21 # 64-byte Reload
@@ -11882,17 +11882,17 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm27, %zmm15, %zmm8
; AVX512F-NEXT: vmovdqa32 %zmm23, %zmm8 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512F-NEXT: vpermt2d %zmm2, %zmm8, %zmm1
; AVX512F-NEXT: vpermt2d %zmm10, %zmm15, %zmm1
; AVX512F-NEXT: vmovdqa32 %zmm12, %zmm1 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,1,2,3,4,5,23,u,8,9,10,11,12,24,u,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,1,2,3,4,5,23,u,8,9,10,11,12,24,u,15]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm3, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,1,2,3,25,u,6,7,8,9,10,26,u,13,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,25,u,6,7,8,9,10,26,u,13,14,15]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm11, %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm13, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u]
; AVX512F-NEXT: vpermi2d %zmm2, %zmm14, %zmm12
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [22,1,2,3,4,5,6,23,8,9,10,11,12,13,24,15]
; AVX512F-NEXT: vpermi2d %zmm10, %zmm8, %zmm2
@@ -11952,47 +11952,47 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm6
; AVX512BW-NEXT: vmovdqa64 64(%r9), %zmm5
; AVX512BW-NEXT: vmovdqa64 128(%r9), %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,1,17,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [2,18,u,u,u,u,u,3,19,u,u,u,u,u,4,20]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm7, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,u,7,23,u,u,u,u,u,8,24,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm12
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,9,25,u,u,u,u,u,10,26,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm13
; AVX512BW-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,11,27,u,u,u,u,u,12,28,u,u,u,u,u,13]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512BW-NEXT: vpermt2d %zmm6, %zmm1, %zmm14
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [13,u,u,u,u,u,30,14,u,u,u,u,u,31,15,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm0
@@ -12032,23 +12032,23 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermt2d %zmm22, %zmm1, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,5,21,u,u,u,u,u,6,22,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,7,23,u,u,u,u,u,8,24,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm11, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = <u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = [u,9,25,u,u,u,u,u,10,26,u,u,u,u,u,11]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm30, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [11,u,u,u,u,u,28,12,u,u,u,u,u,29,13,u]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm3, %zmm4, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,14,30,u,u,u,u,u,15,31,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm24
@@ -12108,29 +12108,29 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm0, %zmm29
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm22
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,16,u,u,u,u,u,1,17,u,u,u,u,u,2,18]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm9, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,u,u,u,3,19,u,u,u,u,u,4,20,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm20, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm27 = <u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm27 = [u,u,u,5,21,u,u,u,u,u,6,22,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm27, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,7,23,u,u,u,u,u,8,24,u,u,u,u,u,9]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [9,u,u,u,u,u,26,10,u,u,u,u,u,27,11,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm3
; AVX512BW-NEXT: vpermt2d %zmm22, %zmm6, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,12,28,u,u,u,u,u,13,29,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm18
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,14,30,u,u,u,u,u,15,31,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm8, %zmm22
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm5
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm0
@@ -12190,7 +12190,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm7 {%k1}
; AVX512BW-NEXT: movq {{[0-9]+}}(%rsp), %rax
; AVX512BW-NEXT: vmovdqa64 (%rax), %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [22,u,u,u,u,5,6,23,u,u,u,u,12,13,24,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm8 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm8
; AVX512BW-NEXT: movw $28897, %cx # imm = 0x70E1
@@ -12217,7 +12217,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm25 {%k2}
; AVX512BW-NEXT: vmovdqa32 %zmm30, %zmm6 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,3,4,25,u,u,u,u,10,11,26,u,u,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm2
; AVX512BW-NEXT: movw $7224, %cx # imm = 0x1C38
@@ -12232,7 +12232,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa32 %zmm2, %zmm6 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,1,2,27,u,u,u,u,8,9,28,u,u,u,u,15]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm2 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm2
; AVX512BW-NEXT: movw $-30962, %cx # imm = 0x870E
@@ -12259,7 +12259,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa32 %zmm0, %zmm22 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm24, %zmm5 {%k1}
; AVX512BW-NEXT: vmovdqa32 %zmm21, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,29,u,u,u,u,6,7,30,u,u,u,u,13,14,31]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm1
; AVX512BW-NEXT: movw $-7741, %cx # imm = 0xE1C3
@@ -12271,7 +12271,7 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqa32 %zmm1, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,4,5,16,u,u,u,u,11,12,17,u,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm1
; AVX512BW-NEXT: movw $14448, %cx # imm = 0x3870
@@ -12297,12 +12297,12 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa32 %zmm1, %zmm19 {%k3}
; AVX512BW-NEXT: vmovdqa64 192(%r8), %zmm1
; AVX512BW-NEXT: vmovdqa64 192(%r9), %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,0,16,u,u,u,u,u,1,17,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm11
; AVX512BW-NEXT: vmovdqa64 192(%rax), %zmm10
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm0, %zmm11
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm9 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,2,3,18,u,u,u,u,9,10,19,u,u,u,u]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm11 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm0, %zmm11
; AVX512BW-NEXT: movw $3612, %ax # imm = 0xE1C
@@ -12323,23 +12323,23 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa32 %zmm11, %zmm24 {%k2}
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm18 # 64-byte Reload
; AVX512BW-NEXT: vmovdqa32 %zmm20, %zmm18 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,18,u,u,u,u,u,3,19,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm1, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm0, %zmm13
; AVX512BW-NEXT: vmovdqa32 %zmm13, %zmm18 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,1,2,3,4,23,u,u,8,9,10,11,24,u,u,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,1,2,3,4,23,u,u,8,9,10,11,24,u,u,15]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm16, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,2,25,u,u,6,7,8,9,26,u,u,13,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,25,u,u,6,7,8,9,26,u,u,13,14,15]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm14, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,27,u,u,4,5,6,7,28,u,u,11,12,13,14,29]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm14 # 64-byte Reload
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm14, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,2,3,4,5,30,u,u,9,10,11,12,31,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm15, %zmm14
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm15 # 64-byte Reload
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512BW-NEXT: vmovdqa32 %zmm15, %zmm0 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,20,u,u,u,u,7,8,21,u,u,u,u,14,15]
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm16 # 64-byte Reload
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm15, %zmm16
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm21 # 64-byte Reload
@@ -12358,17 +12358,17 @@ define void @store_i32_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm15, %zmm8
; AVX512BW-NEXT: vmovdqa32 %zmm23, %zmm8 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm20
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [4,20,u,u,u,u,u,5,21,u,u,u,u,u,6,22]
; AVX512BW-NEXT: vpermt2d %zmm2, %zmm8, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm10, %zmm15, %zmm1
; AVX512BW-NEXT: vmovdqa32 %zmm12, %zmm1 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,1,2,3,4,5,23,u,8,9,10,11,12,24,u,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,1,2,3,4,5,23,u,8,9,10,11,12,24,u,15]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm3, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,1,2,3,25,u,6,7,8,9,10,26,u,13,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,25,u,6,7,8,9,10,26,u,13,14,15]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm11, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,27,u,4,5,6,7,8,28,u,11,12,13,14,15]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm13, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [29,u,2,3,4,5,6,30,u,9,10,11,12,13,31,u]
; AVX512BW-NEXT: vpermi2d %zmm2, %zmm14, %zmm12
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [22,1,2,3,4,5,6,23,8,9,10,11,12,13,24,15]
; AVX512BW-NEXT: vpermi2d %zmm10, %zmm8, %zmm2
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll
index 5c844cf04e1f509..9b0e526af057dfb 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i32-stride-8.ll
@@ -716,26 +716,26 @@ define void @store_i32_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512F-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
; AVX512F-NEXT: vinserti64x4 $1, (%r10), %zmm3, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,0,8,16,24,u,u,u,u,1,9,17,25>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,0,8,16,24,u,u,u,u,1,9,17,25]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,8,16,24,u,u,u,u,1,9,17,25,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,8,16,24,u,u,u,u,1,9,17,25,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm5
; AVX512F-NEXT: movb $-52, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,2,10,18,26,u,u,u,u,3,11,19,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,2,10,18,26,u,u,u,u,3,11,19,27]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <2,10,18,26,u,u,u,u,3,11,19,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [2,10,18,26,u,u,u,u,3,11,19,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm6
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,4,12,20,28,u,u,u,u,5,13,21,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,4,12,20,28,u,u,u,u,5,13,21,29]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,12,20,28,u,u,u,u,5,13,21,29,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,12,20,28,u,u,u,u,5,13,21,29,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,6,14,22,30,u,u,u,u,7,15,23,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,6,14,22,30,u,u,u,u,7,15,23,31]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,22,30,u,u,u,u,7,15,23,31,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,22,30,u,u,u,u,7,15,23,31,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm2, 192(%rax)
@@ -758,26 +758,26 @@ define void @store_i32_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512BW-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
; AVX512BW-NEXT: vinserti64x4 $1, (%r10), %zmm3, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,0,8,16,24,u,u,u,u,1,9,17,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,0,8,16,24,u,u,u,u,1,9,17,25]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,8,16,24,u,u,u,u,1,9,17,25,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,8,16,24,u,u,u,u,1,9,17,25,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: movb $-52, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,2,10,18,26,u,u,u,u,3,11,19,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,2,10,18,26,u,u,u,u,3,11,19,27]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <2,10,18,26,u,u,u,u,3,11,19,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [2,10,18,26,u,u,u,u,3,11,19,27,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,4,12,20,28,u,u,u,u,5,13,21,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,4,12,20,28,u,u,u,u,5,13,21,29]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,12,20,28,u,u,u,u,5,13,21,29,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,12,20,28,u,u,u,u,5,13,21,29,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,6,14,22,30,u,u,u,u,7,15,23,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,6,14,22,30,u,u,u,u,7,15,23,31]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,22,30,u,u,u,u,7,15,23,31,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,22,30,u,u,u,u,7,15,23,31,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm2, 192(%rax)
@@ -1557,16 +1557,16 @@ define void @store_i32_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm6
; AVX512F-NEXT: vmovdqa64 (%r11), %zmm7
; AVX512F-NEXT: vmovdqa64 (%r10), %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm9
; AVX512F-NEXT: movb $-120, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm9 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm4
; AVX512F-NEXT: movb $34, %cl
; AVX512F-NEXT: kmovw %ecx, %k2
@@ -1574,80 +1574,80 @@ define void @store_i32_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: movb $-52, %cl
; AVX512F-NEXT: kmovw %ecx, %k3
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm4 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm10
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm10 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm9
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm9 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm9 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm11
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm11 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm10
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm10 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm10 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm12
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm12 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm11
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm11 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm11 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm14
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm14 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm14 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm15
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm15 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm15 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm16
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm16 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm13, %zmm16 {%k3}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512F-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512F-NEXT: vpermi2d %zmm6, %zmm5, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm2 {%k2}
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm2 {%k3}
@@ -1675,16 +1675,16 @@ define void @store_i32_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm6
; AVX512BW-NEXT: vmovdqa64 (%r11), %zmm7
; AVX512BW-NEXT: vmovdqa64 (%r10), %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm9
; AVX512BW-NEXT: movb $-120, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm9 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm4
; AVX512BW-NEXT: movb $34, %cl
; AVX512BW-NEXT: kmovd %ecx, %k2
@@ -1692,80 +1692,80 @@ define void @store_i32_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movb $-52, %cl
; AVX512BW-NEXT: kmovd %ecx, %k3
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm4 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm10
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm10 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm9
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm9 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm9 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm11
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm10
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm10 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm10 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm12
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm12 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm11
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm11 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm11 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm14
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm14 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm14 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm15
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm15 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm15 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm13
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm13 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm16
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm16 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm13, %zmm16 {%k3}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512BW-NEXT: vpermi2d %zmm8, %zmm7, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512BW-NEXT: vpermi2d %zmm6, %zmm5, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm3, %zmm2, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm2 {%k2}
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm2 {%k3}
@@ -3252,57 +3252,57 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%r10), %zmm0
; AVX512F-NEXT: vmovdqa64 (%rax), %zmm30
; AVX512F-NEXT: vmovdqa64 64(%rax), %zmm28
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm5
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm4
; AVX512F-NEXT: vpermt2d %zmm30, %zmm3, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm5
; AVX512F-NEXT: vpermt2d %zmm27, %zmm2, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm6
; AVX512F-NEXT: vpermt2d %zmm30, %zmm5, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm7
; AVX512F-NEXT: vpermt2d %zmm27, %zmm6, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm8
; AVX512F-NEXT: vpermt2d %zmm30, %zmm7, %zmm8
; AVX512F-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm9
; AVX512F-NEXT: vpermt2d %zmm27, %zmm8, %zmm9
; AVX512F-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm10
; AVX512F-NEXT: vpermt2d %zmm30, %zmm9, %zmm10
; AVX512F-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm11
; AVX512F-NEXT: vpermt2d %zmm27, %zmm10, %zmm11
; AVX512F-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512F-NEXT: vpermt2d %zmm30, %zmm11, %zmm12
; AVX512F-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512F-NEXT: vpermt2d %zmm27, %zmm12, %zmm13
; AVX512F-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm14
; AVX512F-NEXT: vpermt2d %zmm30, %zmm13, %zmm14
; AVX512F-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm16
; AVX512F-NEXT: vpermt2d %zmm30, %zmm15, %zmm16
; AVX512F-NEXT: vmovdqu64 %zmm16, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512F-NEXT: vpermt2d %zmm30, %zmm31, %zmm4
; AVX512F-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm4
@@ -3324,10 +3324,10 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermt2d %zmm28, %zmm31, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm31
; AVX512F-NEXT: vpermt2d %zmm27, %zmm28, %zmm31
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512F-NEXT: vpermt2d %zmm27, %zmm0, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 64(%r8), %zmm27
@@ -3348,33 +3348,33 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm1, %zmm0, %zmm27
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm16
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm20
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm20
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm21
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm21
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm22
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm22
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm23
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm23
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm24
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm24
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm29 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm29 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm25
; AVX512F-NEXT: vpermt2d %zmm0, %zmm29, %zmm25
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm30 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm30 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm26
; AVX512F-NEXT: vpermt2d %zmm0, %zmm30, %zmm26
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm16
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -3393,28 +3393,28 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm3
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm18
; AVX512F-NEXT: vpermt2d %zmm0, %zmm12, %zmm18
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm17
; AVX512F-NEXT: vpermt2d %zmm0, %zmm10, %zmm17
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm15
; AVX512F-NEXT: vpermt2d %zmm0, %zmm8, %zmm15
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm14
; AVX512F-NEXT: vpermt2d %zmm0, %zmm7, %zmm14
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512F-NEXT: vpermt2d %zmm0, %zmm6, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm11
; AVX512F-NEXT: vpermt2d %zmm0, %zmm4, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm3, %zmm9
; AVX512F-NEXT: vpermt2d %zmm0, %zmm2, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm19, %zmm3
; AVX512F-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512F-NEXT: vmovdqa64 64(%rsi), %zmm0
@@ -3546,57 +3546,57 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%r10), %zmm0
; AVX512BW-NEXT: vmovdqa64 (%rax), %zmm30
; AVX512BW-NEXT: vmovdqa64 64(%rax), %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm5
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm3, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm2, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm5, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm7, %zmm8
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm9
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm8, %zmm9
; AVX512BW-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm10
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm9, %zmm10
; AVX512BW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm11
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm10, %zmm11
; AVX512BW-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm11, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm12, %zmm13
; AVX512BW-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm14
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm13, %zmm14
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm16
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm15, %zmm16
; AVX512BW-NEXT: vmovdqu64 %zmm16, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512BW-NEXT: vpermt2d %zmm30, %zmm31, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
@@ -3618,10 +3618,10 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermt2d %zmm28, %zmm31, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm31
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm28, %zmm31
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512BW-NEXT: vpermt2d %zmm27, %zmm0, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm27
@@ -3642,33 +3642,33 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm0, %zmm27
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm16
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm20
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm20
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm21
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm21
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm22
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm22
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm23
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm23
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm24
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm24
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm29 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm29 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm25
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm29, %zmm25
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm30 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm26
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm30, %zmm26
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm16
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -3687,28 +3687,28 @@ define void @store_i32_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm3
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm18
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm12, %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm17
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm10, %zmm17
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm15
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm8, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm14
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm7, %zmm14
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm6, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm11
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm4, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm3, %zmm9
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm19, %zmm3
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm1
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm0
@@ -6769,41 +6769,41 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 (%rax), %zmm1
; AVX512F-NEXT: vmovdqa64 64(%rax), %zmm0
; AVX512F-NEXT: vmovdqa64 128(%rax), %zmm30
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm6
; AVX512F-NEXT: vpermt2d %zmm1, %zmm3, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512F-NEXT: vmovdqa64 %zmm6, %zmm7
; AVX512F-NEXT: vpermt2d %zmm1, %zmm5, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm8
; AVX512F-NEXT: vpermt2d %zmm1, %zmm6, %zmm8
; AVX512F-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm9
; AVX512F-NEXT: vpermt2d %zmm1, %zmm8, %zmm9
; AVX512F-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm10
; AVX512F-NEXT: vpermt2d %zmm1, %zmm8, %zmm10
; AVX512F-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm11
; AVX512F-NEXT: vpermt2d %zmm1, %zmm8, %zmm11
; AVX512F-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm13
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm12
; AVX512F-NEXT: vpermt2d %zmm1, %zmm8, %zmm13
; AVX512F-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512F-NEXT: vpermt2d %zmm1, %zmm8, %zmm12
; AVX512F-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm1
@@ -6872,42 +6872,42 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 (%r8), %zmm26
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm2
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm3
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm3
; AVX512F-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm4
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm4
; AVX512F-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm5
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm6
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm7
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm7
; AVX512F-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm8
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm8
; AVX512F-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm26
; AVX512F-NEXT: vmovdqa64 64(%r8), %zmm25
; AVX512F-NEXT: vmovdqa64 64(%r9), %zmm0
@@ -6976,42 +6976,42 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm27
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm8
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm2
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm3
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm3
; AVX512F-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm4
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm4
; AVX512F-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm5
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512F-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm8
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm6
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -7080,35 +7080,35 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm17
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm5
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm14, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm11, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm10, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm9, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm4, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512F-NEXT: vpermt2d %zmm0, %zmm3, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512F-NEXT: vpermt2d %zmm0, %zmm31, %zmm5
; AVX512F-NEXT: vmovdqa64 64(%rdi), %zmm12
; AVX512F-NEXT: vmovdqa64 64(%rsi), %zmm0
@@ -7423,41 +7423,41 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 (%rax), %zmm1
; AVX512BW-NEXT: vmovdqa64 64(%rax), %zmm0
; AVX512BW-NEXT: vmovdqa64 128(%rax), %zmm30
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,u,u,u,u,0,16,u,u,u,u,u,u,1,17]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm3, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,2,18,u,u,u,u,u,u,3,19]
; AVX512BW-NEXT: vmovdqa64 %zmm6, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm5, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,u,u,u,u,4,20,u,u,u,u,u,u,5,21]
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm6, %zmm8
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,6,22,u,u,u,u,u,u,7,23]
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm9
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm8, %zmm9
; AVX512BW-NEXT: vmovdqu64 %zmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,8,24,u,u,u,u,u,u,9,25]
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm10
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm8, %zmm10
; AVX512BW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,10,26,u,u,u,u,u,u,11,27]
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm11
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm8, %zmm11
; AVX512BW-NEXT: vmovdqu64 %zmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,12,28,u,u,u,u,u,u,13,29]
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm13
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm12
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm8, %zmm13
; AVX512BW-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,u,u,u,u,14,30,u,u,u,u,u,u,15,31]
; AVX512BW-NEXT: vpermt2d %zmm1, %zmm8, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm1
@@ -7526,42 +7526,42 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm26
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,0,16,u,u,u,u,u,u,1,17,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm2
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,2,18,u,u,u,u,u,u,3,19,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm3
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,4,20,u,u,u,u,u,u,5,21,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,6,22,u,u,u,u,u,u,7,23,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,8,24,u,u,u,u,u,u,9,25,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,10,26,u,u,u,u,u,u,11,27,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm7
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm7
; AVX512BW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,12,28,u,u,u,u,u,u,13,29,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm8
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm8
; AVX512BW-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,u,u,14,30,u,u,u,u,u,u,15,31,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm26
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm25
; AVX512BW-NEXT: vmovdqa64 64(%r9), %zmm0
@@ -7630,42 +7630,42 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm27
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm8
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,0,16,u,u,u,u,u,u,1,17,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm2
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,2,18,u,u,u,u,u,u,3,19,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm3
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,4,20,u,u,u,u,u,u,5,21,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm4
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm4
; AVX512BW-NEXT: vmovdqu64 %zmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,6,22,u,u,u,u,u,u,7,23,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm5
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,8,24,u,u,u,u,u,u,9,25,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,10,26,u,u,u,u,u,u,11,27,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,12,28,u,u,u,u,u,u,13,29,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm6
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm6
; AVX512BW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,u,14,30,u,u,u,u,u,u,15,31,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm8
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm6
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -7734,35 +7734,35 @@ define void @store_i32_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm1, %zmm17
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,16,u,u,u,u,u,u,1,17,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm14, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [2,18,u,u,u,u,u,u,3,19,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm11, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [4,20,u,u,u,u,u,u,5,21,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm10, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [6,22,u,u,u,u,u,u,7,23,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm9, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [8,24,u,u,u,u,u,u,9,25,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm7, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [10,26,u,u,u,u,u,u,11,27,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm4, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, (%rsp) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [12,28,u,u,u,u,u,u,13,29,u,u,u,u,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm1
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm3, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = <14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = [14,30,u,u,u,u,u,u,15,31,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermt2d %zmm0, %zmm31, %zmm5
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm12
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm0
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-3.ll
index 3ffcf6d6d6f5ad2..5176db91e914e16 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-3.ll
@@ -64,7 +64,7 @@ define void @store_i64_stride3_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovaps (%rdi), %xmm0
; AVX512-NEXT: vinsertf128 $1, (%rsi), %ymm0, %ymm0
; AVX512-NEXT: vinsertf32x4 $2, (%rdx), %zmm0, %zmm0
-; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = <0,2,4,1,3,5,u,u>
+; AVX512-NEXT: vmovaps {{.*#+}} zmm1 = [0,2,4,1,3,5,u,u]
; AVX512-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512-NEXT: vextractf32x4 $2, %zmm0, 32(%rcx)
; AVX512-NEXT: vmovaps %ymm0, (%rcx)
@@ -308,15 +308,15 @@ define void @store_i64_stride3_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512-NEXT: vmovdqa64 (%rdx), %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,8,u,1,9,u,2,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,8,u,1,9,u,2,10]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,8,3,4,9,6,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,3,11,u,4,12,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,3,11,u,4,12,u,5]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [10,1,2,11,4,5,12,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <5,u,14,6,u,15,7,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [5,u,14,6,u,15,7,u]
; AVX512-NEXT: vpermi2q %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,13,2,3,14,5,6,15]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm0
@@ -605,17 +605,17 @@ define void @store_i64_stride3_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 64(%rsi), %zmm3
; AVX512-NEXT: vmovdqa64 (%rdx), %zmm4
; AVX512-NEXT: vmovdqa64 64(%rdx), %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,8,u,1,9,u,2,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,8,u,1,9,u,2,10]
; AVX512-NEXT: vmovdqa64 %zmm0, %zmm7
; AVX512-NEXT: vpermt2q %zmm2, %zmm6, %zmm7
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,8,3,4,9,6,7]
; AVX512-NEXT: vpermt2q %zmm4, %zmm8, %zmm7
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm9 = <5,u,14,6,u,15,7,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm9 = [5,u,14,6,u,15,7,u]
; AVX512-NEXT: vmovdqa64 %zmm3, %zmm10
; AVX512-NEXT: vpermt2q %zmm1, %zmm9, %zmm10
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,13,2,3,14,5,6,15]
; AVX512-NEXT: vpermt2q %zmm5, %zmm11, %zmm10
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,3,11,u,4,12,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,3,11,u,4,12,u,5]
; AVX512-NEXT: vmovdqa64 %zmm1, %zmm13
; AVX512-NEXT: vpermt2q %zmm3, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [10,1,2,11,4,5,12,7]
@@ -1246,17 +1246,17 @@ define void @store_i64_stride3_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 64(%rdx), %zmm9
; AVX512-NEXT: vmovdqa64 128(%rdx), %zmm10
; AVX512-NEXT: vmovdqa64 192(%rdx), %zmm11
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = <0,8,u,1,9,u,2,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm12 = [0,8,u,1,9,u,2,10]
; AVX512-NEXT: vmovdqa64 %zmm3, %zmm13
; AVX512-NEXT: vpermt2q %zmm4, %zmm12, %zmm13
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,1,8,3,4,9,6,7]
; AVX512-NEXT: vpermt2q %zmm8, %zmm14, %zmm13
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm15 = <5,u,14,6,u,15,7,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm15 = [5,u,14,6,u,15,7,u]
; AVX512-NEXT: vmovdqa64 %zmm7, %zmm16
; AVX512-NEXT: vpermt2q %zmm0, %zmm15, %zmm16
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm17 = [0,13,2,3,14,5,6,15]
; AVX512-NEXT: vpermt2q %zmm11, %zmm17, %zmm16
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,3,11,u,4,12,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,3,11,u,4,12,u,5]
; AVX512-NEXT: vmovdqa64 %zmm0, %zmm19
; AVX512-NEXT: vpermt2q %zmm7, %zmm18, %zmm19
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm20 = [10,1,2,11,4,5,12,7]
@@ -2589,17 +2589,17 @@ define void @store_i64_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512-NEXT: vmovdqa64 64(%rdx), %zmm25
; AVX512-NEXT: vmovdqa64 128(%rdx), %zmm29
; AVX512-NEXT: vmovdqa64 192(%rdx), %zmm31
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,8,u,1,9,u,2,10>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,8,u,1,9,u,2,10]
; AVX512-NEXT: vmovdqa64 %zmm13, %zmm3
; AVX512-NEXT: vpermt2q %zmm0, %zmm14, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,1,8,3,4,9,6,7]
; AVX512-NEXT: vpermt2q %zmm15, %zmm19, %zmm3
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm17 = <u,3,11,u,4,12,u,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm17 = [u,3,11,u,4,12,u,5]
; AVX512-NEXT: vmovdqa64 %zmm13, %zmm10
; AVX512-NEXT: vpermt2q %zmm0, %zmm17, %zmm10
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm21 = [10,1,2,11,4,5,12,7]
; AVX512-NEXT: vpermt2q %zmm15, %zmm21, %zmm10
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm22 = <5,u,14,6,u,15,7,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm22 = [5,u,14,6,u,15,7,u]
; AVX512-NEXT: vpermt2q %zmm13, %zmm22, %zmm0
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,13,2,3,14,5,6,15]
; AVX512-NEXT: vpermt2q %zmm15, %zmm23, %zmm0
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-4.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-4.ll
index a2e7633f69554fd..18392469d52e3f1 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-4.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-4.ll
@@ -376,26 +376,26 @@ define void @store_i64_stride4_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,0,8,u,u,1,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,0,8,u,u,1,9]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,8,u,u,1,9,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,8,u,u,1,9,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm5
; AVX512F-NEXT: movb $-52, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,2,10,u,u,3,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,2,10,u,u,3,11]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <2,10,u,u,3,11,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [2,10,u,u,3,11,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm6
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,4,12,u,u,5,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,4,12,u,u,5,13]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,12,u,u,5,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,12,u,u,5,13,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,6,14,u,u,7,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,6,14,u,u,7,15]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,u,u,7,15,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,u,u,7,15,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm2, 192(%r8)
@@ -411,26 +411,26 @@ define void @store_i64_stride4_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm1
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm2
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,0,8,u,u,1,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,0,8,u,u,1,9]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,8,u,u,1,9,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,8,u,u,1,9,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm5
; AVX512BW-NEXT: movb $-52, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm5 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,2,10,u,u,3,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,2,10,u,u,3,11]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <2,10,u,u,3,11,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [2,10,u,u,3,11,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm6
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm6 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,4,12,u,u,5,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,4,12,u,u,5,13]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <4,12,u,u,5,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [4,12,u,u,5,13,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm7 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,u,6,14,u,u,7,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,u,6,14,u,u,7,15]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,u,u,7,15,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,u,u,7,15,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm2, 192(%r8)
@@ -830,32 +830,32 @@ define void @store_i64_stride4_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm6
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,2,10,u,u,3,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,2,10,u,u,3,11]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm9
; AVX512F-NEXT: vpermt2q %zmm6, %zmm8, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <2,10,u,u,3,11,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [2,10,u,u,3,11,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm11
; AVX512F-NEXT: vpermt2q %zmm2, %zmm10, %zmm11
; AVX512F-NEXT: movb $-52, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm11 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,8,u,u,1,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,8,u,u,1,9]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512F-NEXT: vpermt2q %zmm6, %zmm9, %zmm12
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,8,u,u,1,9,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,8,u,u,1,9,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512F-NEXT: vpermt2q %zmm2, %zmm13, %zmm14
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm14 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,6,14,u,u,7,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,6,14,u,u,7,15]
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm15
; AVX512F-NEXT: vpermt2q %zmm6, %zmm12, %zmm15
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <6,14,u,u,7,15,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [6,14,u,u,7,15,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm17
; AVX512F-NEXT: vpermt2q %zmm2, %zmm16, %zmm17
; AVX512F-NEXT: vmovdqa64 %zmm15, %zmm17 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,4,12,u,u,5,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,4,12,u,u,5,13]
; AVX512F-NEXT: vpermt2q %zmm6, %zmm15, %zmm4
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,12,u,u,5,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,12,u,u,5,13,u,u]
; AVX512F-NEXT: vpermt2q %zmm2, %zmm6, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm0 {%k1}
; AVX512F-NEXT: vpermi2q %zmm7, %zmm5, %zmm8
@@ -891,32 +891,32 @@ define void @store_i64_stride4_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm5
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm6
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,2,10,u,u,3,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,2,10,u,u,3,11]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm9
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm8, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <2,10,u,u,3,11,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [2,10,u,u,3,11,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm11
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm10, %zmm11
; AVX512BW-NEXT: movb $-52, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm11 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,8,u,u,1,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,8,u,u,1,9]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm12
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm9, %zmm12
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <0,8,u,u,1,9,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [0,8,u,u,1,9,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm14
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm13, %zmm14
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm14 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <u,u,6,14,u,u,7,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [u,u,6,14,u,u,7,15]
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm15
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm12, %zmm15
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <6,14,u,u,7,15,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [6,14,u,u,7,15,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm17
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm16, %zmm17
; AVX512BW-NEXT: vmovdqa64 %zmm15, %zmm17 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,4,12,u,u,5,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,4,12,u,u,5,13]
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm15, %zmm4
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <4,12,u,u,5,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [4,12,u,u,5,13,u,u]
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm6, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm0 {%k1}
; AVX512BW-NEXT: vpermi2q %zmm7, %zmm5, %zmm8
@@ -1772,32 +1772,32 @@ define void @store_i64_stride4_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm26
; AVX512F-NEXT: vmovdqa64 128(%rcx), %zmm19
; AVX512F-NEXT: vmovdqa64 192(%rcx), %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,2,10,u,u,3,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,2,10,u,u,3,11]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm8
; AVX512F-NEXT: vpermt2q %zmm21, %zmm14, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <2,10,u,u,3,11,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [2,10,u,u,3,11,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm4
; AVX512F-NEXT: vpermt2q %zmm17, %zmm7, %zmm4
; AVX512F-NEXT: movb $-52, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm4 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,u,0,8,u,u,1,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,u,0,8,u,u,1,9]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm10
; AVX512F-NEXT: vpermt2q %zmm21, %zmm16, %zmm10
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,8,u,u,1,9,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,8,u,u,1,9,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512F-NEXT: vpermt2q %zmm17, %zmm11, %zmm8
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm8 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,u,6,14,u,u,7,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,u,6,14,u,u,7,15]
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm20
; AVX512F-NEXT: vpermt2q %zmm21, %zmm18, %zmm20
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <6,14,u,u,7,15,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [6,14,u,u,7,15,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm10
; AVX512F-NEXT: vpermt2q %zmm17, %zmm15, %zmm10
; AVX512F-NEXT: vmovdqa64 %zmm20, %zmm10 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,4,12,u,u,5,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,4,12,u,u,5,13]
; AVX512F-NEXT: vpermt2q %zmm21, %zmm20, %zmm22
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <4,12,u,u,5,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [4,12,u,u,5,13,u,u]
; AVX512F-NEXT: vpermt2q %zmm17, %zmm21, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm22, %zmm0 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm22
@@ -1885,32 +1885,32 @@ define void @store_i64_stride4_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm26
; AVX512BW-NEXT: vmovdqa64 128(%rcx), %zmm19
; AVX512BW-NEXT: vmovdqa64 192(%rcx), %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,u,2,10,u,u,3,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,u,2,10,u,u,3,11]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm8
; AVX512BW-NEXT: vpermt2q %zmm21, %zmm14, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <2,10,u,u,3,11,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [2,10,u,u,3,11,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm4
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm7, %zmm4
; AVX512BW-NEXT: movb $-52, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm4 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,u,0,8,u,u,1,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,u,0,8,u,u,1,9]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm10
; AVX512BW-NEXT: vpermt2q %zmm21, %zmm16, %zmm10
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,8,u,u,1,9,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,8,u,u,1,9,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm8
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm11, %zmm8
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm8 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,u,6,14,u,u,7,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,u,6,14,u,u,7,15]
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm20
; AVX512BW-NEXT: vpermt2q %zmm21, %zmm18, %zmm20
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <6,14,u,u,7,15,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [6,14,u,u,7,15,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm10
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm15, %zmm10
; AVX512BW-NEXT: vmovdqa64 %zmm20, %zmm10 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <u,u,4,12,u,u,5,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [u,u,4,12,u,u,5,13]
; AVX512BW-NEXT: vpermt2q %zmm21, %zmm20, %zmm22
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <4,12,u,u,5,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [4,12,u,u,5,13,u,u]
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm21, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm22, %zmm0 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm22
@@ -3673,19 +3673,19 @@ define void @store_i64_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm5
; AVX512F-NEXT: vmovdqa64 128(%rcx), %zmm2
; AVX512F-NEXT: vmovdqa64 192(%rcx), %zmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,8,u,u,1,9>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,8,u,u,1,9]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm13
; AVX512F-NEXT: vpermt2q %zmm4, %zmm9, %zmm13
; AVX512F-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,10,u,u,3,11>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,10,u,u,3,11]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm15
; AVX512F-NEXT: vpermt2q %zmm4, %zmm13, %zmm15
; AVX512F-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,4,12,u,u,5,13>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,4,12,u,u,5,13]
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm16
; AVX512F-NEXT: vpermt2q %zmm4, %zmm15, %zmm16
; AVX512F-NEXT: vmovdqu64 %zmm16, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,u,6,14,u,u,7,15>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,u,6,14,u,u,7,15]
; AVX512F-NEXT: vpermt2q %zmm4, %zmm26, %zmm14
; AVX512F-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm4
@@ -3765,19 +3765,19 @@ define void @store_i64_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2q %zmm0, %zmm26, %zmm28
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm26
; AVX512F-NEXT: vmovdqa64 (%rsi), %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,u,u,1,9,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,u,u,1,9,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm7, %zmm0
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm30
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <2,10,u,u,3,11,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [2,10,u,u,3,11,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm5, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <4,12,u,u,5,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [4,12,u,u,5,13,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512F-NEXT: vpermt2q %zmm3, %zmm2, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = <6,14,u,u,7,15,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = [6,14,u,u,7,15,u,u]
; AVX512F-NEXT: vpermt2q %zmm3, %zmm22, %zmm26
; AVX512F-NEXT: vmovdqa64 64(%rdi), %zmm20
; AVX512F-NEXT: vmovdqa64 64(%rsi), %zmm3
@@ -3968,19 +3968,19 @@ define void @store_i64_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm5
; AVX512BW-NEXT: vmovdqa64 128(%rcx), %zmm2
; AVX512BW-NEXT: vmovdqa64 192(%rcx), %zmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,0,8,u,u,1,9>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,0,8,u,u,1,9]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm13
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm9, %zmm13
; AVX512BW-NEXT: vmovdqu64 %zmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,10,u,u,3,11>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,10,u,u,3,11]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm15
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm13, %zmm15
; AVX512BW-NEXT: vmovdqu64 %zmm15, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,4,12,u,u,5,13>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,4,12,u,u,5,13]
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm16
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm15, %zmm16
; AVX512BW-NEXT: vmovdqu64 %zmm16, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,u,6,14,u,u,7,15>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,u,6,14,u,u,7,15]
; AVX512BW-NEXT: vpermt2q %zmm4, %zmm26, %zmm14
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm4
@@ -4060,19 +4060,19 @@ define void @store_i64_stride4_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm26, %zmm28
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm26
; AVX512BW-NEXT: vmovdqa64 (%rsi), %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,u,u,1,9,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,u,u,1,9,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm7, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm30
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <2,10,u,u,3,11,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [2,10,u,u,3,11,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm5, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <4,12,u,u,5,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [4,12,u,u,5,13,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <6,14,u,u,7,15,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [6,14,u,u,7,15,u,u]
; AVX512BW-NEXT: vpermt2q %zmm3, %zmm22, %zmm26
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %zmm20
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm3
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll
index e52fd4013bf4662..4951d3fd8d2e48d 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-5.ll
@@ -220,13 +220,13 @@ define void @store_i64_stride5_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vmovdqa (%r8), %ymm2
; AVX512-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
-; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = <15,3,7,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} ymm3 = [15,3,7,u]
; AVX512-NEXT: vpermi2q %zmm0, %zmm1, %zmm3
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,4,8,12,u,1,5,9>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,4,8,12,u,1,5,9]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm4
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,2,3,8,5,6,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm4, %zmm5
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = <13,u,2,6,10,14,u,3>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [13,u,2,6,10,14,u,3]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm4
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,9,2,3,4,5,10,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm4, %zmm0
@@ -470,9 +470,9 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm4
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm5
; AVX512F-NEXT: vmovdqa64 (%r8), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <3,u,u,u,12,4,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [3,u,u,u,12,4,u,u]
; AVX512F-NEXT: vpermi2q %zmm2, %zmm3, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,3,11,u,u,u,4,12>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,3,11,u,u,u,4,12]
; AVX512F-NEXT: vpermi2q %zmm5, %zmm4, %zmm1
; AVX512F-NEXT: movb $49, %al
; AVX512F-NEXT: kmovw %eax, %k1
@@ -480,18 +480,18 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: movb $8, %al
; AVX512F-NEXT: kmovw %eax, %k2
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm1 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,0,8,u,u,u,1>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,0,8,u,u,u,1]
; AVX512F-NEXT: vpermi2q %zmm5, %zmm4, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,u,u,u,1,9,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,u,u,u,1,9,u]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm7
; AVX512F-NEXT: movb $-116, %al
; AVX512F-NEXT: kmovw %eax, %k2
; AVX512F-NEXT: vmovdqa64 %zmm6, %zmm7 {%k2}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,8,5,6,7]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm7, %zmm6
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <1,u,u,u,10,2,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [1,u,u,u,10,2,u,u]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm5, %zmm7
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,2,10,u,u,u,3>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,2,10,u,u,u,3]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,9,2,3,4,5,10,7]
@@ -499,7 +499,7 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm8 = [13,5,13,5,13,5,13,5]
; AVX512F-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm5, %zmm4, %zmm8
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,5,13,u,u,u,6,14>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,5,13,u,u,u,6,14]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm9
; AVX512F-NEXT: movb $24, %al
; AVX512F-NEXT: kmovw %eax, %k1
@@ -509,7 +509,7 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm9 = [15,7,15,7,15,7,15,7]
; AVX512F-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm9
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,u,u,u,7,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,u,u,u,7,15,u]
; AVX512F-NEXT: vpermi2q %zmm5, %zmm4, %zmm2
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm2 {%k1}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,14,3,4,5,6,15]
@@ -529,9 +529,9 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm4
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm5
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <3,u,u,u,12,4,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [3,u,u,u,12,4,u,u]
; AVX512BW-NEXT: vpermi2q %zmm2, %zmm3, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,3,11,u,u,u,4,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,3,11,u,u,u,4,12]
; AVX512BW-NEXT: vpermi2q %zmm5, %zmm4, %zmm1
; AVX512BW-NEXT: movb $49, %al
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -539,18 +539,18 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: movb $8, %al
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm1 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,u,0,8,u,u,u,1>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,u,0,8,u,u,u,1]
; AVX512BW-NEXT: vpermi2q %zmm5, %zmm4, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,8,u,u,u,1,9,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,8,u,u,u,1,9,u]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm7
; AVX512BW-NEXT: movb $-116, %al
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqa64 %zmm6, %zmm7 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,1,2,3,8,5,6,7]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm7, %zmm6
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <1,u,u,u,10,2,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [1,u,u,u,10,2,u,u]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm5, %zmm7
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,u,2,10,u,u,u,3>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,u,2,10,u,u,u,3]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm8 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,9,2,3,4,5,10,7]
@@ -558,7 +558,7 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm8 = [13,5,13,5,13,5,13,5]
; AVX512BW-NEXT: # zmm8 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm5, %zmm4, %zmm8
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,5,13,u,u,u,6,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,5,13,u,u,u,6,14]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm9
; AVX512BW-NEXT: movb $24, %al
; AVX512BW-NEXT: kmovd %eax, %k1
@@ -568,7 +568,7 @@ define void @store_i64_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm9 = [15,7,15,7,15,7,15,7]
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm9
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <6,14,u,u,u,7,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [6,14,u,u,u,7,15,u]
; AVX512BW-NEXT: vpermi2q %zmm5, %zmm4, %zmm2
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm2 {%k1}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,14,3,4,5,6,15]
@@ -1107,10 +1107,10 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm12
; AVX512F-NEXT: vmovdqa64 (%r8), %zmm4
; AVX512F-NEXT: vmovdqa64 64(%r8), %zmm11
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <3,u,u,u,12,4,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [3,u,u,u,12,4,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm9, %zmm13
; AVX512F-NEXT: vpermt2q %zmm0, %zmm8, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,3,11,u,u,u,4,12>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,3,11,u,u,u,4,12]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm3
; AVX512F-NEXT: vpermt2q %zmm12, %zmm7, %zmm3
; AVX512F-NEXT: movb $49, %al
@@ -1123,10 +1123,10 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermi2q %zmm5, %zmm6, %zmm7
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm7 {%k1}
; AVX512F-NEXT: vmovdqa64 %zmm4, %zmm7 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,0,8,u,u,u,1>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,0,8,u,u,u,1]
; AVX512F-NEXT: vmovdqa64 %zmm6, %zmm13
; AVX512F-NEXT: vpermt2q %zmm5, %zmm15, %zmm13
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <0,8,u,u,u,1,9,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [0,8,u,u,u,1,9,u]
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm8
; AVX512F-NEXT: vpermt2q %zmm2, %zmm16, %zmm8
; AVX512F-NEXT: movb $-116, %al
@@ -1138,7 +1138,7 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm19
; AVX512F-NEXT: vpermt2q %zmm9, %zmm18, %zmm19
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <6,14,u,u,u,7,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [6,14,u,u,u,7,15,u]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm13
; AVX512F-NEXT: vpermt2q %zmm12, %zmm14, %zmm13
; AVX512F-NEXT: movb $24, %al
@@ -1150,16 +1150,16 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm10, %zmm21
; AVX512F-NEXT: vpermt2q %zmm12, %zmm20, %zmm21
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = <u,5,13,u,u,u,6,14>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm22 = [u,5,13,u,u,u,6,14]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm23
; AVX512F-NEXT: vpermt2q %zmm9, %zmm22, %zmm23
; AVX512F-NEXT: vmovdqa64 %zmm21, %zmm23 {%k2}
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [12,1,2,3,4,13,6,7]
; AVX512F-NEXT: vpermt2q %zmm11, %zmm21, %zmm23
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm24 = <1,u,u,u,10,2,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm24 = [1,u,u,u,10,2,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm12, %zmm25
; AVX512F-NEXT: vpermt2q %zmm10, %zmm24, %zmm25
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,u,2,10,u,u,u,3>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,u,2,10,u,u,u,3]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm27
; AVX512F-NEXT: vpermt2q %zmm9, %zmm26, %zmm27
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm27 {%k1}
@@ -1206,10 +1206,10 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm12
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm4
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm11
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <3,u,u,u,12,4,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [3,u,u,u,12,4,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm9, %zmm13
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm8, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,3,11,u,u,u,4,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,3,11,u,u,u,4,12]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm3
; AVX512BW-NEXT: vpermt2q %zmm12, %zmm7, %zmm3
; AVX512BW-NEXT: movb $49, %al
@@ -1222,10 +1222,10 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2q %zmm5, %zmm6, %zmm7
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm7 {%k1}
; AVX512BW-NEXT: vmovdqa64 %zmm4, %zmm7 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,0,8,u,u,u,1>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,0,8,u,u,u,1]
; AVX512BW-NEXT: vmovdqa64 %zmm6, %zmm13
; AVX512BW-NEXT: vpermt2q %zmm5, %zmm15, %zmm13
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <0,8,u,u,u,1,9,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [0,8,u,u,u,1,9,u]
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm8
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm16, %zmm8
; AVX512BW-NEXT: movb $-116, %al
@@ -1237,7 +1237,7 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm18 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm19
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm18, %zmm19
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <6,14,u,u,u,7,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [6,14,u,u,u,7,15,u]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm13
; AVX512BW-NEXT: vpermt2q %zmm12, %zmm14, %zmm13
; AVX512BW-NEXT: movb $24, %al
@@ -1249,16 +1249,16 @@ define void @store_i64_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: # zmm20 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm10, %zmm21
; AVX512BW-NEXT: vpermt2q %zmm12, %zmm20, %zmm21
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <u,5,13,u,u,u,6,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [u,5,13,u,u,u,6,14]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm23
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm22, %zmm23
; AVX512BW-NEXT: vmovdqa64 %zmm21, %zmm23 {%k2}
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [12,1,2,3,4,13,6,7]
; AVX512BW-NEXT: vpermt2q %zmm11, %zmm21, %zmm23
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = <1,u,u,u,10,2,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [1,u,u,u,10,2,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm12, %zmm25
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm24, %zmm25
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <u,u,2,10,u,u,u,3>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [u,u,2,10,u,u,u,3]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm27
; AVX512BW-NEXT: vpermt2q %zmm9, %zmm26, %zmm27
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm27 {%k1}
@@ -2445,11 +2445,11 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm23
; AVX512F-NEXT: vmovdqa64 128(%rcx), %zmm25
; AVX512F-NEXT: vmovdqa64 192(%rcx), %zmm18
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = <3,u,u,u,12,4,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm11 = [3,u,u,u,12,4,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm16, %zmm0
; AVX512F-NEXT: vpermt2q %zmm20, %zmm11, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = <u,3,11,u,u,u,4,12>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = [u,3,11,u,u,u,4,12]
; AVX512F-NEXT: vmovdqa64 %zmm27, %zmm0
; AVX512F-NEXT: vpermt2q %zmm18, %zmm19, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2460,17 +2460,17 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2q %zmm23, %zmm19, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermi2q %zmm1, %zmm17, %zmm19
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,0,8,u,u,u,1>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,0,8,u,u,u,1]
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm0
; AVX512F-NEXT: vpermt2q %zmm1, %zmm28, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,8,u,u,u,1,9,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,8,u,u,u,1,9,u]
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm29 = [15,7,15,7,15,7,15,7]
; AVX512F-NEXT: # zmm29 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm20, %zmm0
; AVX512F-NEXT: vpermt2q %zmm16, %zmm29, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <6,14,u,u,u,7,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [6,14,u,u,u,7,15,u]
; AVX512F-NEXT: vmovdqa64 %zmm27, %zmm26
; AVX512F-NEXT: vpermt2q %zmm18, %zmm12, %zmm26
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm21 = [13,5,13,5,13,5,13,5]
@@ -2478,7 +2478,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm27, %zmm0
; AVX512F-NEXT: vpermt2q %zmm18, %zmm21, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = <1,u,u,u,10,2,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm31 = [1,u,u,u,10,2,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512F-NEXT: vpermt2q %zmm27, %zmm31, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2497,7 +2497,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm23, %zmm30
; AVX512F-NEXT: vpermt2q %zmm24, %zmm31, %zmm30
; AVX512F-NEXT: vpermt2q %zmm23, %zmm28, %zmm24
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,5,13,u,u,u,6,14>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,5,13,u,u,u,6,14]
; AVX512F-NEXT: vpermt2q %zmm23, %zmm12, %zmm25
; AVX512F-NEXT: vpermt2q %zmm23, %zmm21, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2507,7 +2507,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermi2q %zmm1, %zmm17, %zmm21
; AVX512F-NEXT: vpermt2q %zmm17, %zmm31, %zmm1
; AVX512F-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,2,10,u,u,u,3>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,2,10,u,u,u,3]
; AVX512F-NEXT: vmovdqa64 %zmm20, %zmm17
; AVX512F-NEXT: vpermt2q %zmm16, %zmm10, %zmm17
; AVX512F-NEXT: vpermt2q %zmm16, %zmm14, %zmm20
@@ -2647,11 +2647,11 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm23
; AVX512BW-NEXT: vmovdqa64 128(%rcx), %zmm25
; AVX512BW-NEXT: vmovdqa64 192(%rcx), %zmm18
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = <3,u,u,u,12,4,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm11 = [3,u,u,u,12,4,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm16, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm20, %zmm11, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <u,3,11,u,u,u,4,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [u,3,11,u,u,u,4,12]
; AVX512BW-NEXT: vmovdqa64 %zmm27, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm18, %zmm19, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2662,17 +2662,17 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2q %zmm23, %zmm19, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm17, %zmm19
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,u,0,8,u,u,u,1>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,u,0,8,u,u,u,1]
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm28, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <0,8,u,u,u,1,9,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [0,8,u,u,u,1,9,u]
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm29 = [15,7,15,7,15,7,15,7]
; AVX512BW-NEXT: # zmm29 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm20, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm16, %zmm29, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <6,14,u,u,u,7,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [6,14,u,u,u,7,15,u]
; AVX512BW-NEXT: vmovdqa64 %zmm27, %zmm26
; AVX512BW-NEXT: vpermt2q %zmm18, %zmm12, %zmm26
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm21 = [13,5,13,5,13,5,13,5]
@@ -2680,7 +2680,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm27, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm18, %zmm21, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = <1,u,u,u,10,2,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm31 = [1,u,u,u,10,2,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm18, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm27, %zmm31, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2699,7 +2699,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm23, %zmm30
; AVX512BW-NEXT: vpermt2q %zmm24, %zmm31, %zmm30
; AVX512BW-NEXT: vpermt2q %zmm23, %zmm28, %zmm24
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = <u,5,13,u,u,u,6,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm28 = [u,5,13,u,u,u,6,14]
; AVX512BW-NEXT: vpermt2q %zmm23, %zmm12, %zmm25
; AVX512BW-NEXT: vpermt2q %zmm23, %zmm21, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -2709,7 +2709,7 @@ define void @store_i64_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm17, %zmm21
; AVX512BW-NEXT: vpermt2q %zmm17, %zmm31, %zmm1
; AVX512BW-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,2,10,u,u,u,3>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,2,10,u,u,u,3]
; AVX512BW-NEXT: vmovdqa64 %zmm20, %zmm17
; AVX512BW-NEXT: vpermt2q %zmm16, %zmm10, %zmm17
; AVX512BW-NEXT: vpermt2q %zmm16, %zmm14, %zmm20
@@ -5252,7 +5252,7 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rsi), %zmm4
; AVX512F-NEXT: vmovdqa64 128(%rsi), %zmm3
; AVX512F-NEXT: vmovdqa64 192(%rsi), %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <3,u,u,u,12,4,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [3,u,u,u,12,4,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm12
; AVX512F-NEXT: vpermt2q %zmm11, %zmm20, %zmm12
; AVX512F-NEXT: vmovdqu64 %zmm12, (%rsp) # 64-byte Spill
@@ -5268,15 +5268,15 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm12
; AVX512F-NEXT: vpermt2q %zmm7, %zmm20, %zmm12
; AVX512F-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,8,u,u,u,1,9,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,8,u,u,u,1,9,u]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512F-NEXT: vpermt2q %zmm5, %zmm21, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,10,u,u,u,3>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,10,u,u,u,3]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512F-NEXT: vpermt2q %zmm5, %zmm13, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,5,13,u,u,u,6,14>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,5,13,u,u,u,6,14]
; AVX512F-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512F-NEXT: vpermt2q %zmm5, %zmm14, %zmm0
; AVX512F-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -5371,20 +5371,20 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 (%rdx), %zmm25
; AVX512F-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,3,11,u,u,u,4,12>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,3,11,u,u,u,4,12]
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm3
; AVX512F-NEXT: vpermt2q %zmm0, %zmm6, %zmm3
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,0,8,u,u,u,1>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,0,8,u,u,u,1]
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm22
; AVX512F-NEXT: vpermt2q %zmm0, %zmm2, %zmm22
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = <1,u,u,u,10,2,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm12 = [1,u,u,u,10,2,u,u]
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm27
; AVX512F-NEXT: vpermt2q %zmm25, %zmm12, %zmm27
; AVX512F-NEXT: vbroadcasti32x4 {{.*#+}} zmm10 = [13,5,13,5,13,5,13,5]
; AVX512F-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vmovdqa64 %zmm25, %zmm28
; AVX512F-NEXT: vpermt2q %zmm0, %zmm10, %zmm28
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = <6,14,u,u,u,7,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm16 = [6,14,u,u,u,7,15,u]
; AVX512F-NEXT: vpermt2q %zmm0, %zmm16, %zmm25
; AVX512F-NEXT: vmovdqa64 64(%rdx), %zmm26
; AVX512F-NEXT: vmovdqa64 64(%rcx), %zmm0
@@ -5715,7 +5715,7 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rsi), %zmm4
; AVX512BW-NEXT: vmovdqa64 128(%rsi), %zmm3
; AVX512BW-NEXT: vmovdqa64 192(%rsi), %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <3,u,u,u,12,4,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [3,u,u,u,12,4,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm12
; AVX512BW-NEXT: vpermt2q %zmm11, %zmm20, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, (%rsp) # 64-byte Spill
@@ -5731,15 +5731,15 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm12
; AVX512BW-NEXT: vpermt2q %zmm7, %zmm20, %zmm12
; AVX512BW-NEXT: vmovdqu64 %zmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,8,u,u,u,1,9,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,8,u,u,u,1,9,u]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm5, %zmm21, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = <u,u,2,10,u,u,u,3>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm13 = [u,u,2,10,u,u,u,3]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm5, %zmm13, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = <u,5,13,u,u,u,6,14>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm14 = [u,5,13,u,u,u,6,14]
; AVX512BW-NEXT: vmovdqa64 %zmm11, %zmm0
; AVX512BW-NEXT: vpermt2q %zmm5, %zmm14, %zmm0
; AVX512BW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -5834,20 +5834,20 @@ define void @store_i64_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 (%rdx), %zmm25
; AVX512BW-NEXT: vmovdqa64 (%rcx), %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,3,11,u,u,u,4,12>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,3,11,u,u,u,4,12]
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm3
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm6, %zmm3
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,u,0,8,u,u,u,1>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,u,0,8,u,u,u,1]
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm22
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm2, %zmm22
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = <1,u,u,u,10,2,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm12 = [1,u,u,u,10,2,u,u]
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm27
; AVX512BW-NEXT: vpermt2q %zmm25, %zmm12, %zmm27
; AVX512BW-NEXT: vbroadcasti32x4 {{.*#+}} zmm10 = [13,5,13,5,13,5,13,5]
; AVX512BW-NEXT: # zmm10 = mem[0,1,2,3,0,1,2,3,0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vmovdqa64 %zmm25, %zmm28
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm10, %zmm28
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <6,14,u,u,u,7,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [6,14,u,u,u,7,15,u]
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm16, %zmm25
; AVX512BW-NEXT: vmovdqa64 64(%rdx), %zmm26
; AVX512BW-NEXT: vmovdqa64 64(%rcx), %zmm0
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll
index ab51f0bf9135cc9..37f197b1895b8c1 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-6.ll
@@ -271,11 +271,11 @@ define void @store_i64_stride6_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512-NEXT: vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512-NEXT: vinserti64x4 $1, (%r9), %zmm2, %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,4,8,12,u,u,1,5>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,4,8,12,u,u,1,5]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,8,12,6,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm4
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <1,5,u,u,10,14,2,6>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [1,5,u,u,10,14,2,6]
; AVX512-NEXT: vpermi2q %zmm0, %zmm1, %zmm3
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,9,13,4,5,6,7]
; AVX512-NEXT: vpermi2q %zmm2, %zmm3, %zmm5
@@ -626,7 +626,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: movb $48, %r9b
; AVX512F-NEXT: kmovw %r9d, %k2
; AVX512F-NEXT: vmovdqa64 %zmm7, %zmm8 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,1,9,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,9,u,4,5,6,7]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm8, %zmm9
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,9,4,5,6,7]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm9, %zmm7
@@ -637,7 +637,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm6, %zmm5, %zmm9
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm9 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,13,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,13,u,4,5,6,7]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm9, %zmm8
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,13,4,5,6,7]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm8, %zmm9
@@ -649,7 +649,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vpermt2q %zmm6, %zmm8, %zmm5
; AVX512F-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm5 = zmm8[0,1,2,3],zmm5[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <14,u,2,3,4,5,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [14,u,2,3,4,5,15,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm5, %zmm6
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,14,2,3,4,5,6,15]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm6, %zmm5
@@ -666,7 +666,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vmovdqa (%rdi), %ymm2
; AVX512F-NEXT: vpunpckhqdq {{.*#+}} ymm2 = ymm2[1],mem[1],ymm2[3],mem[3]
; AVX512F-NEXT: vinserti64x4 $0, %ymm2, %zmm10, %zmm2
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <10,u,2,3,4,5,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [10,u,2,3,4,5,11,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm2, %zmm6
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,10,2,3,4,5,6,11]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm6, %zmm1
@@ -711,7 +711,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: movb $48, %r9b
; AVX512BW-NEXT: kmovd %r9d, %k2
; AVX512BW-NEXT: vmovdqa64 %zmm7, %zmm8 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,1,9,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,9,u,4,5,6,7]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm8, %zmm9
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,1,2,9,4,5,6,7]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm9, %zmm7
@@ -722,7 +722,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: # zmm9 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm6, %zmm5, %zmm9
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm9 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,13,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,13,u,4,5,6,7]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm9, %zmm8
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,1,2,13,4,5,6,7]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm8, %zmm9
@@ -734,7 +734,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm8, %zmm5
; AVX512BW-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm5 = zmm8[0,1,2,3],zmm5[4,5,6,7]
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <14,u,2,3,4,5,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [14,u,2,3,4,5,15,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm5, %zmm6
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,14,2,3,4,5,6,15]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm6, %zmm5
@@ -751,7 +751,7 @@ define void @store_i64_stride6_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vmovdqa (%rdi), %ymm2
; AVX512BW-NEXT: vpunpckhqdq {{.*#+}} ymm2 = ymm2[1],mem[1],ymm2[3],mem[3]
; AVX512BW-NEXT: vinserti64x4 $0, %ymm2, %zmm10, %zmm2
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <10,u,2,3,4,5,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [10,u,2,3,4,5,11,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm2, %zmm6
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,10,2,3,4,5,6,11]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm6, %zmm1
@@ -1478,7 +1478,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm8, %zmm13
; AVX512F-NEXT: vpermt2q %zmm11, %zmm19, %zmm13
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm9 = zmm13[0,1,2,3],zmm9[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <14,u,2,3,4,5,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [14,u,2,3,4,5,15,u]
; AVX512F-NEXT: vpermt2q %zmm6, %zmm20, %zmm9
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,14,2,3,4,5,6,15]
; AVX512F-NEXT: vpermt2q %zmm14, %zmm21, %zmm9
@@ -1493,7 +1493,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: movb $48, %r9b
; AVX512F-NEXT: kmovw %r9d, %k2
; AVX512F-NEXT: vmovdqa64 %zmm17, %zmm13 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm23 = <0,1,13,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,13,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm6, %zmm23, %zmm13
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm24 = [0,1,2,13,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm14, %zmm24, %zmm13
@@ -1506,7 +1506,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 %zmm5, %zmm17
; AVX512F-NEXT: vpermt2q %zmm7, %zmm18, %zmm17
; AVX512F-NEXT: vmovdqa64 %zmm26, %zmm17 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = <0,1,9,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm26 = [0,1,9,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm6, %zmm26, %zmm17
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,1,2,9,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm14, %zmm27, %zmm17
@@ -1552,7 +1552,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 64(%rdi), %ymm20
; AVX512F-NEXT: vpunpckhqdq {{.*#+}} ymm20 = ymm20[1],mem[1],ymm20[3],mem[3]
; AVX512F-NEXT: vinserti64x4 $0, %ymm20, %zmm5, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = <10,u,2,3,4,5,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm20 = [10,u,2,3,4,5,11,u]
; AVX512F-NEXT: vpermt2q %zmm6, %zmm20, %zmm5
; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,10,2,3,4,5,6,11]
; AVX512F-NEXT: vpermt2q %zmm14, %zmm6, %zmm5
@@ -1617,7 +1617,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm8, %zmm13
; AVX512BW-NEXT: vpermt2q %zmm11, %zmm19, %zmm13
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm9 = zmm13[0,1,2,3],zmm9[4,5,6,7]
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <14,u,2,3,4,5,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [14,u,2,3,4,5,15,u]
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm20, %zmm9
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,14,2,3,4,5,6,15]
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm21, %zmm9
@@ -1632,7 +1632,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movb $48, %r9b
; AVX512BW-NEXT: kmovd %r9d, %k2
; AVX512BW-NEXT: vmovdqa64 %zmm17, %zmm13 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm23 = <0,1,13,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm23 = [0,1,13,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm23, %zmm13
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [0,1,2,13,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm24, %zmm13
@@ -1645,7 +1645,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 %zmm5, %zmm17
; AVX512BW-NEXT: vpermt2q %zmm7, %zmm18, %zmm17
; AVX512BW-NEXT: vmovdqa64 %zmm26, %zmm17 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = <0,1,9,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm26 = [0,1,9,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm26, %zmm17
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm27 = [0,1,2,9,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm27, %zmm17
@@ -1691,7 +1691,7 @@ define void @store_i64_stride6_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 64(%rdi), %ymm20
; AVX512BW-NEXT: vpunpckhqdq {{.*#+}} ymm20 = ymm20[1],mem[1],ymm20[3],mem[3]
; AVX512BW-NEXT: vinserti64x4 $0, %ymm20, %zmm5, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = <10,u,2,3,4,5,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm20 = [10,u,2,3,4,5,11,u]
; AVX512BW-NEXT: vpermt2q %zmm6, %zmm20, %zmm5
; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,10,2,3,4,5,6,11]
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm6, %zmm5
@@ -3294,7 +3294,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm3 = zmm28[0,1,2,3],zmm17[4,5,6,7]
; AVX512F-NEXT: vmovdqa64 (%r9), %zmm8
; AVX512F-NEXT: vmovdqa64 %zmm14, %zmm25 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,9,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,9,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm0, %zmm5, %zmm20
; AVX512F-NEXT: vpermt2q %zmm1, %zmm5, %zmm18
; AVX512F-NEXT: vpermt2q %zmm2, %zmm5, %zmm13
@@ -3308,7 +3308,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqa64 192(%r9), %zmm14
; AVX512F-NEXT: vpermt2q %zmm14, %zmm5, %zmm25
; AVX512F-NEXT: vmovdqa64 %zmm15, %zmm29 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,13,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,13,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm0, %zmm5, %zmm23
; AVX512F-NEXT: vpermt2q %zmm1, %zmm5, %zmm31
; AVX512F-NEXT: vpermt2q %zmm2, %zmm5, %zmm4
@@ -3319,7 +3319,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2q %zmm30, %zmm5, %zmm4
; AVX512F-NEXT: vpermt2q %zmm14, %zmm5, %zmm29
; AVX512F-NEXT: vshufi64x2 {{.*#+}} zmm15 = zmm16[0,1,2,3],zmm10[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = <14,u,2,3,4,5,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm5 = [14,u,2,3,4,5,15,u]
; AVX512F-NEXT: vpermt2q %zmm0, %zmm5, %zmm12
; AVX512F-NEXT: vpermt2q %zmm1, %zmm5, %zmm22
; AVX512F-NEXT: vpermt2q %zmm2, %zmm5, %zmm3
@@ -3363,7 +3363,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: movb $16, %al
; AVX512F-NEXT: kmovw %eax, %k1
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm27 {%k1}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = <10,u,2,3,4,5,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm19 = [10,u,2,3,4,5,11,u]
; AVX512F-NEXT: vpermt2q %zmm0, %zmm19, %zmm6
; AVX512F-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX512F-NEXT: vpunpckhqdq {{.*#+}} ymm0 = ymm0[1],mem[1],ymm0[3],mem[3]
@@ -3572,7 +3572,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm3 = zmm28[0,1,2,3],zmm17[4,5,6,7]
; AVX512BW-NEXT: vmovdqa64 (%r9), %zmm8
; AVX512BW-NEXT: vmovdqa64 %zmm14, %zmm25 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,9,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,9,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm5, %zmm20
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm5, %zmm18
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm5, %zmm13
@@ -3586,7 +3586,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqa64 192(%r9), %zmm14
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm5, %zmm25
; AVX512BW-NEXT: vmovdqa64 %zmm15, %zmm29 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,13,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,13,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm5, %zmm23
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm5, %zmm31
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm5, %zmm4
@@ -3597,7 +3597,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2q %zmm30, %zmm5, %zmm4
; AVX512BW-NEXT: vpermt2q %zmm14, %zmm5, %zmm29
; AVX512BW-NEXT: vshufi64x2 {{.*#+}} zmm15 = zmm16[0,1,2,3],zmm10[4,5,6,7]
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <14,u,2,3,4,5,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [14,u,2,3,4,5,15,u]
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm5, %zmm12
; AVX512BW-NEXT: vpermt2q %zmm1, %zmm5, %zmm22
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm5, %zmm3
@@ -3641,7 +3641,7 @@ define void @store_i64_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: movb $16, %al
; AVX512BW-NEXT: kmovd %eax, %k1
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm27 {%k1}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = <10,u,2,3,4,5,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [10,u,2,3,4,5,11,u]
; AVX512BW-NEXT: vpermt2q %zmm0, %zmm19, %zmm6
; AVX512BW-NEXT: vmovdqa 64(%rdi), %ymm0
; AVX512BW-NEXT: vpunpckhqdq {{.*#+}} ymm0 = ymm0[1],mem[1],ymm0[3],mem[3]
@@ -7013,7 +7013,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512F-NEXT: vmovdqa64 %zmm0, %zmm27 {%k2}
; AVX512F-NEXT: vmovdqa64 (%r8), %zmm29
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,9,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,9,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm29, %zmm1, %zmm2
; AVX512F-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa64 64(%r8), %zmm2
@@ -7038,7 +7038,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vpermt2q %zmm10, %zmm1, %zmm27
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-NEXT: vmovdqa64 %zmm1, %zmm28 {%k2}
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,13,u,4,5,6,7>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,13,u,4,5,6,7]
; AVX512F-NEXT: vpermt2q %zmm29, %zmm1, %zmm3
; AVX512F-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vpermt2q %zmm2, %zmm1, %zmm5
@@ -7056,7 +7056,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512F-NEXT: vshufi64x2 $228, {{[-0-9]+}}(%r{{[sb]}}p), %zmm1, %zmm3 # 64-byte Folded Reload
; AVX512F-NEXT: # zmm3 = zmm1[0,1,2,3],mem[4,5,6,7]
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <14,u,2,3,4,5,15,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [14,u,2,3,4,5,15,u]
; AVX512F-NEXT: vpermt2q %zmm29, %zmm1, %zmm14
; AVX512F-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqu64 (%rsp), %zmm14 # 64-byte Reload
@@ -7084,7 +7084,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-NEXT: kmovw %eax, %k2
; AVX512F-NEXT: vmovdqa64 %zmm2, %zmm31 {%k2}
; AVX512F-NEXT: vmovdqu64 %zmm31, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <10,u,2,3,4,5,11,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [10,u,2,3,4,5,11,u]
; AVX512F-NEXT: vpermt2q %zmm2, %zmm1, %zmm5
; AVX512F-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-NEXT: vmovdqa 128(%rdi), %ymm2
@@ -7709,7 +7709,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm0 # 64-byte Reload
; AVX512BW-NEXT: vmovdqa64 %zmm0, %zmm27 {%k2}
; AVX512BW-NEXT: vmovdqa64 (%r8), %zmm29
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,9,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,9,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm29, %zmm1, %zmm2
; AVX512BW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa64 64(%r8), %zmm2
@@ -7734,7 +7734,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vpermt2q %zmm10, %zmm1, %zmm27
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-NEXT: vmovdqa64 %zmm1, %zmm28 {%k2}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,13,u,4,5,6,7>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,13,u,4,5,6,7]
; AVX512BW-NEXT: vpermt2q %zmm29, %zmm1, %zmm3
; AVX512BW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm1, %zmm5
@@ -7752,7 +7752,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; AVX512BW-NEXT: vshufi64x2 $228, {{[-0-9]+}}(%r{{[sb]}}p), %zmm1, %zmm3 # 64-byte Folded Reload
; AVX512BW-NEXT: # zmm3 = zmm1[0,1,2,3],mem[4,5,6,7]
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <14,u,2,3,4,5,15,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [14,u,2,3,4,5,15,u]
; AVX512BW-NEXT: vpermt2q %zmm29, %zmm1, %zmm14
; AVX512BW-NEXT: vmovdqu64 %zmm14, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqu64 (%rsp), %zmm14 # 64-byte Reload
@@ -7780,7 +7780,7 @@ define void @store_i64_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-NEXT: kmovd %eax, %k2
; AVX512BW-NEXT: vmovdqa64 %zmm2, %zmm31 {%k2}
; AVX512BW-NEXT: vmovdqu64 %zmm31, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <10,u,2,3,4,5,11,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [10,u,2,3,4,5,11,u]
; AVX512BW-NEXT: vpermt2q %zmm2, %zmm1, %zmm5
; AVX512BW-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-NEXT: vmovdqa 128(%rdi), %ymm2
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-7.ll
index 6fb5e43a2df68a9..18c0a59329612f3 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i64-stride-7.ll
@@ -114,7 +114,7 @@ define void @store_i64_stride7_vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512-NEXT: vinserti32x4 $2, (%r10), %zmm1, %zmm1
; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,2,4,6,8,10,12,1]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
-; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = <3,5,7,9,11,13,u,u>
+; AVX512-NEXT: vmovdqa64 {{.*#+}} zmm3 = [3,5,7,9,11,13,u,u]
; AVX512-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
; AVX512-NEXT: vextracti32x4 $2, %zmm3, 96(%rax)
; AVX512-NEXT: vmovdqa64 %zmm2, (%rax)
@@ -307,12 +307,12 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vinserti64x4 $1, (%rsi), %zmm1, %zmm4
; AVX512F-NEXT: vinserti64x4 $1, (%rcx), %zmm2, %zmm2
; AVX512F-NEXT: vinserti64x4 $1, (%r9), %zmm3, %zmm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = <15,3,7,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm1 = [15,3,7,u]
; AVX512F-NEXT: vpermi2q %zmm2, %zmm3, %zmm1
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [0,4,8,0,0,4,8,0]
; AVX512F-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,4,8,12,u,u,u,1>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,4,8,12,u,u,u,1]
; AVX512F-NEXT: vpermi2q %zmm2, %zmm4, %zmm6
; AVX512F-NEXT: movb $112, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -320,7 +320,7 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [5,9,0,1,5,9,0,1]
; AVX512F-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = <5,9,13,u,u,u,2,6>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm7 = [5,9,13,u,u,u,2,6]
; AVX512F-NEXT: vpermi2q %zmm2, %zmm4, %zmm7
; AVX512F-NEXT: movb $56, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -328,7 +328,7 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [10,0,2,6,10,0,2,6]
; AVX512F-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512F-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = <2,6,u,u,u,11,15,3>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm3 = [2,6,u,u,u,11,15,3]
; AVX512F-NEXT: vpermi2q %zmm4, %zmm2, %zmm3
; AVX512F-NEXT: movb $28, %cl
; AVX512F-NEXT: kmovw %ecx, %k1
@@ -352,12 +352,12 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vinserti64x4 $1, (%rsi), %zmm1, %zmm4
; AVX512BW-NEXT: vinserti64x4 $1, (%rcx), %zmm2, %zmm2
; AVX512BW-NEXT: vinserti64x4 $1, (%r9), %zmm3, %zmm3
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <15,3,7,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [15,3,7,u]
; AVX512BW-NEXT: vpermi2q %zmm2, %zmm3, %zmm1
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [0,4,8,0,0,4,8,0]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,4,8,12,u,u,u,1>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,4,8,12,u,u,u,1]
; AVX512BW-NEXT: vpermi2q %zmm2, %zmm4, %zmm6
; AVX512BW-NEXT: movb $112, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -365,7 +365,7 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [5,9,0,1,5,9,0,1]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <5,9,13,u,u,u,2,6>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [5,9,13,u,u,u,2,6]
; AVX512BW-NEXT: vpermi2q %zmm2, %zmm4, %zmm7
; AVX512BW-NEXT: movb $56, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -373,7 +373,7 @@ define void @store_i64_stride7_vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-NEXT: vbroadcasti64x4 {{.*#+}} zmm5 = [10,0,2,6,10,0,2,6]
; AVX512BW-NEXT: # zmm5 = mem[0,1,2,3,0,1,2,3]
; AVX512BW-NEXT: vpermi2q %zmm0, %zmm3, %zmm5
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <2,6,u,u,u,11,15,3>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [2,6,u,u,u,11,15,3]
; AVX512BW-NEXT: vpermi2q %zmm4, %zmm2, %zmm3
; AVX512BW-NEXT: movb $28, %cl
; AVX512BW-NEXT: kmovd %ecx, %k1
@@ -934,7 +934,7 @@ define void @store_i64_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} zmm10 {%k2} = zmm0[0],zmm5[0],zmm0[2],zmm5[2],zmm0[4],zmm5[4],zmm0[6],zmm5[6]
; AVX512F-ONLY-FAST-NEXT: vmovdqa (%r9), %ymm11
; AVX512F-ONLY-FAST-NEXT: vmovdqa (%r8), %ymm12
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <1,3,7,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [1,3,7,u]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %ymm11, %ymm12, %ymm13
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm13 = ymm13[0,1,2,3,4,5],mem[6,7]
; AVX512F-ONLY-FAST-NEXT: movb $14, %sil
@@ -1147,7 +1147,7 @@ define void @store_i64_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512DQ-FAST-NEXT: vpunpcklqdq {{.*#+}} zmm8 {%k1} = zmm0[0],zmm6[0],zmm0[2],zmm6[2],zmm0[4],zmm6[4],zmm0[6],zmm6[6]
; AVX512DQ-FAST-NEXT: vmovdqa (%r9), %ymm10
; AVX512DQ-FAST-NEXT: vmovdqa (%r8), %ymm11
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <1,3,7,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [1,3,7,u]
; AVX512DQ-FAST-NEXT: vpermi2q %ymm10, %ymm11, %ymm9
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],mem[6,7]
; AVX512DQ-FAST-NEXT: movb $14, %sil
@@ -1423,7 +1423,7 @@ define void @store_i64_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} zmm10 {%k2} = zmm0[0],zmm5[0],zmm0[2],zmm5[2],zmm0[4],zmm5[4],zmm0[6],zmm5[6]
; AVX512BW-ONLY-FAST-NEXT: vmovdqa (%r9), %ymm11
; AVX512BW-ONLY-FAST-NEXT: vmovdqa (%r8), %ymm12
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <1,3,7,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [1,3,7,u]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %ymm11, %ymm12, %ymm13
; AVX512BW-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm13 = ymm13[0,1,2,3,4,5],mem[6,7]
; AVX512BW-ONLY-FAST-NEXT: movb $14, %sil
@@ -1636,7 +1636,7 @@ define void @store_i64_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512DQBW-FAST-NEXT: vpunpcklqdq {{.*#+}} zmm8 {%k1} = zmm0[0],zmm6[0],zmm0[2],zmm6[2],zmm0[4],zmm6[4],zmm0[6],zmm6[6]
; AVX512DQBW-FAST-NEXT: vmovdqa (%r9), %ymm10
; AVX512DQBW-FAST-NEXT: vmovdqa (%r8), %ymm11
-; AVX512DQBW-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <1,3,7,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [1,3,7,u]
; AVX512DQBW-FAST-NEXT: vpermi2q %ymm10, %ymm11, %ymm9
; AVX512DQBW-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],mem[6,7]
; AVX512DQBW-FAST-NEXT: movb $14, %sil
@@ -2556,13 +2556,13 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: movb $24, %sil
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k1
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm21, %zmm22 {%k1}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm22, %zmm18
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm31, %zmm18, %zmm21
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [14,1,2,3,4,5,6,15]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm28, %zmm21, %zmm18
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm31, %zmm19, %zmm21
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,13,2,3,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm28, %zmm21, %zmm19
@@ -2737,14 +2737,14 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: movb $24, %sil
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k1
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm7, %zmm9 {%k1}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm6, %zmm9, %zmm7
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm23, %zmm7, %zmm9
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [14,1,2,3,4,5,6,15]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm12, %zmm9, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm23, %zmm8, %zmm9
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,13,2,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm12, %zmm9, %zmm8
@@ -2968,13 +2968,13 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: movb $24, %sil
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k1
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm21, %zmm22 {%k1}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,1,2,3,4,15,u,u]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm16, %zmm22, %zmm18
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,5,15,u]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm30, %zmm18, %zmm21
; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [14,1,2,3,4,5,6,15]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm27, %zmm21, %zmm18
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [13,u,2,3,4,5,6,14]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm30, %zmm19, %zmm21
; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,13,2,3,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm27, %zmm21, %zmm19
@@ -3150,7 +3150,7 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: movb $24, %sil
; AVX512DQ-FAST-NEXT: kmovw %esi, %k1
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm8, %zmm16 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [u,1,2,3,4,15,u,u]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm21, %zmm16, %zmm31
; AVX512DQ-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm22 = [0,8,0,1,0,8,0,1]
; AVX512DQ-FAST-NEXT: # zmm22 = mem[0,1,2,3,0,1,2,3]
@@ -3293,9 +3293,9 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: kmovw %eax, %k1
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm27, %zmm19 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm7, %zmm2 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,1,2,3,4,5,15,u]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm6, %zmm31, %zmm4
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,2,3,4,5,6,14]
; AVX512DQ-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm9 # 64-byte Reload
; AVX512DQ-FAST-NEXT: vpermi2q %zmm6, %zmm9, %zmm7
; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [14,1,2,3,4,5,6,15]
@@ -3380,13 +3380,13 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: movb $24, %sil
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k1
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm21, %zmm22 {%k1}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm22, %zmm18
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm31, %zmm18, %zmm21
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [14,1,2,3,4,5,6,15]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm28, %zmm21, %zmm18
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm31, %zmm19, %zmm21
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,13,2,3,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm28, %zmm21, %zmm19
@@ -3561,14 +3561,14 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: movb $24, %sil
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k1
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm7, %zmm9 {%k1}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm6, %zmm9, %zmm7
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm23, %zmm7, %zmm9
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [14,1,2,3,4,5,6,15]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm12, %zmm9, %zmm0
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm23, %zmm8, %zmm9
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,13,2,3,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm12, %zmm9, %zmm8
@@ -3792,13 +3792,13 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: movb $24, %sil
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k1
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm21, %zmm22 {%k1}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm16, %zmm22, %zmm18
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm30, %zmm18, %zmm21
; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm18 = [14,1,2,3,4,5,6,15]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm27, %zmm21, %zmm18
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm30, %zmm19, %zmm21
; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm19 = [0,13,2,3,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm27, %zmm21, %zmm19
@@ -3974,7 +3974,7 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: movb $24, %sil
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k1
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm8, %zmm16 {%k1}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm21, %zmm16, %zmm31
; AVX512DQBW-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm22 = [0,8,0,1,0,8,0,1]
; AVX512DQBW-FAST-NEXT: # zmm22 = mem[0,1,2,3,0,1,2,3]
@@ -4117,9 +4117,9 @@ define void @store_i64_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: kmovd %eax, %k1
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm27, %zmm19 {%k1}
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm7, %zmm2 {%k1}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm6, %zmm31, %zmm4
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm9 # 64-byte Reload
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm6, %zmm9, %zmm7
; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [14,1,2,3,4,5,6,15]
@@ -5952,15 +5952,15 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm10
; AVX512F-ONLY-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm31 = zmm25[0,1,2,3],zmm9[4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 192(%r8), %zmm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,11,u,u,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,11,u,u,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm24, %zmm9
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm8, %zmm5 {%k1}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm25 = <12,u,u,3,4,5,6,13>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [12,u,u,3,4,5,6,13]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm25
; AVX512F-ONLY-SLOW-NEXT: movb $24, %sil
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k3
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm22, %zmm6 {%k3}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm5
; AVX512F-ONLY-SLOW-NEXT: vmovdqa (%rdx), %xmm6
; AVX512F-ONLY-SLOW-NEXT: vpunpcklqdq {{.*#+}} xmm6 = xmm6[0],mem[0]
@@ -6072,7 +6072,7 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm8 # 64-byte Reload
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm6, %zmm8, %zmm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm0, %zmm11 {%k3}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,1,11,u,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,11,u,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm9, %zmm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm7, %zmm10 {%k1}
; AVX512F-ONLY-SLOW-NEXT: vmovdqa 192(%r8), %ymm7
@@ -6102,11 +6102,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,2,3,3]
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1,2,3,4,5],mem[6,7]
; AVX512F-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm7, %zmm0, %zmm28 {%k1}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,12,u,3,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,12,u,3,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm25, %zmm7
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm31, %zmm8
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm5, %zmm9
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,12,3,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm6, %zmm7, %zmm1
@@ -6392,19 +6392,19 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm8[0,1,2,3],zmm15[4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 192(%r8), %zmm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,11,u,u,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,11,u,u,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm24, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 192(%r9), %zmm8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,11,u,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,11,u,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm0, %zmm15
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm5, %zmm10 {%k1}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <12,u,u,3,4,5,6,13>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [12,u,u,3,4,5,6,13]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm10, %zmm0
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: movb $24, %sil
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k3
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm18, %zmm9 {%k3}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm9, %zmm24
; AVX512F-ONLY-FAST-NEXT: vmovdqa (%rdx), %xmm0
; AVX512F-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],mem[0]
@@ -6540,12 +6540,12 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vmovdqa 192(%r8), %ymm2
; AVX512F-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} ymm2 = ymm2[0],mem[0],ymm2[2],mem[2]
; AVX512F-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm7 {%k2} = zmm2[2,3,2,3],zmm3[2,3,2,3]
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,12,u,3,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,12,u,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm4 # 64-byte Reload
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm4, %zmm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm6, %zmm4
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm24, %zmm6
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,12,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
@@ -6837,15 +6837,15 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm21, %zmm15, %zmm12
; AVX512DQ-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm24 = zmm24[0,1,2,3],zmm10[4,5,6,7]
; AVX512DQ-SLOW-NEXT: vmovdqa64 192(%r8), %zmm19
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,11,u,u,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,11,u,u,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm19, %zmm22, %zmm10
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm9, %zmm3 {%k1}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <12,u,u,3,4,5,6,13>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [12,u,u,3,4,5,6,13]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm19, %zmm3, %zmm22
; AVX512DQ-SLOW-NEXT: movb $24, %sil
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k3
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm11, %zmm4 {%k3}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,15,u,u]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm19, %zmm4, %zmm21
; AVX512DQ-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm3 # 64-byte Reload
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm3, %zmm25 {%k1}
@@ -6952,7 +6952,7 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm5 # 64-byte Reload
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm0, %zmm5, %zmm19
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm19, %zmm14 {%k3}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,11,u,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,11,u,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm2, %zmm10, %zmm5
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm8, %zmm12 {%k1}
; AVX512DQ-SLOW-NEXT: vmovdqa 192(%r8), %ymm6
@@ -6983,11 +6983,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,2,3,3]
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2,3,4,5],mem[6,7]
; AVX512DQ-SLOW-NEXT: vinserti64x4 $0, %ymm6, %zmm0, %zmm31 {%k1}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,12,u,3,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,12,u,3,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm2, %zmm22, %zmm6
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [13,u,2,3,4,5,6,14]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm2, %zmm24, %zmm8
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm2, %zmm21, %zmm9
; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,12,3,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm2
@@ -7273,18 +7273,18 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpermt2q %zmm1, %zmm19, %zmm10
; AVX512DQ-FAST-NEXT: vshufi64x2 {{.*#+}} zmm23 = zmm17[0,1,2,3],zmm23[4,5,6,7]
; AVX512DQ-FAST-NEXT: vmovdqa64 192(%r8), %zmm1
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = <0,11,u,u,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = [0,11,u,u,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm1, %zmm4, %zmm17
; AVX512DQ-FAST-NEXT: vmovdqa64 192(%r9), %zmm4
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <0,1,11,u,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,1,11,u,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm4, %zmm17, %zmm20
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm8, %zmm6 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <12,u,u,3,4,5,6,13>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [12,u,u,3,4,5,6,13]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm1, %zmm6, %zmm8
; AVX512DQ-FAST-NEXT: movb $24, %sil
; AVX512DQ-FAST-NEXT: kmovw %esi, %k3
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm9, %zmm21 {%k3}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,1,2,3,4,15,u,u]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm1, %zmm21, %zmm6
; AVX512DQ-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],mem[6,7]
@@ -7416,11 +7416,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vmovdqa 192(%r8), %ymm1
; AVX512DQ-FAST-NEXT: vpunpcklqdq {{.*#+}} ymm1 = ymm1[0],mem[0],ymm1[2],mem[2]
; AVX512DQ-FAST-NEXT: vshufi64x2 {{.*#+}} zmm10 {%k2} = zmm1[2,3,2,3],zmm2[2,3,2,3]
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,12,u,3,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,12,u,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm4, %zmm8, %zmm1
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [13,u,2,3,4,5,6,14]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm4, %zmm23, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,1,2,3,4,5,15,u]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm4, %zmm6, %zmm5
; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,12,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm2, %zmm1, %zmm4
@@ -7710,15 +7710,15 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm10
; AVX512BW-ONLY-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm31 = zmm25[0,1,2,3],zmm9[4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 192(%r8), %zmm0
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,11,u,u,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,11,u,u,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm24, %zmm9
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm8, %zmm5 {%k1}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm25 = <12,u,u,3,4,5,6,13>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [12,u,u,3,4,5,6,13]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm25
; AVX512BW-ONLY-SLOW-NEXT: movb $24, %sil
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k3
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm22, %zmm6 {%k3}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm5
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa (%rdx), %xmm6
; AVX512BW-ONLY-SLOW-NEXT: vpunpcklqdq {{.*#+}} xmm6 = xmm6[0],mem[0]
@@ -7830,7 +7830,7 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm8 # 64-byte Reload
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm6, %zmm8, %zmm0
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm0, %zmm11 {%k3}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,1,11,u,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,11,u,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm9, %zmm0
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm7, %zmm10 {%k1}
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa 192(%r8), %ymm7
@@ -7860,11 +7860,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,2,3,3]
; AVX512BW-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm7 = ymm7[0,1,2,3,4,5],mem[6,7]
; AVX512BW-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm7, %zmm0, %zmm28 {%k1}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,12,u,3,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,12,u,3,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm25, %zmm7
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm31, %zmm8
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm5, %zmm9
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,12,3,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm6, %zmm7, %zmm1
@@ -8150,19 +8150,19 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm8[0,1,2,3],zmm15[4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 192(%r8), %zmm2
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <0,11,u,u,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,11,u,u,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm24, %zmm0
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 192(%r9), %zmm8
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = <0,1,11,u,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [0,1,11,u,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm0, %zmm15
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm5, %zmm10 {%k1}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <12,u,u,3,4,5,6,13>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [12,u,u,3,4,5,6,13]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm10, %zmm0
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-ONLY-FAST-NEXT: movb $24, %sil
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k3
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm18, %zmm9 {%k3}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm24 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm2, %zmm9, %zmm24
; AVX512BW-ONLY-FAST-NEXT: vmovdqa (%rdx), %xmm0
; AVX512BW-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],mem[0]
@@ -8298,12 +8298,12 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: vmovdqa 192(%r8), %ymm2
; AVX512BW-ONLY-FAST-NEXT: vpunpcklqdq {{.*#+}} ymm2 = ymm2[0],mem[0],ymm2[2],mem[2]
; AVX512BW-ONLY-FAST-NEXT: vshufi64x2 {{.*#+}} zmm7 {%k2} = zmm2[2,3,2,3],zmm3[2,3,2,3]
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,12,u,3,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,12,u,3,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm4 # 64-byte Reload
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm4, %zmm2
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm6, %zmm4
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm8, %zmm24, %zmm6
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,12,3,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm8
@@ -8595,15 +8595,15 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm21, %zmm15, %zmm12
; AVX512DQBW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm24 = zmm24[0,1,2,3],zmm10[4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vmovdqa64 192(%r8), %zmm19
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,11,u,u,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,11,u,u,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm19, %zmm22, %zmm10
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm9, %zmm3 {%k1}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = <12,u,u,3,4,5,6,13>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm22 = [12,u,u,3,4,5,6,13]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm19, %zmm3, %zmm22
; AVX512DQBW-SLOW-NEXT: movb $24, %sil
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k3
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm11, %zmm4 {%k3}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm19, %zmm4, %zmm21
; AVX512DQBW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm3 # 64-byte Reload
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm3, %zmm25 {%k1}
@@ -8710,7 +8710,7 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm5 # 64-byte Reload
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm0, %zmm5, %zmm19
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm19, %zmm14 {%k3}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,1,11,u,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,1,11,u,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm2, %zmm10, %zmm5
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm8, %zmm12 {%k1}
; AVX512DQBW-SLOW-NEXT: vmovdqa 192(%r8), %ymm6
@@ -8741,11 +8741,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,2,3,3]
; AVX512DQBW-SLOW-NEXT: vpblendd {{.*#+}} ymm6 = ymm6[0,1,2,3,4,5],mem[6,7]
; AVX512DQBW-SLOW-NEXT: vinserti64x4 $0, %ymm6, %zmm0, %zmm31 {%k1}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <0,12,u,3,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,12,u,3,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm2, %zmm22, %zmm6
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm2, %zmm24, %zmm8
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm2, %zmm21, %zmm9
; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,12,3,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm2
@@ -9031,18 +9031,18 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm1, %zmm19, %zmm10
; AVX512DQBW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm23 = zmm17[0,1,2,3],zmm23[4,5,6,7]
; AVX512DQBW-FAST-NEXT: vmovdqa64 192(%r8), %zmm1
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = <0,11,u,u,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm17 = [0,11,u,u,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm1, %zmm4, %zmm17
; AVX512DQBW-FAST-NEXT: vmovdqa64 192(%r9), %zmm4
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <0,1,11,u,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,1,11,u,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm4, %zmm17, %zmm20
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm8, %zmm6 {%k1}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <12,u,u,3,4,5,6,13>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [12,u,u,3,4,5,6,13]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm1, %zmm6, %zmm8
; AVX512DQBW-FAST-NEXT: movb $24, %sil
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k3
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm9, %zmm21 {%k3}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm1, %zmm21, %zmm6
; AVX512DQBW-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm9 # 32-byte Reload
; AVX512DQBW-FAST-NEXT: vpblendd {{.*#+}} ymm9 = ymm9[0,1,2,3,4,5],mem[6,7]
@@ -9174,11 +9174,11 @@ define void @store_i64_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: vmovdqa 192(%r8), %ymm1
; AVX512DQBW-FAST-NEXT: vpunpcklqdq {{.*#+}} ymm1 = ymm1[0],mem[0],ymm1[2],mem[2]
; AVX512DQBW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm10 {%k2} = zmm1[2,3,2,3],zmm2[2,3,2,3]
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,12,u,3,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,12,u,3,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm4, %zmm8, %zmm1
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm4, %zmm23, %zmm3
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm4, %zmm6, %zmm5
; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,12,3,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm2, %zmm1, %zmm4
@@ -12843,10 +12843,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm1, %zmm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm10
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm1, %zmm3, %zmm10
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
@@ -13080,20 +13080,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm2 = zmm2[0,1,2,3],zmm12[4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm3, %zmm6 {%k1}
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,11,u,u,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,11,u,u,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm9, %zmm5
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm13, %zmm1 {%k1}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <12,u,u,3,4,5,6,13>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [12,u,u,3,4,5,6,13]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm1, %zmm9
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm8, %zmm17 {%k3}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm3, %zmm1, %zmm17
; AVX512F-ONLY-SLOW-NEXT: movb $6, %sil
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k4
; AVX512F-ONLY-SLOW-NEXT: vpbroadcastq 456(%rcx), %ymm1
; AVX512F-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = mem[0,1,2,3],ymm1[4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm0 {%k4}
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,2,9,u,u,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,9,u,u,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm0, %zmm8
; AVX512F-ONLY-SLOW-NEXT: movb $64, %sil
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k5
@@ -13102,16 +13102,16 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k5
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm3, %zmm6 {%k5}
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,11,u,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,11,u,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm1
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm3
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,12,u,3,4,5,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,12,u,3,4,5,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm9, %zmm5
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm6
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm17
; AVX512F-ONLY-SLOW-NEXT: movb $12, %sil
; AVX512F-ONLY-SLOW-NEXT: kmovw %esi, %k5
@@ -13120,9 +13120,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512F-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm4 {%k5}
; AVX512F-ONLY-SLOW-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm2
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm4
-; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,9,u,6,7>
+; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,9,u,6,7]
; AVX512F-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm8, %zmm2
; AVX512F-ONLY-SLOW-NEXT: vmovdqa64 384(%rax), %zmm0
; AVX512F-ONLY-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm9 # 64-byte Reload
@@ -13827,10 +13827,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm14, %zmm20, %zmm1
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm24
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm1, %zmm2, %zmm24
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm20
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
@@ -14065,21 +14065,21 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: # zmm1 = zmm10[0,1,2,3],mem[4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm3, %zmm8 {%k1}
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,11,u,u,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,11,u,u,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm11, %zmm10
; AVX512F-ONLY-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm6 # 64-byte Reload
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm6, %zmm2 {%k1}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <12,u,u,3,4,5,6,13>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [12,u,u,3,4,5,6,13]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm11
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm30, %zmm15 {%k3}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm3, %zmm2, %zmm15
; AVX512F-ONLY-FAST-NEXT: movb $6, %sil
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k5
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq 456(%rcx), %ymm2
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm2 = mem[0,1,2,3],ymm2[4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm0 {%k5}
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,9,u,u,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,9,u,u,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm0, %zmm2
; AVX512F-ONLY-FAST-NEXT: movb $64, %sil
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k4
@@ -14088,15 +14088,15 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k4
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %zmm3, %zmm8 {%k4}
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <0,1,11,u,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [0,1,11,u,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm10, %zmm31
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm8, %zmm3
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,12,u,3,4,5,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,12,u,3,4,5,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm11, %zmm8
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <13,u,2,3,4,5,6,14>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [13,u,2,3,4,5,6,14]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm10
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512F-ONLY-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm15
; AVX512F-ONLY-FAST-NEXT: movb $12, %sil
; AVX512F-ONLY-FAST-NEXT: kmovw %esi, %k4
@@ -14105,9 +14105,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512F-ONLY-FAST-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm4 {%k4}
; AVX512F-ONLY-FAST-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm1
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm4
-; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,2,3,9,u,6,7>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,3,9,u,6,7]
; AVX512F-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm2, %zmm11
; AVX512F-ONLY-FAST-NEXT: vpbroadcastq 8(%rcx), %ymm0
; AVX512F-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5,6,7]
@@ -14761,10 +14761,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm11, %zmm1, %zmm2
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-SLOW-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm0, %zmm2, %zmm7
; AVX512DQ-SLOW-NEXT: vmovdqa64 384(%r9), %zmm2
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,1,2,3,4,5,15,u]
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm2, %zmm8, %zmm7
; AVX512DQ-SLOW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm2, %zmm0, %zmm1
@@ -14998,20 +14998,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm2 = zmm2[0,1,2,3],zmm12[4,5,6,7]
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm3, %zmm7 {%k1}
; AVX512DQ-SLOW-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,11,u,u,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,11,u,u,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm3, %zmm8, %zmm5
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm11, %zmm1 {%k1}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <12,u,u,3,4,5,6,13>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [12,u,u,3,4,5,6,13]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm3, %zmm1, %zmm8
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm10, %zmm13 {%k3}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm3, %zmm1, %zmm13
; AVX512DQ-SLOW-NEXT: movb $6, %sil
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k4
; AVX512DQ-SLOW-NEXT: vpbroadcastq 456(%rcx), %ymm1
; AVX512DQ-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = mem[0,1,2,3],ymm1[4,5,6,7]
; AVX512DQ-SLOW-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm0 {%k4}
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,2,9,u,u,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,9,u,u,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm3, %zmm0, %zmm10
; AVX512DQ-SLOW-NEXT: movb $64, %sil
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k5
@@ -15020,16 +15020,16 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k5
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm3, %zmm7 {%k5}
; AVX512DQ-SLOW-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,11,u,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,11,u,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm1
; AVX512DQ-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm7, %zmm3
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,12,u,3,4,5,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,12,u,3,4,5,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm8, %zmm5
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,2,3,4,5,6,14]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm7
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512DQ-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm13
; AVX512DQ-SLOW-NEXT: movb $12, %sil
; AVX512DQ-SLOW-NEXT: kmovw %esi, %k5
@@ -15038,9 +15038,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-SLOW-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm4 {%k5}
; AVX512DQ-SLOW-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm2
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm4
-; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,9,u,6,7>
+; AVX512DQ-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,9,u,6,7]
; AVX512DQ-SLOW-NEXT: vpermi2q %zmm0, %zmm10, %zmm2
; AVX512DQ-SLOW-NEXT: vmovdqa64 384(%rax), %zmm0
; AVX512DQ-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm10 # 64-byte Reload
@@ -15734,10 +15734,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vpermt2q %zmm7, %zmm18, %zmm8
; AVX512DQ-FAST-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQ-FAST-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm30
; AVX512DQ-FAST-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,5,15,u]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm1, %zmm2, %zmm30
; AVX512DQ-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm18
; AVX512DQ-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm22
@@ -15976,20 +15976,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: # zmm1 = zmm19[0,1,2,3],mem[4,5,6,7]
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm3, %zmm10 {%k1}
; AVX512DQ-FAST-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,11,u,u,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,11,u,u,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm3, %zmm13, %zmm9
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm17, %zmm2 {%k1}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm12 = <12,u,u,3,4,5,6,13>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm12 = [12,u,u,3,4,5,6,13]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm12
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm7, %zmm16 {%k3}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm3, %zmm2, %zmm16
; AVX512DQ-FAST-NEXT: movb $6, %sil
; AVX512DQ-FAST-NEXT: kmovw %esi, %k5
; AVX512DQ-FAST-NEXT: vpbroadcastq 456(%rcx), %ymm2
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm2 = mem[0,1,2,3],ymm2[4,5,6,7]
; AVX512DQ-FAST-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm0 {%k5}
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,9,u,u,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,9,u,u,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm3, %zmm0, %zmm2
; AVX512DQ-FAST-NEXT: movb $64, %sil
; AVX512DQ-FAST-NEXT: kmovw %esi, %k4
@@ -15998,15 +15998,15 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: kmovw %esi, %k4
; AVX512DQ-FAST-NEXT: vmovdqa64 %zmm3, %zmm10 {%k4}
; AVX512DQ-FAST-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,1,11,u,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,1,11,u,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm9, %zmm21
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm10, %zmm3
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,12,u,3,4,5,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,12,u,3,4,5,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm12, %zmm7
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <13,u,2,3,4,5,6,14>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [13,u,2,3,4,5,6,14]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm9
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512DQ-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm16
; AVX512DQ-FAST-NEXT: movb $12, %sil
; AVX512DQ-FAST-NEXT: kmovw %esi, %k4
@@ -16015,9 +16015,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512DQ-FAST-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm4 {%k4}
; AVX512DQ-FAST-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm1
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm4
-; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,2,3,9,u,6,7>
+; AVX512DQ-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,9,u,6,7]
; AVX512DQ-FAST-NEXT: vpermi2q %zmm0, %zmm2, %zmm10
; AVX512DQ-FAST-NEXT: vpbroadcastq 8(%rcx), %ymm0
; AVX512DQ-FAST-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5,6,7]
@@ -16672,10 +16672,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm1, %zmm2
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm10
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm1, %zmm3, %zmm10
; AVX512BW-ONLY-SLOW-NEXT: vmovdqu64 %zmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
@@ -16909,20 +16909,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm2 = zmm2[0,1,2,3],zmm12[4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm3, %zmm6 {%k1}
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,11,u,u,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,11,u,u,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm9, %zmm5
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm13, %zmm1 {%k1}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <12,u,u,3,4,5,6,13>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [12,u,u,3,4,5,6,13]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm1, %zmm9
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm8, %zmm17 {%k3}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm3, %zmm1, %zmm17
; AVX512BW-ONLY-SLOW-NEXT: movb $6, %sil
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k4
; AVX512BW-ONLY-SLOW-NEXT: vpbroadcastq 456(%rcx), %ymm1
; AVX512BW-ONLY-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = mem[0,1,2,3],ymm1[4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm0 {%k4}
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,1,2,9,u,u,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,1,2,9,u,u,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm3, %zmm0, %zmm8
; AVX512BW-ONLY-SLOW-NEXT: movb $64, %sil
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k5
@@ -16931,16 +16931,16 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k5
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm3, %zmm6 {%k5}
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,11,u,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,11,u,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm1
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm6, %zmm3
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,12,u,3,4,5,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,12,u,3,4,5,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm9, %zmm5
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm6
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm17
; AVX512BW-ONLY-SLOW-NEXT: movb $12, %sil
; AVX512BW-ONLY-SLOW-NEXT: kmovd %esi, %k5
@@ -16949,9 +16949,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512BW-ONLY-SLOW-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm4 {%k5}
; AVX512BW-ONLY-SLOW-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm2
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm4
-; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,9,u,6,7>
+; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,9,u,6,7]
; AVX512BW-ONLY-SLOW-NEXT: vpermi2q %zmm0, %zmm8, %zmm2
; AVX512BW-ONLY-SLOW-NEXT: vmovdqa64 384(%rax), %zmm0
; AVX512BW-ONLY-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm9 # 64-byte Reload
@@ -17656,10 +17656,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: vpermt2q %zmm14, %zmm20, %zmm1
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm24
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-FAST-NEXT: vpermt2q %zmm1, %zmm2, %zmm24
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm20
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm3
@@ -17894,21 +17894,21 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: # zmm1 = zmm10[0,1,2,3],mem[4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm3, %zmm8 {%k1}
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,11,u,u,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,11,u,u,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm11, %zmm10
; AVX512BW-ONLY-FAST-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm6 # 64-byte Reload
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm6, %zmm2 {%k1}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <12,u,u,3,4,5,6,13>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [12,u,u,3,4,5,6,13]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm11
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm30, %zmm15 {%k3}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512BW-ONLY-FAST-NEXT: vpermt2q %zmm3, %zmm2, %zmm15
; AVX512BW-ONLY-FAST-NEXT: movb $6, %sil
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k5
; AVX512BW-ONLY-FAST-NEXT: vpbroadcastq 456(%rcx), %ymm2
; AVX512BW-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm2 = mem[0,1,2,3],ymm2[4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm0 {%k5}
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,9,u,u,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,9,u,u,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm3, %zmm0, %zmm2
; AVX512BW-ONLY-FAST-NEXT: movb $64, %sil
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k4
@@ -17917,15 +17917,15 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k4
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 %zmm3, %zmm8 {%k4}
; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = <0,1,11,u,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm31 = [0,1,11,u,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm10, %zmm31
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm8, %zmm3
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,12,u,3,4,5,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,12,u,3,4,5,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm11, %zmm8
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <13,u,2,3,4,5,6,14>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [13,u,2,3,4,5,6,14]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm10
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512BW-ONLY-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm15
; AVX512BW-ONLY-FAST-NEXT: movb $12, %sil
; AVX512BW-ONLY-FAST-NEXT: kmovd %esi, %k4
@@ -17934,9 +17934,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512BW-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512BW-ONLY-FAST-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm4 {%k4}
; AVX512BW-ONLY-FAST-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm1
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm4
-; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = <0,1,2,3,9,u,6,7>
+; AVX512BW-ONLY-FAST-NEXT: vmovdqa64 {{.*#+}} zmm11 = [0,1,2,3,9,u,6,7]
; AVX512BW-ONLY-FAST-NEXT: vpermi2q %zmm0, %zmm2, %zmm11
; AVX512BW-ONLY-FAST-NEXT: vpbroadcastq 8(%rcx), %ymm0
; AVX512BW-ONLY-FAST-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5,6,7]
@@ -18590,10 +18590,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm11, %zmm1, %zmm2
; AVX512DQBW-SLOW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQBW-SLOW-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm0, %zmm2, %zmm7
; AVX512DQBW-SLOW-NEXT: vmovdqa64 384(%r9), %zmm2
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm2, %zmm8, %zmm7
; AVX512DQBW-SLOW-NEXT: vmovdqu64 %zmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm2, %zmm0, %zmm1
@@ -18827,20 +18827,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm2 = zmm2[0,1,2,3],zmm12[4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm3, %zmm7 {%k1}
; AVX512DQBW-SLOW-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,11,u,u,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,11,u,u,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm3, %zmm8, %zmm5
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm11, %zmm1 {%k1}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <12,u,u,3,4,5,6,13>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [12,u,u,3,4,5,6,13]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm3, %zmm1, %zmm8
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm10, %zmm13 {%k3}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm3, %zmm1, %zmm13
; AVX512DQBW-SLOW-NEXT: movb $6, %sil
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k4
; AVX512DQBW-SLOW-NEXT: vpbroadcastq 456(%rcx), %ymm1
; AVX512DQBW-SLOW-NEXT: vpblendd {{.*#+}} ymm1 = mem[0,1,2,3],ymm1[4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm0 {%k4}
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,2,9,u,u,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,9,u,u,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm3, %zmm0, %zmm10
; AVX512DQBW-SLOW-NEXT: movb $64, %sil
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k5
@@ -18849,16 +18849,16 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k5
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm3, %zmm7 {%k5}
; AVX512DQBW-SLOW-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,1,11,u,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,1,11,u,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm5, %zmm1
; AVX512DQBW-SLOW-NEXT: vmovdqa64 %zmm1, %zmm11
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm7, %zmm3
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <0,12,u,3,4,5,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [0,12,u,3,4,5,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm8, %zmm5
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm7
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-SLOW-NEXT: vpermt2q %zmm0, %zmm1, %zmm13
; AVX512DQBW-SLOW-NEXT: movb $12, %sil
; AVX512DQBW-SLOW-NEXT: kmovd %esi, %k5
@@ -18867,9 +18867,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-SLOW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQBW-SLOW-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm4 {%k5}
; AVX512DQBW-SLOW-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm2
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm2, %zmm4
-; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,3,9,u,6,7>
+; AVX512DQBW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,3,9,u,6,7]
; AVX512DQBW-SLOW-NEXT: vpermi2q %zmm0, %zmm10, %zmm2
; AVX512DQBW-SLOW-NEXT: vmovdqa64 384(%rax), %zmm0
; AVX512DQBW-SLOW-NEXT: vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm10 # 64-byte Reload
@@ -19563,10 +19563,10 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm7, %zmm18, %zmm8
; AVX512DQBW-FAST-NEXT: vmovdqu64 %zmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512DQBW-FAST-NEXT: vmovdqa64 384(%r8), %zmm0
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm30
; AVX512DQBW-FAST-NEXT: vmovdqa64 384(%r9), %zmm1
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm1, %zmm2, %zmm30
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm18
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm1, %zmm0, %zmm22
@@ -19805,20 +19805,20 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: # zmm1 = zmm19[0,1,2,3],mem[4,5,6,7]
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm3, %zmm10 {%k1}
; AVX512DQBW-FAST-NEXT: vmovdqa64 448(%r8), %zmm3
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <0,11,u,u,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [0,11,u,u,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm3, %zmm13, %zmm9
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm17, %zmm2 {%k1}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm12 = <12,u,u,3,4,5,6,13>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm12 = [12,u,u,3,4,5,6,13]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm3, %zmm2, %zmm12
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm7, %zmm16 {%k3}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <u,1,2,3,4,15,u,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [u,1,2,3,4,15,u,u]
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm3, %zmm2, %zmm16
; AVX512DQBW-FAST-NEXT: movb $6, %sil
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k5
; AVX512DQBW-FAST-NEXT: vpbroadcastq 456(%rcx), %ymm2
; AVX512DQBW-FAST-NEXT: vpblendd {{.*#+}} ymm2 = mem[0,1,2,3],ymm2[4,5,6,7]
; AVX512DQBW-FAST-NEXT: vinserti64x4 $0, %ymm2, %zmm0, %zmm0 {%k5}
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,1,2,9,u,u,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,1,2,9,u,u,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm3, %zmm0, %zmm2
; AVX512DQBW-FAST-NEXT: movb $64, %sil
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k4
@@ -19827,15 +19827,15 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k4
; AVX512DQBW-FAST-NEXT: vmovdqa64 %zmm3, %zmm10 {%k4}
; AVX512DQBW-FAST-NEXT: vmovdqa64 448(%r9), %zmm0
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = <0,1,11,u,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = [0,1,11,u,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm9, %zmm21
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,1,2,10,u,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,1,2,10,u,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm10, %zmm3
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <0,12,u,3,4,5,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [0,12,u,3,4,5,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm12, %zmm7
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <13,u,2,3,4,5,6,14>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [13,u,2,3,4,5,6,14]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm9
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <u,1,2,3,4,5,15,u>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [u,1,2,3,4,5,15,u]
; AVX512DQBW-FAST-NEXT: vpermt2q %zmm0, %zmm1, %zmm16
; AVX512DQBW-FAST-NEXT: movb $12, %sil
; AVX512DQBW-FAST-NEXT: kmovd %esi, %k4
@@ -19844,9 +19844,9 @@ define void @store_i64_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.ve
; AVX512DQBW-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm1
; AVX512DQBW-FAST-NEXT: vinserti64x4 $0, %ymm1, %zmm0, %zmm4 {%k4}
; AVX512DQBW-FAST-NEXT: vinserti32x4 $2, 448(%r8), %zmm4, %zmm1
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = <0,1,2,3,4,8,u,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm4 = [0,1,2,3,4,8,u,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm1, %zmm4
-; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,1,2,3,9,u,6,7>
+; AVX512DQBW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,1,2,3,9,u,6,7]
; AVX512DQBW-FAST-NEXT: vpermi2q %zmm0, %zmm2, %zmm10
; AVX512DQBW-FAST-NEXT: vpbroadcastq 8(%rcx), %ymm0
; AVX512DQBW-FAST-NEXT: vpblendd {{.*#+}} ymm0 = mem[0,1,2,3],ymm0[4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-3.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-3.ll
index f727622682cf237..c6c74acc618fc4c 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-3.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-3.ll
@@ -855,9 +855,9 @@ define void @store_i8_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm8, %xmm6, %xmm0
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm8, %xmm2, %xmm8
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,u,u,128,128,128,128,128,128,6,7,8,9,10>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,u,u,128,128,128,128,128,128,6,7,8,9,10]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm2, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,u,u,5,6,7,8,9,10,128,128,128,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,u,u,5,6,7,8,9,10,128,128,128,128,128]
; AVX1-ONLY-NEXT: vmovdqa 16(%rsi), %xmm12
; AVX1-ONLY-NEXT: vmovdqa 32(%rsi), %xmm13
; AVX1-ONLY-NEXT: vmovdqa 48(%rsi), %xmm14
@@ -878,7 +878,7 @@ define void @store_i8_stride3_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpor %xmm7, %xmm10, %xmm0
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm7 = xmm3[8],xmm15[8],xmm3[9],xmm15[9],xmm3[10],xmm15[10],xmm3[11],xmm15[11],xmm3[12],xmm15[12],xmm3[13],xmm15[13],xmm3[14],xmm15[14],xmm3[15],xmm15[15]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,u,u,4,6,8,10,12,14,7,9,11,13,15>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,u,u,4,6,8,10,12,14,7,9,11,13,15]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm7, %xmm6
; AVX1-ONLY-NEXT: vmovdqa %xmm1, %xmm0
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm7 = xmm1[8],xmm12[8],xmm1[9],xmm12[9],xmm1[10],xmm12[10],xmm1[11],xmm12[11],xmm1[12],xmm12[12],xmm1[13],xmm12[13],xmm1[14],xmm12[14],xmm1[15],xmm12[15]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll
index 9c09bb7dca60cb4..c4bf0c3630e835b 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-5.ll
@@ -459,7 +459,7 @@ define void @store_i8_stride5_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,0,1]
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,ymm2[0,8,u],zero,zero,ymm2[1,9,u],zero,zero,ymm2[2,10,u],zero,ymm2[27],zero,zero,ymm2[u,20,28],zero,zero,ymm2[u,21,29],zero,zero,ymm2[u,22,30]
; AVX512F-FAST-NEXT: vpor %ymm2, %ymm4, %ymm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,0,0,0,0,u,1,1>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,0,0,0,0,u,1,1]
; AVX512F-FAST-NEXT: vpermd %ymm3, %ymm4, %ymm3
; AVX512F-FAST-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm3
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,xmm0[u,7,15],zero,zero,xmm0[u,u,u,u,u,u,u,u,u]
@@ -730,7 +730,7 @@ define void @store_i8_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = xmm3[u,u,u],zero,xmm3[7,u,u,u],zero,xmm3[8,u,u,u],zero,xmm3[9,u]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = xmm2[u,u,u,7],zero,xmm2[u,u,u,8],zero,xmm2[u,u,u,9],zero,xmm2[u]
; AVX1-ONLY-NEXT: vpor %xmm6, %xmm7, %xmm6
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255]
; AVX1-ONLY-NEXT: vpblendvb %xmm7, %xmm5, %xmm6, %xmm5
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[0,1],zero,xmm5[3,4,5,6],zero,xmm5[8,9,10,11],zero,xmm5[13,14,15]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = zero,zero,xmm0[6],zero,zero,zero,zero,xmm0[7],zero,zero,zero,zero,xmm0[8],zero,zero,zero
@@ -785,7 +785,7 @@ define void @store_i8_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm5[2,3,0,1]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm9 = zero,zero,zero,zero,ymm9[7],zero,zero,zero,zero,ymm9[8],zero,zero,zero,zero,ymm9[9],zero,zero,zero,ymm9[26],zero,zero,zero,zero,ymm9[27],zero,zero,zero,zero,ymm9[28],zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm8 = xmm2[1,1,2,2]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,1,1,1]
@@ -872,7 +872,7 @@ define void @store_i8_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm5[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = zero,zero,zero,zero,ymm9[7],zero,zero,zero,zero,ymm9[8],zero,zero,zero,zero,ymm9[9],zero,zero,zero,ymm9[26],zero,zero,zero,zero,ymm9[27],zero,zero,zero,zero,ymm9[28],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm8, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm8 = xmm2[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,1,1,1]
@@ -925,7 +925,7 @@ define void @store_i8_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm6 = ymm6[0,8],zero,zero,ymm6[u,1,9],zero,zero,ymm6[u,2,10],zero,zero,ymm6[u,3,19],zero,zero,ymm6[u,28,20],zero,zero,ymm6[u,29,21],zero,zero,ymm6[u,30,22]
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm5, %zmm6, %zmm5
; AVX512F-SLOW-NEXT: vporq %zmm7, %zmm5, %zmm5
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = <u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm6 = [u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u]
; AVX512F-SLOW-NEXT: vpermd %zmm1, %zmm6, %zmm6
; AVX512F-SLOW-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm5, %zmm6
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm3[8],xmm4[8],xmm3[9],xmm4[9],xmm3[10],xmm4[10],xmm3[11],xmm4[11],xmm3[12],xmm4[12],xmm3[13],xmm4[13],xmm3[14],xmm4[14],xmm3[15],xmm4[15]
@@ -964,7 +964,7 @@ define void @store_i8_stride5_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm6, %zmm5, %zmm5
; AVX512F-FAST-NEXT: vporq %zmm7, %zmm5, %zmm5
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm6
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u]
; AVX512F-FAST-NEXT: vpermd %zmm6, %zmm7, %zmm6
; AVX512F-FAST-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm5, %zmm6
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm3[8],xmm4[8],xmm3[9],xmm4[9],xmm3[10],xmm4[10],xmm3[11],xmm4[11],xmm3[12],xmm4[12],xmm3[13],xmm4[13],xmm3[14],xmm4[14],xmm3[15],xmm4[15]
@@ -1594,7 +1594,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: # ymm8 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm6
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm6 = ymm0[2,2,3,3,6,6,7,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,3,3]
@@ -1610,7 +1610,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm10[0],xmm9[0],xmm10[1],xmm9[1],xmm10[2],xmm9[2],xmm10[3],xmm9[3],xmm10[4],xmm9[4],xmm10[5],xmm9[5],xmm10[6],xmm9[6],xmm10[7],xmm9[7]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm11 = xmm11[2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,1,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm6, %ymm11, %ymm6
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm11
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm11[0,0,1,1]
@@ -1625,7 +1625,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm7 = zero,xmm7[8,u],zero,xmm7[7],zero,xmm7[9,u,11,u],zero,xmm7[10],zero,xmm7[12,u],zero
; AVX2-SLOW-NEXT: vpor %xmm7, %xmm8, %xmm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,0,1,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm9, %ymm7, %ymm7
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm8 = xmm11[1,1,2,2]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[0,1,1,1]
@@ -1641,23 +1641,23 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm10 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm1[19],zero,ymm1[21],zero,zero,ymm1[20],zero,ymm1[22],zero,ymm1[24],zero,zero,ymm1[23],zero,ymm1[25]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,2,3,3]
; AVX2-SLOW-NEXT: vpor %ymm9, %ymm10, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm9 = ymm0[0,2,1,1,4,6,5,5]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,3,3,2]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <3,3,3,u,4,4,4,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [3,3,3,u,4,4,4,4]
; AVX2-SLOW-NEXT: vpermd %ymm3, %ymm9, %ymm3
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm4 = ymm4[u,u,13,u,u,u,u,14,u,u,u,u,15,u,u,u,u,16,u,u,u,u,17,u,u,u,u,18,u,u,u,u]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm3, %ymm4, %ymm3
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,ymm2[13],zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,ymm2[17],zero,zero,zero,zero,ymm2[18],zero,zero
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = zero,zero,zero,ymm1[13],zero,zero,zero,zero,ymm1[14],zero,zero,zero,zero,ymm1[15],zero,zero,zero,zero,ymm1[16],zero,zero,zero,zero,ymm1[17],zero,zero,zero,zero,ymm1[18],zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm2, %ymm1, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm3, %ymm1, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <3,3,3,3,u,4,4,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [3,3,3,3,u,4,4,4]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
@@ -1686,7 +1686,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm10 = xmm9[0],xmm8[0],xmm9[1],xmm8[1],xmm9[2],xmm8[2],xmm9[3],xmm8[3],xmm9[4],xmm8[4],xmm9[5],xmm8[5],xmm9[6],xmm8[6],xmm9[7],xmm8[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,0,1,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm5, %ymm10, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [0,0,0,0,0,0,1,1]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm10, %ymm10
@@ -1700,7 +1700,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm6 = zero,xmm6[8,u],zero,xmm6[7],zero,xmm6[9,u,11,u],zero,xmm6[10],zero,xmm6[12,u],zero
; AVX2-FAST-NEXT: vpor %xmm7, %xmm6, %xmm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,1,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm8, %ymm6, %ymm6
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [1,1,2,2,2,2,2,2]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm7, %ymm7
@@ -1716,7 +1716,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm9 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm1[19],zero,ymm1[21],zero,zero,ymm1[20],zero,ymm1[22],zero,ymm1[24],zero,zero,ymm1[23],zero,ymm1[25]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,3]
; AVX2-FAST-NEXT: vpor %ymm8, %ymm9, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [4,6,5,5,5,5,4,6]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm8, %ymm8
@@ -1732,23 +1732,23 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,27],zero,zero,ymm4[26],zero,ymm4[28],zero,zero,zero,zero,ymm4[29],zero,ymm4[31],zero,zero,ymm4[30]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,2,3,3]
; AVX2-FAST-NEXT: vpor %ymm9, %ymm10, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [6,6,6,6,7,7,7,7]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm9, %ymm9
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <3,3,3,u,4,4,4,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [3,3,3,u,4,4,4,4]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm9, %ymm3
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm4 = ymm4[u,u,13,u,u,u,u,14,u,u,u,u,15,u,u,u,u,16,u,u,u,u,17,u,u,u,u,18,u,u,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,ymm2[13],zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,ymm2[17],zero,zero,zero,zero,ymm2[18],zero,zero
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = zero,zero,zero,ymm1[13],zero,zero,zero,zero,ymm1[14],zero,zero,zero,zero,ymm1[15],zero,zero,zero,zero,ymm1[16],zero,zero,zero,zero,ymm1[17],zero,zero,zero,zero,ymm1[18],zero,zero,zero
; AVX2-FAST-NEXT: vpor %ymm2, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm3, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <3,3,3,3,u,4,4,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [3,3,3,3,u,4,4,4]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
@@ -1777,7 +1777,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm10 = xmm9[0],xmm8[0],xmm9[1],xmm8[1],xmm9[2],xmm8[2],xmm9[3],xmm8[3],xmm9[4],xmm8[4],xmm9[5],xmm8[5],xmm9[6],xmm8[6],xmm9[7],xmm8[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm10 = xmm10[2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,0,1,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm5, %ymm10, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %xmm10
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm11 = xmm10[0,0,1,1]
@@ -1792,7 +1792,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm6 = zero,xmm6[8,u],zero,xmm6[7],zero,xmm6[9,u,11,u],zero,xmm6[10],zero,xmm6[12,u],zero
; AVX2-FAST-PERLANE-NEXT: vpor %xmm7, %xmm6, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,1,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm8, %ymm6, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} xmm7 = xmm10[1,1,2,2]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,1,1]
@@ -1808,7 +1808,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm9 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm1[19],zero,ymm1[21],zero,zero,ymm1[20],zero,ymm1[22],zero,ymm1[24],zero,zero,ymm1[23],zero,ymm1[25]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm8, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm8 = ymm0[0,2,1,1,4,6,5,5]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,3,2]
@@ -1824,23 +1824,23 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm10 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,27],zero,zero,ymm4[26],zero,ymm4[28],zero,zero,zero,zero,ymm4[29],zero,ymm4[31],zero,zero,ymm4[30]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm9, %ymm10, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm9 = ymm0[2,2,3,3,6,6,7,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm8, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <3,3,3,u,4,4,4,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [3,3,3,u,4,4,4,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm3, %ymm9, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm4 = ymm4[u,u,13,u,u,u,u,14,u,u,u,u,15,u,u,u,u,16,u,u,u,u,17,u,u,u,u,18,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm3, %ymm4, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,ymm2[13],zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,ymm2[17],zero,zero,zero,zero,ymm2[18],zero,zero
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = zero,zero,zero,ymm1[13],zero,zero,zero,zero,ymm1[14],zero,zero,zero,zero,ymm1[15],zero,zero,zero,zero,ymm1[16],zero,zero,zero,zero,ymm1[17],zero,zero,zero,zero,ymm1[18],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm2, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm3, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <3,3,3,3,u,4,4,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [3,3,3,3,u,4,4,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
@@ -1884,7 +1884,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpternlogq $226, %ymm8, %ymm7, %ymm5
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm5 = zmm5[0,1,2,3],zmm6[4,5,6,7]
; AVX512F-SLOW-NEXT: vmovdqa (%r8), %xmm6
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u]
; AVX512F-SLOW-NEXT: vpermd %zmm6, %zmm8, %zmm6
; AVX512F-SLOW-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm5, %zmm6
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm5 = ymm4[u,u,u],zero,ymm4[13,u,u,u],zero,ymm4[14,u,u,u],zero,ymm4[15,u,u,u],zero,ymm4[16,u,u,u],zero,ymm4[17,u,u,u],zero,ymm4[18,u,u]
@@ -1966,7 +1966,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpternlogq $226, %ymm7, %ymm6, %ymm4
; AVX512F-FAST-NEXT: vshufi64x2 {{.*#+}} zmm5 = zmm4[0,1,2,3],zmm5[4,5,6,7]
; AVX512F-FAST-NEXT: vbroadcasti64x4 {{.*#+}} zmm4 = mem[0,1,2,3,0,1,2,3]
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u]
; AVX512F-FAST-NEXT: vpermd %zmm4, %zmm7, %zmm7
; AVX512F-FAST-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm5, %zmm7
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm5 = ymm3[u,u,u],zero,ymm3[13,u,u,u],zero,ymm3[14,u,u,u],zero,ymm3[15,u,u,u],zero,ymm3[16,u,u,u],zero,ymm3[17,u,u,u],zero,ymm3[18,u,u]
@@ -1990,7 +1990,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm9, %ymm5
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm5, %zmm0, %zmm5
; AVX512F-FAST-NEXT: vshufi64x2 {{.*#+}} zmm5 = zmm8[0,1,2,3],zmm5[4,5,6,7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <4,u,5,5,5,5,u,6>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [4,u,5,5,5,5,u,6]
; AVX512F-FAST-NEXT: vpermd %ymm4, %ymm8, %ymm8
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255]
; AVX512F-FAST-NEXT: vpandn %ymm8, %ymm9, %ymm8
@@ -2008,7 +2008,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,3,3]
; AVX512F-FAST-NEXT: vpternlogq $248, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm0
; AVX512F-FAST-NEXT: vpternlogq $184, %ymm2, %ymm6, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <6,6,6,u,7,7,7,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [6,6,6,u,7,7,7,7]
; AVX512F-FAST-NEXT: vpermd %ymm4, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm1
; AVX512F-FAST-NEXT: vmovdqa %ymm1, 128(%r9)
@@ -2065,7 +2065,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm4[21],zero,zero,ymm4[20],zero,ymm4[22],zero,ymm4[24],zero,zero,ymm4[23],zero,ymm4[25],zero,zero
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,2,3,3]
; AVX512BW-SLOW-NEXT: vpor %ymm7, %ymm8, %ymm7
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <3,3,3,u,4,4,4,4>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [3,3,3,u,4,4,4,4]
; AVX512BW-SLOW-NEXT: vpermd %ymm4, %ymm8, %ymm8
; AVX512BW-SLOW-NEXT: movl $138547332, %eax # imm = 0x8421084
; AVX512BW-SLOW-NEXT: kmovd %eax, %k1
@@ -2074,7 +2074,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: movabsq $-8330787646191410408, %rax # imm = 0x8C6318C6318C6318
; AVX512BW-SLOW-NEXT: kmovq %rax, %k1
; AVX512BW-SLOW-NEXT: vmovdqu8 %zmm6, %zmm7 {%k1}
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <3,3,3,3,u,4,4,4>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [3,3,3,3,u,4,4,4]
; AVX512BW-SLOW-NEXT: vpermd %ymm0, %ymm6, %ymm6
; AVX512BW-SLOW-NEXT: vpshufd {{.*#+}} ymm8 = ymm0[0,2,1,1,4,6,5,5]
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,3,2]
@@ -2157,7 +2157,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm1[21],zero,zero,ymm1[20],zero,ymm1[22],zero,ymm1[24],zero,zero,ymm1[23],zero,ymm1[25],zero,zero
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,2,3,3]
; AVX512BW-FAST-NEXT: vpor %ymm7, %ymm8, %ymm7
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <3,3,3,u,4,4,4,4>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [3,3,3,u,4,4,4,4]
; AVX512BW-FAST-NEXT: vpermd %ymm1, %ymm8, %ymm8
; AVX512BW-FAST-NEXT: movl $138547332, %eax # imm = 0x8421084
; AVX512BW-FAST-NEXT: kmovd %eax, %k1
@@ -2166,7 +2166,7 @@ define void @store_i8_stride5_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: movabsq $-8330787646191410408, %rax # imm = 0x8C6318C6318C6318
; AVX512BW-FAST-NEXT: kmovq %rax, %k1
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm6, %zmm7 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = <3,3,3,3,u,4,4,4,12,14,13,13,13,13,12,14>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [3,3,3,3,u,4,4,4,12,14,13,13,13,13,12,14]
; AVX512BW-FAST-NEXT: vpermd %zmm5, %zmm6, %zmm6
; AVX512BW-FAST-NEXT: movabsq $1190112520884487201, %rax # imm = 0x1084210842108421
; AVX512BW-FAST-NEXT: kmovq %rax, %k1
@@ -3235,26 +3235,26 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-SLOW-NEXT: vmovdqa 32(%rcx), %xmm11
; AVX2-SLOW-NEXT: vmovdqa %xmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vmovdqa (%rdx), %xmm3
; AVX2-SLOW-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-SLOW-NEXT: vmovdqa 32(%rdx), %xmm12
; AVX2-SLOW-NEXT: vmovdqa %xmm12, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-SLOW-NEXT: vpor %xmm1, %xmm3, %xmm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,1]
; AVX2-SLOW-NEXT: vmovdqa (%rdi), %xmm4
; AVX2-SLOW-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm4, %xmm4
; AVX2-SLOW-NEXT: vmovdqa (%rsi), %xmm14
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX2-SLOW-NEXT: vpshufb %xmm5, %xmm14, %xmm8
; AVX2-SLOW-NEXT: vpor %xmm4, %xmm8, %xmm4
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm4, %ymm1
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm4
; AVX2-SLOW-NEXT: vmovdqa %xmm4, (%rsp) # 16-byte Spill
@@ -3304,7 +3304,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb %ymm4, %ymm13, %ymm12
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,2,3,3]
; AVX2-SLOW-NEXT: vpor %ymm9, %ymm12, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm1, %ymm9, %ymm9
; AVX2-SLOW-NEXT: vmovdqa (%rdi), %ymm6
; AVX2-SLOW-NEXT: vpshufb %ymm15, %ymm6, %ymm1
@@ -3333,11 +3333,11 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,3,3,2]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm0, %ymm4, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <3,3,3,u,4,4,4,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [3,3,3,u,4,4,4,4]
; AVX2-SLOW-NEXT: vpermd %ymm11, %ymm3, %ymm4
; AVX2-SLOW-NEXT: vpbroadcastq {{.*#+}} ymm5 = [0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14]
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm2, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm4, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpermd %ymm6, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm15, %ymm4
@@ -3348,13 +3348,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128,128]
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm13, %ymm11
; AVX2-SLOW-NEXT: vpor %ymm4, %ymm11, %ymm4
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm0, %ymm4, %ymm0
; AVX2-SLOW-NEXT: vpshufb %ymm2, %ymm10, %ymm2
; AVX2-SLOW-NEXT: vpshufb %ymm5, %ymm8, %ymm4
; AVX2-SLOW-NEXT: vpor %ymm2, %ymm4, %ymm2
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <3,3,3,3,u,4,4,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [3,3,3,3,u,4,4,4]
; AVX2-SLOW-NEXT: vpermd %ymm12, %ymm2, %ymm4
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm0, %ymm4, %ymm0
@@ -3367,13 +3367,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm1 # 16-byte Folded Reload
; AVX2-SLOW-NEXT: # xmm1 = xmm1[0],mem[0],xmm1[1],mem[1],xmm1[2],mem[2],xmm1[3],mem[3],xmm1[4],mem[4],xmm1[5],mem[5],xmm1[6],mem[6],xmm1[7],mem[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX2-SLOW-NEXT: vpshufb %xmm14, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-SLOW-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm0, %ymm1, %ymm6
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm0, %xmm0 # 16-byte Folded Reload
@@ -3422,7 +3422,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm2, %ymm4, %ymm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,3,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,3,3]
@@ -3464,27 +3464,27 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vmovdqa 32(%rcx), %xmm6
; AVX2-FAST-NEXT: vmovdqa %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-NEXT: vmovdqa (%rdx), %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vmovdqa 32(%rdx), %xmm8
; AVX2-FAST-NEXT: vmovdqa %xmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-NEXT: vpor %xmm1, %xmm3, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,1]
; AVX2-FAST-NEXT: vmovdqa (%rdi), %xmm4
; AVX2-FAST-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm4, %xmm4
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm7
; AVX2-FAST-NEXT: vmovdqa %xmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm7, %xmm7
; AVX2-FAST-NEXT: vpor %xmm4, %xmm7, %xmm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm1, %ymm4, %ymm1
; AVX2-FAST-NEXT: vmovdqa 32(%rdi), %xmm4
; AVX2-FAST-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -3528,7 +3528,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm15, %ymm2, %ymm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,2,3,3]
; AVX2-FAST-NEXT: vpor %ymm4, %ymm5, %ymm4
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm1, %ymm4, %ymm4
; AVX2-FAST-NEXT: vmovdqa (%rdx), %ymm10
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm10, %ymm0
@@ -3569,7 +3569,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm13, %ymm14
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,2,3,3]
; AVX2-FAST-NEXT: vpor %ymm15, %ymm14, %ymm14
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm7, %ymm14, %ymm7
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm1, %ymm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,3,3]
@@ -3588,11 +3588,11 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm7, %ymm8, %ymm9
; AVX2-FAST-NEXT: vpermd %ymm11, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm4, %ymm6, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <3,3,3,u,4,4,4,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [3,3,3,u,4,4,4,4]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm4, %ymm3
; AVX2-FAST-NEXT: vpbroadcastq {{.*#+}} ymm6 = [0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14]
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm3, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm4, %ymm1
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm0, %ymm0
@@ -3603,13 +3603,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128,128]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm13, %ymm6
; AVX2-FAST-NEXT: vpor %ymm3, %ymm6, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpshufb %ymm1, %ymm5, %ymm1
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm10, %ymm3
; AVX2-FAST-NEXT: vpor %ymm1, %ymm3, %ymm1
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm1, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <3,3,3,3,u,4,4,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [3,3,3,3,u,4,4,4]
; AVX2-FAST-NEXT: vpermd %ymm12, %ymm3, %ymm0
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm2, %ymm0, %ymm0
@@ -3621,13 +3621,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm3, %xmm3 # 16-byte Folded Reload
; AVX2-FAST-NEXT: # xmm3 = xmm3[0],mem[0],xmm3[1],mem[1],xmm3[2],mem[2],xmm3[3],mem[3],xmm3[4],mem[4],xmm3[5],mem[5],xmm3[6],mem[6],xmm3[7],mem[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX2-FAST-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-FAST-NEXT: vpshufb %xmm5, %xmm3, %xmm3
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm3, %xmm3 # 16-byte Folded Reload
@@ -3672,27 +3672,27 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm1, (%rsp) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rcx), %xmm8
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdx), %xmm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdx), %xmm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpor %xmm1, %xmm3, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,1,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdi), %xmm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm4, %xmm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsi), %xmm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm6, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpor %xmm4, %xmm6, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm1, %ymm4, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %xmm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -3741,7 +3741,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm2, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm6, %ymm7, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u,0,0,255,255,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm1, %ymm6, %ymm3
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdx), %ymm13
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm13, %ymm0
@@ -3785,7 +3785,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm11, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm14, %ymm12, %ymm12
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm9, %ymm12, %ymm9
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm8, %ymm1, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,2,3,3]
@@ -3805,11 +3805,11 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufd {{.*#+}} ymm8 = ymm3[0,2,1,1,4,6,5,5]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,3,2]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm6, %ymm8, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <3,3,3,u,4,4,4,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [3,3,3,u,4,4,4,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm4, %ymm6, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpbroadcastq {{.*#+}} ymm8 = [0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm8, %ymm2, %ymm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255,0,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm4, %ymm2, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm1, %ymm6, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm8, %ymm0, %ymm0
@@ -3820,13 +3820,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm11, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpor %ymm4, %ymm8, %ymm4
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm2, %ymm4, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm1, %ymm7, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm13, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpor %ymm1, %ymm4, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm0, %ymm1, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <3,3,3,3,u,4,4,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [3,3,3,3,u,4,4,4]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm5, %ymm4, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255,255,255,255,0,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm2, %ymm0, %ymm0
@@ -3838,13 +3838,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsp), %xmm3 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm3, %xmm3 # 16-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # xmm3 = xmm3[0],mem[0],xmm3[1],mem[1],xmm3[2],mem[2],xmm3[3],mem[3],xmm3[4],mem[4],xmm3[5],mem[5],xmm3[6],mem[6],xmm3[7],mem[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm4 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,1,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm5 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm5, %xmm3, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255,0,0,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm2, %ymm3, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm3, %xmm3 # 16-byte Folded Reload
@@ -3890,16 +3890,16 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm15 = [128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128,128,128]
; AVX512F-SLOW-NEXT: vpshufb %ymm15, %ymm3, %ymm0
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdi), %ymm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <12,13,128,15,12,13,14,128,12,13,14,15,128,u,u,u,16,128,18,19,16,17,128,19,16,17,18,128,16,17,18,19>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [12,13,128,15,12,13,14,128,12,13,14,15,128,u,u,u,16,128,18,19,16,17,128,19,16,17,18,128,16,17,18,19]
; AVX512F-SLOW-NEXT: vpshufb %ymm9, %ymm2, %ymm1
; AVX512F-SLOW-NEXT: vpor %ymm0, %ymm1, %ymm0
; AVX512F-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdi), %xmm1
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm5 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX512F-SLOW-NEXT: vpshufb %xmm5, %xmm1, %xmm0
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm1, %xmm16
; AVX512F-SLOW-NEXT: vmovdqa 32(%rsi), %xmm4
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm4, %xmm1
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm4, %xmm31
; AVX512F-SLOW-NEXT: vpor %xmm0, %xmm1, %xmm0
@@ -3908,16 +3908,16 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128]
; AVX512F-SLOW-NEXT: vpshufb %ymm0, %ymm8, %ymm4
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdx), %ymm11
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,12,13,128,u,u,u,14,128,u,u,14,15,128,u,u,u,16,128,u,u,16,17,128,u,u,u,18,128,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,12,13,128,u,u,u,14,128,u,u,14,15,128,u,u,u,16,128,u,u,16,17,128,u,u,u,18,128,u,u]
; AVX512F-SLOW-NEXT: vpshufb %ymm1, %ymm11, %ymm10
; AVX512F-SLOW-NEXT: vpor %ymm4, %ymm10, %ymm4
; AVX512F-SLOW-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 32(%rcx), %xmm13
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX512F-SLOW-NEXT: vpshufb %xmm6, %xmm13, %xmm4
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm6, %xmm25
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdx), %xmm10
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX512F-SLOW-NEXT: vpshufb %xmm6, %xmm10, %xmm12
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm6, %xmm26
; AVX512F-SLOW-NEXT: vporq %xmm4, %xmm12, %xmm20
@@ -3986,14 +3986,14 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm16, %xmm1
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm31, %xmm4
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <4,u,5,5,5,5,u,6,6,6,6,u,7,7,7,7>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [4,u,5,5,5,5,u,6,6,6,6,u,7,7,7,7]
; AVX512F-SLOW-NEXT: vpermd %zmm15, %zmm1, %zmm31
; AVX512F-SLOW-NEXT: vmovdqa64 (%r8), %zmm16
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <6,6,6,u,7,7,7,7,u,16,16,16,16,u,17,17>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [6,6,6,u,7,7,7,7,u,16,16,16,16,u,17,17]
; AVX512F-SLOW-NEXT: vpermi2d %zmm15, %zmm16, %zmm1
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm17, %xmm15
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm15[0],xmm5[0],xmm15[1],xmm5[1],xmm15[2],xmm5[2],xmm15[3],xmm5[3],xmm15[4],xmm5[4],xmm15[5],xmm5[5],xmm15[6],xmm5[6],xmm15[7],xmm5[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX512F-SLOW-NEXT: vpshufb %xmm15, %xmm4, %xmm4
; AVX512F-SLOW-NEXT: vpshufb %xmm15, %xmm5, %xmm5
; AVX512F-SLOW-NEXT: vinserti32x4 $2, %xmm27, %zmm5, %zmm5
@@ -4007,7 +4007,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,3,3]
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm18, %xmm15
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm9 = xmm15[0],xmm9[0],xmm15[1],xmm9[1],xmm15[2],xmm9[2],xmm15[3],xmm9[3],xmm15[4],xmm9[4],xmm15[5],xmm9[5],xmm15[6],xmm9[6],xmm15[7],xmm9[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX512F-SLOW-NEXT: vpshufb %xmm15, %xmm10, %xmm10
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,0,1,1]
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,2,3,3]
@@ -4049,7 +4049,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} zmm0 = zmm5[0,0,1,1,4,4,5,5]
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} zmm2 = zmm9[0,0,1,1,4,4,5,5]
; AVX512F-SLOW-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm0, %zmm2
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,0,0,0,0,u,1,1,1,1,u,2,2,2,2,u]
; AVX512F-SLOW-NEXT: vpermd %zmm16, %zmm0, %zmm0
; AVX512F-SLOW-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm2, %zmm0
; AVX512F-SLOW-NEXT: vmovdqa64 %zmm25, 64(%r9)
@@ -4067,16 +4067,16 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128,128,128]
; AVX512F-FAST-NEXT: vpshufb %ymm8, %ymm3, %ymm0
; AVX512F-FAST-NEXT: vmovdqa 32(%rdi), %ymm1
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <12,13,128,15,12,13,14,128,12,13,14,15,128,u,u,u,16,128,18,19,16,17,128,19,16,17,18,128,16,17,18,19>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [12,13,128,15,12,13,14,128,12,13,14,15,128,u,u,u,16,128,18,19,16,17,128,19,16,17,18,128,16,17,18,19]
; AVX512F-FAST-NEXT: vpshufb %ymm11, %ymm1, %ymm2
; AVX512F-FAST-NEXT: vpor %ymm0, %ymm2, %ymm0
; AVX512F-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 32(%rdi), %xmm0
; AVX512F-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX512F-FAST-NEXT: vpshufb %xmm5, %xmm0, %xmm0
; AVX512F-FAST-NEXT: vmovdqa 32(%rsi), %xmm4
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX512F-FAST-NEXT: vpshufb %xmm13, %xmm4, %xmm2
; AVX512F-FAST-NEXT: vmovdqa64 %xmm4, %xmm19
; AVX512F-FAST-NEXT: vpor %xmm0, %xmm2, %xmm0
@@ -4085,16 +4085,16 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [128,128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128]
; AVX512F-FAST-NEXT: vpshufb %ymm0, %ymm7, %ymm2
; AVX512F-FAST-NEXT: vmovdqa 32(%rdx), %ymm9
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,12,13,128,u,u,u,14,128,u,u,14,15,128,u,u,u,16,128,u,u,16,17,128,u,u,u,18,128,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,12,13,128,u,u,u,14,128,u,u,14,15,128,u,u,u,16,128,u,u,16,17,128,u,u,u,18,128,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm15, %ymm9, %ymm4
; AVX512F-FAST-NEXT: vpor %ymm2, %ymm4, %ymm2
; AVX512F-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 32(%rcx), %xmm12
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX512F-FAST-NEXT: vpshufb %xmm4, %xmm12, %xmm2
; AVX512F-FAST-NEXT: vmovdqa64 %xmm4, %xmm25
; AVX512F-FAST-NEXT: vmovdqa 32(%rdx), %xmm10
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX512F-FAST-NEXT: vpshufb %xmm6, %xmm10, %xmm4
; AVX512F-FAST-NEXT: vmovdqa64 %xmm6, %xmm26
; AVX512F-FAST-NEXT: vpor %xmm2, %xmm4, %xmm2
@@ -4167,7 +4167,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; AVX512F-FAST-NEXT: vmovdqa64 %xmm16, %xmm1
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm3 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3],xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX512F-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-FAST-NEXT: vpshufb %xmm1, %xmm3, %xmm1
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm28, %zmm1, %zmm28
@@ -4182,7 +4182,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqa64 %xmm18, %xmm3
; AVX512F-FAST-NEXT: vmovdqa64 %xmm17, %xmm5
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm3[0],xmm5[0],xmm3[1],xmm5[1],xmm3[2],xmm5[2],xmm3[3],xmm5[3],xmm3[4],xmm5[4],xmm3[5],xmm5[5],xmm3[6],xmm5[6],xmm3[7],xmm5[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX512F-FAST-NEXT: vpshufb %xmm3, %xmm10, %xmm10
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,0,1,1]
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm15[2,2,3,3]
@@ -4225,13 +4225,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm6, %zmm0
; AVX512F-FAST-NEXT: vpternlogq $226, %zmm1, %zmm7, %zmm0
; AVX512F-FAST-NEXT: vpternlogq $248, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm2, %zmm27
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <6,6,6,u,7,7,7,7,u,8,8,8,8,u,9,9>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [6,6,6,u,7,7,7,7,u,8,8,8,8,u,9,9]
; AVX512F-FAST-NEXT: vpermd %zmm5, %zmm1, %zmm1
; AVX512F-FAST-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm0, %zmm1
; AVX512F-FAST-NEXT: vpermq {{.*#+}} zmm0 = zmm28[0,0,1,1,4,4,5,5]
; AVX512F-FAST-NEXT: vpermq {{.*#+}} zmm2 = zmm3[0,0,1,1,4,4,5,5]
; AVX512F-FAST-NEXT: vpternlogq $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm0, %zmm2
-; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u>
+; AVX512F-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,0,0,0,0,u,1,1,9,9,u,10,10,10,10,u]
; AVX512F-FAST-NEXT: vpermd %zmm11, %zmm0, %zmm0
; AVX512F-FAST-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm2, %zmm0
; AVX512F-FAST-NEXT: vmovdqa64 %zmm27, 64(%r9)
@@ -4258,7 +4258,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovdqa 32(%rdx), %xmm6
; AVX512BW-SLOW-NEXT: vmovdqa 32(%rcx), %xmm12
; AVX512BW-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm12[0],xmm6[0],xmm12[1],xmm6[1],xmm12[2],xmm6[2],xmm12[3],xmm6[3],xmm12[4],xmm6[4],xmm12[5],xmm6[5],xmm12[6],xmm6[6],xmm12[7],xmm6[7]
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX512BW-SLOW-NEXT: vpshufb %xmm7, %xmm4, %xmm4
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,1,1]
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm4, %zmm2, %zmm10
@@ -4275,7 +4275,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovdqa 32(%rsi), %xmm13
; AVX512BW-SLOW-NEXT: vmovdqa 32(%rdi), %xmm14
; AVX512BW-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm14[0],xmm13[0],xmm14[1],xmm13[1],xmm14[2],xmm13[2],xmm14[3],xmm13[3],xmm14[4],xmm13[4],xmm14[5],xmm13[5],xmm14[6],xmm13[6],xmm14[7],xmm13[7]
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX512BW-SLOW-NEXT: vpshufb %xmm9, %xmm11, %xmm11
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,1,1]
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm11, %zmm2, %zmm2
@@ -4295,21 +4295,21 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [128,128,128,128,13,128,128,128,128,14,128,128,128,128,15,128,128,128,128,16,128,128,128,128,17,128,128,128,128,18,128,128]
; AVX512BW-SLOW-NEXT: vpshufb %ymm11, %ymm24, %ymm18
; AVX512BW-SLOW-NEXT: vporq %ymm17, %ymm18, %ymm17
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm20 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm20 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX512BW-SLOW-NEXT: vpshufb %xmm20, %xmm12, %xmm12
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm22 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm22 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX512BW-SLOW-NEXT: vpshufb %xmm22, %xmm6, %xmm6
; AVX512BW-SLOW-NEXT: vpor %xmm6, %xmm12, %xmm6
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,1,1]
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm17, %zmm6, %zmm6
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm19 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm19 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX512BW-SLOW-NEXT: vpshufb %xmm19, %xmm14, %xmm12
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm21 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm21 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX512BW-SLOW-NEXT: vpshufb %xmm21, %xmm13, %xmm13
; AVX512BW-SLOW-NEXT: vpor %xmm12, %xmm13, %xmm12
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm13 = ymm12[0,0,1,1]
; AVX512BW-SLOW-NEXT: vmovdqa64 32(%rdi), %ymm25
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <3,3,3,u,4,4,4,4>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [3,3,3,u,4,4,4,4]
; AVX512BW-SLOW-NEXT: vpermd %ymm25, %ymm12, %ymm17
; AVX512BW-SLOW-NEXT: vmovdqa64 32(%rsi), %ymm26
; AVX512BW-SLOW-NEXT: vpbroadcastq {{.*#+}} ymm14 = [0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14]
@@ -4320,7 +4320,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: movabsq $-8330787646191410408, %rax # imm = 0x8C6318C6318C6318
; AVX512BW-SLOW-NEXT: kmovq %rax, %k2
; AVX512BW-SLOW-NEXT: vmovdqu8 %zmm13, %zmm6 {%k2}
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <3,3,3,3,u,4,4,4>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [3,3,3,3,u,4,4,4]
; AVX512BW-SLOW-NEXT: vpermd %ymm16, %ymm13, %ymm17
; AVX512BW-SLOW-NEXT: vpshufd {{.*#+}} xmm18 = mem[1,1,2,2]
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm18 = ymm18[0,1,1,1]
@@ -4434,27 +4434,27 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpor %ymm0, %ymm3, %ymm0
; AVX512BW-FAST-NEXT: vmovdqa (%rcx), %xmm6
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rcx), %xmm18
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = <128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = [128,6,128,8,u,128,7,128,9,128,11,u,128,10,128,12]
; AVX512BW-FAST-NEXT: vpshufb %xmm8, %xmm18, %xmm3
; AVX512BW-FAST-NEXT: vmovdqa (%rdx), %xmm7
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rdx), %xmm20
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [6,128,8,128,u,7,128,9,128,11,128,u,10,128,12,128]
; AVX512BW-FAST-NEXT: vpshufb %xmm10, %xmm20, %xmm4
; AVX512BW-FAST-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,1]
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm3, %zmm0
; AVX512BW-FAST-NEXT: vmovdqa (%rdi), %xmm9
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rdi), %xmm17
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = <8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm12 = [8,128,u,7,128,9,128,u,128,u,10,128,12,128,u,11]
; AVX512BW-FAST-NEXT: vpshufb %xmm12, %xmm17, %xmm3
; AVX512BW-FAST-NEXT: vmovdqa (%rsi), %xmm11
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rsi), %xmm19
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [128,8,u,128,7,128,9,u,11,u,128,10,128,12,u,128]
; AVX512BW-FAST-NEXT: vpshufb %xmm14, %xmm19, %xmm4
; AVX512BW-FAST-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm3[0,0,1,1]
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rdi), %ymm16
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <3,3,3,u,4,4,4,4>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [3,3,3,u,4,4,4,4]
; AVX512BW-FAST-NEXT: vpermd %ymm16, %ymm3, %ymm22
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rsi), %ymm23
; AVX512BW-FAST-NEXT: vpbroadcastq {{.*#+}} ymm4 = [0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14,0,0,13,2,15,0,1,14]
@@ -4466,13 +4466,13 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: kmovq %rax, %k2
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm15, %zmm0 {%k2}
; AVX512BW-FAST-NEXT: vmovdqa64 32(%r8), %ymm24
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = <1,1,2,2,2,2,2,2,27,27,27,27,u,28,28,28>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [1,1,2,2,2,2,2,2,27,27,27,27,u,28,28,28]
; AVX512BW-FAST-NEXT: vpermi2d %zmm5, %zmm24, %zmm15
; AVX512BW-FAST-NEXT: movabsq $4760450083537948804, %rax # imm = 0x4210842108421084
; AVX512BW-FAST-NEXT: kmovq %rax, %k3
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm15, %zmm0 {%k3}
; AVX512BW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm22 = zmm23[0,1,2,3],mem[4,5,6,7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,19,128,21,128,128,20,128,22,128,24,128,128,23,128,25,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,27,128,128,26,128,28,128,128,128,128,29,128,31,128,128,30>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm15 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,19,128,21,128,128,20,128,22,128,24,128,128,23,128,25,128,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,27,128,128,26,128,28,128,128,128,128,29,128,31,128,128,30]
; AVX512BW-FAST-NEXT: vpshufb %zmm15, %zmm22, %zmm22
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm22 = zmm22[2,2,3,3,6,6,7,7]
; AVX512BW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm23 = zmm16[0,1,2,3],mem[4,5,6,7]
@@ -4481,11 +4481,11 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm23 = zmm23[2,2,3,3,6,6,7,7]
; AVX512BW-FAST-NEXT: vporq %zmm22, %zmm23, %zmm23
; AVX512BW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm22 = zmm21[0,1,2,3],mem[4,5,6,7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,18,19,128,21,128,21,20,128,22,128,24,128,22,23,128,25,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,25,128,27,128,128,26,128,28,128,30,128,128,29,128,31,128>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm21 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,18,19,128,21,128,21,20,128,22,128,24,128,22,23,128,25,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,25,128,27,128,128,26,128,28,128,30,128,128,29,128,31,128]
; AVX512BW-FAST-NEXT: vpshufb %zmm21, %zmm22, %zmm22
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm25 = zmm22[2,2,3,3,6,6,7,7]
; AVX512BW-FAST-NEXT: vshufi64x2 {{.*#+}} zmm13 = zmm13[0,1,2,3],mem[4,5,6,7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,19,128,21,128,128,20,128,22,128,24,128,128,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,27,128,128,26,128,28,128,30,128,128,29,128,31,128,128>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm22 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,19,128,21,128,128,20,128,22,128,24,128,128,23,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,27,128,128,26,128,28,128,30,128,128,29,128,31,128,128]
; AVX512BW-FAST-NEXT: vpshufb %zmm22, %zmm13, %zmm13
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm13 = zmm13[2,2,3,3,6,6,7,7]
; AVX512BW-FAST-NEXT: vporq %zmm25, %zmm13, %zmm13
@@ -4505,7 +4505,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm26 = ymm26[2,2,3,3]
; AVX512BW-FAST-NEXT: vporq %ymm25, %ymm26, %ymm25
; AVX512BW-FAST-NEXT: vpunpcklbw {{.*#+}} xmm18 = xmm18[0],xmm20[0],xmm18[1],xmm20[1],xmm18[2],xmm20[2],xmm18[3],xmm20[3],xmm18[4],xmm20[4],xmm18[5],xmm20[5],xmm18[6],xmm20[6],xmm18[7],xmm20[7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm26 = <2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm26 = [2,u,1,0,5,4,u,3,u,7,6,11,10,u,9,8]
; AVX512BW-FAST-NEXT: vpshufb %xmm26, %xmm18, %xmm18
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm18 = ymm18[0,0,1,1]
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm18, %zmm25, %zmm25
@@ -4517,7 +4517,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm28 = ymm28[2,2,3,3]
; AVX512BW-FAST-NEXT: vporq %ymm27, %ymm28, %ymm27
; AVX512BW-FAST-NEXT: vpunpcklbw {{.*#+}} xmm17 = xmm17[0],xmm19[0],xmm17[1],xmm19[1],xmm17[2],xmm19[2],xmm17[3],xmm19[3],xmm17[4],xmm19[4],xmm17[5],xmm19[5],xmm17[6],xmm19[6],xmm17[7],xmm19[7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm19 = <0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm19 = [0,1,4,5,u,2,3,6,7,10,11,u,8,9,12,13]
; AVX512BW-FAST-NEXT: vpshufb %xmm19, %xmm17, %xmm17
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm17 = ymm17[0,0,1,1]
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm17, %zmm27, %zmm17
@@ -4568,7 +4568,7 @@ define void @store_i8_stride5_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpshufb %ymm4, %ymm18, %ymm3 {%k1}
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm2, %zmm3, %zmm2
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm1, %zmm2 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = <3,3,3,3,u,4,4,4,12,14,13,13,13,13,12,14>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm1 = [3,3,3,3,u,4,4,4,12,14,13,13,13,13,12,14]
; AVX512BW-FAST-NEXT: vpermd %zmm5, %zmm1, %zmm1
; AVX512BW-FAST-NEXT: movabsq $1190112520884487201, %rax # imm = 0x1084210842108421
; AVX512BW-FAST-NEXT: kmovq %rax, %k1
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll
index 5554ff3b6dca051..de0c3ccbf51d567 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-6.ll
@@ -1180,7 +1180,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm0, %xmm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u]
; AVX1-ONLY-NEXT: vpshufb %xmm2, %xmm12, %xmm1
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4],xmm0[5,6],xmm1[7]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm0, %xmm0
@@ -1238,7 +1238,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm10 = zero,zero,zero,zero,zero,xmm13[0],zero,zero,zero,zero,zero,xmm13[1],zero,zero,zero,zero
; AVX1-ONLY-NEXT: vpor %xmm10, %xmm11, %xmm11
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm15, %xmm10
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <2,u,128,128,128,128,3,u,128,128,128,128,4,u,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [2,u,128,128,128,128,3,u,128,128,128,128,4,u,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm12, %xmm12
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm10 = xmm12[0],xmm10[1,2],xmm12[3],xmm10[4,5],xmm12[6],xmm10[7]
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [0,128,2,3,4,5,6,128,8,9,10,11,12,128,14,15]
@@ -1339,7 +1339,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb %xmm9, %xmm13, %xmm9
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm9 = xmm9[8],xmm10[8],xmm9[9],xmm10[9],xmm9[10],xmm10[10],xmm9[11],xmm10[11],xmm9[12],xmm10[12],xmm9[13],xmm10[13],xmm9[14],xmm10[14],xmm9[15],xmm10[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm5, %ymm9, %ymm9
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm5
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm12 = xmm5[6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
@@ -1379,7 +1379,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm15 = xmm15[1,0,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm15 = xmm15[0,1,2,3,4,4,4,4]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm14, %ymm15, %ymm14
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} ymm15 = ymm1[0],ymm3[0],ymm1[1],ymm3[1],ymm1[2],ymm3[2],ymm1[3],ymm3[3],ymm1[4],ymm3[4],ymm1[5],ymm3[5],ymm1[6],ymm3[6],ymm1[7],ymm3[7],ymm1[16],ymm3[16],ymm1[17],ymm3[17],ymm1[18],ymm3[18],ymm1[19],ymm3[19],ymm1[20],ymm3[20],ymm1[21],ymm3[21],ymm1[22],ymm3[22],ymm1[23],ymm3[23]
; AVX2-SLOW-NEXT: vmovdqa %ymm3, %ymm12
@@ -1413,7 +1413,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm6 = xmm8[8],xmm6[8],xmm8[9],xmm6[9],xmm8[10],xmm6[10],xmm8[11],xmm6[11],xmm8[12],xmm6[12],xmm8[13],xmm6[13],xmm8[14],xmm6[14],xmm8[15],xmm6[15]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm0, %ymm6, %ymm0
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} ymm2 = ymm2[8],ymm12[8],ymm2[9],ymm12[9],ymm2[10],ymm12[10],ymm2[11],ymm12[11],ymm2[12],ymm12[12],ymm2[13],ymm12[13],ymm2[14],ymm12[14],ymm2[15],ymm12[15],ymm2[24],ymm12[24],ymm2[25],ymm12[25],ymm2[26],ymm12[26],ymm2[27],ymm12[27],ymm2[28],ymm12[28],ymm2[29],ymm12[29],ymm2[30],ymm12[30],ymm2[31],ymm12[31]
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} ymm1 = ymm3[8],ymm1[8],ymm3[9],ymm1[9],ymm3[10],ymm1[10],ymm3[11],ymm1[11],ymm3[12],ymm1[12],ymm3[13],ymm1[13],ymm3[14],ymm1[14],ymm3[15],ymm1[15],ymm3[24],ymm1[24],ymm3[25],ymm1[25],ymm3[26],ymm1[26],ymm3[27],ymm1[27],ymm3[28],ymm1[28],ymm3[29],ymm1[29],ymm3[30],ymm1[30],ymm3[31],ymm1[31]
@@ -1475,7 +1475,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm5, %xmm11
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm11 = xmm11[8],xmm12[8],xmm11[9],xmm12[9],xmm11[10],xmm12[10],xmm11[11],xmm12[11],xmm11[12],xmm12[12],xmm11[13],xmm12[13],xmm11[14],xmm12[14],xmm11[15],xmm12[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm7, %ymm11, %ymm11
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm13 = xmm7[6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
@@ -1517,7 +1517,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm15 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm15 = xmm15[10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm14, %ymm15, %ymm14
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} ymm15 = ymm8[8],ymm6[8],ymm8[9],ymm6[9],ymm8[10],ymm6[10],ymm8[11],ymm6[11],ymm8[12],ymm6[12],ymm8[13],ymm6[13],ymm8[14],ymm6[14],ymm8[15],ymm6[15],ymm8[24],ymm6[24],ymm8[25],ymm6[25],ymm8[26],ymm6[26],ymm8[27],ymm6[27],ymm8[28],ymm6[28],ymm8[29],ymm6[29],ymm8[30],ymm6[30],ymm8[31],ymm6[31]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm15 = ymm15[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,22,23,28,29,26,27,30,31,30,31,30,31,30,31]
@@ -1549,7 +1549,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3],xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[2,3,0,1,6,7,4,5,8,9,8,9,8,9,8,9]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm0, %ymm5, %ymm0
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} ymm2 = ymm8[0],ymm6[0],ymm8[1],ymm6[1],ymm8[2],ymm6[2],ymm8[3],ymm6[3],ymm8[4],ymm6[4],ymm8[5],ymm6[5],ymm8[6],ymm6[6],ymm8[7],ymm6[7],ymm8[16],ymm6[16],ymm8[17],ymm6[17],ymm8[18],ymm6[18],ymm8[19],ymm6[19],ymm8[20],ymm6[20],ymm8[21],ymm6[21],ymm8[22],ymm6[22],ymm8[23],ymm6[23]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} ymm1 = ymm3[0],ymm4[0],ymm3[1],ymm4[1],ymm3[2],ymm4[2],ymm3[3],ymm4[3],ymm3[4],ymm4[4],ymm3[5],ymm4[5],ymm3[6],ymm4[6],ymm3[7],ymm4[7],ymm3[16],ymm4[16],ymm3[17],ymm4[17],ymm3[18],ymm4[18],ymm3[19],ymm4[19],ymm3[20],ymm4[20],ymm3[21],ymm4[21],ymm3[22],ymm4[22],ymm3[23],ymm4[23]
@@ -1611,7 +1611,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm5, %xmm11
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm11 = xmm11[8],xmm12[8],xmm11[9],xmm12[9],xmm11[10],xmm12[10],xmm11[11],xmm12[11],xmm11[12],xmm12[12],xmm11[13],xmm12[13],xmm11[14],xmm12[14],xmm11[15],xmm12[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm7, %ymm11, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %xmm7
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm13 = xmm7[6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
@@ -1653,7 +1653,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm15 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm15 = xmm15[10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm14, %ymm15, %ymm14
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} ymm15 = ymm8[8],ymm6[8],ymm8[9],ymm6[9],ymm8[10],ymm6[10],ymm8[11],ymm6[11],ymm8[12],ymm6[12],ymm8[13],ymm6[13],ymm8[14],ymm6[14],ymm8[15],ymm6[15],ymm8[24],ymm6[24],ymm8[25],ymm6[25],ymm8[26],ymm6[26],ymm8[27],ymm6[27],ymm8[28],ymm6[28],ymm8[29],ymm6[29],ymm8[30],ymm6[30],ymm8[31],ymm6[31]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm15 = ymm15[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,22,23,28,29,26,27,30,31,30,31,30,31,30,31]
@@ -1685,7 +1685,7 @@ define void @store_i8_stride6_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3],xmm1[4],xmm2[4],xmm1[5],xmm2[5],xmm1[6],xmm2[6],xmm1[7],xmm2[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[2,3,0,1,6,7,4,5,8,9,8,9,8,9,8,9]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm0, %ymm5, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} ymm2 = ymm8[0],ymm6[0],ymm8[1],ymm6[1],ymm8[2],ymm6[2],ymm8[3],ymm6[3],ymm8[4],ymm6[4],ymm8[5],ymm6[5],ymm8[6],ymm6[6],ymm8[7],ymm6[7],ymm8[16],ymm6[16],ymm8[17],ymm6[17],ymm8[18],ymm6[18],ymm8[19],ymm6[19],ymm8[20],ymm6[20],ymm8[21],ymm6[21],ymm8[22],ymm6[22],ymm8[23],ymm6[23]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} ymm1 = ymm3[0],ymm4[0],ymm3[1],ymm4[1],ymm3[2],ymm4[2],ymm3[3],ymm4[3],ymm3[4],ymm4[4],ymm3[5],ymm4[5],ymm3[6],ymm4[6],ymm3[7],ymm4[7],ymm3[16],ymm4[16],ymm3[17],ymm4[17],ymm3[18],ymm4[18],ymm3[19],ymm4[19],ymm3[20],ymm4[20],ymm3[21],ymm4[21],ymm3[22],ymm4[22],ymm3[23],ymm4[23]
@@ -2765,7 +2765,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vandnps %ymm4, %ymm10, %ymm4
; AVX1-ONLY-NEXT: vorps %ymm4, %ymm1, %ymm1
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm1, %xmm4
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm2, %xmm6
; AVX1-ONLY-NEXT: vmovdqa %xmm7, %xmm15
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm4 = xmm4[0],xmm6[1],xmm4[2,3],xmm6[4],xmm4[5,6],xmm6[7]
@@ -2833,7 +2833,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm13, %xmm5, %xmm9
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm9, %xmm3
; AVX1-ONLY-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <128,128,128,128,0,u,128,128,128,128,1,u,128,128,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [128,128,128,128,0,u,128,128,128,128,1,u,128,128,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm2, %xmm2
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2],xmm0[3,4],xmm2[5],xmm0[6,7]
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,2,3,4,128,6,7,8,9,10,128,12,13,14,15]
@@ -2884,7 +2884,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovaps %ymm13, %ymm15
; AVX1-ONLY-NEXT: vorps %ymm0, %ymm2, %ymm0
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm0, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,8,u,128,128,128,128,9,u,128,128,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,8,u,128,128,128,128,9,u,128,128,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm1, %xmm3
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1],xmm3[2],xmm2[3,4],xmm3[5],xmm2[6,7]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm2, %xmm2
@@ -2893,7 +2893,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm6, %xmm8, %xmm3
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,5,u,128,128,128,128,6,u,128,128,128,128,7,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,5,u,128,128,128,128,6,u,128,128,128,128,7,u]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4],xmm0[5,6],xmm1[7]
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = [0,1,2,128,4,5,6,7,8,128,10,11,12,13,14,128]
@@ -2948,7 +2948,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vandnps %ymm1, %ymm15, %ymm1
; AVX1-ONLY-NEXT: vorps %ymm1, %ymm0, %ymm0
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [128,128,13,u,128,128,128,128,14,u,128,128,128,128,15,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm3, %xmm4
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm4[1],xmm1[2,3],xmm4[4],xmm1[5,6],xmm4[7]
; AVX1-ONLY-NEXT: vpshufb %xmm13, %xmm1, %xmm1
@@ -2956,7 +2956,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm2, %xmm4
; AVX1-ONLY-NEXT: vpor %xmm4, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = <10,u,128,128,128,128,11,u,128,128,128,128,12,u,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = [10,u,128,128,128,128,11,u,128,128,128,128,12,u,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm13, %xmm3, %xmm1
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3],xmm0[4,5],xmm1[6],xmm0[7]
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm13 = [0,128,2,3,4,5,6,128,8,9,10,11,12,128,14,15]
@@ -3012,7 +3012,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vandnps %ymm10, %ymm13, %ymm10
; AVX1-ONLY-NEXT: vorps %ymm10, %ymm11, %ymm11
; AVX1-ONLY-NEXT: vextractf128 $1, %ymm11, %xmm10
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <2,u,128,128,128,128,3,u,128,128,128,128,4,u,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [2,u,128,128,128,128,3,u,128,128,128,128,4,u,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm3, %xmm15
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm10 = xmm15[0],xmm10[1,2],xmm15[3],xmm10[4,5],xmm15[6],xmm10[7]
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm10, %xmm10
@@ -3020,7 +3020,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm2, %xmm15
; AVX1-ONLY-NEXT: vpor %xmm15, %xmm10, %xmm4
; AVX1-ONLY-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <128,128,128,128,0,u,128,128,128,128,1,u,128,128,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [128,128,128,128,0,u,128,128,128,128,1,u,128,128,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vpblendw {{.*#+}} xmm3 = xmm11[0,1],xmm3[2],xmm11[3,4],xmm3[5],xmm11[6,7]
; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [0,1,2,3,4,128,6,7,8,9,10,128,12,13,14,15]
@@ -3158,7 +3158,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa %xmm10, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm4[8],xmm2[8],xmm4[9],xmm2[9],xmm4[10],xmm2[10],xmm4[11],xmm2[11],xmm4[12],xmm2[12],xmm4[13],xmm2[13],xmm4[14],xmm2[14],xmm4[15],xmm2[15]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm2[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm4, %ymm1
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm5, %xmm4
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm9, %xmm5
@@ -3204,7 +3204,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm5, %ymm0
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm6
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm6, %xmm5
; AVX2-SLOW-NEXT: vmovdqa %xmm6, %xmm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
@@ -3229,7 +3229,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm0
; AVX2-SLOW-NEXT: vmovdqa (%r9), %xmm5
; AVX2-SLOW-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm5, %xmm5
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255]
@@ -3264,7 +3264,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm0[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm11
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm1 = xmm13[0],xmm14[0],xmm13[1],xmm14[1],xmm13[2],xmm14[2],xmm13[3],xmm14[3],xmm13[4],xmm14[4],xmm13[5],xmm14[5],xmm13[6],xmm14[6],xmm13[7],xmm14[7]
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,3,2,1,4,5,6,7]
@@ -3303,7 +3303,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} ymm3 = ymm3[0,1,2,3,4,4,4,4,8,9,10,11,12,12,12,12]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm2, %ymm3, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u]
; AVX2-SLOW-NEXT: vmovdqa %xmm7, %xmm14
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm7, %xmm3
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,0,1]
@@ -3323,7 +3323,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm8, %ymm6
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm0, %ymm6, %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4]
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm11 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm11, %xmm6
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,0,1]
@@ -3358,7 +3358,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm1[0,0,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm2, %ymm0, %ymm6
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX2-SLOW-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm0, %xmm0 # 16-byte Folded Reload
@@ -3394,7 +3394,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm0, %ymm2, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u]
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm7, %xmm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,0,1]
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0]
@@ -3411,7 +3411,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb %ymm4, %ymm8, %ymm4
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,2,2,3]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm0, %ymm4, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15]
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm5, %xmm5
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
@@ -3483,7 +3483,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa %xmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm4[8],xmm2[8],xmm4[9],xmm2[9],xmm4[10],xmm2[10],xmm4[11],xmm2[11],xmm4[12],xmm2[12],xmm4[13],xmm2[13],xmm4[14],xmm2[14],xmm4[15],xmm2[15]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm2[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm4, %ymm1
; AVX2-FAST-NEXT: vpshufb %xmm0, %xmm5, %xmm4
; AVX2-FAST-NEXT: vmovdqa %xmm5, %xmm14
@@ -3533,7 +3533,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm5, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%r8), %xmm5
; AVX2-FAST-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
@@ -3557,7 +3557,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm6
; AVX2-FAST-NEXT: vmovdqa (%r9), %xmm0
; AVX2-FAST-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm0, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255]
@@ -3591,7 +3591,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm5
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm13[8],xmm12[8],xmm13[9],xmm12[9],xmm13[10],xmm12[10],xmm13[11],xmm12[11],xmm13[12],xmm12[12],xmm13[13],xmm12[13],xmm13[14],xmm12[14],xmm13[15],xmm12[15]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
@@ -3627,7 +3627,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u]
; AVX2-FAST-NEXT: vmovdqa %xmm7, %xmm15
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm7, %xmm4
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,0,1]
@@ -3647,7 +3647,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm2, %ymm11, %ymm2
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm2, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15]
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,0,1]
@@ -3683,7 +3683,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,0,1,6,7,4,5,8,9,8,9,8,9,8,9]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm4, %ymm5, %ymm4
; AVX2-FAST-NEXT: vmovdqa (%rsp), %xmm2 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm2, %xmm5 # 16-byte Folded Reload
@@ -3720,7 +3720,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm7, %xmm2
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,0,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
@@ -3736,7 +3736,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm11, %ymm3
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,2,3]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm8, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255]
@@ -3808,7 +3808,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm13, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm4[8],xmm2[8],xmm4[9],xmm2[9],xmm4[10],xmm2[10],xmm4[11],xmm2[11],xmm4[12],xmm2[12],xmm4[13],xmm2[13],xmm4[14],xmm2[14],xmm4[15],xmm2[15]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm2[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255,u,u,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm1, %ymm4, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm5, %xmm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm5, %xmm14
@@ -3858,7 +3858,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm5, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %xmm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255]
@@ -3882,7 +3882,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %xmm0
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm0, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255]
@@ -3916,7 +3916,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [10,11,8,9,6,7,12,13,14,15,14,15,14,15,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm13[8],xmm12[8],xmm13[9],xmm12[9],xmm13[10],xmm12[10],xmm13[11],xmm12[11],xmm13[12],xmm12[12],xmm13[13],xmm12[13],xmm13[14],xmm12[14],xmm13[15],xmm12[15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm0, %xmm0
@@ -3952,7 +3952,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm4, %ymm1, %ymm0, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm7, %xmm15
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm7, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,0,0,1]
@@ -3972,7 +3972,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm2, %ymm11, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm0, %ymm2, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,0,1]
@@ -4008,7 +4008,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [2,3,0,1,6,7,4,5,8,9,8,9,8,9,8,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255,0,0,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsp), %xmm2 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm2, %xmm5 # 16-byte Folded Reload
@@ -4045,7 +4045,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm1, %ymm2, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm7, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255,255,255,0,0,255,255]
@@ -4061,7 +4061,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm3, %ymm11, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm3 = ymm3[2,2,2,3]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm0, %ymm3, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm8, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,0,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255,255,255,255,0,255,255]
@@ -4128,10 +4128,10 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,5,6,5]
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9]
; AVX512F-SLOW-NEXT: vmovdqa %xmm7, %xmm3
; AVX512F-SLOW-NEXT: vpshufb %xmm0, %xmm7, %xmm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4]
; AVX512F-SLOW-NEXT: vpshufb %xmm1, %xmm7, %xmm7
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm3, %xmm26
; AVX512F-SLOW-NEXT: vinserti32x4 $2, %xmm2, %zmm7, %zmm2
@@ -4228,7 +4228,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm5, %ymm21
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} ymm5 = ymm12[0],ymm6[0],ymm12[1],ymm6[1],ymm12[2],ymm6[2],ymm12[3],ymm6[3],ymm12[4],ymm6[4],ymm12[5],ymm6[5],ymm12[6],ymm6[6],ymm12[7],ymm6[7],ymm12[16],ymm6[16],ymm12[17],ymm6[17],ymm12[18],ymm6[18],ymm12[19],ymm6[19],ymm12[20],ymm6[20],ymm12[21],ymm6[21],ymm12[22],ymm6[22],ymm12[23],ymm6[23]
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm5, %ymm20
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15]
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm31, %xmm5
; AVX512F-SLOW-NEXT: vpshufb %xmm6, %xmm5, %xmm13
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm26, %xmm5
@@ -4252,14 +4252,14 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm10[0],xmm2[0],xmm10[1],xmm2[1],xmm10[2],xmm2[2],xmm10[3],xmm2[3],xmm10[4],xmm2[4],xmm10[5],xmm2[5],xmm10[6],xmm2[6],xmm10[7],xmm2[7]
; AVX512F-SLOW-NEXT: vmovdqa 32(%r8), %xmm3
; AVX512F-SLOW-NEXT: vprold $16, %xmm2, %xmm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u]
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm26
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm1, %xmm6
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm3, %xmm15
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
; AVX512F-SLOW-NEXT: vpshufb %xmm2, %xmm3, %xmm10
; AVX512F-SLOW-NEXT: vmovdqa64 %ymm10, %ymm17
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u]
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm3, %xmm10
; AVX512F-SLOW-NEXT: vpshufb %xmm2, %xmm1, %xmm14
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm1, %xmm0
@@ -4471,9 +4471,9 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm1, %zmm2, %zmm1
; AVX512F-FAST-NEXT: vmovdqu64 %zmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 32(%r9), %xmm4
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [u,6,u,5,u,8,u,7,u,9,u,9,u,9,u,9]
; AVX512F-FAST-NEXT: vpshufb %xmm13, %xmm4, %xmm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [u,2,u,1,u,0,u,3,u,4,u,4,u,4,u,4]
; AVX512F-FAST-NEXT: vpshufb %xmm1, %xmm4, %xmm3
; AVX512F-FAST-NEXT: vmovdqa64 %xmm4, %xmm19
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm2, %zmm3, %zmm26
@@ -4536,7 +4536,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm10 = [0,2,0,1,0,0,0,3,0,0,0,0,0,4,0,0,0,2,0,1,0,0,0,3,0,0,0,0,0,4,0,0]
; AVX512F-FAST-NEXT: # ymm10 = mem[0,1,0,1]
; AVX512F-FAST-NEXT: vpshufb %ymm10, %ymm9, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [u,10,u,13,u,12,u,11,u,14,u,13,u,14,u,15]
; AVX512F-FAST-NEXT: vpshufb %xmm5, %xmm15, %xmm3
; AVX512F-FAST-NEXT: vmovdqa64 %xmm5, %xmm17
; AVX512F-FAST-NEXT: vpermt2q %zmm0, %zmm7, %zmm3
@@ -4561,7 +4561,7 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm6 = xmm6[8],xmm8[8],xmm6[9],xmm8[9],xmm6[10],xmm8[10],xmm6[11],xmm8[11],xmm6[12],xmm8[12],xmm6[13],xmm8[13],xmm6[14],xmm8[14],xmm6[15],xmm8[15]
; AVX512F-FAST-NEXT: vmovdqa 32(%r8), %xmm15
; AVX512F-FAST-NEXT: vpshufb %xmm12, %xmm6, %xmm8
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [6,u,5,u,8,u,7,u,9,u,9,u,9,u,9,u]
; AVX512F-FAST-NEXT: vpermt2q %zmm4, %zmm7, %zmm8
; AVX512F-FAST-NEXT: vpshufb %xmm6, %xmm15, %xmm4
; AVX512F-FAST-NEXT: vpternlogq $226, %zmm2, %zmm16, %zmm8
@@ -4569,11 +4569,11 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX512F-FAST-NEXT: vpshufb %ymm2, %ymm5, %ymm12
; AVX512F-FAST-NEXT: vpshufb %ymm2, %ymm13, %ymm9
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [2,u,1,u,0,u,3,u,4,u,4,u,4,u,4,u]
; AVX512F-FAST-NEXT: vpshufb %xmm6, %xmm14, %xmm6
; AVX512F-FAST-NEXT: vpshufb %xmm2, %xmm15, %xmm5
; AVX512F-FAST-NEXT: vpshufb %xmm2, %xmm14, %xmm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [10,u,13,u,12,u,11,u,14,u,13,u,14,u,15,u]
; AVX512F-FAST-NEXT: vpshufb %xmm13, %xmm14, %xmm14
; AVX512F-FAST-NEXT: vpshufb %xmm13, %xmm15, %xmm13
; AVX512F-FAST-NEXT: vmovdqa64 %ymm31, %ymm15
@@ -4972,14 +4972,14 @@ define void @store_i8_stride6_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm10, %zmm16, %zmm10
; AVX512BW-FAST-NEXT: vmovdqu16 %zmm12, %zmm10 {%k2}
; AVX512BW-FAST-NEXT: vmovdqa 32(%r8), %xmm12
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm27 = <8,u,9,u,u,u,u,u,u,u,5,u,6,u,7,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm27 = [8,u,9,u,u,u,u,u,u,u,5,u,6,u,7,u]
; AVX512BW-FAST-NEXT: vpshufb %xmm27, %xmm12, %xmm16
; AVX512BW-FAST-NEXT: vpmovzxbw {{.*#+}} xmm28 = xmm12[0],zero,xmm12[1],zero,xmm12[2],zero,xmm12[3],zero,xmm12[4],zero,xmm12[5],zero,xmm12[6],zero,xmm12[7],zero
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm16, %zmm28, %zmm16
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm28 = [2,1,0,3,2,1,0,3,2,1,0,3,4,4,4,4,22,21,16,23,22,21,16,23,22,21,16,23,17,17,17,17]
; AVX512BW-FAST-NEXT: vpermw %zmm16, %zmm28, %zmm10 {%k1}
; AVX512BW-FAST-NEXT: vmovdqa64 32(%r9), %xmm16
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm29 = <u,8,u,9,u,10,u,11,u,4,u,5,u,6,u,7>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm29 = [u,8,u,9,u,10,u,11,u,4,u,5,u,6,u,7]
; AVX512BW-FAST-NEXT: vpshufb %xmm29, %xmm16, %xmm30
; AVX512BW-FAST-NEXT: vpunpcklbw {{.*#+}} xmm31 = xmm16[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; AVX512BW-FAST-NEXT: vinserti32x4 $2, %xmm30, %zmm31, %zmm30
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-7.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-7.ll
index b7baf7b40d714f7..482da013d741be4 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-7.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-7.ll
@@ -525,7 +525,7 @@ define void @store_i8_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm5 = zero,zero,ymm0[4],zero,zero,zero,zero,zero,zero,ymm0[5],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm0[23,31],zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm6 = ymm1[4,12],zero,zero,zero,zero,zero,ymm1[5,13],zero,zero,zero,zero,zero,ymm1[6,14,22],zero,zero,zero,zero,zero,zero,ymm1[23],zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm5, %ymm6, %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,u,u,u,u,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm3, %ymm5, %ymm3
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[0,8],zero,zero,zero,zero,zero,ymm2[1,9],zero,zero,zero,zero,zero,ymm2[2,10,18,26],zero,zero,zero,zero,zero,ymm2[19,27],zero,zero,zero,zero,zero,ymm2[20,28]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm4 = zero,zero,ymm4[0,8],zero,zero,zero,zero,zero,ymm4[1,9],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[19,27],zero,zero,zero,zero,zero,ymm4[20,28],zero,zero
@@ -563,7 +563,7 @@ define void @store_i8_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX2-FAST-NEXT: # ymm1 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermd %ymm2, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = zero,zero,zero,ymm1[1,5,9,13],zero,zero,zero,ymm1[2,6,10,14],zero,zero,zero,ymm1[19,23,27,31],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,3,5,u,5,1,3,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,3,5,u,5,1,3,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[0,4,8],zero,zero,zero,zero,ymm3[1,5,9],zero,zero,zero,zero,ymm3[2,6,18],zero,zero,zero,zero,ymm3[23,27,19],zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT: vpor %ymm3, %ymm1, %ymm1
@@ -609,7 +609,7 @@ define void @store_i8_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm5 = zero,zero,ymm0[4],zero,zero,zero,zero,zero,zero,ymm0[5],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm0[23,31],zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm6 = ymm1[4,12],zero,zero,zero,zero,zero,ymm1[5,13],zero,zero,zero,zero,zero,ymm1[6,14,22],zero,zero,zero,zero,zero,zero,ymm1[23],zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm5, %ymm6, %ymm5
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,u,u,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm3, %ymm5, %ymm3
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[0,8],zero,zero,zero,zero,zero,ymm2[1,9],zero,zero,zero,zero,zero,ymm2[2,10,18,26],zero,zero,zero,zero,zero,ymm2[19,27],zero,zero,zero,zero,zero,ymm2[20,28]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm4 = zero,zero,ymm4[0,8],zero,zero,zero,zero,zero,ymm4[1,9],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[19,27],zero,zero,zero,zero,zero,ymm4[20,28],zero,zero
@@ -682,13 +682,13 @@ define void @store_i8_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2
; AVX512F-FAST-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u]
; AVX512F-FAST-NEXT: vpermi2q %ymm3, %ymm0, %ymm1
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm0 = [0,2,4,6,0,2,4,6]
; AVX512F-FAST-NEXT: # ymm0 = mem[0,1,0,1]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm0, %ymm0
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,zero,ymm0[0,4,8],zero,zero,zero,zero,ymm0[1,5,9],zero,zero,zero,zero,ymm0[18,22,26],zero,zero,zero,zero,ymm0[19,23,27],zero,zero,zero,zero
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,3,5,u,5,1,3,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,3,5,u,5,1,3,u]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm1
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,4,8],zero,zero,zero,zero,ymm1[1,5,9],zero,zero,zero,zero,ymm1[2,6,18],zero,zero,zero,zero,ymm1[23,27,19],zero,zero,zero,zero,zero,zero,zero,zero
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
@@ -761,13 +761,13 @@ define void @store_i8_stride7_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX512BW-FAST-NEXT: vinserti128 $1, %xmm3, %ymm2, %ymm2
; AVX512BW-FAST-NEXT: vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX512BW-FAST-NEXT: vinserti128 $1, %xmm1, %ymm0, %ymm0
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u]
; AVX512BW-FAST-NEXT: vpermi2q %ymm3, %ymm0, %ymm1
; AVX512BW-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm0 = [1,3,5,7,1,3,5,7]
; AVX512BW-FAST-NEXT: # ymm0 = mem[0,1,0,1]
; AVX512BW-FAST-NEXT: vpermd %ymm2, %ymm0, %ymm0
; AVX512BW-FAST-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,ymm0[1,5,9,13],zero,zero,zero,ymm0[2,6,10,14],zero,zero,zero,ymm0[19,23,27,31],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <1,3,5,u,5,1,3,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [1,3,5,u,5,1,3,u]
; AVX512BW-FAST-NEXT: vpermd %ymm1, %ymm3, %ymm3
; AVX512BW-FAST-NEXT: vpshufb {{.*#+}} ymm3 = ymm3[0,4,8],zero,zero,zero,zero,ymm3[1,5,9],zero,zero,zero,zero,ymm3[2,6,18],zero,zero,zero,zero,ymm3[23,27,19],zero,zero,zero,zero,zero,zero,zero,zero
; AVX512BW-FAST-NEXT: vpor %ymm0, %ymm3, %ymm0
@@ -1272,7 +1272,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm8[2,3,0,1]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm12 = zero,zero,zero,zero,zero,zero,ymm12[5],zero,zero,zero,zero,zero,zero,ymm12[6],zero,zero,zero,zero,zero,ymm12[23],zero,zero,zero,zero,zero,zero,ymm12[24],zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm12, %ymm11, %ymm11
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm11 = ymm10[4],zero,zero,zero,zero,zero,zero,ymm10[5],zero,zero,zero,zero,zero,zero,ymm10[6],zero,zero,zero,zero,zero,zero,zero,ymm10[23],zero,zero,zero,zero,zero,zero,ymm10[24],zero,zero
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm10[2,3,0,1]
@@ -1280,7 +1280,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpor %ymm12, %ymm11, %ymm11
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm12 = xmm2[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,0,1,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
@@ -1289,7 +1289,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm12 = xmm2[1,1,0,0,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm12[0,1,2,0]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,0,1,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm8[0,2,0,2]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm12 = zero,zero,ymm12[0,8],zero,zero,zero,zero,zero,ymm12[1,9],zero,zero,zero,zero,zero,ymm12[18,26],zero,zero,zero,zero,zero,ymm12[19,27],zero,zero,zero,zero,zero,ymm12[20,28]
@@ -1303,7 +1303,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm12 = xmm2[0,1,2,3,4,5,5,6]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm12 = xmm12[2,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm13, %ymm10, %ymm12, %ymm10
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[3,1,1,3]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm7 = ymm7[1],zero,zero,zero,zero,zero,ymm7[10,2],zero,zero,zero,zero,zero,ymm7[11,3],zero,zero,zero,zero,zero,ymm7[20,28],zero,zero,zero,zero,zero,ymm7[21,29],zero,zero,zero
@@ -1321,7 +1321,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[10,u,u,u,u,u,13,12,u,u,u,u,u,15,14,u]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,6,7,7,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[2,1,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %xmm3, %xmm1, %xmm2, %xmm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0]
; AVX2-SLOW-NEXT: vpblendvb %xmm2, %xmm0, %xmm1, %xmm0
@@ -1351,7 +1351,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermd %ymm7, %ymm11, %ymm7
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm10[0,2,0,2]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm11 = ymm11[u,u,u,u,0,8,u,u,u,u,u,1,9,u,u,u,u,u,18,26,u,u,u,u,u,19,27,u,u,u,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm11, %ymm7, %ymm7
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm9[0,2,0,2]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm11 = zero,zero,ymm11[0,8],zero,zero,zero,zero,zero,ymm11[1,9],zero,zero,zero,zero,zero,ymm11[18,26],zero,zero,zero,zero,zero,ymm11[19,27],zero,zero,zero,zero,zero,ymm11[20,28]
@@ -1366,7 +1366,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: # ymm12 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermd %ymm10, %ymm12, %ymm13
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm13[0,4,u,u,u,u,u,1,5,u,u,u,u,u,2,6,u,u,u,u,u,19,23,u,u,u,u,u,24,28,u,u]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm13, %ymm11, %ymm11
; AVX2-FAST-NEXT: vpermd %ymm9, %ymm12, %ymm13
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = zero,zero,zero,zero,zero,ymm13[1,5],zero,zero,zero,zero,zero,ymm13[2,6],zero,zero,zero,zero,zero,ymm13[19,23],zero,zero,zero,zero,zero,ymm13[24,28],zero,zero,zero,zero
@@ -1381,7 +1381,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermd %ymm12, %ymm13, %ymm12
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm10 = ymm10[1,3,1,3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,1,9,u,u,u,u,u,2,10,u,u,u,u,u,19,27,u,u,u,u,u,20,28,u,u,u,u,u,21]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm10, %ymm12, %ymm10
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[3,1,1,3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm8[1],zero,zero,zero,zero,zero,ymm8[10,2],zero,zero,zero,zero,zero,ymm8[11,3],zero,zero,zero,zero,zero,ymm8[20,28],zero,zero,zero,zero,zero,ymm8[21,29],zero,zero,zero
@@ -1429,7 +1429,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm7[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm11 = zero,zero,zero,zero,zero,zero,ymm11[5],zero,zero,zero,zero,zero,zero,ymm11[6],zero,zero,zero,zero,zero,ymm11[23],zero,zero,zero,zero,zero,zero,ymm11[24],zero,zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm11, %ymm9, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm8, %ymm9, %ymm9
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm8 = ymm10[4],zero,zero,zero,zero,zero,zero,ymm10[5],zero,zero,zero,zero,zero,zero,ymm10[6],zero,zero,zero,zero,zero,zero,zero,ymm10[23],zero,zero,zero,zero,zero,zero,ymm10[24],zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm10[2,3,0,1]
@@ -1438,7 +1438,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r10), %xmm8
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm12 = xmm8[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,0,1,0]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm9, %ymm11, %ymm9
@@ -1446,7 +1446,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,0,1,0]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm10[0,2,0,2]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm12 = ymm12[u,u,u,u,0,8,u,u,u,u,u,1,9,u,u,u,u,u,18,26,u,u,u,u,u,19,27,u,u,u,u,u]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm12, %ymm11, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm7[0,2,0,2]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm12 = zero,zero,ymm12[0,8],zero,zero,zero,zero,zero,ymm12[1,9],zero,zero,zero,zero,zero,ymm12[18,26],zero,zero,zero,zero,zero,ymm12[19,27],zero,zero,zero,zero,zero,ymm12[20,28]
@@ -1459,7 +1459,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[1,3,1,3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm10 = ymm10[u,u,u,1,9,u,u,u,u,u,2,10,u,u,u,u,u,19,27,u,u,u,u,u,20,28,u,u,u,u,u,21]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm10, %ymm12, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[3,1,1,3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm6 = ymm6[1],zero,zero,zero,zero,zero,ymm6[10,2],zero,zero,zero,zero,zero,ymm6[11,3],zero,zero,zero,zero,zero,ymm6[20,28],zero,zero,zero,zero,zero,ymm6[21,29],zero,zero,zero
@@ -1617,7 +1617,7 @@ define void @store_i8_stride7_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm7, %zmm6, %zmm6
; AVX512F-FAST-NEXT: vporq %zmm4, %zmm6, %zmm4
; AVX512F-FAST-NEXT: vpshuflw {{.*#+}} xmm6 = xmm0[1,1,0,0,4,5,6,7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,1,u,1,u,0,0,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,1,u,1,u,0,0,u]
; AVX512F-FAST-NEXT: vpermd %ymm6, %ymm7, %ymm6
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,1,0]
@@ -2537,24 +2537,24 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,2,3],zero,xmm1[u,u,u,u,4,5],zero,xmm1[u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm1, %xmm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,u,u,128,7,u,u,u,u,u,128,8,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,u,u,128,7,u,u,u,u,u,128,8,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm11, %xmm2, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,u,u,u,7,128,u,u,u,u,u,8,128,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,u,u,u,7,128,u,u,u,u,u,8,128,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm3, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = <128,u,u,u,u,5,6,128,u,u,u,u,12,13,128,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm8 = [128,u,u,u,u,5,6,128,u,u,u,u,12,13,128,u]
; AVX1-ONLY-NEXT: vpshufb %xmm8, %xmm1, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <6,u,u,u,u,128,128,7,u,u,u,u,128,128,8,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [6,u,u,u,u,128,128,7,u,u,u,u,128,128,8,u]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm14, %xmm3
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm7
; AVX1-ONLY-NEXT: vmovdqa 16(%rcx), %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,128,7,u,u,u,u,u,128,8,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,128,7,u,u,u,u,u,128,8,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm1, %xmm0
; AVX1-ONLY-NEXT: vmovdqa 16(%rdx), %xmm2
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,7,128,u,u,u,u,u,8,128,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,7,128,u,u,u,u,u,8,128,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm2, %xmm3
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm3, %xmm0
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
@@ -2562,7 +2562,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,2,3,u,u,u,u,u,4,5,u,u,u,u,u,6]
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm15
; AVX1-ONLY-NEXT: vmovdqa 16(%rsi), %xmm10
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,128,7,u,u,u,u,u,128,8,u,u,u,u,u,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,128,7,u,u,u,u,u,128,8,u,u,u,u,u,128]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm10, %xmm0
; AVX1-ONLY-NEXT: vmovdqa 16(%rdi), %xmm6
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm6[u,7],zero,xmm6[u,u,u,u,u,8],zero,xmm6[u,u,u,u,u,9]
@@ -2592,7 +2592,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa %xmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm8, %xmm2
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm0, %xmm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,4,u,u,u,u,128,128,5,u,u,u,u,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,4,u,u,u,u,128,128,5,u,u,u,u,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm8, %xmm2
; AVX1-ONLY-NEXT: vmovdqa %xmm5, %xmm8
; AVX1-ONLY-NEXT: vpunpcklbw {{.*#+}} xmm3 = xmm3[0],xmm7[0],xmm3[1],xmm7[1],xmm3[2],xmm7[2],xmm3[3],xmm7[3],xmm3[4],xmm7[4],xmm3[5],xmm7[5],xmm3[6],xmm7[6],xmm3[7],xmm7[7]
@@ -2617,7 +2617,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm11, %xmm2
; AVX1-ONLY-NEXT: vpunpcklbw {{.*#+}} xmm9 = xmm3[0],xmm5[0],xmm3[1],xmm5[1],xmm3[2],xmm5[2],xmm3[3],xmm5[3],xmm3[4],xmm5[4],xmm3[5],xmm5[5],xmm3[6],xmm5[6],xmm3[7],xmm5[7]
; AVX1-ONLY-NEXT: vmovdqa %xmm9, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,u,10,11,u,u,u,u,u,12,13,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,u,10,11,u,u,u,u,u,12,13,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm9, %xmm11
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm2, %ymm11, %ymm11
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm4 = [255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255]
@@ -2661,13 +2661,13 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm6[8],xmm10[8],xmm6[9],xmm10[9],xmm6[10],xmm10[10],xmm6[11],xmm10[11],xmm6[12],xmm10[12],xmm6[13],xmm10[13],xmm6[14],xmm10[14],xmm6[15],xmm10[15]
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm4 = xmm0[u,u,12,13,u,u,u,u,u,14,15,u,u,u,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,u,u,8,9,u,u,u,u,u,10,11,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,u,u,8,9,u,u,u,u,u,10,11,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm4, %ymm0, %ymm0
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm6 = xmm1[u,u,u,u,12,13,u,u,u,u,u,14,15,u,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u]
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm4, %xmm4
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm6, %ymm4, %ymm6
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm4 = [255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255]
@@ -2723,17 +2723,17 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vandnps %ymm2, %ymm5, %ymm2
; AVX1-ONLY-NEXT: vorps %ymm2, %ymm1, %ymm6
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm3[u,u,u,u,12,13,u,u,u,u,u,14,15,u,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,0,1,u,u,u,u,u,2,3,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,0,1,u,u,u,u,u,2,3,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm2, %xmm9, %xmm3
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm3, %ymm1, %ymm1
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm5 = xmm12[u,u,12,13,u,u,u,u,u,14,15,u,u,u,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <0,1,u,u,u,u,u,2,3,u,u,u,u,u,4,5>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,u,u,u,u,u,2,3,u,u,u,u,u,4,5]
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm11, %xmm7
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm7, %ymm5, %ymm5
; AVX1-ONLY-NEXT: vandnps %ymm1, %ymm4, %ymm1
; AVX1-ONLY-NEXT: vandps %ymm4, %ymm5, %ymm4
; AVX1-ONLY-NEXT: vorps %ymm1, %ymm4, %ymm4
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,0,1,u,u,u,u,u,2,3,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,0,1,u,u,u,u,u,2,3,u,u,u]
; AVX1-ONLY-NEXT: vmovdqa (%rsp), %xmm5 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm5, %xmm5
; AVX1-ONLY-NEXT: vpalignr {{.*#+}} xmm5 = xmm5[4,5,6,7,8,9,10,11,12,13,14,15],xmm14[0,1,2,3]
@@ -2741,7 +2741,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = xmm7[10],zero,xmm7[u,u,u,u,13,12],zero,xmm7[u,u,u,u,15,14],zero
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm9 = zero,xmm8[13,u,u,u,u],zero,zero,xmm8[14,u,u,u,u],zero,zero,xmm8[15]
; AVX1-ONLY-NEXT: vpor %xmm7, %xmm9, %xmm7
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,0,1,12,u,u,u,u,7,8,13,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,0,1,12,u,u,u,u,7,8,13,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm5, %xmm5
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm5, %ymm7, %ymm5
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm7 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255]
@@ -2816,14 +2816,14 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: # ymm10 = mem[0,1,0,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm10, %ymm9, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm7[27],zero,ymm7[27,28,29,30],zero,ymm7[28],zero,ymm7[26,27,30,31],zero,ymm7[29]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm9 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29],zero,ymm2[27],zero,zero,zero,zero,ymm2[30],zero,ymm2[28],zero,zero,zero,zero,ymm2[31],zero
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm8, %ymm9, %ymm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm0, %ymm8, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
@@ -2840,7 +2840,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm9 = xmm0[8],xmm14[8],xmm0[9],xmm14[9],xmm0[10],xmm14[10],xmm0[11],xmm14[11],xmm0[12],xmm14[12],xmm0[13],xmm14[13],xmm0[14],xmm14[14],xmm0[15],xmm14[15]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm8, %ymm9, %ymm8
; AVX2-SLOW-NEXT: vmovdqa (%r9), %xmm12
; AVX2-SLOW-NEXT: vmovdqa (%r8), %xmm13
@@ -2851,7 +2851,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm7 = xmm15[0,1,2,3,4,5,5,6]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[2,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm1, %ymm9, %ymm7, %ymm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm8, %ymm1, %ymm1
@@ -2864,7 +2864,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpor %xmm7, %xmm9, %xmm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm1, %ymm7, %ymm1
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm7 = zero,xmm12[4,u,u,u],zero,xmm12[7],zero,xmm12[5,u,u,u],zero,xmm12[8],zero,xmm12[6]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm9 = xmm13[4],zero,xmm13[u,u,u,7],zero,xmm13[5],zero,xmm13[u,u,u,8],zero,xmm13[6],zero
@@ -2872,7 +2872,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm9 = xmm15[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,0,1,0]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm7, %ymm9, %ymm7
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm9
@@ -2882,7 +2882,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm1 = xmm10[0],xmm11[0],xmm10[1],xmm11[1],xmm10[2],xmm11[2],xmm10[3],xmm11[3],xmm10[4],xmm11[4],xmm10[5],xmm11[5],xmm10[6],xmm11[6],xmm10[7],xmm11[7]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm1 = xmm13[0],xmm12[0],xmm13[1],xmm12[1],xmm13[2],xmm12[2],xmm13[3],xmm12[3],xmm13[4],xmm12[4],xmm13[5],xmm12[5],xmm13[6],xmm12[6],xmm13[7],xmm12[7]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
@@ -2890,7 +2890,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm7 = xmm15[1,1,0,0,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm7 = xmm7[0,1,2,0]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,0,1,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm10
@@ -2906,7 +2906,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm7 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,18],zero,zero,zero,zero,ymm3[21],zero,ymm3[19],zero,zero,zero,zero,ymm3[22],zero,ymm3[20],zero
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm1, %ymm7, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm2[20],zero,ymm2[18],zero,zero,zero,zero,ymm2[21],zero,ymm2[19],zero,zero,zero,zero,ymm2[22]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
@@ -2918,7 +2918,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} ymm7 = ymm11[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm7 = ymm7[0,1,1,3,4,5,5,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,2,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm0, %ymm1, %ymm0
@@ -2932,7 +2932,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm3[25],zero,ymm3[23],zero,zero,zero,zero,ymm3[26],zero,ymm3[24],zero,zero,zero,zero,ymm3[27]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm7, %ymm8, %ymm7
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm1
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm7 = ymm2[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm2[25],zero,ymm2[23],zero,zero,zero,zero,ymm2[26],zero,ymm2[24],zero,zero
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,3,2,3]
@@ -2943,7 +2943,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm11[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-SLOW-NEXT: vmovdqa %ymm11, %ymm12
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm8 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm8, %ymm1, %ymm7, %ymm1
@@ -2953,13 +2953,13 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm5 = zero,zero,zero,zero,zero,ymm5[14],zero,zero,zero,zero,zero,zero,ymm5[15],zero,zero,zero,zero,zero,zero,ymm5[16],zero,zero,zero,zero,zero,zero,ymm5[17],zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm3 = zero,zero,zero,zero,ymm3[14],zero,zero,zero,zero,zero,zero,ymm3[15],zero,zero,zero,zero,zero,zero,ymm3[16],zero,zero,zero,zero,zero,zero,ymm3[17],zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm5, %ymm3, %ymm3
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm4, %ymm3, %ymm3
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm4 = zero,ymm13[1,2,3,0,1,14],zero,ymm13[0,1,0,1,14,15],zero,ymm13[15,16,17,18,19,16],zero,ymm13[30,31,16,17,16,17],zero,ymm13[31,30,31]
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[13],zero,zero,zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,zero,zero,ymm2[17],zero,zero,zero
; AVX2-SLOW-NEXT: vpor %ymm4, %ymm2, %ymm2
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm4 = ymm12[12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm2, %ymm4, %ymm2
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm3, %ymm2, %ymm2
@@ -2994,7 +2994,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm5 = xmm15[8],xmm13[8],xmm15[9],xmm13[9],xmm15[10],xmm13[10],xmm15[11],xmm13[11],xmm15[12],xmm13[12],xmm15[13],xmm13[13],xmm15[14],xmm13[14],xmm15[15],xmm13[15]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-NEXT: vmovdqa (%rax), %xmm11
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm5 = xmm11[0,1,2,3,4,5,5,6]
@@ -3006,7 +3006,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm6 = xmm14[8],xmm12[8],xmm14[9],xmm12[9],xmm14[10],xmm12[10],xmm14[11],xmm12[11],xmm14[12],xmm12[12],xmm14[13],xmm12[13],xmm14[14],xmm12[14],xmm14[15],xmm12[15]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm6, %ymm5, %ymm5
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
@@ -3018,7 +3018,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm7 = xmm9[u,u,u,7],zero,xmm9[5],zero,xmm9[u,u,u,8],zero,xmm9[6],zero,xmm9[u,u]
; AVX2-FAST-NEXT: vpor %xmm6, %xmm7, %xmm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm6 = xmm11[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,1,0]
@@ -3026,7 +3026,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm8 = xmm14[4],zero,xmm14[u,u,u,7],zero,xmm14[5],zero,xmm14[u,u,u,8],zero,xmm14[6],zero
; AVX2-FAST-NEXT: vpor %xmm7, %xmm8, %xmm7
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm7, %ymm6, %ymm7
; AVX2-FAST-NEXT: vmovdqa (%r8), %ymm6
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255]
@@ -3039,7 +3039,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm9 = xmm9[0],xmm10[0],xmm9[1],xmm10[1],xmm9[2],xmm10[2],xmm9[3],xmm10[3],xmm9[4],xmm10[4],xmm9[5],xmm10[5],xmm9[6],xmm10[6],xmm9[7],xmm10[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm10, %ymm13, %ymm9, %ymm9
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} xmm10 = xmm11[1,1,0,0,4,5,6,7]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [0,1,0,1,2,0,0,1]
@@ -3047,7 +3047,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm14[0],xmm12[0],xmm14[1],xmm12[1],xmm14[2],xmm12[2],xmm14[3],xmm12[3],xmm14[4],xmm12[4],xmm14[5],xmm12[5],xmm14[6],xmm12[6],xmm14[7],xmm12[7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm11 = xmm11[u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm11, %ymm10, %ymm10
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm9, %ymm10, %ymm9
@@ -3061,7 +3061,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,ymm1[21],zero,ymm1[19],zero,zero,zero,zero,ymm1[22],zero,ymm1[20],zero,zero,zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm11, %ymm12, %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} ymm11 = ymm7[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [4,5,4,5,5,7,4,5]
@@ -3071,7 +3071,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,20],zero,ymm6[18],zero,zero,zero,zero,ymm6[21],zero,ymm6[19],zero,zero,zero,zero,ymm6[22],zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm12, %ymm13, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm12, %ymm11, %ymm11
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
@@ -3085,14 +3085,14 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,ymm1[30],zero,ymm1[28],zero,zero,zero,zero,ymm1[31],zero,ymm1[29],zero,zero,zero,zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm12, %ymm13, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm6[27],zero,ymm6[27,28,29,30],zero,ymm6[28],zero,ymm6[26,27,30,31],zero,ymm6[29]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29],zero,ymm8[27],zero,zero,zero,zero,ymm8[30],zero,ymm8[28],zero,zero,zero,zero,ymm8[31],zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm12, %ymm13, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u]
; AVX2-FAST-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm12 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
@@ -3108,7 +3108,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm14 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm0[25],zero,ymm0[23],zero,zero,zero,zero,ymm0[26],zero,ymm0[24],zero,zero,zero,zero,ymm0[27]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm13, %ymm14, %ymm13
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm12, %ymm13, %ymm12
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm13 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm8[25],zero,ymm8[23],zero,zero,zero,zero,ymm8[26],zero,ymm8[24],zero,zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
@@ -3117,7 +3117,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpor %ymm13, %ymm14, %ymm13
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm14 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm13, %ymm14, %ymm13
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm12, %ymm13, %ymm12
@@ -3127,13 +3127,13 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,zero,zero,ymm2[17],zero,zero,zero,zero,zero
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,zero,ymm0[14],zero,zero,zero,zero,zero,zero,ymm0[15],zero,zero,zero,zero,zero,zero,ymm0[16],zero,zero,zero,zero,zero,zero,ymm0[17],zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT: vpor %ymm2, %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = zero,ymm6[1,2,3,0,1,14],zero,ymm6[0,1,0,1,14,15],zero,ymm6[15,16,17,18,19,16],zero,ymm6[30,31,16,17,16,17],zero,ymm6[31,30,31]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = ymm8[13],zero,zero,zero,zero,zero,zero,ymm8[14],zero,zero,zero,zero,zero,zero,ymm8[15],zero,zero,zero,zero,zero,zero,ymm8[16],zero,zero,zero,zero,zero,zero,ymm8[17],zero,zero,zero
; AVX2-FAST-NEXT: vpor %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = ymm7[12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
@@ -3165,7 +3165,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm5 = xmm15[8],xmm13[8],xmm15[9],xmm13[9],xmm15[10],xmm13[10],xmm15[11],xmm13[11],xmm15[12],xmm13[12],xmm15[13],xmm13[13],xmm15[14],xmm13[14],xmm15[15],xmm13[15]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm5 = xmm5[2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rax), %xmm11
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm5 = xmm11[8,9,10,11,8,9,10,11,10,11,12,13,10,11,12,13]
@@ -3175,7 +3175,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm6 = xmm14[8],xmm12[8],xmm14[9],xmm12[9],xmm14[10],xmm12[10],xmm14[11],xmm12[11],xmm14[12],xmm12[12],xmm14[13],xmm12[13],xmm14[14],xmm12[14],xmm14[15],xmm12[15]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm6 = xmm6[u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm6, %ymm5, %ymm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm4
@@ -3187,7 +3187,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm7 = xmm9[u,u,u,7],zero,xmm9[5],zero,xmm9[u,u,u,8],zero,xmm9[6],zero,xmm9[u,u]
; AVX2-FAST-PERLANE-NEXT: vpor %xmm6, %xmm7, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm6 = xmm11[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,0,1,0]
@@ -3195,7 +3195,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm8 = xmm14[4],zero,xmm14[u,u,u,7],zero,xmm14[5],zero,xmm14[u,u,u,8],zero,xmm14[6],zero
; AVX2-FAST-PERLANE-NEXT: vpor %xmm7, %xmm8, %xmm7
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm7, %ymm6, %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255]
@@ -3208,14 +3208,14 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm9 = xmm9[0],xmm10[0],xmm9[1],xmm10[1],xmm9[2],xmm10[2],xmm9[3],xmm10[3],xmm9[4],xmm10[4],xmm9[5],xmm10[5],xmm9[6],xmm10[6],xmm9[7],xmm10[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm9 = xmm9[4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm9 = ymm9[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm10, %ymm13, %ymm9, %ymm9
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm10 = xmm11[2,3,2,3,0,1,0,1,8,9,10,11,2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm10 = ymm10[0,0,1,0]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm14[0],xmm12[0],xmm14[1],xmm12[1],xmm14[2],xmm12[2],xmm14[3],xmm12[3],xmm14[4],xmm12[4],xmm14[5],xmm12[5],xmm14[6],xmm12[6],xmm14[7],xmm12[7]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm11 = xmm11[u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm11, %ymm10, %ymm10
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm9, %ymm10, %ymm9
@@ -3229,14 +3229,14 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm12 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,ymm1[30],zero,ymm1[28],zero,zero,zero,zero,ymm1[31],zero,ymm1[29],zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm11, %ymm12, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm11 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm6[27],zero,ymm6[27,28,29,30],zero,ymm6[28],zero,ymm6[26,27,30,31],zero,ymm6[29]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm12 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29],zero,ymm8[27],zero,zero,zero,zero,ymm8[30],zero,ymm8[28],zero,zero,zero,zero,ymm8[31],zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm11, %ymm12, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u,255,255,255,255,0,0,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm10, %ymm11, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm11 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,3,2,3]
@@ -3252,7 +3252,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm13 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm0[25],zero,ymm0[23],zero,zero,zero,zero,ymm0[26],zero,ymm0[24],zero,zero,zero,zero,ymm0[27]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm12, %ymm13, %ymm12
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm12 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm8[25],zero,ymm8[23],zero,zero,zero,zero,ymm8[26],zero,ymm8[24],zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[2,3,2,3]
@@ -3261,7 +3261,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm12, %ymm13, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm13 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm12, %ymm13, %ymm12
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm13 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm13, %ymm11, %ymm12, %ymm11
@@ -3275,7 +3275,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm14 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,ymm1[21],zero,ymm1[19],zero,zero,zero,zero,ymm1[22],zero,ymm1[20],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm13, %ymm14, %ymm13
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm12, %ymm13, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm13 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm8[20],zero,ymm8[18],zero,zero,zero,zero,ymm8[21],zero,ymm8[19],zero,zero,zero,zero,ymm8[22]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm13 = ymm13[2,3,2,3]
@@ -3284,7 +3284,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm13, %ymm14, %ymm13
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm14 = ymm7[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,20,21,18,19,18,19,20,21,18,19,20,21,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm14 = ymm14[2,2,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm13, %ymm14, %ymm13
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm14, %ymm12, %ymm13, %ymm12
@@ -3294,13 +3294,13 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,ymm2[14],zero,zero,zero,zero,zero,zero,ymm2[15],zero,zero,zero,zero,zero,zero,ymm2[16],zero,zero,zero,zero,zero,zero,ymm2[17],zero,zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,zero,ymm0[14],zero,zero,zero,zero,zero,zero,ymm0[15],zero,zero,zero,zero,zero,zero,ymm0[16],zero,zero,zero,zero,zero,zero,ymm0[17],zero,zero,zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm2, %ymm0, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = zero,ymm6[1,2,3,0,1,14],zero,ymm6[0,1,0,1,14,15],zero,ymm6[15,16,17,18,19,16],zero,ymm6[30,31,16,17,16,17],zero,ymm6[31,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm8[13],zero,zero,zero,zero,zero,zero,ymm8[14],zero,zero,zero,zero,zero,zero,ymm8[15],zero,zero,zero,zero,zero,zero,ymm8[16],zero,zero,zero,zero,zero,zero,ymm8[17],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpor %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm7[12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm3 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
@@ -3328,7 +3328,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqa (%r10), %ymm4
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm7 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,u,u,u,u,26,27,24,25]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} ymm8 = ymm4[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,5,4,u,5,u,4,u,20,21,u,23,u,21,u,23>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,5,4,u,5,u,4,u,20,21,u,23,u,21,u,23]
; AVX512F-SLOW-NEXT: vpermi2d %zmm7, %zmm8, %zmm9
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm7 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm6[20],zero,ymm6[18],zero,zero,zero,zero,ymm6[21],zero,ymm6[19],zero,zero,zero,zero,ymm6[22]
; AVX512F-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm6[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm6[25],zero,ymm6[23],zero,zero,zero,zero,ymm6[26],zero,ymm6[24],zero,zero
@@ -3501,7 +3501,7 @@ define void @store_i8_stride7_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpternlogd $216, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm7, %zmm15
; AVX512F-FAST-NEXT: vmovdqa (%r10), %xmm10
; AVX512F-FAST-NEXT: vpshuflw {{.*#+}} xmm7 = xmm10[1,1,0,0,4,5,6,7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = <u,1,u,1,u,0,0,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm13 = [u,1,u,1,u,0,0,u]
; AVX512F-FAST-NEXT: vpermd %ymm7, %ymm13, %ymm7
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm13 = xmm10[4,5,4,5,4,5,8,9,6,7,6,7,6,7,6,7]
; AVX512F-FAST-NEXT: vpermq {{.*#+}} ymm13 = ymm13[0,0,1,0]
@@ -5397,38 +5397,38 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa %xmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm6[u,u,u],zero,zero,xmm6[9,u,u,u,u],zero,zero,xmm6[10,u,u,u]
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm0, %xmm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,128,7,u,u,u,u,u,128,8,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,128,7,u,u,u,u,u,128,8,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm8, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm3, %xmm8
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,u,u,7,128,u,u,u,u,u,8,128,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,u,u,7,128,u,u,u,u,u,8,128,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm10, %xmm3
; AVX1-ONLY-NEXT: vmovdqa %xmm4, %xmm10
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm3, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,u,u,u,u,5,6,128,u,u,u,u,12,13,128,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,u,u,u,u,5,6,128,u,u,u,u,12,13,128,u]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm1, %xmm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <6,u,u,u,u,128,128,7,u,u,u,u,128,128,8,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [6,u,u,u,u,128,128,7,u,u,u,u,128,128,8,u]
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm6, %xmm3
; AVX1-ONLY-NEXT: vmovdqa %xmm4, %xmm6
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm1, %xmm1
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,128,7,u,u,u,u,u,128,8,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,128,7,u,u,u,u,u,128,8,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm12, %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm3, %xmm12
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,7,128,u,u,u,u,u,8,128,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,7,128,u,u,u,u,u,8,128,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm13, %xmm3
; AVX1-ONLY-NEXT: vpor %xmm1, %xmm3, %xmm1
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,2,3,u,u,u,u,u,4,5,u,u,u,u,u,6]
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm2, %ymm1, %ymm1
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <u,128,7,u,u,u,u,u,128,8,u,u,u,u,u,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [u,128,7,u,u,u,u,u,128,8,u,u,u,u,u,128]
; AVX1-ONLY-NEXT: vmovdqa %xmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm7, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <u,7,128,u,u,u,u,u,8,128,u,u,u,u,u,9>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [u,7,128,u,u,u,u,u,8,128,u,u,u,u,u,9]
; AVX1-ONLY-NEXT: vmovdqa %xmm11, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm11, %xmm3
; AVX1-ONLY-NEXT: vmovdqa %xmm4, %xmm13
; AVX1-ONLY-NEXT: vpor %xmm2, %xmm3, %xmm2
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm7[8],xmm11[8],xmm7[9],xmm11[9],xmm7[10],xmm11[10],xmm7[11],xmm11[11],xmm7[12],xmm11[12],xmm7[13],xmm11[13],xmm7[14],xmm11[14],xmm7[15],xmm11[15]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <2,u,u,u,u,u,5,4,u,u,u,u,u,7,6,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [2,u,u,u,u,u,5,4,u,u,u,u,u,7,6,u]
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm3, %ymm2, %ymm2
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = [255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0]
@@ -5476,7 +5476,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpor %xmm6, %xmm7, %xmm6
; AVX1-ONLY-NEXT: vpunpcklbw {{.*#+}} xmm7 = xmm3[0],xmm1[0],xmm3[1],xmm1[1],xmm3[2],xmm1[2],xmm3[3],xmm1[3],xmm3[4],xmm1[4],xmm3[5],xmm1[5],xmm3[6],xmm1[6],xmm3[7],xmm1[7]
; AVX1-ONLY-NEXT: vmovdqa %xmm7, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = <u,u,u,10,11,u,u,u,u,u,12,13,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm12 = [u,u,u,10,11,u,u,u,u,u,12,13,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm12, %xmm7, %xmm7
; AVX1-ONLY-NEXT: vmovdqa %xmm12, %xmm9
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm6, %ymm7, %ymm6
@@ -5507,7 +5507,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm0[8],xmm2[8],xmm0[9],xmm2[9],xmm0[10],xmm2[10],xmm0[11],xmm2[11],xmm0[12],xmm2[12],xmm0[13],xmm2[13],xmm0[14],xmm2[14],xmm0[15],xmm2[15]
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm2[8],xmm0[8],xmm2[9],xmm0[9],xmm2[10],xmm0[10],xmm2[11],xmm0[11],xmm2[12],xmm0[12],xmm2[13],xmm0[13],xmm2[14],xmm0[14],xmm2[15],xmm0[15]
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u]
; AVX1-ONLY-NEXT: vpshufb %xmm0, %xmm3, %xmm0
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,2,3,u,u,u,u,u,4,5,u,u,u,u,u,6]
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm0, %ymm2, %ymm0
@@ -5525,11 +5525,11 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vmovdqa 48(%r9), %xmm1
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,4,u,u,u,u,128,128,5,u,u,u,u,128,128>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,4,u,u,u,u,128,128,5,u,u,u,u,128,128]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm12, %xmm0
; AVX1-ONLY-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3],xmm2[4],xmm1[4],xmm2[5],xmm1[5],xmm2[6],xmm1[6],xmm2[7],xmm1[7]
; AVX1-ONLY-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <8,9,128,u,u,u,u,10,11,128,u,u,u,u,12,13>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [8,9,128,u,u,u,u,10,11,128,u,u,u,u,12,13]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm2, %xmm1
; AVX1-ONLY-NEXT: vpor %xmm0, %xmm1, %xmm0
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm1 = xmm12[u,u],zero,zero,xmm12[2,u,u,u,u],zero,zero,xmm12[3,u,u,u,u]
@@ -5549,7 +5549,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa 48(%rdx), %xmm13
; AVX1-ONLY-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm13[0],xmm2[0],xmm13[1],xmm2[1],xmm13[2],xmm2[2],xmm13[3],xmm2[3],xmm13[4],xmm2[4],xmm13[5],xmm2[5],xmm13[6],xmm2[6],xmm13[7],xmm2[7]
; AVX1-ONLY-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,u,10,11,u,u,u,u,u,12,13,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,u,10,11,u,u,u,u,u,12,13,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm9, %xmm4, %xmm3
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm11 = xmm4[4,5,u,u,u,u,u,6,7,u,u,u,u,u,8,9]
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm3, %ymm11, %ymm3
@@ -5617,17 +5617,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm2[8],xmm14[8],xmm2[9],xmm14[9],xmm2[10],xmm14[10],xmm2[11],xmm14[11],xmm2[12],xmm14[12],xmm2[13],xmm14[13],xmm2[14],xmm14[14],xmm2[15],xmm14[15]
; AVX1-ONLY-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm14[8],xmm2[8],xmm14[9],xmm2[9],xmm14[10],xmm2[10],xmm14[11],xmm2[11],xmm14[12],xmm2[12],xmm14[13],xmm2[13],xmm14[14],xmm2[14],xmm14[15],xmm2[15]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <2,u,u,u,u,u,5,4,u,u,u,u,u,7,6,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [2,u,u,u,u,u,5,4,u,u,u,u,u,7,6,u]
; AVX1-ONLY-NEXT: vpshufb %xmm6, %xmm0, %xmm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = <u,u,u,u,8,9,u,u,u,u,u,10,11,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm14 = [u,u,u,u,8,9,u,u,u,u,u,10,11,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm4, %xmm4
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm4, %ymm0, %ymm0
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm5[8],xmm3[8],xmm5[9],xmm3[9],xmm5[10],xmm3[10],xmm5[11],xmm3[11],xmm5[12],xmm3[12],xmm5[13],xmm3[13],xmm5[14],xmm3[14],xmm5[15],xmm3[15]
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm3[8],xmm5[8],xmm3[9],xmm5[9],xmm3[10],xmm5[10],xmm3[11],xmm5[11],xmm3[12],xmm5[12],xmm3[13],xmm5[13],xmm3[14],xmm5[14],xmm3[15],xmm5[15]
; AVX1-ONLY-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [6,u,u,u,u,u,9,8,u,u,u,u,u,11,10,u]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm4, %xmm2
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = <u,2,3,u,u,u,u,u,4,5,u,u,u,u,u,6>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm5 = [u,2,3,u,u,u,u,u,4,5,u,u,u,u,u,6]
; AVX1-ONLY-NEXT: vpshufb %xmm5, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm2, %ymm3, %ymm2
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm3 = [0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255]
@@ -5663,7 +5663,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: # xmm1 = xmm1[0],mem[0],xmm1[1],mem[1],xmm1[2],mem[2],xmm1[3],mem[3],xmm1[4],mem[4],xmm1[5],mem[5],xmm1[6],mem[6],xmm1[7],mem[7]
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm4 = xmm1[u,u,u,u,u,10,11,u,u,u,u,u,12,13,u,u]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = <4,5,u,u,u,u,u,6,7,u,u,u,u,u,8,9>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm2 = [4,5,u,u,u,u,u,6,7,u,u,u,u,u,8,9]
; AVX1-ONLY-NEXT: vpshufb %xmm2, %xmm1, %xmm11
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm4, %ymm11, %ymm4
; AVX1-ONLY-NEXT: vandnps %ymm3, %ymm9, %ymm3
@@ -5718,12 +5718,12 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm3 = xmm10[8],xmm8[8],xmm10[9],xmm8[9],xmm10[10],xmm8[10],xmm10[11],xmm8[11],xmm10[12],xmm8[12],xmm10[13],xmm8[13],xmm10[14],xmm8[14],xmm10[15],xmm8[15]
; AVX1-ONLY-NEXT: vpshufb %xmm14, %xmm3, %xmm4
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = <u,u,12,13,u,u,u,u,u,14,15,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm10 = [u,u,12,13,u,u,u,u,u,14,15,u,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm10, %xmm3, %xmm3
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm3, %ymm4, %ymm3
; AVX1-ONLY-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm9[8],xmm13[8],xmm9[9],xmm13[9],xmm9[10],xmm13[10],xmm9[11],xmm13[11],xmm9[12],xmm13[12],xmm9[13],xmm13[13],xmm9[14],xmm13[14],xmm9[15],xmm13[15]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm4, %xmm4
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,12,13,u,u,u,u,u,14,15,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,12,13,u,u,u,u,u,14,15,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm0, %ymm4, %ymm0
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm1 = [255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255,0,0,255,255,255,255,255]
@@ -5750,14 +5750,14 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm3 = zero,xmm12[13],zero,zero,zero,zero,zero,zero,xmm12[14],zero,zero,zero,zero,zero,zero,xmm12[15]
; AVX1-ONLY-NEXT: vpor %xmm3, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = <u,u,0,1,u,u,u,u,u,2,3,u,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm15 = [u,u,0,1,u,u,u,u,u,2,3,u,u,u,u,u]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm0, %xmm0
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm7, %xmm1, %xmm3
; AVX1-ONLY-NEXT: vmovdqa %xmm7, %xmm9
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm0, %ymm3, %ymm0
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = <0,1,u,u,u,u,u,2,3,u,u,u,u,u,4,5>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm3 = [0,1,u,u,u,u,u,2,3,u,u,u,u,u,4,5]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm1, %xmm4
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -5773,11 +5773,11 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm14 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb {{.*#+}} xmm7 = zero,xmm14[13,u,u,u,u],zero,zero,xmm14[14,u,u,u,u],zero,zero,xmm14[15]
; AVX1-ONLY-NEXT: vpor %xmm7, %xmm4, %xmm7
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,u,0,1,u,u,u,u,u,2,3,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,u,0,1,u,u,u,u,u,2,3,u,u,u]
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm1, %xmm10
; AVX1-ONLY-NEXT: vpalignr {{.*#+}} xmm5 = xmm10[4,5,6,7,8,9,10,11,12,13,14,15],xmm12[0,1,2,3]
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,u,0,1,12,u,u,u,u,7,8,13,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,u,0,1,12,u,u,u,u,7,8,13,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm11, %xmm5, %xmm5
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm5, %ymm7, %ymm5
; AVX1-ONLY-NEXT: vmovaps {{.*#+}} ymm7 = [0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255]
@@ -5790,7 +5790,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vpshufb %xmm15, %xmm8, %xmm10
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm5, %ymm10, %ymm5
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,u,u,u,6,7,u,u,u,u,u,8,9,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,u,u,u,6,7,u,u,u,u,u,8,9,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm6, %xmm8, %xmm10
; AVX1-ONLY-NEXT: vpshufb %xmm3, %xmm8, %xmm12
; AVX1-ONLY-NEXT: vinsertf128 $1, %xmm10, %ymm12, %ymm10
@@ -5798,10 +5798,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX1-ONLY-NEXT: vandnps %ymm5, %ymm12, %ymm5
; AVX1-ONLY-NEXT: vandps %ymm12, %ymm10, %ymm10
; AVX1-ONLY-NEXT: vorps %ymm5, %ymm10, %ymm5
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,128,128,2,u,u,u,u,128,128,3,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,128,128,2,u,u,u,u,128,128,3,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm0, %xmm14, %xmm10
; AVX1-ONLY-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
-; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,4,5,128,u,u,u,u,6,7,128,u,u,u,u>
+; AVX1-ONLY-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,4,5,128,u,u,u,u,6,7,128,u,u,u,u]
; AVX1-ONLY-NEXT: vpshufb %xmm1, %xmm8, %xmm13
; AVX1-ONLY-NEXT: vpor %xmm10, %xmm13, %xmm10
; AVX1-ONLY-NEXT: vpshufb %xmm4, %xmm8, %xmm13
@@ -5926,15 +5926,15 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm6[25],zero,ymm6[23],zero,zero,zero,zero,ymm6[26],zero,ymm6[24],zero,zero,zero,zero,ymm6[27]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm1, %ymm2, %ymm1
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,24,25,26,27,24,25,30,31]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,25,24,23,u,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
@@ -5954,7 +5954,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vmovdqa (%rdx), %ymm1
; AVX2-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -5980,7 +5980,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm2, %ymm3, %ymm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
@@ -5990,16 +5990,16 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm2[0],xmm13[0],xmm2[1],xmm13[1],xmm2[2],xmm13[2],xmm2[3],xmm13[3],xmm2[4],xmm13[4],xmm2[5],xmm13[5],xmm2[6],xmm13[6],xmm2[7],xmm13[7]
; AVX2-SLOW-NEXT: vmovdqa %xmm2, %xmm14
; AVX2-SLOW-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-SLOW-NEXT: vmovdqa 32(%rcx), %xmm11
; AVX2-SLOW-NEXT: vmovdqa 32(%rdx), %xmm9
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm9[0],xmm11[0],xmm9[1],xmm11[1],xmm9[2],xmm11[2],xmm9[3],xmm11[3],xmm9[4],xmm11[4],xmm9[5],xmm11[5],xmm9[6],xmm11[6],xmm9[7],xmm11[7]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm0, %ymm2, %ymm10
; AVX2-SLOW-NEXT: vmovdqa (%rsi), %xmm0
; AVX2-SLOW-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -6021,7 +6021,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa 32(%r8), %xmm2
; AVX2-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm1 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3],xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
; AVX2-SLOW-NEXT: vmovdqa %xmm2, %xmm8
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-SLOW-NEXT: vmovdqa 32(%rax), %xmm2
@@ -6029,7 +6029,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshuflw {{.*#+}} xmm3 = xmm2[1,1,0,0,4,5,6,7]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm3 = xmm3[0,1,2,0]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,0,1,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm1, %ymm3, %ymm1
; AVX2-SLOW-NEXT: vmovdqa (%r9), %xmm3
; AVX2-SLOW-NEXT: vmovdqa %xmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -6049,19 +6049,19 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm4, %ymm2, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm13, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm14, %xmm12
; AVX2-SLOW-NEXT: vpor %xmm2, %xmm12, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm15, %xmm11, %xmm12
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX2-SLOW-NEXT: vpshufb %xmm0, %xmm9, %xmm14
; AVX2-SLOW-NEXT: vpor %xmm12, %xmm14, %xmm12
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm2, %ymm12, %ymm2
; AVX2-SLOW-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm2 # 16-byte Reload
@@ -6075,10 +6075,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm14, %ymm1, %ymm0, %ymm12
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm1 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm10 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm10, %xmm2
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX2-SLOW-NEXT: vmovdqa %xmm8, %xmm3
; AVX2-SLOW-NEXT: vpshufb %xmm4, %xmm8, %xmm14
; AVX2-SLOW-NEXT: vpor %xmm2, %xmm14, %xmm2
@@ -6087,7 +6087,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm5 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm14, %xmm5, %xmm15
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,1,0]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm0, %ymm2, %ymm15, %ymm2
; AVX2-SLOW-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
; AVX2-SLOW-NEXT: vpshufb %xmm1, %xmm8, %xmm1
@@ -6107,13 +6107,13 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm11[8],xmm9[8],xmm11[9],xmm9[9],xmm11[10],xmm9[10],xmm11[11],xmm9[11],xmm11[12],xmm9[12],xmm11[13],xmm9[13],xmm11[14],xmm9[14],xmm11[15],xmm9[15]
; AVX2-SLOW-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm13, %xmm1 # 16-byte Folded Reload
; AVX2-SLOW-NEXT: # xmm1 = xmm13[8],mem[8],xmm13[9],mem[9],xmm13[10],mem[10],xmm13[11],mem[11],xmm13[12],mem[12],xmm13[13],mem[13],xmm13[14],mem[14],xmm13[15],mem[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX2-SLOW-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-SLOW-NEXT: vpshufb %xmm9, %xmm1, %xmm1
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm4 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm0, %ymm1, %ymm0
; AVX2-SLOW-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm6, %xmm1 # 16-byte Folded Reload
; AVX2-SLOW-NEXT: # xmm1 = xmm6[8],mem[8],xmm6[9],mem[9],xmm6[10],mem[10],xmm6[11],mem[11],xmm6[12],mem[12],xmm6[13],mem[13],xmm6[14],mem[14],xmm6[15],mem[15]
@@ -6126,13 +6126,13 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
; AVX2-SLOW-NEXT: vpblendvb %ymm4, %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm3[8],xmm10[8],xmm3[9],xmm10[9],xmm3[10],xmm10[10],xmm3[11],xmm10[11],xmm3[12],xmm10[12],xmm3[13],xmm10[13],xmm3[14],xmm10[14],xmm3[15],xmm10[15]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
; AVX2-SLOW-NEXT: vpshufhw {{.*#+}} xmm4 = xmm5[0,1,2,3,4,5,5,6]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[2,2,3,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,0,1]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm5, %ymm2, %ymm4, %ymm2
; AVX2-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm15[8],xmm8[8],xmm15[9],xmm8[9],xmm15[10],xmm8[10],xmm15[11],xmm8[11],xmm15[12],xmm8[12],xmm15[13],xmm8[13],xmm15[14],xmm8[14],xmm15[15],xmm8[15]
; AVX2-SLOW-NEXT: vpshufb %xmm3, %xmm4, %xmm3
@@ -6160,7 +6160,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: # ymm5 = mem[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} ymm5 = ymm5[0,1,1,3,4,5,5,7]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,2,3,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm2, %ymm5, %ymm0
; AVX2-SLOW-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm14 # 32-byte Reload
@@ -6202,7 +6202,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm9 = ymm9[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm7, %ymm9, %ymm7
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm4 = ymm4[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm7, %ymm4, %ymm4
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm15 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm15, %ymm6
@@ -6231,7 +6231,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm6 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm12[25],zero,ymm12[23],zero,zero,zero,zero,ymm12[26],zero,ymm12[24],zero,zero,zero,zero,ymm12[27]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,3,2,3]
; AVX2-SLOW-NEXT: vpor %ymm5, %ymm6, %ymm5
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-SLOW-NEXT: vpblendvb %ymm3, %ymm4, %ymm5, %ymm4
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm5 = ymm14[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm14[25],zero,ymm14[23],zero,zero,zero,zero,ymm14[26],zero,ymm14[24],zero,zero
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm5 = ymm5[2,3,2,3]
@@ -6241,7 +6241,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm6 = ymm8[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-SLOW-NEXT: vmovdqa %ymm8, %ymm14
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm7 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm7, %ymm5, %ymm6, %ymm5
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm6 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm6, %ymm4, %ymm5, %ymm1
@@ -6258,7 +6258,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [128,128,0,128,14,128,128,128,128,1,128,15,128,128,128,128,128,128,16,128,30,128,128,128,128,17,128,31,128,128,128,128]
; AVX2-SLOW-NEXT: vpshufb %ymm10, %ymm2, %ymm11
; AVX2-SLOW-NEXT: vpor %ymm9, %ymm11, %ymm9
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-SLOW-NEXT: vpblendvb %ymm11, %ymm5, %ymm9, %ymm5
; AVX2-SLOW-NEXT: vpshufb %ymm6, %ymm0, %ymm6
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -6278,7 +6278,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm10 = [12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-SLOW-NEXT: vpshufb %ymm10, %ymm1, %ymm11
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm12 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm12, %ymm8, %ymm11, %ymm8
; AVX2-SLOW-NEXT: vpshufb %ymm7, %ymm13, %ymm7
; AVX2-SLOW-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm11 # 32-byte Reload
@@ -6302,15 +6302,15 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm9, %ymm8, %ymm8
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,3,2,3]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm2 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm2, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,26,27,30,31,30,31,28,29,28,29,28,29,28,29]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm15[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29,28,27,u,u,u,31,30,u,u,u,u,u,u,u,u]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,2,2,2]
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = <255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm9 = [255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u]
; AVX2-SLOW-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-SLOW-NEXT: vpshufb {{.*#+}} ymm8 = ymm1[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
@@ -6374,17 +6374,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,24,25,26,27,24,25,30,31]
; AVX2-FAST-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,25,24,23,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -6402,15 +6402,15 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm8[30],zero,ymm8[28],zero,zero,zero,zero,ymm8[31],zero,ymm8[29],zero,zero,zero,zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,26,27,30,31,30,31,28,29,28,29,28,29,28,29]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29,28,27,u,u,u,31,30,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u]
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm1 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
@@ -6423,16 +6423,16 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; AVX2-FAST-NEXT: vmovdqa %xmm1, %xmm14
; AVX2-FAST-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-FAST-NEXT: vmovdqa 32(%rcx), %xmm10
; AVX2-FAST-NEXT: vmovdqa 32(%rdx), %xmm7
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm7[0],xmm10[0],xmm7[1],xmm10[1],xmm7[2],xmm10[2],xmm7[3],xmm10[3],xmm7[4],xmm10[4],xmm7[5],xmm10[5],xmm7[6],xmm10[6],xmm7[7],xmm10[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm4, %ymm0, %ymm2, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vmovdqa (%rsi), %xmm11
@@ -6458,10 +6458,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa 32(%r9), %xmm8
; AVX2-FAST-NEXT: vmovdqa 32(%r8), %xmm13
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm13[0],xmm8[0],xmm13[1],xmm8[1],xmm13[2],xmm8[2],xmm13[3],xmm8[3],xmm13[4],xmm8[4],xmm13[5],xmm8[5],xmm13[6],xmm8[6],xmm13[7],xmm8[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm5, %xmm5
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm1, %ymm5, %ymm4, %ymm0
; AVX2-FAST-NEXT: vmovdqa (%rax), %xmm4
; AVX2-FAST-NEXT: vmovdqa %xmm4, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -6479,20 +6479,20 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpblendvb %ymm2, %ymm6, %ymm1, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm14, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm3 # 16-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm3, %xmm3
; AVX2-FAST-NEXT: vpor %xmm0, %xmm3, %xmm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm10, %xmm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm15 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm15, %xmm7, %xmm14
; AVX2-FAST-NEXT: vpor %xmm6, %xmm14, %xmm6
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm0, %ymm6, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb %xmm1, %xmm11, %xmm1
@@ -6506,10 +6506,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
; AVX2-FAST-NEXT: vpblendvb %ymm14, %ymm1, %ymm2, %ymm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm8, %xmm3
; AVX2-FAST-NEXT: vmovdqa %xmm8, %xmm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX2-FAST-NEXT: vpshufb %xmm6, %xmm13, %xmm14
; AVX2-FAST-NEXT: vpor %xmm3, %xmm14, %xmm3
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm3 = ymm3[0,1,0,1]
@@ -6517,7 +6517,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
; AVX2-FAST-NEXT: vpshufb %xmm14, %xmm8, %xmm15
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,1,0]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm3, %ymm15, %ymm3
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm9, %xmm2
; AVX2-FAST-NEXT: vmovdqa %xmm9, %xmm12
@@ -6538,13 +6538,13 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-FAST-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm1 # 16-byte Folded Reload
; AVX2-FAST-NEXT: # xmm1 = xmm1[8],mem[8],xmm1[9],mem[9],xmm1[10],mem[10],xmm1[11],mem[11],xmm1[12],mem[12],xmm1[13],mem[13],xmm1[14],mem[14],xmm1[15],mem[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-FAST-NEXT: vpblendvb %ymm6, %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm5[8],xmm4[8],xmm5[9],xmm4[9],xmm5[10],xmm4[10],xmm5[11],xmm4[11],xmm5[12],xmm4[12],xmm5[13],xmm4[13],xmm5[14],xmm4[14],xmm5[15],xmm4[15]
; AVX2-FAST-NEXT: vpshufb %xmm2, %xmm1, %xmm1
@@ -6560,10 +6560,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [2,2,3,3,2,2,3,3]
; AVX2-FAST-NEXT: # ymm6 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm6, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX2-FAST-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-NEXT: vpblendvb %ymm5, %ymm2, %ymm3, %ymm2
; AVX2-FAST-NEXT: vpshufhw {{.*#+}} xmm3 = xmm15[0,1,2,3,4,5,5,6]
; AVX2-FAST-NEXT: vpermd %ymm3, %ymm6, %ymm3
@@ -6593,7 +6593,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm12
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm3, %ymm6, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm2, %ymm3, %ymm6
; AVX2-FAST-NEXT: vmovdqa (%r8), %ymm0
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm2 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm0[27],zero,ymm0[27,28,29,30],zero,ymm0[28],zero,ymm0[26,27,30,31],zero,ymm0[29]
@@ -6611,7 +6611,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa %ymm0, %ymm11
; AVX2-FAST-NEXT: vmovdqu %ymm0, (%rsp) # 32-byte Spill
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm0
@@ -6632,7 +6632,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[25],zero,ymm4[23],zero,zero,zero,zero,ymm4[26],zero,ymm4[24],zero,zero,zero,zero,ymm4[27]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm7, %ymm8, %ymm7
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm6, %ymm7, %ymm6
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm7 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm3[25],zero,ymm3[23],zero,zero,zero,zero,ymm3[26],zero,ymm3[24],zero,zero
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,3,2,3]
@@ -6641,7 +6641,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpor %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vpshufb {{.*#+}} ymm8 = ymm11[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm9 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm8 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm0
@@ -6666,7 +6666,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm1, %ymm15
; AVX2-FAST-NEXT: vpermq {{.*#+}} ymm15 = ymm15[2,3,2,3]
; AVX2-FAST-NEXT: vpor %ymm11, %ymm15, %ymm11
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm15, %ymm7, %ymm11, %ymm0
; AVX2-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-NEXT: vpshufb %ymm8, %ymm14, %ymm8
@@ -6694,7 +6694,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshuflw {{.*#+}} ymm12 = ymm13[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [4,5,4,5,5,7,4,5]
; AVX2-FAST-NEXT: vpermd %ymm12, %ymm15, %ymm12
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm0, %ymm10, %ymm12, %ymm10
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; AVX2-FAST-NEXT: vpshufb %ymm9, %ymm2, %ymm9
@@ -6722,7 +6722,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm15 = [128,128,0,128,14,128,128,128,128,1,128,15,128,128,128,128,128,128,16,128,30,128,128,128,128,17,128,31,128,128,128,128]
; AVX2-FAST-NEXT: vpshufb %ymm15, %ymm6, %ymm6
; AVX2-FAST-NEXT: vpor %ymm6, %ymm12, %ymm6
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-FAST-NEXT: vpblendvb %ymm12, %ymm9, %ymm6, %ymm9
; AVX2-FAST-NEXT: vpshufb %ymm0, %ymm5, %ymm0
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
@@ -6741,7 +6741,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpor %ymm4, %ymm6, %ymm4
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm13, %ymm10
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-FAST-NEXT: vpblendvb %ymm11, %ymm4, %ymm10, %ymm4
; AVX2-FAST-NEXT: vpshufb %ymm1, %ymm2, %ymm1
; AVX2-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
@@ -6810,17 +6810,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm6, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm1, %ymm2, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,24,25,24,25,22,23,22,23,24,25,26,27,24,25,30,31]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm4, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm5, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,25,24,23,u,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
@@ -6838,15 +6838,15 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm8[30],zero,ymm8[28],zero,zero,zero,zero,ymm8[31],zero,ymm8[29],zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm1, %ymm2, %ymm1
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm4[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,26,27,26,27,30,31,30,31,28,29,28,29,28,29,28,29]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u,255,255,255,255,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm5[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,29,28,27,u,u,u,31,30,u,u,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,2,2,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u,255,255,255,255,255,0,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm1 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,28,29,26,27,28,29,30,31,30,31,28,29,28,29,30,31]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,2,3]
@@ -6860,17 +6860,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm1, %xmm15
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rcx), %xmm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%rdx), %xmm5
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm5[0],xmm9[0],xmm5[1],xmm9[1],xmm5[2],xmm9[2],xmm5[3],xmm9[3],xmm5[4],xmm9[4],xmm5[5],xmm9[5],xmm5[6],xmm9[6],xmm5[7],xmm9[7]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm5, %xmm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm5, %ymm0, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsi), %xmm0
@@ -6898,10 +6898,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm0, (%rsp) # 16-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa 32(%r8), %xmm14
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm14[0],xmm0[0],xmm14[1],xmm0[1],xmm14[2],xmm0[2],xmm14[3],xmm0[3],xmm14[4],xmm0[4],xmm14[5],xmm0[5],xmm14[6],xmm0[6],xmm14[7],xmm0[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm5, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm5 = ymm5[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm2, %ymm5, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r9), %xmm5
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -6919,20 +6919,20 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm3, %ymm2, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm15, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm7, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpor %xmm2, %xmm6, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm9, %xmm12
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm0 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm11, %xmm7
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm0, %xmm11, %xmm15
; AVX2-FAST-PERLANE-NEXT: vpor %xmm12, %xmm15, %xmm12
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm12 = ymm12[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm2, %ymm12, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm12 # 16-byte Reload
@@ -6946,10 +6946,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm1, %ymm0, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm1 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rsp), %xmm8 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm8, %xmm2
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm14, %xmm6
; AVX2-FAST-PERLANE-NEXT: vpor %xmm2, %xmm6, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
@@ -6957,7 +6957,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm4 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm4, %xmm15
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm15[0,0,1,0]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm2, %ymm15, %ymm2
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm10 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm1, %xmm10, %xmm1
@@ -6977,13 +6977,13 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm1 # 16-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # xmm1 = xmm1[8],mem[8],xmm1[9],mem[9],xmm1[10],mem[10],xmm1[11],mem[11],xmm1[12],mem[12],xmm1[13],mem[13],xmm1[14],mem[14],xmm1[15],mem[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = <6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm2 = [6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = <0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm6, %ymm0, %ymm1, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{[-0-9]+}}(%r{{[sb]}}p), %xmm5, %xmm1 # 16-byte Folded Reload
; AVX2-FAST-PERLANE-NEXT: # xmm1 = xmm5[8],mem[8],xmm5[9],mem[9],xmm5[10],mem[10],xmm5[11],mem[11],xmm5[12],mem[12],xmm5[13],mem[13],xmm5[14],mem[14],xmm5[15],mem[15]
@@ -6998,10 +6998,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [8,9,10,11,8,9,10,11,10,11,12,13,10,11,12,13]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm4, %xmm4
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm4 = ymm4[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = <u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm6 = [u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm6, %xmm2, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm2 = ymm2[0,1,0,1]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm7, %ymm2, %ymm4, %ymm2
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm13, %xmm3
; AVX2-FAST-PERLANE-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm15[8],xmm10[8],xmm15[9],xmm10[9],xmm15[10],xmm10[10],xmm15[11],xmm10[11],xmm15[12],xmm10[12],xmm15[13],xmm10[13],xmm15[14],xmm10[14],xmm15[15],xmm10[15]
@@ -7031,7 +7031,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm6 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm5[30],zero,ymm5[28],zero,zero,zero,zero,ymm5[31],zero,ymm5[29],zero,zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm6 = ymm6[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm3, %ymm6, %ymm3
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm2, %ymm3, %ymm6
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%r8), %ymm0
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm2 = ymm0[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,ymm0[27],zero,ymm0[27,28,29,30],zero,ymm0[28],zero,ymm0[26,27,30,31],zero,ymm0[29]
@@ -7049,7 +7049,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm0, %ymm11
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm0
@@ -7070,7 +7070,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm8 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,ymm4[25],zero,ymm4[23],zero,zero,zero,zero,ymm4[26],zero,ymm4[24],zero,zero,zero,zero,ymm4[27]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm7, %ymm8, %ymm7
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm6, %ymm7, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm7 = ymm3[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u],zero,zero,zero,zero,ymm3[25],zero,ymm3[23],zero,zero,zero,zero,ymm3[26],zero,ymm3[24],zero,zero
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm7 = ymm7[2,3,2,3]
@@ -7079,7 +7079,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} ymm8 = ymm11[u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,22,23,26,27,24,25,22,23,24,25,26,27,26,27,24,25]
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = <0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm9 = [0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm9, %ymm7, %ymm8, %ymm7
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm8 = [0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255,255,0,0,0,255,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm8, %ymm6, %ymm7, %ymm0
@@ -7104,7 +7104,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm12, %ymm1, %ymm15
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm15[2,3,2,3]
; AVX2-FAST-PERLANE-NEXT: vpor %ymm11, %ymm15, %ymm11
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = <255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm15 = [255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u,u,0,0,255,255,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm15, %ymm7, %ymm11, %ymm0
; AVX2-FAST-PERLANE-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm8, %ymm13, %ymm8
@@ -7133,7 +7133,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm13 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm12, %ymm13, %ymm15
; AVX2-FAST-PERLANE-NEXT: vpermq {{.*#+}} ymm15 = ymm15[2,2,3,2]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = <u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm0 = [u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm0, %ymm10, %ymm15, %ymm10
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm15 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm9, %ymm15, %ymm9
@@ -7161,7 +7161,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [128,128,0,128,14,128,128,128,128,1,128,15,128,128,128,128,128,128,16,128,30,128,128,128,128,17,128,31,128,128,128,128]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm6, %ymm6
; AVX2-FAST-PERLANE-NEXT: vpor %ymm6, %ymm12, %ymm6
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = <u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255,0,0,u,u,u,255,255]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm12, %ymm9, %ymm6, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm0, %ymm1, %ymm0
@@ -7180,7 +7180,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpor %ymm4, %ymm6, %ymm4
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [12,13,2,3,12,13,0,1,14,15,2,3,0,1,14,15,28,29,18,19,28,29,16,17,30,31,18,19,16,17,30,31]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm13, %ymm10
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u,u,u,255,255,0,u,u]
; AVX2-FAST-PERLANE-NEXT: vpblendvb %ymm11, %ymm4, %ymm10, %ymm4
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm5, %ymm15, %ymm3
@@ -7307,16 +7307,16 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa (%rax), %ymm1
; AVX512F-SLOW-NEXT: vpshufb %ymm2, %ymm1, %ymm0
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm16 = <u,5,4,u,5,u,4,u,20,21,u,23,u,21,u,23>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm16 = [u,5,4,u,5,u,4,u,20,21,u,23,u,21,u,23]
; AVX512F-SLOW-NEXT: vpshuflw {{.*#+}} ymm2 = ymm1[2,1,1,2,4,5,6,7,10,9,9,10,12,13,14,15]
; AVX512F-SLOW-NEXT: vpermi2d %zmm0, %zmm2, %zmm16
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdx), %xmm3
; AVX512F-SLOW-NEXT: vmovdqa 32(%rcx), %xmm0
; AVX512F-SLOW-NEXT: vmovdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX512F-SLOW-NEXT: vpshufb %xmm2, %xmm0, %xmm0
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm2, %xmm29
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm3, %xmm2
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm4, %xmm31
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm3, %xmm30
@@ -7324,14 +7324,14 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 32(%rdi), %xmm3
; AVX512F-SLOW-NEXT: vmovdqa 32(%rsi), %xmm4
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm12 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX512F-SLOW-NEXT: vpshufb %xmm12, %xmm4, %xmm0
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm15 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX512F-SLOW-NEXT: vpshufb %xmm15, %xmm3, %xmm2
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm3, %xmm21
; AVX512F-SLOW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
-; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = <0,u,0,u,2,3,u,1,u,18,u,19,18,u,19,u>
+; AVX512F-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm3 = [0,u,0,u,2,3,u,1,u,18,u,19,18,u,19,u]
; AVX512F-SLOW-NEXT: vmovdqa 32(%rax), %xmm2
; AVX512F-SLOW-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX512F-SLOW-NEXT: vpshufhw {{.*#+}} xmm0 = xmm2[0,1,2,3,4,5,5,6]
@@ -7340,10 +7340,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vmovdqa 32(%r9), %xmm0
; AVX512F-SLOW-NEXT: vmovdqa 32(%r8), %xmm2
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm6 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX512F-SLOW-NEXT: vpshufb %xmm6, %xmm0, %xmm8
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm0, %xmm28
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm2, %xmm10
; AVX512F-SLOW-NEXT: vmovdqa %xmm2, %xmm3
; AVX512F-SLOW-NEXT: vporq %xmm8, %xmm10, %xmm26
@@ -7390,7 +7390,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm6, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm24, %zmm0, %zmm6
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm11 = xmm5[8],xmm2[8],xmm5[9],xmm2[9],xmm5[10],xmm2[10],xmm5[11],xmm2[11],xmm5[12],xmm2[12],xmm5[13],xmm2[13],xmm5[14],xmm2[14],xmm5[15],xmm2[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = <u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm7 = [u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX512F-SLOW-NEXT: vpshufb %xmm7, %xmm11, %xmm11
; AVX512F-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm2 = zmm11[0,1,0,1],zmm6[4,5,6,7]
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -7442,7 +7442,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm1[8],xmm0[8],xmm1[9],xmm0[9],xmm1[10],xmm0[10],xmm1[11],xmm0[11],xmm1[12],xmm0[12],xmm1[13],xmm0[13],xmm1[14],xmm0[14],xmm1[15],xmm0[15]
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm15 = xmm13[8],xmm9[8],xmm13[9],xmm9[9],xmm13[10],xmm9[10],xmm13[11],xmm9[11],xmm13[12],xmm9[12],xmm13[13],xmm9[13],xmm13[14],xmm9[14],xmm13[15],xmm9[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX512F-SLOW-NEXT: vpshufb %xmm0, %xmm15, %xmm3
; AVX512F-SLOW-NEXT: vmovdqu %ymm3, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-SLOW-NEXT: vpshufb %xmm0, %xmm1, %xmm0
@@ -7453,7 +7453,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm4[8],xmm0[8],xmm4[9],xmm0[9],xmm4[10],xmm0[10],xmm4[11],xmm0[11],xmm4[12],xmm0[12],xmm4[13],xmm0[13],xmm4[14],xmm0[14],xmm4[15],xmm0[15]
; AVX512F-SLOW-NEXT: vmovdqa64 %xmm25, %xmm3
; AVX512F-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm3[8],xmm12[8],xmm3[9],xmm12[9],xmm3[10],xmm12[10],xmm3[11],xmm12[11],xmm3[12],xmm12[12],xmm3[13],xmm12[13],xmm3[14],xmm12[14],xmm3[15],xmm12[15]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = <2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm4 = [2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm1, %xmm1
; AVX512F-SLOW-NEXT: vmovdqu %ymm1, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-SLOW-NEXT: vpshufb %xmm4, %xmm0, %xmm0
@@ -7506,13 +7506,13 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-SLOW-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm14, %ymm14
; AVX512F-SLOW-NEXT: vinserti64x4 $1, %ymm14, %zmm2, %zmm2
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm12 = xmm12[0],xmm3[0],xmm12[1],xmm3[1],xmm12[2],xmm3[2],xmm12[3],xmm3[3],xmm12[4],xmm3[4],xmm12[5],xmm3[5],xmm12[6],xmm3[6],xmm12[7],xmm3[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm15, %xmm15
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm12, %xmm12
; AVX512F-SLOW-NEXT: vinserti32x4 $2, %xmm22, %zmm12, %zmm0
; AVX512F-SLOW-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm13 = xmm9[0],xmm13[0],xmm9[1],xmm13[1],xmm9[2],xmm13[2],xmm9[3],xmm13[3],xmm9[4],xmm13[4],xmm9[5],xmm13[5],xmm9[6],xmm13[6],xmm9[7],xmm13[7]
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} xmm14 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm11, %xmm9
; AVX512F-SLOW-NEXT: vpshufb %xmm14, %xmm13, %xmm11
; AVX512F-SLOW-NEXT: vpermq {{.*#+}} ymm13 = ymm6[2,3,2,3]
@@ -7721,10 +7721,10 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vporq %ymm3, %ymm5, %ymm24
; AVX512F-FAST-NEXT: vmovdqa 32(%rdx), %xmm3
; AVX512F-FAST-NEXT: vmovdqa 32(%rcx), %xmm6
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX512F-FAST-NEXT: vpshufb %xmm8, %xmm6, %xmm5
; AVX512F-FAST-NEXT: vmovdqa64 %xmm6, %xmm28
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX512F-FAST-NEXT: vpshufb %xmm9, %xmm3, %xmm6
; AVX512F-FAST-NEXT: vmovdqa64 %xmm9, %xmm19
; AVX512F-FAST-NEXT: vmovdqa64 %xmm3, %xmm29
@@ -7732,18 +7732,18 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqu64 %zmm3, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 32(%rdi), %xmm10
; AVX512F-FAST-NEXT: vmovdqa 32(%rsi), %xmm6
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX512F-FAST-NEXT: vpshufb %xmm11, %xmm6, %xmm5
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX512F-FAST-NEXT: vpshufb %xmm3, %xmm10, %xmm9
; AVX512F-FAST-NEXT: vmovdqa64 %xmm10, %xmm27
; AVX512F-FAST-NEXT: vpor %xmm5, %xmm9, %xmm5
; AVX512F-FAST-NEXT: vmovdqu64 %zmm5, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vmovdqa 32(%r9), %xmm15
; AVX512F-FAST-NEXT: vmovdqa 32(%r8), %xmm10
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX512F-FAST-NEXT: vpshufb %xmm5, %xmm15, %xmm9
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX512F-FAST-NEXT: vpshufb %xmm0, %xmm10, %xmm12
; AVX512F-FAST-NEXT: vmovdqa64 %xmm0, %xmm21
; AVX512F-FAST-NEXT: vporq %xmm9, %xmm12, %xmm22
@@ -7811,7 +7811,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm24, %zmm0, %zmm0
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm4[8],xmm3[8],xmm4[9],xmm3[9],xmm4[10],xmm3[10],xmm4[11],xmm3[11],xmm4[12],xmm3[12],xmm4[13],xmm3[13],xmm4[14],xmm3[14],xmm4[15],xmm3[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [u,6,7,2,3,u,u,u,8,9,4,5,u,u,u,10]
; AVX512F-FAST-NEXT: vpshufb %xmm3, %xmm1, %xmm1
; AVX512F-FAST-NEXT: vshufi64x2 {{.*#+}} zmm0 = zmm1[0,1,0,1],zmm0[4,5,6,7]
; AVX512F-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
@@ -7849,7 +7849,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm7 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],xmm2[5],xmm0[6],xmm2[6],xmm0[7],xmm2[7]
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm2[8],xmm0[8],xmm2[9],xmm0[9],xmm2[10],xmm0[10],xmm2[11],xmm0[11],xmm2[12],xmm0[12],xmm2[13],xmm0[13],xmm2[14],xmm0[14],xmm2[15],xmm0[15]
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm14[8],xmm8[8],xmm14[9],xmm8[9],xmm14[10],xmm8[10],xmm14[11],xmm8[11],xmm14[12],xmm8[12],xmm14[13],xmm8[13],xmm14[14],xmm8[14],xmm14[15],xmm8[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX512F-FAST-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX512F-FAST-NEXT: vmovdqu %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vpshufb %xmm4, %xmm0, %xmm0
@@ -7860,7 +7860,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm6 = xmm0[0],xmm6[0],xmm0[1],xmm6[1],xmm0[2],xmm6[2],xmm0[3],xmm6[3],xmm0[4],xmm6[4],xmm0[5],xmm6[5],xmm0[6],xmm6[6],xmm0[7],xmm6[7]
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm2 = xmm2[8],xmm0[8],xmm2[9],xmm0[9],xmm2[10],xmm0[10],xmm2[11],xmm0[11],xmm2[12],xmm0[12],xmm2[13],xmm0[13],xmm2[14],xmm0[14],xmm2[15],xmm0[15]
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm13[8],xmm9[8],xmm13[9],xmm9[9],xmm13[10],xmm9[10],xmm13[11],xmm9[11],xmm13[12],xmm9[12],xmm13[13],xmm9[13],xmm13[14],xmm9[14],xmm13[15],xmm9[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = <2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm5 = [2,u,u,u,9,8,5,4,u,u,u,11,10,7,6,u]
; AVX512F-FAST-NEXT: vpshufb %xmm5, %xmm4, %xmm0
; AVX512F-FAST-NEXT: vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX512F-FAST-NEXT: vpshufb %xmm5, %xmm2, %xmm2
@@ -7902,7 +7902,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm0, %ymm0
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm3, %zmm16
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm9[0],xmm13[0],xmm9[1],xmm13[1],xmm9[2],xmm13[2],xmm9[3],xmm13[3],xmm9[4],xmm13[4],xmm9[5],xmm13[5],xmm9[6],xmm13[6],xmm9[7],xmm13[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX512F-FAST-NEXT: vpshufb %xmm9, %xmm6, %xmm6
; AVX512F-FAST-NEXT: vpshufb %xmm9, %xmm0, %xmm0
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm9 = [28,29,30,128,28,128,30,31,30,31,128,29,128,31,28,29,28,29,30,128,28,128,30,31,30,31,128,29,128,31,28,29]
@@ -7917,7 +7917,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vinserti32x4 $2, %xmm31, %zmm0, %zmm0
; AVX512F-FAST-NEXT: vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; AVX512F-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm8[0],xmm14[0],xmm8[1],xmm14[1],xmm8[2],xmm14[2],xmm8[3],xmm14[3],xmm8[4],xmm14[4],xmm8[5],xmm14[5],xmm8[6],xmm14[6],xmm8[7],xmm14[7]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX512F-FAST-NEXT: vpshufb %xmm8, %xmm7, %xmm7
; AVX512F-FAST-NEXT: vpshufb %xmm8, %xmm0, %xmm13
; AVX512F-FAST-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
@@ -8152,7 +8152,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovdqa64 32(%rdx), %ymm29
; AVX512BW-SLOW-NEXT: vpshufb %ymm20, %ymm29, %ymm8
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm8 = ymm8[2,3,2,3]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,18,128,128,128,128,21,128,19,128,128,128,128,22,128,20,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,25,128,23,128,128,128,128,26,128,24,128,128,128,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm21 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,18,128,128,128,128,21,128,19,128,128,128,128,22,128,20,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,25,128,23,128,128,128,128,26,128,24,128,128,128,128]
; AVX512BW-SLOW-NEXT: vmovdqa64 32(%rcx), %ymm30
; AVX512BW-SLOW-NEXT: vpshufb %ymm21, %ymm30, %ymm25
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm25 = ymm25[2,3,2,3]
@@ -8181,7 +8181,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vmovdqa64 32(%r8), %ymm31
; AVX512BW-SLOW-NEXT: vpshufb %ymm23, %ymm31, %ymm0
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm26 = ymm0[2,3,2,3]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,20,128,18,128,128,128,128,21,128,19,128,128,128,128,22,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,25,128,23,128,128,128,128,26,128,24,128,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm24 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,20,128,18,128,128,128,128,21,128,19,128,128,128,128,22,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,25,128,23,128,128,128,128,26,128,24,128,128]
; AVX512BW-SLOW-NEXT: vmovdqa 32(%r9), %ymm0
; AVX512BW-SLOW-NEXT: vpshufb %ymm24, %ymm0, %ymm11
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[2,3,2,3]
@@ -8295,12 +8295,12 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vpshufd {{.*#+}} ymm26 = ymm18[2,2,3,3,6,6,7,7]
; AVX512BW-SLOW-NEXT: vpshufb %ymm27, %ymm19, %ymm26 {%k4}
; AVX512BW-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm19 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm18 = <0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm18 = [0,1,u,u,u,6,7,2,3,u,u,u,8,9,4,5]
; AVX512BW-SLOW-NEXT: vpshufb %xmm18, %xmm19, %xmm19
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm19 = zmm26[2,3,2,3],zmm19[0,1,0,1]
; AVX512BW-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm26 = xmm24[0],xmm25[0],xmm24[1],xmm25[1],xmm24[2],xmm25[2],xmm24[3],xmm25[3],xmm24[4],xmm25[4],xmm24[5],xmm25[5],xmm24[6],xmm25[6],xmm24[7],xmm25[7]
; AVX512BW-SLOW-NEXT: vpshufb %ymm11, %ymm16, %ymm11
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm16 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm16 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX512BW-SLOW-NEXT: vpshufb %xmm16, %xmm26, %xmm26
; AVX512BW-SLOW-NEXT: vpshufhw {{.*#+}} ymm14 = ymm14[0,1,2,3,7,6,6,7,8,9,10,11,15,14,14,15]
; AVX512BW-SLOW-NEXT: vpshufd {{.*#+}} ymm14 = ymm14[0,2,3,3,4,6,7,7]
@@ -8313,7 +8313,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm10 = ymm10[2,3,2,3]
; AVX512BW-SLOW-NEXT: vpor %ymm9, %ymm10, %ymm10
; AVX512BW-SLOW-NEXT: vpunpcklbw {{.*#+}} xmm11 = xmm23[0],xmm21[0],xmm23[1],xmm21[1],xmm23[2],xmm21[2],xmm23[3],xmm21[3],xmm23[4],xmm21[4],xmm23[5],xmm21[5],xmm23[6],xmm21[6],xmm23[7],xmm21[7]
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %xmm9, %xmm11, %xmm11
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} ymm11 = ymm11[0,1,0,1]
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm11, %zmm10, %zmm10
@@ -8325,17 +8325,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: movabsq $1016749673354069774, %rax # imm = 0xE1C3870E1C3870E
; AVX512BW-SLOW-NEXT: kmovq %rax, %k1
; AVX512BW-SLOW-NEXT: vmovdqu8 %zmm10, %zmm14 {%k1}
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %xmm11, %xmm25, %xmm10
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm19 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm19 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %xmm19, %xmm24, %xmm26
; AVX512BW-SLOW-NEXT: vporq %xmm10, %xmm26, %xmm10
; AVX512BW-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm24 = xmm25[8],xmm24[8],xmm25[9],xmm24[9],xmm25[10],xmm24[10],xmm25[11],xmm24[11],xmm25[12],xmm24[12],xmm25[13],xmm24[13],xmm25[14],xmm24[14],xmm25[15],xmm24[15]
; AVX512BW-SLOW-NEXT: vpshufb {{.*#+}} xmm24 = xmm24[6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX512BW-SLOW-NEXT: vinserti32x4 $2, %xmm24, %zmm10, %zmm10
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm24 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm24 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX512BW-SLOW-NEXT: vpshufb %xmm24, %xmm1, %xmm25
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm26 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm26 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX512BW-SLOW-NEXT: vpshufb %xmm26, %xmm0, %xmm27
; AVX512BW-SLOW-NEXT: vporq %xmm25, %xmm27, %xmm25
; AVX512BW-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm1[8],xmm0[8],xmm1[9],xmm0[9],xmm1[10],xmm0[10],xmm1[11],xmm0[11],xmm1[12],xmm0[12],xmm1[13],xmm0[13],xmm1[14],xmm0[14],xmm1[15],xmm0[15]
@@ -8344,11 +8344,11 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vinserti32x4 $2, %xmm0, %zmm25, %zmm0
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} zmm10 = zmm0[0,1,0,1,4,5,4,5]
; AVX512BW-SLOW-NEXT: vmovdqu8 %zmm1, %zmm10 {%k3}
-; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX512BW-SLOW-NEXT: vmovdqa {{.*#+}} xmm0 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX512BW-SLOW-NEXT: vpshufb %xmm0, %xmm21, %xmm1
; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm25 = [2,2,2,4,2,2,2,4,3,3,3,3,2,2,2,4,52,53,52,53,53,54,53,54,52,53,52,53,53,54,53,54]
; AVX512BW-SLOW-NEXT: vpermi2w %zmm22, %zmm17, %zmm25
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm17 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} xmm17 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX512BW-SLOW-NEXT: vpshufb %xmm17, %xmm23, %xmm22
; AVX512BW-SLOW-NEXT: vporq %xmm1, %xmm22, %xmm1
; AVX512BW-SLOW-NEXT: vpunpckhbw {{.*#+}} xmm21 = xmm23[8],xmm21[8],xmm23[9],xmm21[9],xmm23[10],xmm21[10],xmm23[11],xmm21[11],xmm23[12],xmm21[12],xmm23[13],xmm21[13],xmm23[14],xmm21[14],xmm23[15],xmm21[15]
@@ -8477,7 +8477,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa 32(%rdx), %ymm2
; AVX512BW-FAST-NEXT: vpshufb %ymm0, %ymm2, %ymm3
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm25 = ymm3[2,3,2,3]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,18,128,128,128,128,21,128,19,128,128,128,128,22,128,20,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,25,128,23,128,128,128,128,26,128,24,128,128,128,128>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,18,128,128,128,128,21,128,19,128,128,128,128,22,128,20,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,25,128,23,128,128,128,128,26,128,24,128,128,128,128]
; AVX512BW-FAST-NEXT: vmovdqa 32(%rcx), %ymm3
; AVX512BW-FAST-NEXT: vpshufb %ymm0, %ymm3, %ymm26
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm26 = ymm26[2,3,2,3]
@@ -8486,11 +8486,11 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpshufb %ymm11, %ymm3, %ymm11
; AVX512BW-FAST-NEXT: vpor %ymm4, %ymm11, %ymm4
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm25, %zmm4, %zmm4
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,18,19,20,21,128,19,128,21,20,21,22,128,20,128,22,23,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,25,128,23,128,128,128,128,26,128,24,128,128,128,128,27,128>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,18,19,20,21,128,19,128,21,20,21,22,128,20,128,22,23,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,25,128,23,128,128,128,128,26,128,24,128,128,128,128,27,128]
; AVX512BW-FAST-NEXT: vmovdqa 32(%rdi), %ymm11
; AVX512BW-FAST-NEXT: vpshufb %ymm0, %ymm11, %ymm25
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm25 = ymm25[2,3,2,3]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,21,128,19,128,128,128,128,22,128,20,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,23,128,128,128,128,26,128,24,128,128,128,128,27,128,25>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm0 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,21,128,19,128,128,128,128,22,128,20,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,23,128,128,128,128,26,128,24,128,128,128,128,27,128,25]
; AVX512BW-FAST-NEXT: vmovdqa64 32(%rsi), %ymm27
; AVX512BW-FAST-NEXT: vpshufb %ymm0, %ymm27, %ymm26
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm26 = ymm26[2,3,2,3]
@@ -8506,7 +8506,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa 32(%r8), %ymm4
; AVX512BW-FAST-NEXT: vpshufb %ymm0, %ymm4, %ymm21
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm25 = ymm21[2,3,2,3]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,20,128,18,128,128,128,128,21,128,19,128,128,128,128,22,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,25,128,23,128,128,128,128,26,128,24,128,128>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,20,128,18,128,128,128,128,21,128,19,128,128,128,128,22,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,128,128,128,128,25,128,23,128,128,128,128,26,128,24,128,128]
; AVX512BW-FAST-NEXT: vmovdqa 32(%r9), %ymm0
; AVX512BW-FAST-NEXT: vpshufb %ymm5, %ymm0, %ymm26
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm26 = ymm26[2,3,2,3]
@@ -8584,7 +8584,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa 32(%rdx), %xmm2
; AVX512BW-FAST-NEXT: vmovdqa 32(%rcx), %xmm1
; AVX512BW-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3],xmm2[4],xmm1[4],xmm2[5],xmm1[5],xmm2[6],xmm1[6],xmm2[7],xmm1[7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm18 = <4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm18 = [4,5,0,1,u,u,u,6,7,2,3,u,u,u,8,9]
; AVX512BW-FAST-NEXT: vpshufb %xmm18, %xmm0, %xmm0
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm4, %zmm9
@@ -8598,7 +8598,7 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa 32(%r9), %xmm5
; AVX512BW-FAST-NEXT: vmovdqa 32(%r8), %xmm4
; AVX512BW-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm4[0],xmm5[0],xmm4[1],xmm5[1],xmm4[2],xmm5[2],xmm4[3],xmm5[3],xmm4[4],xmm5[4],xmm4[5],xmm5[5],xmm4[6],xmm5[6],xmm4[7],xmm5[7]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm19 = <u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} xmm19 = [u,u,4,5,0,1,u,u,u,6,7,2,3,u,u,u]
; AVX512BW-FAST-NEXT: vpshufb %xmm19, %xmm0, %xmm0
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,0,1]
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm3, %zmm0
@@ -8610,17 +8610,17 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: movabsq $1016749673354069774, %rax # imm = 0xE1C3870E1C3870E
; AVX512BW-FAST-NEXT: kmovq %rax, %k2
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm0, %zmm9 {%k2}
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [u,u,u,128,7,128,5,u,u,u,128,8,128,6,u,u]
; AVX512BW-FAST-NEXT: vpshufb %xmm11, %xmm1, %xmm3
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,u,7,128,5,128,u,u,u,8,128,6,128,u,u]
; AVX512BW-FAST-NEXT: vpshufb %xmm6, %xmm2, %xmm23
; AVX512BW-FAST-NEXT: vporq %xmm3, %xmm23, %xmm3
; AVX512BW-FAST-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
; AVX512BW-FAST-NEXT: vpshufb {{.*#+}} xmm1 = xmm1[6,3,2,u,u,u,9,8,5,4,u,u,u,11,10,7]
; AVX512BW-FAST-NEXT: vinserti32x4 $2, %xmm1, %zmm3, %zmm1
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = <u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm2 = [u,128,7,128,5,u,u,u,128,8,128,6,u,u,u,128]
; AVX512BW-FAST-NEXT: vpshufb %xmm2, %xmm7, %xmm3
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = <u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm0 = [u,7,128,5,128,u,u,u,8,128,6,128,u,u,u,9]
; AVX512BW-FAST-NEXT: vpshufb %xmm0, %xmm8, %xmm23
; AVX512BW-FAST-NEXT: vporq %xmm3, %xmm23, %xmm3
; AVX512BW-FAST-NEXT: vpunpckhbw {{.*#+}} xmm7 = xmm7[8],xmm8[8],xmm7[9],xmm8[9],xmm7[10],xmm8[10],xmm7[11],xmm8[11],xmm7[12],xmm8[12],xmm7[13],xmm8[13],xmm7[14],xmm8[14],xmm7[15],xmm8[15]
@@ -8629,9 +8629,9 @@ define void @store_i8_stride7_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm1 = zmm1[0,1,0,1,4,5,4,5]
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm23 = zmm3[0,1,0,1,4,5,4,5]
; AVX512BW-FAST-NEXT: vmovdqu8 %zmm1, %zmm23 {%k1}
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = <128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm1 = [128,4,u,u,u,128,7,128,5,u,u,u,128,8,128,6]
; AVX512BW-FAST-NEXT: vpshufb %xmm1, %xmm5, %xmm3
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [4,128,u,u,u,7,128,5,128,u,u,u,8,128,6,128]
; AVX512BW-FAST-NEXT: vpshufb %xmm7, %xmm4, %xmm8
; AVX512BW-FAST-NEXT: vpor %xmm3, %xmm8, %xmm3
; AVX512BW-FAST-NEXT: vpunpckhbw {{.*#+}} xmm4 = xmm4[8],xmm5[8],xmm4[9],xmm5[9],xmm4[10],xmm5[10],xmm4[11],xmm5[11],xmm4[12],xmm5[12],xmm4[13],xmm5[13],xmm4[14],xmm5[14],xmm4[15],xmm5[15]
diff --git a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll
index 7bb387481ff7b6b..e60034f7d6b75f0 100644
--- a/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll
+++ b/llvm/test/CodeGen/X86/vector-interleaved-store-i8-stride-8.ll
@@ -474,10 +474,10 @@ define void @store_i8_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [0,2,4,6,0,2,4,6]
; AVX2-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX2-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm3
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2],ymm3[3],ymm2[4],ymm3[5],ymm2[6],ymm3[7]
; AVX2-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [1,3,5,7,1,3,5,7]
@@ -594,10 +594,10 @@ define void @store_i8_stride8_vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecp
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm2 = [1,3,5,7,1,3,5,7]
; AVX512F-FAST-NEXT: # ymm2 = mem[0,1,0,1]
; AVX512F-FAST-NEXT: vpermd %ymm1, %ymm2, %ymm3
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <u,u,u,u,0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [u,u,u,u,0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15]
; AVX512F-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [0,4,8,12,u,u,u,u,1,5,9,13,u,u,u,u,2,6,10,14,u,u,u,u,3,7,11,15,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm2, %ymm2
; AVX512F-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2],ymm3[3],ymm2[4],ymm3[5],ymm2[6],ymm3[7]
; AVX512F-FAST-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [0,2,4,6,0,2,4,6]
@@ -1030,28 +1030,28 @@ define void @store_i8_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-ONLY-NEXT: vinserti128 $1, (%r9), %ymm2, %ymm4
; AVX2-ONLY-NEXT: vinserti128 $1, (%r10), %ymm3, %ymm3
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm5 = ymm3[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = <u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm6 = [u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11]
; AVX2-ONLY-NEXT: vpshufb %ymm6, %ymm5, %ymm2
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm7 = ymm4[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm8, %ymm7, %ymm9
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm2 = ymm9[0,1,2],ymm2[3],ymm9[4,5,6],ymm2[7],ymm9[8,9,10],ymm2[11],ymm9[12,13,14],ymm2[15]
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm9 = ymm1[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm10, %ymm9, %ymm11
; AVX2-ONLY-NEXT: vpermq {{.*#+}} ymm12 = ymm0[0,2,0,2]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = <0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm13 = [0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm13, %ymm12, %ymm14
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2,3,4],ymm11[5],ymm14[6,7,8],ymm11[9],ymm14[10,11,12],ymm11[13],ymm14[14,15]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm2 = ymm11[0],ymm2[1],ymm11[2],ymm2[3],ymm11[4],ymm2[5],ymm11[6],ymm2[7]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15]
; AVX2-ONLY-NEXT: vpshufb %ymm11, %ymm5, %ymm5
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm14, %ymm7, %ymm7
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm5 = ymm7[0,1,2],ymm5[3],ymm7[4,5,6],ymm5[7],ymm7[8,9,10],ymm5[11],ymm7[12,13,14],ymm5[15]
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm7, %ymm9, %ymm9
-; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm15 = <4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u>
+; AVX2-ONLY-NEXT: vmovdqa {{.*#+}} ymm15 = [4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u]
; AVX2-ONLY-NEXT: vpshufb %ymm15, %ymm12, %ymm12
; AVX2-ONLY-NEXT: vpblendw {{.*#+}} ymm9 = ymm12[0],ymm9[1],ymm12[2,3,4],ymm9[5],ymm12[6,7,8],ymm9[9],ymm12[10,11,12],ymm9[13],ymm12[14,15]
; AVX2-ONLY-NEXT: vpblendd {{.*#+}} ymm5 = ymm9[0],ymm5[1],ymm9[2],ymm5[3],ymm9[4],ymm5[5],ymm9[6],ymm5[7]
@@ -1094,28 +1094,28 @@ define void @store_i8_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-NEXT: vinserti128 $1, (%r9), %ymm2, %ymm2
; AVX512F-NEXT: vinserti128 $1, (%r10), %ymm3, %ymm3
; AVX512F-NEXT: vpermq {{.*#+}} ymm4 = ymm3[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15]
; AVX512F-NEXT: vpshufb %ymm5, %ymm4, %ymm6
; AVX512F-NEXT: vpermq {{.*#+}} ymm7 = ymm2[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = <u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm8 = [u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u]
; AVX512F-NEXT: vpshufb %ymm8, %ymm7, %ymm9
; AVX512F-NEXT: vpblendw {{.*#+}} ymm6 = ymm9[0,1,2],ymm6[3],ymm9[4,5,6],ymm6[7],ymm9[8,9,10],ymm6[11],ymm9[12,13,14],ymm6[15]
; AVX512F-NEXT: vpermq {{.*#+}} ymm9 = ymm1[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = <u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm10 = [u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm10, %ymm9, %ymm11
; AVX512F-NEXT: vpermq {{.*#+}} ymm12 = ymm0[0,2,0,2]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = <4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm13 = [4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm13, %ymm12, %ymm14
; AVX512F-NEXT: vpblendw {{.*#+}} ymm11 = ymm14[0],ymm11[1],ymm14[2,3,4],ymm11[5],ymm14[6,7,8],ymm11[9],ymm14[10,11,12],ymm11[13],ymm14[14,15]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm6 = ymm11[0],ymm6[1],ymm11[2],ymm6[3],ymm11[4],ymm6[5],ymm11[6],ymm6[7]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = <u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm11 = [u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11]
; AVX512F-NEXT: vpshufb %ymm11, %ymm4, %ymm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = <u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm14 = [u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u]
; AVX512F-NEXT: vpshufb %ymm14, %ymm7, %ymm7
; AVX512F-NEXT: vpblendw {{.*#+}} ymm4 = ymm7[0,1,2],ymm4[3],ymm7[4,5,6],ymm4[7],ymm7[8,9,10],ymm4[11],ymm7[12,13,14],ymm4[15]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = <u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm7 = [u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm7, %ymm9, %ymm9
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = <0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm15 = [0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %ymm15, %ymm12, %ymm12
; AVX512F-NEXT: vpblendw {{.*#+}} ymm9 = ymm12[0],ymm9[1],ymm12[2,3,4],ymm9[5],ymm12[6,7,8],ymm9[9],ymm12[10,11,12],ymm9[13],ymm12[14,15]
; AVX512F-NEXT: vpblendd {{.*#+}} ymm4 = ymm9[0],ymm4[1],ymm9[2],ymm4[3],ymm9[4],ymm4[5],ymm9[6],ymm4[7]
@@ -1161,22 +1161,22 @@ define void @store_i8_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm3, %zmm2, %zmm3
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm2, %zmm2, %zmm2
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} zmm4 = zmm2[0,2,0,2,4,6,4,6]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = <u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm5 = [u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %zmm5, %zmm4, %zmm4
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm3 = zmm3[4,5,6,7,4,5,6,7]
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} zmm6 = zmm3[0,2,0,2,4,6,4,6]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15]
; AVX512BW-SLOW-NEXT: vpshufb %zmm7, %zmm6, %zmm6
; AVX512BW-SLOW-NEXT: movl $8913032, %ecx # imm = 0x880088
; AVX512BW-SLOW-NEXT: kmovd %ecx, %k1
; AVX512BW-SLOW-NEXT: vmovdqu16 %zmm6, %zmm4 {%k1}
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm0
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} zmm6 = zmm0[0,2,0,2,4,6,4,6]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = <0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm8 = [0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %zmm8, %zmm6, %zmm6
; AVX512BW-SLOW-NEXT: vshufi64x2 {{.*#+}} zmm1 = zmm1[4,5,6,7,4,5,6,7]
; AVX512BW-SLOW-NEXT: vpermq {{.*#+}} zmm9 = zmm1[0,2,0,2,4,6,4,6]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = <u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} zmm10 = [u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u]
; AVX512BW-SLOW-NEXT: vpshufb %zmm10, %zmm9, %zmm9
; AVX512BW-SLOW-NEXT: movl $2228258, %ecx # imm = 0x220022
; AVX512BW-SLOW-NEXT: kmovd %ecx, %k2
@@ -1217,21 +1217,21 @@ define void @store_i8_stride8_vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm3, %zmm2, %zmm5
; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm6 = [0,2,0,2,12,14,12,14]
; AVX512BW-FAST-NEXT: vpermt2q %zmm5, %zmm6, %zmm3
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = <u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm7 = [u,u,u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15]
; AVX512BW-FAST-NEXT: vpshufb %zmm7, %zmm3, %zmm3
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm2, %zmm2, %zmm2
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm8 = zmm2[0,2,0,2,4,6,4,6]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = <u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm9 = [u,u,u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u]
; AVX512BW-FAST-NEXT: vpshufb %zmm9, %zmm8, %zmm8
; AVX512BW-FAST-NEXT: movl $8913032, %ecx # imm = 0x880088
; AVX512BW-FAST-NEXT: kmovd %ecx, %k1
; AVX512BW-FAST-NEXT: vmovdqu16 %zmm3, %zmm8 {%k1}
; AVX512BW-FAST-NEXT: vpermt2q %zmm4, %zmm6, %zmm1
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = <u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm3 = [u,u,0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u]
; AVX512BW-FAST-NEXT: vpshufb %zmm3, %zmm1, %zmm1
; AVX512BW-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm0, %zmm0
; AVX512BW-FAST-NEXT: vpermq {{.*#+}} zmm6 = zmm0[0,2,0,2,4,6,4,6]
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = <0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm10 = [0,8,u,u,u,u,u,u,1,9,u,u,u,u,u,u,2,10,u,u,u,u,u,u,3,11,u,u,u,u,u,u,4,12,u,u,u,u,u,u,5,13,u,u,u,u,u,u,6,14,u,u,u,u,u,u,7,15,u,u,u,u,u,u]
; AVX512BW-FAST-NEXT: vpshufb %zmm10, %zmm6, %zmm6
; AVX512BW-FAST-NEXT: movl $2228258, %ecx # imm = 0x220022
; AVX512BW-FAST-NEXT: kmovd %ecx, %k2
@@ -2235,7 +2235,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm4, %ymm0, %ymm0
; AVX2-FAST-NEXT: vinserti128 $1, %xmm12, %ymm12, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm11, %ymm9, %ymm9
; AVX2-FAST-NEXT: vmovdqa %ymm11, %ymm12
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm9[0],ymm0[1],ymm9[2,3,4],ymm0[5],ymm9[6,7,8],ymm0[9],ymm9[10,11,12],ymm0[13],ymm9[14,15]
@@ -2303,7 +2303,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,2,3,0,1,2,3,8,9,10,11,2,3,6,7,4,5,2,3,4,5,2,3,8,9,10,11,6,7,6,7]
; AVX2-FAST-NEXT: vpshufb %ymm6, %ymm5, %ymm5
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0,1,2],ymm4[3],ymm5[4,5,6],ymm4[7],ymm5[8,9,10],ymm4[11],ymm5[12,13,14],ymm4[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm8 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm8, %xmm1, %xmm5
; AVX2-FAST-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm1, %ymm1
@@ -2388,7 +2388,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm4 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm4, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm12, %ymm12, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm9, %ymm9
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm11, %ymm12
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm9[0],ymm0[1],ymm9[2,3,4],ymm0[5],ymm9[6,7,8],ymm0[9],ymm9[10,11,12],ymm0[13],ymm9[14,15]
@@ -2456,7 +2456,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm6 = [0,1,2,3,0,1,2,3,8,9,10,11,2,3,6,7,4,5,2,3,4,5,2,3,8,9,10,11,6,7,6,7]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm6, %ymm5, %ymm5
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0,1,2],ymm4[3],ymm5[4,5,6],ymm4[7],ymm5[8,9,10],ymm4[11],ymm5[12,13,14],ymm4[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm8 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm8, %xmm1, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm1, %ymm1
@@ -2759,7 +2759,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-ONLY-FAST-NEXT: vmovdqa (%rdi), %xmm13
; AVX512F-ONLY-FAST-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm13[8],xmm12[8],xmm13[9],xmm12[9],xmm13[10],xmm12[10],xmm13[11],xmm12[11],xmm13[12],xmm12[12],xmm13[13],xmm12[13],xmm13[14],xmm12[14],xmm13[15],xmm12[15]
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm1, %ymm2
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm2, %ymm2
; AVX512F-ONLY-FAST-NEXT: vmovdqa %ymm3, %ymm14
; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
@@ -2770,7 +2770,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-ONLY-FAST-NEXT: vpshufb %ymm3, %ymm0, %ymm0
; AVX512F-ONLY-FAST-NEXT: vmovdqa64 %ymm3, %ymm29
; AVX512F-ONLY-FAST-NEXT: vpmovzxwq {{.*#+}} xmm3 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
-; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX512F-ONLY-FAST-NEXT: vmovdqa {{.*#+}} xmm9 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX512F-ONLY-FAST-NEXT: vpshufb %xmm9, %xmm1, %xmm1
; AVX512F-ONLY-FAST-NEXT: vinserti128 $1, %xmm1, %ymm3, %ymm1
; AVX512F-ONLY-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2,3,4],ymm0[5],ymm1[6,7,8],ymm0[9],ymm1[10,11,12],ymm0[13],ymm1[14,15]
@@ -3117,14 +3117,14 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm4, %xmm21
; AVX512DQ-FAST-NEXT: vmovdqa64 %xmm3, %xmm22
; AVX512DQ-FAST-NEXT: vinserti128 $1, %xmm2, %ymm2, %ymm3
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm4 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %ymm4, %ymm3, %ymm3
; AVX512DQ-FAST-NEXT: vmovdqa %ymm4, %ymm5
; AVX512DQ-FAST-NEXT: vpblendw {{.*#+}} ymm1 = ymm3[0],ymm1[1],ymm3[2,3,4],ymm1[5],ymm3[6,7,8],ymm1[9],ymm3[10,11,12],ymm1[13],ymm3[14,15]
; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [0,1,0,1,4,5,2,3,4,5,2,3,12,13,14,15,0,1,4,5,4,5,6,7,4,5,6,7,12,13,14,15]
; AVX512DQ-FAST-NEXT: vpshufb %ymm3, %ymm0, %ymm0
; AVX512DQ-FAST-NEXT: vmovdqa64 %ymm3, %ymm23
-; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX512DQ-FAST-NEXT: vmovdqa {{.*#+}} xmm4 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX512DQ-FAST-NEXT: vpshufb %xmm4, %xmm2, %xmm3
; AVX512DQ-FAST-NEXT: vmovdqa %xmm4, %xmm10
; AVX512DQ-FAST-NEXT: vpmovzxwq {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero
@@ -3299,7 +3299,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-SLOW-NEXT: vpmovzxdq {{.*#+}} xmm22 = xmm22[0],zero,xmm22[1],zero
; AVX512BW-SLOW-NEXT: vinserti32x4 $1, %xmm22, %ymm21, %ymm4
; AVX512BW-SLOW-NEXT: vpblendw {{.*#+}} ymm4 = ymm10[0],ymm4[1],ymm10[2,3,4],ymm4[5],ymm10[6,7,8],ymm4[9],ymm10[10,11,12],ymm4[13],ymm10[14,15]
-; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} ymm22 = <4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u>
+; AVX512BW-SLOW-NEXT: vmovdqa64 {{.*#+}} ymm22 = [4,20,u,u,5,21,u,u,6,22,u,u,7,23,u,u]
; AVX512BW-SLOW-NEXT: vpermt2w %ymm20, %ymm22, %ymm7
; AVX512BW-SLOW-NEXT: vinserti64x4 $1, %ymm7, %zmm4, %zmm7
; AVX512BW-SLOW-NEXT: movw $-21846, %cx # imm = 0xAAAA
@@ -3403,7 +3403,7 @@ define void @store_i8_stride8_vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa64 16(%rdi), %xmm19
; AVX512BW-FAST-NEXT: vpunpckhbw {{.*#+}} xmm6 = xmm18[8],xmm16[8],xmm18[9],xmm16[9],xmm18[10],xmm16[10],xmm18[11],xmm16[11],xmm18[12],xmm16[12],xmm18[13],xmm16[13],xmm18[14],xmm16[14],xmm18[15],xmm16[15]
; AVX512BW-FAST-NEXT: vinserti32x4 $2, %xmm6, %zmm6, %zmm6
-; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = <0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa64 {{.*#+}} zmm20 = [0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u]
; AVX512BW-FAST-NEXT: vpermt2w %zmm15, %zmm20, %zmm6
; AVX512BW-FAST-NEXT: movw $-21846, %cx # imm = 0xAAAA
; AVX512BW-FAST-NEXT: kmovd %ecx, %k1
@@ -5271,7 +5271,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpshufb %ymm3, %ymm1, %ymm1
; AVX2-FAST-NEXT: vmovdqa %ymm3, %ymm10
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3],ymm1[4,5,6],ymm0[7],ymm1[8,9,10],ymm0[11],ymm1[12,13,14],ymm0[15]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm3 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm3, %xmm4, %xmm1
; AVX2-FAST-NEXT: vmovdqa %xmm3, %xmm13
; AVX2-FAST-NEXT: vpmovzxwq {{.*#+}} xmm3 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero
@@ -5314,7 +5314,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm7 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm7, %ymm15, %ymm6
; AVX2-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm0, %ymm0
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm6[1],ymm0[2,3,4],ymm6[5],ymm0[6,7,8],ymm6[9],ymm0[10,11,12],ymm6[13],ymm0[14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
@@ -5330,7 +5330,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0,1,2],ymm3[3],ymm4[4,5,6],ymm3[7],ymm4[8,9,10],ymm3[11],ymm4[12,13,14],ymm3[15]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm2[0],xmm13[0],xmm2[1],xmm13[1],xmm2[2],xmm13[2],xmm2[3],xmm13[3],xmm2[4],xmm13[4],xmm2[5],xmm13[5],xmm2[6],xmm13[6],xmm2[7],xmm13[7]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm9[0],xmm10[0],xmm9[1],xmm10[1],xmm9[2],xmm10[2],xmm9[3],xmm10[3],xmm9[4],xmm10[4],xmm9[5],xmm10[5],xmm9[6],xmm10[6],xmm9[7],xmm10[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm10 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm10, %xmm4, %xmm5
; AVX2-FAST-NEXT: vpmovzxwq {{.*#+}} xmm6 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm6, %ymm5
@@ -5381,7 +5381,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm12, %ymm14, %ymm7
; AVX2-FAST-NEXT: vinserti128 $1, %xmm9, %ymm9, %ymm8
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm14 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm14, %ymm8, %ymm8
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm7 = ymm8[0],ymm7[1],ymm8[2,3,4],ymm7[5],ymm8[6,7,8],ymm7[9],ymm8[10,11,12],ymm7[13],ymm8[14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0],ymm6[1],ymm7[2],ymm6[3],ymm7[4],ymm6[5],ymm7[6],ymm6[7]
@@ -5397,7 +5397,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0,1,2],ymm4[3],ymm5[4,5,6],ymm4[7],ymm5[8,9,10],ymm4[11],ymm5[12,13,14],ymm4[15]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm13[0],xmm11[0],xmm13[1],xmm11[1],xmm13[2],xmm11[2],xmm13[3],xmm11[3],xmm13[4],xmm11[4],xmm13[5],xmm11[5],xmm13[6],xmm11[6],xmm13[7],xmm11[7]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3],xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm11 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm11, %xmm0, %xmm2
; AVX2-FAST-NEXT: vpmovzxwq {{.*#+}} xmm6 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX2-FAST-NEXT: vinserti128 $1, %xmm2, %ymm6, %ymm2
@@ -5450,7 +5450,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm11 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-NEXT: vpshufb %ymm11, %ymm15, %ymm8
; AVX2-FAST-NEXT: vinserti128 $1, %xmm9, %ymm9, %ymm9
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm10 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %ymm10, %ymm9, %ymm9
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1],ymm9[2,3,4],ymm8[5],ymm9[6,7,8],ymm8[9],ymm9[10,11,12],ymm8[13],ymm9[14,15]
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm6 = ymm8[0],ymm6[1],ymm8[2],ymm6[3],ymm8[4],ymm6[5],ymm8[6],ymm6[7]
@@ -5466,7 +5466,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-NEXT: vpblendw {{.*#+}} ymm0 = ymm4[0,1,2],ymm0[3],ymm4[4,5,6],ymm0[7],ymm4[8,9,10],ymm0[11],ymm4[12,13,14],ymm0[15]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm14[0],xmm13[0],xmm14[1],xmm13[1],xmm14[2],xmm13[2],xmm14[3],xmm13[3],xmm14[4],xmm13[4],xmm14[5],xmm13[5],xmm14[6],xmm13[6],xmm14[7],xmm13[7]
; AVX2-FAST-NEXT: vpunpcklbw {{.*#+}} xmm3 = xmm3[0],xmm5[0],xmm3[1],xmm5[1],xmm3[2],xmm5[2],xmm3[3],xmm5[3],xmm3[4],xmm5[4],xmm3[5],xmm5[5],xmm3[6],xmm5[6],xmm3[7],xmm5[7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} xmm14 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-NEXT: vpshufb %xmm14, %xmm3, %xmm5
; AVX2-FAST-NEXT: vpmovzxwq {{.*#+}} xmm8 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero
; AVX2-FAST-NEXT: vinserti128 $1, %xmm5, %ymm8, %ymm5
@@ -5595,7 +5595,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm3, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %ymm3, %ymm10
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3],ymm1[4,5,6],ymm0[7],ymm1[8,9,10],ymm0[11],ymm1[12,13,14],ymm0[15]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm3 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm3, %xmm4, %xmm1
; AVX2-FAST-PERLANE-NEXT: vmovdqa %xmm3, %xmm13
; AVX2-FAST-PERLANE-NEXT: vpmovzxwq {{.*#+}} xmm3 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero
@@ -5638,7 +5638,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm7 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm7, %ymm15, %ymm6
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm0[0],ymm6[1],ymm0[2,3,4],ymm6[5],ymm0[6,7,8],ymm6[9],ymm0[10,11,12],ymm6[13],ymm0[14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
@@ -5654,7 +5654,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm3 = ymm4[0,1,2],ymm3[3],ymm4[4,5,6],ymm3[7],ymm4[8,9,10],ymm3[11],ymm4[12,13,14],ymm3[15]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm2 = xmm2[0],xmm13[0],xmm2[1],xmm13[1],xmm2[2],xmm13[2],xmm2[3],xmm13[3],xmm2[4],xmm13[4],xmm2[5],xmm13[5],xmm2[6],xmm13[6],xmm2[7],xmm13[7]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm9[0],xmm10[0],xmm9[1],xmm10[1],xmm9[2],xmm10[2],xmm9[3],xmm10[3],xmm9[4],xmm10[4],xmm9[5],xmm10[5],xmm9[6],xmm10[6],xmm9[7],xmm10[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm10 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm10, %xmm4, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpmovzxwq {{.*#+}} xmm6 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm6, %ymm5
@@ -5705,7 +5705,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm12 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm12, %ymm14, %ymm7
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm9, %ymm9, %ymm8
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm14 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm14, %ymm8, %ymm8
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm7 = ymm8[0],ymm7[1],ymm8[2,3,4],ymm7[5],ymm8[6,7,8],ymm7[9],ymm8[10,11,12],ymm7[13],ymm8[14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm6 = ymm7[0],ymm6[1],ymm7[2],ymm6[3],ymm7[4],ymm6[5],ymm7[6],ymm6[7]
@@ -5721,7 +5721,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm4 = ymm5[0,1,2],ymm4[3],ymm5[4,5,6],ymm4[7],ymm5[8,9,10],ymm4[11],ymm5[12,13,14],ymm4[15]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm5 = xmm13[0],xmm11[0],xmm13[1],xmm11[1],xmm13[2],xmm11[2],xmm13[3],xmm11[3],xmm13[4],xmm11[4],xmm13[5],xmm11[5],xmm13[6],xmm11[6],xmm13[7],xmm11[7]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm0 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3],xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm11 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm11, %xmm0, %xmm2
; AVX2-FAST-PERLANE-NEXT: vpmovzxwq {{.*#+}} xmm6 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm2, %ymm6, %ymm2
@@ -5774,7 +5774,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm11 = [8,9,8,9,4,5,6,7,12,13,10,11,12,13,10,11,8,9,12,13,4,5,6,7,12,13,14,15,12,13,14,15]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm11, %ymm15, %ymm8
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm9, %ymm9, %ymm9
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm10 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %ymm10, %ymm9, %ymm9
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm8 = ymm9[0],ymm8[1],ymm9[2,3,4],ymm8[5],ymm9[6,7,8],ymm8[9],ymm9[10,11,12],ymm8[13],ymm9[14,15]
; AVX2-FAST-PERLANE-NEXT: vpblendd {{.*#+}} ymm6 = ymm8[0],ymm6[1],ymm8[2],ymm6[3],ymm8[4],ymm6[5],ymm8[6],ymm6[7]
@@ -5790,7 +5790,7 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX2-FAST-PERLANE-NEXT: vpblendw {{.*#+}} ymm0 = ymm4[0,1,2],ymm0[3],ymm4[4,5,6],ymm0[7],ymm4[8,9,10],ymm0[11],ymm4[12,13,14],ymm0[15]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm4 = xmm14[0],xmm13[0],xmm14[1],xmm13[1],xmm14[2],xmm13[2],xmm14[3],xmm13[3],xmm14[4],xmm13[4],xmm14[5],xmm13[5],xmm14[6],xmm13[6],xmm14[7],xmm13[7]
; AVX2-FAST-PERLANE-NEXT: vpunpcklbw {{.*#+}} xmm3 = xmm3[0],xmm5[0],xmm3[1],xmm5[1],xmm3[2],xmm5[2],xmm3[3],xmm5[3],xmm3[4],xmm5[4],xmm3[5],xmm5[5],xmm3[6],xmm5[6],xmm3[7],xmm5[7]
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm14 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} xmm14 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT: vpshufb %xmm14, %xmm3, %xmm5
; AVX2-FAST-PERLANE-NEXT: vpmovzxwq {{.*#+}} xmm8 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero
; AVX2-FAST-PERLANE-NEXT: vinserti128 $1, %xmm5, %ymm8, %ymm5
@@ -6511,12 +6511,12 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512F-FAST-NEXT: vmovdqa (%rsi), %xmm12
; AVX512F-FAST-NEXT: vmovdqa (%rdi), %xmm9
; AVX512F-FAST-NEXT: vpunpckhbw {{.*#+}} xmm0 = xmm9[8],xmm12[8],xmm9[9],xmm12[9],xmm9[10],xmm12[10],xmm9[11],xmm12[11],xmm9[12],xmm12[12],xmm9[13],xmm12[13],xmm9[14],xmm12[14],xmm9[15],xmm12[15]
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} xmm7 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpmovzxwq {{.*#+}} xmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX512F-FAST-NEXT: vpshufb %xmm7, %xmm0, %xmm2
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm2, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm5 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX512F-FAST-NEXT: vpshufb %ymm5, %ymm0, %ymm0
; AVX512F-FAST-NEXT: vinserti64x4 $1, %ymm0, %zmm1, %zmm4
; AVX512F-FAST-NEXT: vpbroadcastq {{.*#+}} zmm3 = [65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535,65535,0,65535,65535]
@@ -6918,9 +6918,9 @@ define void @store_i8_stride8_vf64(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vec
; AVX512BW-FAST-NEXT: vmovdqa64 48(%rdi), %xmm28
; AVX512BW-FAST-NEXT: vpunpckhbw {{.*#+}} xmm8 = xmm11[8],xmm10[8],xmm11[9],xmm10[9],xmm11[10],xmm10[10],xmm11[11],xmm10[11],xmm11[12],xmm10[12],xmm11[13],xmm10[13],xmm11[14],xmm10[14],xmm11[15],xmm10[15]
; AVX512BW-FAST-NEXT: vinserti128 $1, %xmm8, %ymm8, %ymm13
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = <8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} ymm12 = [8,9,u,u,u,u,u,u,10,11,u,u,u,u,u,u,12,13,u,u,u,u,u,u,14,15,u,u,u,u,u,u]
; AVX512BW-FAST-NEXT: vpshufb %ymm12, %ymm13, %ymm26
-; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = <4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u>
+; AVX512BW-FAST-NEXT: vmovdqa {{.*#+}} xmm13 = [4,5,u,u,u,u,u,u,6,7,u,u,u,u,u,u]
; AVX512BW-FAST-NEXT: vpshufb %xmm13, %xmm8, %xmm27
; AVX512BW-FAST-NEXT: vpmovzxwq {{.*#+}} xmm8 = xmm8[0],zero,zero,zero,xmm8[1],zero,zero,zero
; AVX512BW-FAST-NEXT: vinserti32x4 $1, %xmm27, %ymm8, %ymm8
diff --git a/llvm/test/CodeGen/X86/vector-mulfix-legalize.ll b/llvm/test/CodeGen/X86/vector-mulfix-legalize.ll
index 008188b52c2005e..82afa15079182f6 100644
--- a/llvm/test/CodeGen/X86/vector-mulfix-legalize.ll
+++ b/llvm/test/CodeGen/X86/vector-mulfix-legalize.ll
@@ -13,7 +13,7 @@ declare <4 x i16> @llvm.umul.fix.sat.v4i16(<4 x i16>, <4 x i16>, i32 immarg)
define <4 x i16> @smulfix(<4 x i16> %a) {
; CHECK-LABEL: smulfix:
; CHECK: # %bb.0:
-; CHECK-NEXT: movdqa {{.*#+}} xmm1 = <1,2,3,4,u,u,u,u>
+; CHECK-NEXT: movdqa {{.*#+}} xmm1 = [1,2,3,4,u,u,u,u]
; CHECK-NEXT: movdqa %xmm0, %xmm2
; CHECK-NEXT: pmullw %xmm1, %xmm2
; CHECK-NEXT: psrlw $15, %xmm2
@@ -28,7 +28,7 @@ define <4 x i16> @smulfix(<4 x i16> %a) {
define <4 x i16> @umulfix(<4 x i16> %a) {
; CHECK-LABEL: umulfix:
; CHECK: # %bb.0:
-; CHECK-NEXT: movdqa {{.*#+}} xmm1 = <1,2,3,4,u,u,u,u>
+; CHECK-NEXT: movdqa {{.*#+}} xmm1 = [1,2,3,4,u,u,u,u]
; CHECK-NEXT: movdqa %xmm0, %xmm2
; CHECK-NEXT: pmullw %xmm1, %xmm2
; CHECK-NEXT: psrlw $15, %xmm2
diff --git a/llvm/test/CodeGen/X86/vector-partial-undef.ll b/llvm/test/CodeGen/X86/vector-partial-undef.ll
index ffd95b38dc31bbd..fd41fd53e3be143 100644
--- a/llvm/test/CodeGen/X86/vector-partial-undef.ll
+++ b/llvm/test/CodeGen/X86/vector-partial-undef.ll
@@ -137,7 +137,7 @@ define <8 x i32> @xor_undef_elts_alt(<4 x i32> %x) {
; SSE-LABEL: xor_undef_elts_alt:
; SSE: # %bb.0:
; SSE-NEXT: movaps %xmm0, %xmm1
-; SSE-NEXT: movaps {{.*#+}} xmm2 = <u,u,44,12>
+; SSE-NEXT: movaps {{.*#+}} xmm2 = [u,u,44,12]
; SSE-NEXT: xorps %xmm0, %xmm2
; SSE-NEXT: xorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
; SSE-NEXT: movaps %xmm1, %xmm0
diff --git a/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll b/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll
index 03e39e71aaaf162..dbdc3e09fcef081 100644
--- a/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll
+++ b/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll
@@ -864,7 +864,7 @@ define i16 @test_v4i16_v4i8(<4 x i16> %a0) {
;
; SSE41-LABEL: test_v4i16_v4i8:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,32768,16384,8192,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,32768,16384,8192,u,u,u,u]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: pshufd {{.*#+}} xmm0 = xmm1[1,1,1,1]
diff --git a/llvm/test/CodeGen/X86/vector-replicaton-i1-mask.ll b/llvm/test/CodeGen/X86/vector-replicaton-i1-mask.ll
index 7f828fc293caac3..a3d3ed6a4b243bd 100644
--- a/llvm/test/CodeGen/X86/vector-replicaton-i1-mask.ll
+++ b/llvm/test/CodeGen/X86/vector-replicaton-i1-mask.ll
@@ -481,7 +481,7 @@ define void @mask_replication_factor3_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpcmpeqd %ymm0, %ymm0, %ymm0
; AVX512F-ONLY-NEXT: vmovdqa32 %ymm0, %ymm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,0,1,1,1,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,1,1,u,u]
; AVX512F-ONLY-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX512F-ONLY-NEXT: vpslld $31, %ymm0, %ymm0
; AVX512F-ONLY-NEXT: movb $63, %al
@@ -498,7 +498,7 @@ define void @mask_replication_factor3_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovb (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %ymm0
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,0,1,1,1,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,1,1,u,u]
; AVX512DQ-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512DQ-NEXT: movb $63, %al
@@ -516,7 +516,7 @@ define void @mask_replication_factor3_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW-NEXT: kmovw (%rdi), %k1
; AVX512BW-NEXT: vpcmpeqd %ymm0, %ymm0, %ymm0
; AVX512BW-NEXT: vmovdqa32 %ymm0, %ymm0 {%k1} {z}
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,0,1,1,1,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,1,1,u,u]
; AVX512BW-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX512BW-NEXT: vpslld $31, %ymm0, %ymm0
; AVX512BW-NEXT: movb $63, %al
@@ -542,7 +542,7 @@ define void @mask_replication_factor3_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY: # %bb.0:
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u]
; AVX512F-ONLY-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512F-ONLY-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512F-ONLY-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -558,7 +558,7 @@ define void @mask_replication_factor3_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovw (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u]
; AVX512DQ-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512DQ-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -574,7 +574,7 @@ define void @mask_replication_factor3_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovw (%rdi), %k1
; AVX512BW-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,1,1,1,2,2,2,3,3,3,u,u,u,u]
; AVX512BW-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512BW-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -646,7 +646,7 @@ define void @mask_replication_factor3_vf8(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovw (%rdi), %k0
; AVX512BW-NEXT: vpmovm2w %k0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,1,1,1,2,2,2,3,3,3,4,4,4,5,5,5,6,6,6,7,7,7,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,1,1,1,2,2,2,3,3,3,4,4,4,5,5,5,6,6,6,7,7,7,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: movl $16777215, %eax # imm = 0xFFFFFF
@@ -3121,7 +3121,7 @@ define void @mask_replication_factor5_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY: # %bb.0:
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u]
; AVX512F-ONLY-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512F-ONLY-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512F-ONLY-NEXT: movw $1023, %ax # imm = 0x3FF
@@ -3138,7 +3138,7 @@ define void @mask_replication_factor5_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovw (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512DQ-NEXT: movw $1023, %ax # imm = 0x3FF
@@ -3155,7 +3155,7 @@ define void @mask_replication_factor5_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovw (%rdi), %k1
; AVX512BW-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,1,1,1,1,1,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512BW-NEXT: movw $1023, %ax # imm = 0x3FF
@@ -3219,7 +3219,7 @@ define void @mask_replication_factor5_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovd (%rdi), %k0
; AVX512BW-NEXT: vpmovm2w %k0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: movl $1048575, %eax # imm = 0xFFFFF
@@ -3324,7 +3324,7 @@ define void @mask_replication_factor5_vf8(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512VBMI-ONLY: # %bb.0:
; AVX512VBMI-ONLY-NEXT: kmovw (%rdi), %k0
; AVX512VBMI-ONLY-NEXT: vpmovm2b %k0, %zmm0
-; AVX512VBMI-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4,5,5,5,5,5,6,6,6,6,6,7,7,7,7,7,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4,5,5,5,5,5,6,6,6,6,6,7,7,7,7,7,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-ONLY-NEXT: vpermb %zmm0, %zmm1, %zmm0
; AVX512VBMI-ONLY-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512VBMI-ONLY-NEXT: movabsq $1099511627775, %rax # imm = 0xFFFFFFFFFF
@@ -5857,7 +5857,7 @@ define void @mask_replication_factor6_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY: # %bb.0:
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u]
; AVX512F-ONLY-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512F-ONLY-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512F-ONLY-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -5873,7 +5873,7 @@ define void @mask_replication_factor6_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovw (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u]
; AVX512DQ-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512DQ-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -5889,7 +5889,7 @@ define void @mask_replication_factor6_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovw (%rdi), %k1
; AVX512BW-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,1,1,1,1,1,1,u,u,u,u]
; AVX512BW-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512BW-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -5994,7 +5994,7 @@ define void @mask_replication_factor6_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovd (%rdi), %k0
; AVX512BW-NEXT: vpmovm2w %k0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,1,1,1,1,1,1,2,2,2,2,2,2,3,3,3,3,3,3,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,1,1,1,1,1,1,2,2,2,2,2,2,3,3,3,3,3,3,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: movl $16777215, %eax # imm = 0xFFFFFF
@@ -9071,7 +9071,7 @@ define void @mask_replication_factor7_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY: # %bb.0:
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u]
; AVX512F-ONLY-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512F-ONLY-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512F-ONLY-NEXT: movw $16383, %ax # imm = 0x3FFF
@@ -9089,7 +9089,7 @@ define void @mask_replication_factor7_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovw (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u]
; AVX512DQ-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512DQ-NEXT: movw $16383, %ax # imm = 0x3FFF
@@ -9107,7 +9107,7 @@ define void @mask_replication_factor7_vf2(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovw (%rdi), %k1
; AVX512BW-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,0,1,1,1,1,1,1,1,u,u]
; AVX512BW-NEXT: vpermd %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpslld $31, %zmm0, %zmm0
; AVX512BW-NEXT: movw $16383, %ax # imm = 0x3FFF
@@ -9134,7 +9134,7 @@ define void @mask_replication_factor7_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512F-ONLY: # %bb.0:
; AVX512F-ONLY-NEXT: kmovw (%rdi), %k1
; AVX512F-ONLY-NEXT: vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
-; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u>
+; AVX512F-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u]
; AVX512F-ONLY-NEXT: vpermd %zmm0, %zmm1, %zmm1
; AVX512F-ONLY-NEXT: vpslld $31, %zmm1, %zmm1
; AVX512F-ONLY-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -9155,7 +9155,7 @@ define void @mask_replication_factor7_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: kmovw (%rdi), %k0
; AVX512DQ-NEXT: vpmovm2d %k0, %zmm0
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = <2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm1 = [2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u]
; AVX512DQ-NEXT: vpermd %zmm0, %zmm1, %zmm1
; AVX512DQ-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX512DQ-NEXT: movw $4095, %ax # imm = 0xFFF
@@ -9176,7 +9176,7 @@ define void @mask_replication_factor7_vf4(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: kmovd (%rdi), %k0
; AVX512BW-NEXT: vpmovm2w %k0, %zmm0
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,3,3,3,u,u,u,u]
; AVX512BW-NEXT: vpermw %zmm0, %zmm1, %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: movl $268435455, %eax # imm = 0xFFFFFFF
@@ -9362,7 +9362,7 @@ define void @mask_replication_factor7_vf8(ptr %in.maskvec, ptr %in.vec, ptr %out
; AVX512VBMI-ONLY: # %bb.0:
; AVX512VBMI-ONLY-NEXT: kmovw (%rdi), %k0
; AVX512VBMI-ONLY-NEXT: vpmovm2b %k0, %zmm0
-; AVX512VBMI-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = <0,0,0,0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,3,3,3,4,4,4,4,4,4,4,5,5,5,5,5,5,5,6,6,6,6,6,6,6,7,7,7,7,7,7,7,u,u,u,u,u,u,u,u>
+; AVX512VBMI-ONLY-NEXT: vmovdqa64 {{.*#+}} zmm1 = [0,0,0,0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,2,2,2,3,3,3,3,3,3,3,4,4,4,4,4,4,4,5,5,5,5,5,5,5,6,6,6,6,6,6,6,7,7,7,7,7,7,7,u,u,u,u,u,u,u,u]
; AVX512VBMI-ONLY-NEXT: vpermb %zmm0, %zmm1, %zmm0
; AVX512VBMI-ONLY-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512VBMI-ONLY-NEXT: movabsq $72057594037927935, %rax # imm = 0xFFFFFFFFFFFFFF
diff --git a/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll b/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll
index e6b5998d965f067..cdc2bbe56aee4ca 100644
--- a/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll
@@ -1420,7 +1420,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) nounwind {
;
; SSE41-LABEL: constant_shift_v8i16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,u,16384,8192,4096,2048,1024,512>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,u,16384,8192,4096,2048,1024,512]
; SSE41-NEXT: pmulhw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: psraw $1, %xmm0
diff --git a/llvm/test/CodeGen/X86/vector-shift-ashr-sub128.ll b/llvm/test/CodeGen/X86/vector-shift-ashr-sub128.ll
index dfba0d985c1f04f..e15be224bec8cd5 100644
--- a/llvm/test/CodeGen/X86/vector-shift-ashr-sub128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-ashr-sub128.ll
@@ -1786,7 +1786,7 @@ define <4 x i16> @constant_shift_v4i16(<4 x i16> %a) nounwind {
;
; SSE41-LABEL: constant_shift_v4i16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,u,16384,8192,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,u,16384,8192,u,u,u,u]
; SSE41-NEXT: pmulhw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: psraw $1, %xmm0
diff --git a/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll b/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll
index 7a2dcd1c8ca8c99..43da0f15abaeb90 100644
--- a/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll
@@ -1173,7 +1173,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) nounwind {
;
; SSE41-LABEL: constant_shift_v8i16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,32768,16384,8192,4096,2048,1024,512>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,32768,16384,8192,4096,2048,1024,512]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/vector-shift-lshr-512.ll b/llvm/test/CodeGen/X86/vector-shift-lshr-512.ll
index f647208a8000ea4..efd73b4ca132bbf 100644
--- a/llvm/test/CodeGen/X86/vector-shift-lshr-512.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-lshr-512.ll
@@ -288,7 +288,7 @@ define <32 x i16> @constant_shift_v32i16(<32 x i16> %a) nounwind {
; AVX512DQ-LABEL: constant_shift_v32i16:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm1
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm2 = <u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm2 = [u,32768,16384,8192,4096,2048,1024,512,256,128,64,32,16,8,4,2]
; AVX512DQ-NEXT: vpmulhuw %ymm2, %ymm1, %ymm3
; AVX512DQ-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm3[1,2,3,4,5,6,7]
; AVX512DQ-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0,1,2,3],ymm3[4,5,6,7]
diff --git a/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll b/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll
index 510ae15ba0960a1..f340615464cfa75 100644
--- a/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll
@@ -1486,7 +1486,7 @@ define <4 x i16> @constant_shift_v4i16(<4 x i16> %a) nounwind {
;
; SSE41-LABEL: constant_shift_v4i16:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm1 = <u,32768,16384,8192,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [u,32768,16384,8192,u,u,u,u]
; SSE41-NEXT: pmulhuw %xmm0, %xmm1
; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4,5,6,7]
; SSE41-NEXT: retq
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
index ec75631a9b5ed29..b40be9452ddd77a 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
@@ -1003,7 +1003,7 @@ define <16 x i8> @shuffle_v16i8_01_03_05_07_09_11_13_15_17_19_21_23_25_27_29_31(
;
; SSSE3-LABEL: shuffle_v16i8_01_03_05_07_09_11_13_15_17_19_21_23_25_27_29_31:
; SSSE3: # %bb.0:
-; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSSE3-NEXT: pshufb %xmm2, %xmm1
; SSSE3-NEXT: pshufb %xmm2, %xmm0
; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
@@ -1011,7 +1011,7 @@ define <16 x i8> @shuffle_v16i8_01_03_05_07_09_11_13_15_17_19_21_23_25_27_29_31(
;
; SSE41-LABEL: shuffle_v16i8_01_03_05_07_09_11_13_15_17_19_21_23_25_27_29_31:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm2 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pshufb %xmm2, %xmm1
; SSE41-NEXT: pshufb %xmm2, %xmm0
; SSE41-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
@@ -1819,7 +1819,7 @@ define <16 x i8> @shuffle_v16i8_uu_10_02_07_22_14_07_02_18_03_01_14_18_09_11_00(
;
; AVX512VLVBMI-LABEL: shuffle_v16i8_uu_10_02_07_22_14_07_02_18_03_01_14_18_09_11_00:
; AVX512VLVBMI: # %bb.0: # %entry
-; AVX512VLVBMI-NEXT: vmovdqa {{.*#+}} xmm2 = <u,10,2,7,22,14,7,2,18,3,1,14,18,9,11,0>
+; AVX512VLVBMI-NEXT: vmovdqa {{.*#+}} xmm2 = [u,10,2,7,22,14,7,2,18,3,1,14,18,9,11,0]
; AVX512VLVBMI-NEXT: vpermt2b %xmm1, %xmm2, %xmm0
; AVX512VLVBMI-NEXT: retq
;
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
index f73081cfc404feb..5b4f15b51ec00ae 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
@@ -3510,7 +3510,7 @@ define <16 x i16> @shuffle_v16i16_00_00_00_00_04_04_04_12_08_08_08_08_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_00_00_00_04_04_04_12_08_08_08_08_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,6,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,6,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,0,1,0,1,4,5,4,5,4,5,8,9,16,17,16,17,16,17,16,17,20,21,20,21,20,21,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -3569,7 +3569,7 @@ define <16 x i16> @shuffle_v16i16_uu_00_uu_01_uu_02_uu_11_uu_08_uu_09_uu_10_uu_1
;
; AVX512VL-LABEL: shuffle_v16i16_uu_00_uu_01_uu_02_uu_11_uu_08_uu_09_uu_10_uu_11:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <u,0,u,1,u,2,u,11,u,8,u,9,u,10,u,11>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [u,0,u,1,u,2,u,11,u,8,u,9,u,10,u,11]
; AVX512VL-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512VL-NEXT: retq
;
@@ -3609,7 +3609,7 @@ define <16 x i16> @shuffle_v16i16_uu_04_uu_05_uu_06_uu_15_uu_12_uu_13_uu_14_uu_1
;
; AVX512VL-LABEL: shuffle_v16i16_uu_04_uu_05_uu_06_uu_15_uu_12_uu_13_uu_14_uu_15:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <u,4,u,5,u,6,u,15,u,12,u,13,u,14,u,15>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [u,4,u,5,u,6,u,15,u,12,u,13,u,14,u,15]
; AVX512VL-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512VL-NEXT: retq
;
@@ -3703,7 +3703,7 @@ define <16 x i16> @shuffle_v16i16_04_04_04_04_00_00_00_08_12_12_12_12_08_08_08_0
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_04_04_04_04_00_00_00_08_12_12_12_12_08_08_08_08:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,4,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,4,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,0,1,0,1,4,5,4,5,4,5,8,9,16,17,16,17,16,17,16,17,20,21,20,21,20,21,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -3940,7 +3940,7 @@ define <16 x i16> @shuffle_v16i16_01_00_05_04_05_04_01_08_09_08_13_12_13_12_09_0
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_01_00_05_04_05_04_01_08_09_08_13_12_13_12_09_08:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5,6,7,4,5,2,3,8,9,18,19,16,17,22,23,20,21,22,23,20,21,18,19,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -3999,7 +3999,7 @@ define <16 x i16> @shuffle_v16i16_05_04_01_00_05_04_01_08_13_12_09_08_13_12_09_0
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_05_04_01_00_05_04_01_08_13_12_09_08_13_12_09_08:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,4,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,4,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5,2,3,0,1,6,7,8,9,18,19,16,17,22,23,20,21,18,19,16,17,22,23,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -4058,7 +4058,7 @@ define <16 x i16> @shuffle_v16i16_05_04_01_00_01_00_05_12_13_12_09_08_09_08_13_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_05_04_01_00_01_00_05_12_13_12_09_08_09_08_13_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,6,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,6,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5,6,7,4,5,2,3,8,9,18,19,16,17,22,23,20,21,22,23,20,21,18,19,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4117,7 +4117,7 @@ define <16 x i16> @shuffle_v16i16_00_04_04_00_00_04_04_08_08_12_12_08_08_12_12_0
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_04_04_00_00_04_04_08_08_12_12_08_08_12_12_08:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,4,5,0,1,0,1,4,5,4,5,8,9,16,17,20,21,20,21,16,17,16,17,20,21,20,21,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4176,7 +4176,7 @@ define <16 x i16> @shuffle_v16i16_04_00_00_04_04_00_00_12_12_08_08_12_12_08_08_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_04_00_00_04_04_00_00_12_12_08_08_12_12_08_08_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,6,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,6,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,4,5,0,1,0,1,4,5,4,5,8,9,16,17,20,21,20,21,16,17,16,17,20,21,20,21,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4417,7 +4417,7 @@ define <16 x i16> @shuffle_v16i16_00_00_04_04_04_04_04_12_08_08_12_12_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_00_04_04_04_04_04_12_08_08_12_12_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,6,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,6,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,4,5,4,5,4,5,4,5,4,5,8,9,16,17,16,17,20,21,20,21,20,21,20,21,20,21,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -4476,7 +4476,7 @@ define <16 x i16> @shuffle_v16i16_04_04_00_00_04_04_04_12_12_12_08_08_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_04_04_00_00_04_04_04_12_12_12_08_08_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,6,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,6,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,4,5,4,5,0,1,0,1,0,1,8,9,16,17,16,17,20,21,20,21,16,17,16,17,16,17,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4535,7 +4535,7 @@ define <16 x i16> @shuffle_v16i16_00_04_04_00_04_04_04_12_08_12_12_08_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_04_04_00_04_04_04_12_08_12_12_08_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,6,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,6,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,4,5,0,1,4,5,4,5,4,5,8,9,16,17,20,21,20,21,16,17,20,21,20,21,20,21,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -4594,7 +4594,7 @@ define <16 x i16> @shuffle_v16i16_00_04_04_00_00_00_00_08_08_12_12_08_08_08_08_0
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_04_04_00_00_00_00_08_08_12_12_08_08_08_08_08:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,4,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,4,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,4,5,0,1,0,1,0,1,0,1,8,9,16,17,20,21,20,21,16,17,16,17,16,17,16,17,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4654,7 +4654,7 @@ define <16 x i16> @shuffle_v16i16_00_04_04_00_04_05_06_15_08_12_12_08_12_13_14_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_04_04_00_04_05_06_15_08_12_12_08_12_13_14_15:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,3,7,4,6,7,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,3,7,4,6,7,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,4,5,0,1,4,5,6,7,8,9,14,15,16,17,20,21,20,21,16,17,20,21,22,23,24,25,26,27]
; AVX2-FAST-ALL-NEXT: retq
@@ -4713,7 +4713,7 @@ define <16 x i16> @shuffle_v16i16_00_uu_04_04_04_04_04_12_08_uu_12_12_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_uu_04_04_04_04_04_12_08_uu_12_12_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,2,6,u,4,6,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,2,6,u,4,6,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,u,u,4,5,4,5,4,5,4,5,4,5,8,9,16,17,u,u,20,21,20,21,20,21,20,21,20,21,20,21]
; AVX2-FAST-ALL-NEXT: retq
@@ -4728,7 +4728,7 @@ define <16 x i16> @shuffle_v16i16_00_uu_04_04_04_04_04_12_08_uu_12_12_12_12_12_1
;
; AVX512VL-LABEL: shuffle_v16i16_00_uu_04_04_04_04_04_12_08_uu_12_12_12_12_12_12:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,4,4,4,4,4,12,8,u,12,12,12,12,12,12>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,u,4,4,4,4,4,12,8,u,12,12,12,12,12,12]
; AVX512VL-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512VL-NEXT: retq
;
@@ -4772,7 +4772,7 @@ define <16 x i16> @shuffle_v16i16_04_04_uu_00_04_04_04_12_12_12_uu_08_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_04_04_uu_00_04_04_04_12_12_12_uu_08_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,6,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,6,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,0,1,u,u,4,5,0,1,0,1,0,1,8,9,16,17,16,17,u,u,20,21,16,17,16,17,16,17,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4787,7 +4787,7 @@ define <16 x i16> @shuffle_v16i16_04_04_uu_00_04_04_04_12_12_12_uu_08_12_12_12_1
;
; AVX512VL-LABEL: shuffle_v16i16_04_04_uu_00_04_04_04_12_12_12_uu_08_12_12_12_12:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <4,4,u,0,4,4,4,12,12,12,u,8,12,12,12,12>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [4,4,u,0,4,4,4,12,12,12,u,8,12,12,12,12]
; AVX512VL-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512VL-NEXT: retq
;
@@ -4831,7 +4831,7 @@ define <16 x i16> @shuffle_v16i16_uu_04_04_00_04_04_04_12_uu_12_12_08_12_12_12_1
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_uu_04_04_00_04_04_04_12_uu_12_12_08_12_12_12_12:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = <2,0,6,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,0,6,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,0,1,0,1,4,5,0,1,0,1,0,1,8,9,u,u,16,17,16,17,20,21,16,17,16,17,16,17,16,17]
; AVX2-FAST-ALL-NEXT: retq
@@ -4846,7 +4846,7 @@ define <16 x i16> @shuffle_v16i16_uu_04_04_00_04_04_04_12_uu_12_12_08_12_12_12_1
;
; AVX512VL-LABEL: shuffle_v16i16_uu_04_04_00_04_04_04_12_uu_12_12_08_12_12_12_12:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <u,4,4,0,4,4,4,12,u,12,12,8,12,12,12,12>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [u,4,4,0,4,4,4,12,u,12,12,8,12,12,12,12]
; AVX512VL-NEXT: vpermw %ymm0, %ymm1, %ymm0
; AVX512VL-NEXT: retq
;
@@ -5360,7 +5360,7 @@ define <16 x i16> @shuffle_v16i16_00_16_01_17_06_22_07_31_08_24_09_25_14_30_15_3
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_16_01_17_06_22_07_31_08_24_09_25_14_30_15_31:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = <0,3,7,u,4,7,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = [0,3,7,u,4,7,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5,10,11,u,u,u,u,u,u,u,u,16,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-ALL-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,3,2,3,4,7,6,7]
@@ -5431,7 +5431,7 @@ define <16 x i16> @shuffle_v16i16_00_20_01_21_06_16_07_25_08_28_09_29_14_24_15_2
;
; AVX2-FAST-ALL-LABEL: shuffle_v16i16_00_20_01_21_06_16_07_25_08_28_09_29_14_24_15_25:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = <2,0,4,u,6,4,u,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = [2,0,4,u,6,4,u,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[0,1,2,3,4,5,10,11,u,u,u,u,u,u,u,u,16,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u]
; AVX2-FAST-ALL-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,3,2,3,4,7,6,7]
@@ -5721,7 +5721,7 @@ define <16 x i16> @shuffle_v16i16_04_04_03_18_uu_uu_uu_uu_12_12_11_26_uu_uu_uu_u
;
; AVX512VL-LABEL: shuffle_v16i16_04_04_03_18_uu_uu_uu_uu_12_12_11_26_uu_uu_uu_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <4,4,3,18,u,u,u,u,12,12,11,26,u,u,u,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [4,4,3,18,u,u,u,u,12,12,11,26,u,u,u,u]
; AVX512VL-NEXT: vpermt2w %ymm1, %ymm2, %ymm0
; AVX512VL-NEXT: retq
;
@@ -5766,7 +5766,7 @@ define <16 x i16> @shuffle_v16i16_00_03_02_21_uu_uu_uu_uu_08_11_10_29_uu_uu_uu_u
;
; AVX512VL-LABEL: shuffle_v16i16_00_03_02_21_uu_uu_uu_uu_08_11_10_29_uu_uu_uu_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <0,3,2,21,u,u,u,u,8,11,10,29,u,u,u,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [0,3,2,21,u,u,u,u,8,11,10,29,u,u,u,u]
; AVX512VL-NEXT: vpermt2w %ymm1, %ymm2, %ymm0
; AVX512VL-NEXT: retq
;
@@ -5818,7 +5818,7 @@ define <16 x i16> @shuffle_v16i16_00_01_02_21_uu_uu_uu_uu_08_09_10_29_uu_uu_uu_u
;
; AVX512VL-LABEL: shuffle_v16i16_00_01_02_21_uu_uu_uu_uu_08_09_10_29_uu_uu_uu_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,2,21,u,u,u,u,8,9,10,29,u,u,u,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,2,21,u,u,u,u,8,9,10,29,u,u,u,u]
; AVX512VL-NEXT: vpermt2w %ymm1, %ymm2, %ymm0
; AVX512VL-NEXT: retq
;
@@ -5861,7 +5861,7 @@ define <16 x i16> @shuffle_v16i16_uu_uu_uu_uu_20_21_22_11_uu_uu_uu_uu_28_29_30_1
;
; AVX512VL-LABEL: shuffle_v16i16_uu_uu_uu_uu_20_21_22_11_uu_uu_uu_uu_28_29_30_11:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,u,4,5,6,27,u,u,u,u,12,13,14,27>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,u,4,5,6,27,u,u,u,u,12,13,14,27]
; AVX512VL-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512VL-NEXT: vmovdqa %ymm2, %ymm0
; AVX512VL-NEXT: retq
@@ -5905,7 +5905,7 @@ define <16 x i16> @shuffle_v16i16_20_21_22_03_uu_uu_uu_uu_28_29_30_11_uu_uu_uu_u
;
; AVX512VL-LABEL: shuffle_v16i16_20_21_22_03_uu_uu_uu_uu_28_29_30_11_uu_uu_uu_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <4,5,6,19,u,u,u,u,12,13,14,27,u,u,u,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [4,5,6,19,u,u,u,u,12,13,14,27,u,u,u,u]
; AVX512VL-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512VL-NEXT: vmovdqa %ymm2, %ymm0
; AVX512VL-NEXT: retq
@@ -6063,7 +6063,7 @@ define <16 x i16> @shuffle_v16i16_uu_uu_uu_01_uu_05_07_25_uu_uu_uu_09_uu_13_15_2
;
; AVX512VL-LABEL: shuffle_v16i16_uu_uu_uu_01_uu_05_07_25_uu_uu_uu_09_uu_13_15_25:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,u,1,u,5,7,25,u,u,u,9,u,13,15,25>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,u,1,u,5,7,25,u,u,u,9,u,13,15,25]
; AVX512VL-NEXT: vpermt2w %ymm1, %ymm2, %ymm0
; AVX512VL-NEXT: retq
;
@@ -6115,7 +6115,7 @@ define <16 x i16> @shuffle_v16i16_uu_uu_04_uu_16_18_20_uu_uu_uu_12_uu_24_26_28_u
;
; AVX512VL-LABEL: shuffle_v16i16_uu_uu_04_uu_16_18_20_uu_uu_uu_12_uu_24_26_28_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <u,u,20,u,0,2,4,u,u,u,28,u,8,10,12,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [u,u,20,u,0,2,4,u,u,u,28,u,8,10,12,u]
; AVX512VL-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512VL-NEXT: vmovdqa %ymm2, %ymm0
; AVX512VL-NEXT: retq
@@ -6722,7 +6722,7 @@ define <16 x i16> @shuffle_v16i16_23_uu_03_uu_20_20_05_uu_31_uu_11_uu_28_28_13_u
;
; AVX512VL-LABEL: shuffle_v16i16_23_uu_03_uu_20_20_05_uu_31_uu_11_uu_28_28_13_uu:
; AVX512VL: # %bb.0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = <7,u,19,u,4,4,21,u,15,u,27,u,12,12,29,u>
+; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm2 = [7,u,19,u,4,4,21,u,15,u,27,u,12,12,29,u]
; AVX512VL-NEXT: vpermi2w %ymm0, %ymm1, %ymm2
; AVX512VL-NEXT: vmovdqa %ymm2, %ymm0
; AVX512VL-NEXT: retq
@@ -7458,10 +7458,10 @@ define <16 x i16> @PR24935(<16 x i16> %a, <16 x i16> %b) {
;
; AVX2-FAST-ALL-LABEL: PR24935:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = <0,5,u,u,0,4,6,2>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = [0,5,u,u,0,4,6,2]
; AVX2-FAST-ALL-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm0 = zero,zero,zero,zero,ymm0[2,3],zero,zero,zero,zero,zero,zero,ymm0[6,7],zero,zero,ymm0[18,19,22,23],zero,zero,zero,zero,ymm0[26,27,28,29,16,17],zero,zero
-; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = <5,6,3,0,0,6,4,u>
+; AVX2-FAST-ALL-NEXT: vmovdqa {{.*#+}} ymm2 = [5,6,3,0,0,6,4,u]
; AVX2-FAST-ALL-NEXT: vpermd %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[2,3,0,1],zero,zero,ymm1[6,7,0,1,10,11],zero,zero,ymm1[12,13],zero,zero,zero,zero,ymm1[16,17,20,21],zero,zero,zero,zero,zero,zero,ymm1[24,25]
; AVX2-FAST-ALL-NEXT: vpor %ymm0, %ymm1, %ymm0
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
index f5c5ba663175041..baa942794ccd8a0 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
@@ -3411,7 +3411,7 @@ define <32 x i8> @shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_
; AVX2-LABEL: shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_20_19_52_19_49_54_37_32_48_42_59_07_36_34_36_39:
; AVX2: # %bb.0:
; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,12,12,13,1,6,13,7,u,u,u,u,u,u,u,u,u,u,u,u,17,22,29,23,20,19,u,19,u,u,u,u]
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,5,0,6,u,1,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,5,0,6,u,1,u]
; AVX2-NEXT: vpermd %ymm0, %ymm2, %ymm0
; AVX2-NEXT: vpshufb {{.*#+}} ymm2 = ymm1[10,13],zero,zero,ymm1[3,3],zero,ymm1[8],zero,zero,zero,ymm1[12,1],zero,zero,zero,zero,zero,ymm1[20],zero,ymm1[17,22],zero,zero,ymm1[16],zero,ymm1[27],zero,zero,zero,zero,zero
; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,0,1]
@@ -3424,7 +3424,7 @@ define <32 x i8> @shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_
; AVX512VLBW-LABEL: shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_20_19_52_19_49_54_37_32_48_42_59_07_36_34_36_39:
; AVX512VLBW: # %bb.0:
; AVX512VLBW-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,12,12,13,1,6,13,7,u,u,u,u,u,u,u,u,u,u,u,u,17,22,29,23,20,19,u,19,u,u,u,u]
-; AVX512VLBW-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,5,0,6,u,1,u>
+; AVX512VLBW-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,5,0,6,u,1,u]
; AVX512VLBW-NEXT: vpermd %ymm0, %ymm2, %ymm2
; AVX512VLBW-NEXT: vpshufb {{.*#+}} ymm0 = ymm1[10,13],zero,zero,ymm1[3,3],zero,ymm1[8],zero,zero,zero,ymm1[12,1],zero,zero,zero,zero,zero,ymm1[20],zero,ymm1[17,22],zero,zero,ymm1[16],zero,ymm1[27],zero,zero,zero,zero,zero
; AVX512VLBW-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,0,1]
@@ -3460,7 +3460,7 @@ define <32 x i8> @shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_
; XOPAVX2-LABEL: shuffle_v32i8_42_45_12_13_35_35_60_40_17_22_29_44_33_12_48_51_20_19_52_19_49_54_37_32_48_42_59_07_36_34_36_39:
; XOPAVX2: # %bb.0:
; XOPAVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,12,12,13,1,6,13,7,u,u,u,u,u,u,u,u,u,u,u,u,17,22,29,23,20,19,u,19,u,u,u,u]
-; XOPAVX2-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,5,0,6,u,1,u>
+; XOPAVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,5,0,6,u,1,u]
; XOPAVX2-NEXT: vpermd %ymm0, %ymm2, %ymm0
; XOPAVX2-NEXT: vpshufb {{.*#+}} ymm2 = ymm1[10,13],zero,zero,ymm1[3,3],zero,ymm1[8],zero,zero,zero,ymm1[12,1],zero,zero,zero,zero,zero,ymm1[20],zero,ymm1[17,22],zero,zero,ymm1[16],zero,ymm1[27],zero,zero,zero,zero,zero
; XOPAVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[2,3,0,1]
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v8.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v8.ll
index 30b743cb7bdfdb4..a69a1a18f26e507 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v8.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v8.ll
@@ -445,7 +445,7 @@ define <8 x float> @shuffle_v8f32_08991abb(<8 x float> %a, <8 x float> %b) {
;
; AVX2-SLOW-LABEL: shuffle_v8f32_08991abb:
; AVX2-SLOW: # %bb.0:
-; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,1,u,2,3,3>
+; AVX2-SLOW-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,1,u,2,3,3]
; AVX2-SLOW-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-SLOW-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,1,1,3]
; AVX2-SLOW-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,1,3]
@@ -454,16 +454,16 @@ define <8 x float> @shuffle_v8f32_08991abb(<8 x float> %a, <8 x float> %b) {
;
; AVX2-FAST-ALL-LABEL: shuffle_v8f32_08991abb:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm0
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,1,u,2,3,3>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,1,u,2,3,3]
; AVX2-FAST-ALL-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
; AVX2-FAST-ALL-NEXT: retq
;
; AVX2-FAST-PERLANE-LABEL: shuffle_v8f32_08991abb:
; AVX2-FAST-PERLANE: # %bb.0:
-; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,1,u,2,3,3>
+; AVX2-FAST-PERLANE-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,1,u,2,3,3]
; AVX2-FAST-PERLANE-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-FAST-PERLANE-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,1,1,3]
; AVX2-FAST-PERLANE-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,1,3]
@@ -533,7 +533,7 @@ define <8 x float> @shuffle_v8f32_09ab1def(<8 x float> %a, <8 x float> %b) {
;
; AVX2-FAST-ALL-LABEL: shuffle_v8f32_09ab1def:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm0
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
; AVX2-FAST-ALL-NEXT: retq
@@ -956,7 +956,7 @@ define <8 x float> @shuffle_v8f32_f511235a(<8 x float> %a, <8 x float> %b) {
; AVX2-FAST-ALL: # %bb.0:
; AVX2-FAST-ALL-NEXT: vbroadcastsd {{.*#+}} ymm2 = [7,2,7,2,7,2,7,2]
; AVX2-FAST-ALL-NEXT: vpermps %ymm1, %ymm2, %ymm1
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <u,5,1,1,2,3,5,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [u,5,1,1,2,3,5,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm0
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1,2,3,4,5,6],ymm1[7]
; AVX2-FAST-ALL-NEXT: retq
@@ -1397,7 +1397,7 @@ define <8 x float> @shuffle_v8f32_089abcde(<8 x float> %a, <8 x float> %b) {
;
; AVX2-LABEL: shuffle_v8f32_089abcde:
; AVX2: # %bb.0:
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,2,3,4,5,6>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,2,3,4,5,6]
; AVX2-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
; AVX2-NEXT: retq
@@ -1452,7 +1452,7 @@ define <8 x float> @shuffle_v8f32_01289abc(<8 x float> %a, <8 x float> %b) {
;
; AVX2-LABEL: shuffle_v8f32_01289abc:
; AVX2: # %bb.0:
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <u,u,u,0,1,2,3,4>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [u,u,u,0,1,2,3,4]
; AVX2-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3,4,5,6,7]
; AVX2-NEXT: retq
@@ -2095,9 +2095,9 @@ define <8 x i32> @shuffle_v8i32_08991abb(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-FAST-ALL-LABEL: shuffle_v8i32_08991abb:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm0
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,1,u,2,3,3>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,1,u,2,3,3]
; AVX2-FAST-ALL-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
; AVX2-FAST-ALL-NEXT: retq
@@ -2174,7 +2174,7 @@ define <8 x i32> @shuffle_v8i32_09ab1def(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-FAST-ALL-LABEL: shuffle_v8i32_09ab1def:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm0, %ymm2, %ymm0
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
; AVX2-FAST-ALL-NEXT: retq
@@ -2526,7 +2526,7 @@ define <8 x i32> @shuffle_v8i32_002u6u44(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_002u6u44:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <0,0,2,u,6,u,4,4>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [0,0,2,u,6,u,4,4]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 undef, i32 6, i32 undef, i32 4, i32 4>
@@ -2541,7 +2541,7 @@ define <8 x i32> @shuffle_v8i32_00uu66uu(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_00uu66uu:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <0,0,u,u,6,6,u,u>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [0,0,u,u,6,6,u,u]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 undef, i32 undef, i32 6, i32 6, i32 undef, i32 undef>
@@ -2556,7 +2556,7 @@ define <8 x i32> @shuffle_v8i32_103245uu(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_103245uu:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <1,0,3,2,4,5,u,u>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [1,0,3,2,4,5,u,u]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 4, i32 5, i32 undef, i32 undef>
@@ -2571,7 +2571,7 @@ define <8 x i32> @shuffle_v8i32_1133uu67(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_1133uu67:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <1,1,3,3,u,u,6,7>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [1,1,3,3,u,u,6,7]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 undef, i32 undef, i32 6, i32 7>
@@ -2586,7 +2586,7 @@ define <8 x i32> @shuffle_v8i32_0uu354uu(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_0uu354uu:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <0,u,u,3,5,4,u,u>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [0,u,u,3,5,4,u,u]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 undef, i32 undef, i32 3, i32 5, i32 4, i32 undef, i32 undef>
@@ -2601,7 +2601,7 @@ define <8 x i32> @shuffle_v8i32_uuu3uu66(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2OR512VL-LABEL: shuffle_v8i32_uuu3uu66:
; AVX2OR512VL: # %bb.0:
-; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = <u,u,u,3,u,u,6,6>
+; AVX2OR512VL-NEXT: vmovaps {{.*#+}} ymm1 = [u,u,u,3,u,u,6,6]
; AVX2OR512VL-NEXT: vpermps %ymm0, %ymm1, %ymm0
; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 3, i32 undef, i32 undef, i32 6, i32 6>
@@ -2628,7 +2628,7 @@ define <8 x i32> @shuffle_v8i32_6caa87e5(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-FAST-ALL-LABEL: shuffle_v8i32_6caa87e5:
; AVX2-FAST-ALL: # %bb.0:
-; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = <u,4,2,2,0,u,6,u>
+; AVX2-FAST-ALL-NEXT: vmovaps {{.*#+}} ymm2 = [u,4,2,2,0,u,6,u]
; AVX2-FAST-ALL-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-FAST-ALL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,1,3,2]
; AVX2-FAST-ALL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4],ymm0[5],ymm1[6],ymm0[7]
@@ -2958,7 +2958,7 @@ define <8 x i32> @shuffle_v8i32_089abcde(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-LABEL: shuffle_v8i32_089abcde:
; AVX2: # %bb.0:
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <u,0,1,2,3,4,5,6>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [u,0,1,2,3,4,5,6]
; AVX2-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6,7]
; AVX2-NEXT: retq
@@ -3031,7 +3031,7 @@ define <8 x i32> @shuffle_v8i32_01289abc(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-LABEL: shuffle_v8i32_01289abc:
; AVX2: # %bb.0:
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <u,u,u,0,1,2,3,4>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [u,u,u,0,1,2,3,4]
; AVX2-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1,2],ymm1[3,4,5,6,7]
; AVX2-NEXT: retq
@@ -3244,7 +3244,7 @@ define <8 x i32> @shuffle_v8i32_0dcd3f14(<8 x i32> %a, <8 x i32> %b) {
;
; AVX2-LABEL: shuffle_v8i32_0dcd3f14:
; AVX2: # %bb.0:
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <0,u,u,u,3,u,1,4>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [0,u,u,u,3,u,1,4]
; AVX2-NEXT: vpermps %ymm0, %ymm2, %ymm0
; AVX2-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[2,2,3,3]
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5],ymm0[6,7]
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
index 370a69696d850d4..9c0ffd4558fc80d 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v16.ll
@@ -233,7 +233,7 @@ define <16 x i32> @shuffle_v16i32_01_02_03_16_05_06_07_20_09_10_11_24_13_14_15_2
define <16 x float> @shuffle_v16f32_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01(<16 x float> %a) {
; ALL-LABEL: shuffle_v16f32_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01:
; ALL: # %bb.0:
-; ALL-NEXT: vmovaps {{.*#+}} zmm1 = <2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1>
+; ALL-NEXT: vmovaps {{.*#+}} zmm1 = [2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1]
; ALL-NEXT: vpermps %zmm0, %zmm1, %zmm0
; ALL-NEXT: retq
%c = shufflevector <16 x float> %a, <16 x float> undef, <16 x i32> <i32 2, i32 5, i32 undef, i32 undef, i32 7, i32 undef, i32 10, i32 1, i32 0, i32 5, i32 undef, i32 4, i32 7, i32 undef, i32 10, i32 1>
@@ -243,7 +243,7 @@ define <16 x float> @shuffle_v16f32_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01(<
define <16 x i32> @shuffle_v16i32_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01(<16 x i32> %a) {
; ALL-LABEL: shuffle_v16i32_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01:
; ALL: # %bb.0:
-; ALL-NEXT: vmovaps {{.*#+}} zmm1 = <2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1>
+; ALL-NEXT: vmovaps {{.*#+}} zmm1 = [2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1]
; ALL-NEXT: vpermps %zmm0, %zmm1, %zmm0
; ALL-NEXT: retq
%c = shufflevector <16 x i32> %a, <16 x i32> undef, <16 x i32> <i32 2, i32 5, i32 undef, i32 undef, i32 7, i32 undef, i32 10, i32 1, i32 0, i32 5, i32 undef, i32 4, i32 7, i32 undef, i32 10, i32 1>
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
index 4668d7b6870ef27..fa0ec33bf34080e 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
@@ -42,20 +42,20 @@ define <32 x i16> @shuffle_v32i16_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01_02_
; KNL-NEXT: vpshufb {{.*#+}} ymm1 = ymm0[4,5,10,11,4,5,6,7,14,15,2,3,4,5,2,3,20,21,26,27,20,21,22,23,30,31,18,19,20,21,18,19]
; KNL-NEXT: vpermq {{.*#+}} ymm2 = ymm0[2,3,0,1]
; KNL-NEXT: vpshufb {{.*#+}} ymm3 = ymm2[0,1,10,11,8,9,8,9,14,15,6,7,4,5,14,15,16,17,26,27,24,25,24,25,30,31,22,23,20,21,30,31]
-; KNL-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,u,u,u,u,255,255,u,u,0,0,255,255,0,0,0,0,u,u,0,0,0,0,u,u,255,255,u,u>
+; KNL-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,u,u,u,u,255,255,u,u,0,0,255,255,0,0,0,0,u,u,0,0,0,0,u,u,255,255,u,u]
; KNL-NEXT: vpblendvb %ymm4, %ymm1, %ymm3, %ymm3
; KNL-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; KNL-NEXT: vpblendw {{.*#+}} ymm0 = ymm3[0,1,2,3,4,5,6],ymm0[7],ymm3[8,9,10,11,12,13,14],ymm0[15]
; KNL-NEXT: vpblendd {{.*#+}} ymm0 = ymm3[0,1,2,3],ymm0[4,5,6,7]
; KNL-NEXT: vpshufb {{.*#+}} ymm2 = ymm2[0,1,10,11,8,9,8,9,14,15,2,3,4,5,2,3,16,17,26,27,24,25,24,25,30,31,18,19,20,21,18,19]
-; KNL-NEXT: vmovdqa {{.*#+}} ymm3 = <0,0,0,0,u,u,u,u,0,0,u,u,255,255,0,0,255,255,255,255,u,u,255,255,255,255,u,u,0,0,255,255>
+; KNL-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,u,u,u,u,0,0,u,u,255,255,0,0,255,255,255,255,u,u,255,255,255,255,u,u,0,0,255,255]
; KNL-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; KNL-NEXT: vinserti64x4 $1, %ymm0, %zmm1, %zmm0
; KNL-NEXT: retq
;
; SKX-LABEL: shuffle_v32i16_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_01_02_05_u_u_07_u_0a_01_00_05_u_04_07_u_0a_1f:
; SKX: ## %bb.0:
-; SKX-NEXT: vmovdqa64 {{.*#+}} zmm1 = <2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1,2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,31>
+; SKX-NEXT: vmovdqa64 {{.*#+}} zmm1 = [2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,1,2,5,u,u,7,u,10,1,0,5,u,4,7,u,10,31]
; SKX-NEXT: vpermw %zmm0, %zmm1, %zmm0
; SKX-NEXT: retq
%c = shufflevector <32 x i16> %a, <32 x i16> undef, <32 x i32> <i32 2, i32 5, i32 undef, i32 undef, i32 7, i32 undef, i32 10, i32 1, i32 0, i32 5, i32 undef, i32 4, i32 7, i32 undef, i32 10, i32 1, i32 2, i32 5, i32 undef, i32 undef, i32 7, i32 undef, i32 10, i32 1, i32 0, i32 5, i32 undef, i32 4, i32 7, i32 undef, i32 10, i32 31>
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll
index 843b285ae1c36b7..4df5307316a4269 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v64.ll
@@ -556,14 +556,14 @@ define <64 x i8> @shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_
; AVX512F-LABEL: shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_49_51_55_57_61_63_67_69_73_75_79_81_85_87_91_93_97_99_103_105_109_111_115_117_121_123_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm4, %xmm0, %xmm5
; AVX512F-NEXT: vpor %xmm2, %xmm5, %xmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512F-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,1,5,7,11,13,17,19,23,25,29,31,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpblendvb %ymm5, %ymm2, %ymm0, %ymm0
; AVX512F-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512F-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u,u,u,u],zero,zero,zero,zero,zero,xmm2[1,5,7,11,13]
@@ -583,14 +583,14 @@ define <64 x i8> @shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_
; AVX512BW-LABEL: shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_49_51_55_57_61_63_67_69_73_75_79_81_85_87_91_93_97_99_103_105_109_111_115_117_121_123_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm4, %xmm0, %xmm5
; AVX512BW-NEXT: vpor %xmm2, %xmm5, %xmm2
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,1,5,7,11,13,17,19,23,25,29,31,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpblendvb %ymm5, %ymm2, %ymm0, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512BW-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u,u,u,u],zero,zero,zero,zero,zero,xmm2[1,5,7,11,13]
@@ -610,14 +610,14 @@ define <64 x i8> @shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_
; AVX512DQ-LABEL: shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_49_51_55_57_61_63_67_69_73_75_79_81_85_87_91_93_97_99_103_105_109_111_115_117_121_123_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vextracti128 $1, %ymm0, %xmm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm3, %xmm2, %xmm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm4 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm4 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm4, %xmm0, %xmm5
; AVX512DQ-NEXT: vpor %xmm2, %xmm5, %xmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,1,5,7,11,13,17,19,23,25,29,31,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm5 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpblendvb %ymm5, %ymm2, %ymm0, %ymm0
; AVX512DQ-NEXT: vextracti128 $1, %ymm1, %xmm2
; AVX512DQ-NEXT: vpshufb {{.*#+}} xmm2 = xmm2[u,u,u,u,u,u],zero,zero,zero,zero,zero,xmm2[1,5,7,11,13]
@@ -636,7 +636,7 @@ define <64 x i8> @shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_
;
; AVX512VBMI-LABEL: shuffle_v64i8_01_03_07_09_13_15_19_21_25_27_31_33_37_39_43_45_49_51_55_57_61_63_67_69_73_75_79_81_85_87_91_93_97_99_103_105_109_111_115_117_121_123_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512VBMI: # %bb.0:
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = <1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,67,69,73,75,79,81,85,87,91,93,97,99,103,105,109,111,115,117,121,123,127,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = [1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,67,69,73,75,79,81,85,87,91,93,97,99,103,105,109,111,115,117,121,123,127,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermt2b %zmm1, %zmm2, %zmm0
; AVX512VBMI-NEXT: retq
%r = shufflevector <64 x i8> %a0, <64 x i8> %a1, <64 x i32> <i32 1, i32 3, i32 7, i32 9, i32 13, i32 15, i32 19, i32 21, i32 25, i32 27, i32 31, i32 33, i32 37, i32 39, i32 43, i32 45, i32 49, i32 51, i32 55, i32 57, i32 61, i32 63, i32 67, i32 69, i32 73, i32 75, i32 79, i32 81, i32 85, i32 87, i32 91, i32 93, i32 97, i32 99, i32 103, i32 105, i32 109, i32 111, i32 115, i32 117, i32 121, i32 123, i32 127, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
@@ -647,10 +647,10 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512F-LABEL: shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_49_53_55_59_61_65_67_71_73_77_79_83_85_89_91_95_97_101_103_107_109_113_115_119_121_125_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512F-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512F-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -659,7 +659,7 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512F-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512F-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512F-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512F-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[3,5,9,11,15]
@@ -674,10 +674,10 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512BW-LABEL: shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_49_53_55_59_61_65_67_71_73_77_79_83_85_89_91_95_97_101_103_107_109_113_115_119_121_125_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512BW-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512BW-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -686,7 +686,7 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512BW-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512BW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[3,5,9,11,15]
@@ -701,10 +701,10 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512DQ-LABEL: shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_49_53_55_59_61_65_67_71_73_77_79_83_85_89_91_95_97_101_103_107_109_113_115_119_121_125_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512DQ-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512DQ-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512DQ-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -713,7 +713,7 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
; AVX512DQ-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512DQ-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512DQ-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[3,5,9,11,15]
@@ -727,7 +727,7 @@ define <64 x i8> @shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_
;
; AVX512VBMI-LABEL: shuffle_v64i8_01_05_07_11_13_17_19_23_25_29_31_35_37_41_43_47_49_53_55_59_61_65_67_71_73_77_79_83_85_89_91_95_97_101_103_107_109_113_115_119_121_125_127_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512VBMI: # %bb.0:
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = <65,69,71,75,77,81,83,87,89,93,95,99,101,105,107,111,113,117,119,123,125,1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = [65,69,71,75,77,81,83,87,89,93,95,99,101,105,107,111,113,117,119,123,125,1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b %zmm0, %zmm1, %zmm2
; AVX512VBMI-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512VBMI-NEXT: retq
@@ -740,9 +740,9 @@ define <64 x i8> @shuffle_v64i8_02_04_08_10_14_16_20_22_26_28_32_34_38_40_44_46_
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti64x4 $1, %zmm1, %ymm2
; AVX512F-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm4, %xmm3, %xmm3
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vextracti128 $1, %ymm0, %xmm3
@@ -767,9 +767,9 @@ define <64 x i8> @shuffle_v64i8_02_04_08_10_14_16_20_22_26_28_32_34_38_40_44_46_
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vextracti64x4 $1, %zmm1, %ymm2
; AVX512BW-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm4, %xmm3, %xmm3
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vextracti128 $1, %ymm0, %xmm3
@@ -794,9 +794,9 @@ define <64 x i8> @shuffle_v64i8_02_04_08_10_14_16_20_22_26_28_32_34_38_40_44_46_
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm1, %ymm2
; AVX512DQ-NEXT: vextracti128 $1, %ymm2, %xmm3
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm4, %xmm3, %xmm3
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512DQ-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512DQ-NEXT: vextracti128 $1, %ymm0, %xmm3
@@ -819,7 +819,7 @@ define <64 x i8> @shuffle_v64i8_02_04_08_10_14_16_20_22_26_28_32_34_38_40_44_46_
;
; AVX512VBMI-LABEL: shuffle_v64i8_02_04_08_10_14_16_20_22_26_28_32_34_38_40_44_46_50_52_56_58_62_64_68_70_74_76_80_82_86_88_92_94_98_100_104_106_110_112_116_118_122_124_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512VBMI: # %bb.0:
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = <2,4,8,10,14,16,20,22,26,28,32,34,38,40,44,46,50,52,56,58,62,64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = [2,4,8,10,14,16,20,22,26,28,32,34,38,40,44,46,50,52,56,58,62,64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermt2b %zmm1, %zmm2, %zmm0
; AVX512VBMI-NEXT: retq
%r = shufflevector <64 x i8> %a0, <64 x i8> %a1, <64 x i32> <i32 2, i32 4, i32 8, i32 10, i32 14, i32 16, i32 20, i32 22, i32 26, i32 28, i32 32, i32 34, i32 38, i32 40, i32 44, i32 46, i32 50, i32 52, i32 56, i32 58, i32 62, i32 64, i32 68, i32 70, i32 74, i32 76, i32 80, i32 82, i32 86, i32 88, i32 92, i32 94, i32 98, i32 100, i32 104, i32 106, i32 110, i32 112, i32 116, i32 118, i32 122, i32 124, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
@@ -830,10 +830,10 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512F-LABEL: shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_48_52_54_58_60_64_66_70_72_76_78_82_84_88_90_94_96_100_102_106_108_112_114_118_120_124_126_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512F-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512F-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -842,7 +842,7 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512F-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512F-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512F-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512F-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512F-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512F-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[2,4,8,10,14]
@@ -857,10 +857,10 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512BW-LABEL: shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_48_52_54_58_60_64_66_70_72_76_78_82_84_88_90_94_96_100_102_106_108_112_114_118_120_124_126_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512BW-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512BW-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -869,7 +869,7 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512BW-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512BW-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512BW-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[2,4,8,10,14]
@@ -884,10 +884,10 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512DQ-LABEL: shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_48_52_54_58_60_64_66_70_72_76_78_82_84_88_90_94_96_100_102_106_108_112_114_118_120_124_126_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm1, %ymm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm3 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm3, %xmm2, %xmm4
; AVX512DQ-NEXT: vextracti128 $1, %ymm2, %xmm2
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX512DQ-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX512DQ-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512DQ-NEXT: vpshufb %xmm3, %xmm0, %xmm3
@@ -896,7 +896,7 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
; AVX512DQ-NEXT: vpor %xmm3, %xmm4, %xmm3
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm0
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm4 = <255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,u,u,u,u,u,u,u,u,u,u,u]
; AVX512DQ-NEXT: vpblendvb %ymm4, %ymm3, %ymm0, %ymm0
; AVX512DQ-NEXT: vextracti128 $1, %ymm1, %xmm3
; AVX512DQ-NEXT: vpshufb {{.*#+}} xmm3 = xmm3[u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm3[2,4,8,10,14]
@@ -910,7 +910,7 @@ define <64 x i8> @shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_
;
; AVX512VBMI-LABEL: shuffle_v64i8_00_04_06_10_12_16_18_22_24_28_30_34_36_40_42_46_48_52_54_58_60_64_66_70_72_76_78_82_84_88_90_94_96_100_102_106_108_112_114_118_120_124_126_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u_u:
; AVX512VBMI: # %bb.0:
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = <64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,0,2,6,8,12,14,18,20,24,26,30,32,36,38,42,44,48,50,54,56,60,62,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm2 = [64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,0,2,6,8,12,14,18,20,24,26,30,32,36,38,42,44,48,50,54,56,60,62,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b %zmm0, %zmm1, %zmm2
; AVX512VBMI-NEXT: vmovdqa64 %zmm2, %zmm0
; AVX512VBMI-NEXT: retq
@@ -989,7 +989,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512F-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512F-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512F-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512F-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
@@ -1006,7 +1006,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512BW-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512BW-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
@@ -1023,7 +1023,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512DQ-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512DQ-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,3,5,9,11,15,17,21,23,27,29,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512DQ-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
@@ -1050,7 +1050,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512F-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512F-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,0,2,6,8,12,14,18,20,24,26,30,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0]
; AVX512F-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX512F-NEXT: vextracti32x4 $2, %zmm0, %xmm2
; AVX512F-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3,4],xmm1[5,6,7]
@@ -1067,7 +1067,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512BW-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,0,2,6,8,12,14,18,20,24,26,30,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0]
; AVX512BW-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX512BW-NEXT: vextracti32x4 $2, %zmm0, %xmm2
; AVX512BW-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3,4],xmm1[5,6,7]
@@ -1084,7 +1084,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512DQ-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512DQ-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,0,2,6,8,12,14,18,20,24,26,30,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0]
; AVX512DQ-NEXT: vpblendvb %ymm3, %ymm1, %ymm2, %ymm1
; AVX512DQ-NEXT: vextracti32x4 $2, %zmm0, %xmm2
; AVX512DQ-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3,4],xmm1[5,6,7]
@@ -1111,7 +1111,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512F-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512F-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512F-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512F-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
@@ -1128,7 +1128,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512BW-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512BW-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512BW-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512BW-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
@@ -1145,7 +1145,7 @@ define <64 x i8> @shuffle_v64i8_00_01_02_03_04_05_06_07_08_09_10_11_12_13_14_15_
; AVX512DQ-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX512DQ-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX512DQ-NEXT: vpshufb {{.*#+}} ymm1 = ymm1[u,u,u,u,u,u,u,u,u,u,u,2,4,8,10,14,16,20,22,26,28,u,u,u,u,u,u,u,u,u,u,u]
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = <u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255>
+; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [u,u,u,u,u,u,u,u,u,u,u,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255]
; AVX512DQ-NEXT: vpblendvb %ymm3, %ymm2, %ymm1, %ymm1
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm0, %ymm2
; AVX512DQ-NEXT: vmovdqa {{.*#+}} ymm3 = [0,0,0,0,0,0,0,0,0,0,0,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255]
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
index 2bb7dd90cc05f65..9bfca824fb71a97 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v8.ll
@@ -687,13 +687,13 @@ define <8 x double> @shuffle_v8f64_01235466(<8 x double> %a, <8 x double> %b) {
define <8 x double> @shuffle_v8f64_002u6u44(<8 x double> %a, <8 x double> %b) {
; AVX512F-LABEL: shuffle_v8f64_002u6u44:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,2,u,6,u,4,4>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,2,u,6,u,4,4]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8f64_002u6u44:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,0,0,2,0,u,u,6,0,u,u,4,0,4,0>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,0,0,2,0,u,u,6,0,u,u,4,0,4,0]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x double> %a, <8 x double> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 undef, i32 6, i32 undef, i32 4, i32 4>
@@ -703,13 +703,13 @@ define <8 x double> @shuffle_v8f64_002u6u44(<8 x double> %a, <8 x double> %b) {
define <8 x double> @shuffle_v8f64_00uu66uu(<8 x double> %a, <8 x double> %b) {
; AVX512F-LABEL: shuffle_v8f64_00uu66uu:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,u,u,6,6,u,u>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,u,u,6,6,u,u]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8f64_00uu66uu:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,0,0,u,u,u,u,6,0,6,0,u,u,u,u>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,0,0,u,u,u,u,6,0,6,0,u,u,u,u]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x double> %a, <8 x double> %b, <8 x i32> <i32 0, i32 0, i32 undef, i32 undef, i32 6, i32 6, i32 undef, i32 undef>
@@ -1553,13 +1553,13 @@ define <8 x i64> @shuffle_v8i64_01235466(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_002u6u44(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_002u6u44:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,2,u,6,u,4,4>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,2,u,6,u,4,4]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_002u6u44:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,0,0,2,0,u,u,6,0,u,u,4,0,4,0>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,0,0,2,0,u,u,6,0,u,u,4,0,4,0]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 undef, i32 6, i32 undef, i32 4, i32 4>
@@ -1569,13 +1569,13 @@ define <8 x i64> @shuffle_v8i64_002u6u44(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_00uu66uu(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_00uu66uu:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,u,u,6,6,u,u>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,u,u,6,6,u,u]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_00uu66uu:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,0,0,u,u,u,u,6,0,6,0,u,u,u,u>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,0,0,u,u,u,u,6,0,6,0,u,u,u,u]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 0, i32 0, i32 undef, i32 undef, i32 6, i32 6, i32 undef, i32 undef>
@@ -1585,13 +1585,13 @@ define <8 x i64> @shuffle_v8i64_00uu66uu(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_103245uu(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_103245uu:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <1,0,3,2,4,5,u,u>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [1,0,3,2,4,5,u,u]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_103245uu:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <1,0,0,0,3,0,2,0,4,0,5,0,u,u,u,u>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [1,0,0,0,3,0,2,0,4,0,5,0,u,u,u,u]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 4, i32 5, i32 undef, i32 undef>
@@ -1601,13 +1601,13 @@ define <8 x i64> @shuffle_v8i64_103245uu(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_1133uu67(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_1133uu67:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <1,1,3,3,u,u,6,7>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [1,1,3,3,u,u,6,7]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_1133uu67:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <1,0,1,0,3,0,3,0,u,u,u,u,6,0,7,0>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [1,0,1,0,3,0,3,0,u,u,u,u,6,0,7,0]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 undef, i32 undef, i32 6, i32 7>
@@ -1617,13 +1617,13 @@ define <8 x i64> @shuffle_v8i64_1133uu67(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_0uu354uu(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_0uu354uu:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <0,u,u,3,5,4,u,u>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [0,u,u,3,5,4,u,u]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_0uu354uu:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <0,0,u,u,u,u,3,0,5,0,4,0,u,u,u,u>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [0,0,u,u,u,u,3,0,5,0,4,0,u,u,u,u]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 0, i32 undef, i32 undef, i32 3, i32 5, i32 4, i32 undef, i32 undef>
@@ -1633,13 +1633,13 @@ define <8 x i64> @shuffle_v8i64_0uu354uu(<8 x i64> %a, <8 x i64> %b) {
define <8 x i64> @shuffle_v8i64_uuu3uu66(<8 x i64> %a, <8 x i64> %b) {
; AVX512F-LABEL: shuffle_v8i64_uuu3uu66:
; AVX512F: # %bb.0:
-; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = <u,u,u,3,u,u,6,6>
+; AVX512F-NEXT: vmovaps {{.*#+}} zmm1 = [u,u,u,3,u,u,6,6]
; AVX512F-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-NEXT: retq
;
; AVX512F-32-LABEL: shuffle_v8i64_uuu3uu66:
; AVX512F-32: # %bb.0:
-; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = <u,u,u,u,u,u,3,0,u,u,u,u,6,0,6,0>
+; AVX512F-32-NEXT: vmovaps {{.*#+}} zmm1 = [u,u,u,u,u,u,3,0,u,u,u,u,6,0,6,0]
; AVX512F-32-NEXT: vpermpd %zmm0, %zmm1, %zmm0
; AVX512F-32-NEXT: retl
%shuffle = shufflevector <8 x i64> %a, <8 x i64> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 3, i32 undef, i32 undef, i32 6, i32 6>
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
index 3aafb1af366ba56..917f26dd3cadf7b 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
@@ -362,7 +362,7 @@ define void @PR39483() {
; X86-AVX2: # %bb.0: # %entry
; X86-AVX2-NEXT: vmovups 32, %ymm0
; X86-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],mem[2],ymm0[3,4],mem[5],ymm0[6,7]
-; X86-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; X86-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; X86-AVX2-NEXT: vpermps %ymm0, %ymm1, %ymm0
; X86-AVX2-NEXT: vpermilps {{.*#+}} ymm1 = mem[0,1,0,3,4,5,4,7]
; X86-AVX2-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
@@ -403,7 +403,7 @@ define void @PR39483() {
; X64-AVX2: # %bb.0: # %entry
; X64-AVX2-NEXT: vmovups 32, %ymm0
; X64-AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],mem[2],ymm0[3,4],mem[5],ymm0[6,7]
-; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = <2,5,0,3,6,u,u,u>
+; X64-AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [2,5,0,3,6,u,u,u]
; X64-AVX2-NEXT: vpermps %ymm0, %ymm1, %ymm0
; X64-AVX2-NEXT: vpermilps {{.*#+}} ymm1 = mem[0,1,0,3,4,5,4,7]
; X64-AVX2-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,0,3]
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
index 9b26f95f5147824..f2e42d51a7eb63c 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
@@ -128,7 +128,7 @@ define <8 x i32> @combine_as_vpermd(<8 x i32> %a0) {
define <8 x float> @combine_as_vpermps(<8 x float> %a0) {
; CHECK-LABEL: combine_as_vpermps:
; CHECK: # %bb.0:
-; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = <6,4,7,5,1,u,4,7>
+; CHECK-NEXT: vmovaps {{.*#+}} ymm1 = [6,4,7,5,1,u,4,7]
; CHECK-NEXT: vpermps %ymm0, %ymm1, %ymm0
; CHECK-NEXT: ret{{[l|q]}}
%1 = shufflevector <8 x float> %a0, <8 x float> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 4, i32 5, i32 6, i32 7>
@@ -784,7 +784,7 @@ define <8 x float> @constant_fold_permps() {
define <32 x i8> @constant_fold_pshufb_256() {
; CHECK-LABEL: constant_fold_pshufb_256:
; CHECK: # %bb.0:
-; CHECK-NEXT: vmovaps {{.*#+}} ymm0 = <14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9,255,0,0,0,u,u,0,0,241,0,0,0,0,0,249,250>
+; CHECK-NEXT: vmovaps {{.*#+}} ymm0 = [14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9,255,0,0,0,u,u,0,0,241,0,0,0,0,0,249,250]
; CHECK-NEXT: ret{{[l|q]}}
%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> <i8 15, i8 14, i8 13, i8 12, i8 11, i8 10, i8 9, i8 8, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0, i8 0, i8 -1, i8 -2, i8 -3, i8 -4, i8 -5, i8 -6, i8 -7, i8 -8, i8 -9, i8 -10, i8 -11, i8 -12, i8 -13, i8 -14, i8 -15>, <32 x i8> <i8 1, i8 -1, i8 -1, i8 -1, i8 undef, i8 undef, i8 -1, i8 -1, i8 15, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 7, i8 6, i8 1, i8 -1, i8 -1, i8 -1, i8 undef, i8 undef, i8 -1, i8 -1, i8 15, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 7, i8 6>)
ret <32 x i8> %1
@@ -832,7 +832,7 @@ define internal fastcc <8 x float> @PR34577(<8 x float> %inp0, <8 x float> %inp1
; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[1,1,1,1]
; AVX2-NEXT: vxorps %xmm2, %xmm2, %xmm2
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm2[0,1,2,3],ymm0[4,5,6,7]
-; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = <u,u,7,2,u,u,3,2>
+; AVX2-NEXT: vmovaps {{.*#+}} ymm2 = [u,u,7,2,u,u,3,2]
; AVX2-NEXT: vpermps %ymm1, %ymm2, %ymm1
; AVX2-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3],ymm0[4,5],ymm1[6,7]
; AVX2-NEXT: ret{{[l|q]}}
@@ -843,7 +843,7 @@ define internal fastcc <8 x float> @PR34577(<8 x float> %inp0, <8 x float> %inp1
; AVX512-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[1,1,1,1]
; AVX512-NEXT: vxorps %xmm2, %xmm2, %xmm2
; AVX512-NEXT: vblendps {{.*#+}} ymm2 = ymm2[0,1,2,3],ymm0[4,5],ymm2[6,7]
-; AVX512-NEXT: vmovaps {{.*#+}} ymm0 = <23,18,7,2,20,u,3,2>
+; AVX512-NEXT: vmovaps {{.*#+}} ymm0 = [23,18,7,2,20,u,3,2]
; AVX512-NEXT: vpermi2ps %zmm2, %zmm1, %zmm0
; AVX512-NEXT: # kill: def $ymm0 killed $ymm0 killed $zmm0
; AVX512-NEXT: ret{{[l|q]}}
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-sse41.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-sse41.ll
index 6f128a003439756..5c16fbb7be4fd6d 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-sse41.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-sse41.ll
@@ -31,9 +31,9 @@ define <16 x i8> @PR50049(ptr %p1, ptr %p2) {
; SSE-NEXT: movdqa (%rsi), %xmm4
; SSE-NEXT: movdqa 16(%rsi), %xmm5
; SSE-NEXT: movdqa 32(%rsi), %xmm3
-; SSE-NEXT: movdqa {{.*#+}} xmm6 = <128,128,128,128,128,128,2,5,8,11,14,u,u,u,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm6 = [128,128,128,128,128,128,2,5,8,11,14,u,u,u,u,u]
; SSE-NEXT: pshufb %xmm6, %xmm0
-; SSE-NEXT: movdqa {{.*#+}} xmm7 = <0,3,6,9,12,15,128,128,128,128,128,u,u,u,u,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm7 = [0,3,6,9,12,15,128,128,128,128,128,u,u,u,u,u]
; SSE-NEXT: pshufb %xmm7, %xmm2
; SSE-NEXT: por %xmm0, %xmm2
; SSE-NEXT: pshufb %xmm6, %xmm5
@@ -44,9 +44,9 @@ define <16 x i8> @PR50049(ptr %p1, ptr %p2) {
; SSE-NEXT: pmullw %xmm5, %xmm0
; SSE-NEXT: movdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
; SSE-NEXT: pand %xmm5, %xmm0
-; SSE-NEXT: movdqa {{.*#+}} xmm6 = <8,u,9,u,10,u,128,u,128,u,128,u,128,u,128,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm6 = [8,u,9,u,10,u,128,u,128,u,128,u,128,u,128,u]
; SSE-NEXT: pshufb %xmm6, %xmm4
-; SSE-NEXT: movdqa {{.*#+}} xmm7 = <128,u,128,u,128,u,1,u,4,u,7,u,10,u,13,u>
+; SSE-NEXT: movdqa {{.*#+}} xmm7 = [128,u,128,u,128,u,1,u,4,u,7,u,10,u,13,u]
; SSE-NEXT: pshufb %xmm7, %xmm3
; SSE-NEXT: por %xmm4, %xmm3
; SSE-NEXT: pshufb %xmm6, %xmm2
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-ssse3.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-ssse3.ll
index 83f40046d4b5f91..5c035346415b0ed 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-ssse3.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-ssse3.ll
@@ -766,12 +766,12 @@ define <16 x i8> @combine_and_pshufb_or_pshufb(<16 x i8> %a0, <16 x i8> %a1) {
define <16 x i8> @constant_fold_pshufb() {
; SSE-LABEL: constant_fold_pshufb:
; SSE: # %bb.0:
-; SSE-NEXT: movaps {{.*#+}} xmm0 = <14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9>
+; SSE-NEXT: movaps {{.*#+}} xmm0 = [14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9]
; SSE-NEXT: retq
;
; AVX-LABEL: constant_fold_pshufb:
; AVX: # %bb.0:
-; AVX-NEXT: vmovaps {{.*#+}} xmm0 = <14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9>
+; AVX-NEXT: vmovaps {{.*#+}} xmm0 = [14,0,0,0,u,u,0,0,0,0,0,0,0,0,8,9]
; AVX-NEXT: retq
%1 = tail call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> <i8 15, i8 14, i8 13, i8 12, i8 11, i8 10, i8 9, i8 8, i8 7, i8 6, i8 5, i8 4, i8 3, i8 2, i8 1, i8 0>, <16 x i8> <i8 1, i8 -1, i8 -1, i8 -1, i8 undef, i8 undef, i8 -1, i8 -1, i8 15, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 7, i8 6>)
ret <16 x i8> %1
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining.ll
index d0334dfb66e8af4..1e2ee7b99a608d7 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining.ll
@@ -1744,7 +1744,7 @@ define <4 x i8> @combine_test1c(ptr %a, ptr %b) {
; SSE41: # %bb.0:
; SSE41-NEXT: movd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; SSE41-NEXT: movd {{.*#+}} xmm2 = mem[0],zero,zero,zero
-; SSE41-NEXT: movaps {{.*#+}} xmm0 = <0,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movaps {{.*#+}} xmm0 = [0,255,255,255,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: retq
@@ -1838,7 +1838,7 @@ define <4 x i8> @combine_test4c(ptr %a, ptr %b) {
; SSE41: # %bb.0:
; SSE41-NEXT: movd {{.*#+}} xmm1 = mem[0],zero,zero,zero
; SSE41-NEXT: movd {{.*#+}} xmm2 = mem[0],zero,zero,zero
-; SSE41-NEXT: movaps {{.*#+}} xmm0 = <255,0,255,255,u,u,u,u,u,u,u,u,u,u,u,u>
+; SSE41-NEXT: movaps {{.*#+}} xmm0 = [255,0,255,255,u,u,u,u,u,u,u,u,u,u,u,u]
; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1
; SSE41-NEXT: movdqa %xmm1, %xmm0
; SSE41-NEXT: retq
@@ -2671,14 +2671,14 @@ define void @combine_scalar_load_with_blend_with_zero(ptr %a0, ptr %a1) {
define <4 x float> @combine_constant_insertion_v4f32(float %f) {
; SSE2-LABEL: combine_constant_insertion_v4f32:
; SSE2: # %bb.0:
-; SSE2-NEXT: movaps {{.*#+}} xmm1 = <u,4.0E+0,5.0E+0,3.0E+0>
+; SSE2-NEXT: movaps {{.*#+}} xmm1 = [u,4.0E+0,5.0E+0,3.0E+0]
; SSE2-NEXT: movss {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3]
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-LABEL: combine_constant_insertion_v4f32:
; SSSE3: # %bb.0:
-; SSSE3-NEXT: movaps {{.*#+}} xmm1 = <u,4.0E+0,5.0E+0,3.0E+0>
+; SSSE3-NEXT: movaps {{.*#+}} xmm1 = [u,4.0E+0,5.0E+0,3.0E+0]
; SSSE3-NEXT: movss {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3]
; SSSE3-NEXT: movaps %xmm1, %xmm0
; SSSE3-NEXT: retq
@@ -2701,26 +2701,26 @@ define <4 x i32> @combine_constant_insertion_v4i32(i32 %f) {
; SSE2-LABEL: combine_constant_insertion_v4i32:
; SSE2: # %bb.0:
; SSE2-NEXT: movd %edi, %xmm1
-; SSE2-NEXT: movaps {{.*#+}} xmm0 = <u,4,5,30>
+; SSE2-NEXT: movaps {{.*#+}} xmm0 = [u,4,5,30]
; SSE2-NEXT: movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; SSE2-NEXT: retq
;
; SSSE3-LABEL: combine_constant_insertion_v4i32:
; SSSE3: # %bb.0:
; SSSE3-NEXT: movd %edi, %xmm1
-; SSSE3-NEXT: movaps {{.*#+}} xmm0 = <u,4,5,30>
+; SSSE3-NEXT: movaps {{.*#+}} xmm0 = [u,4,5,30]
; SSSE3-NEXT: movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
; SSSE3-NEXT: retq
;
; SSE41-LABEL: combine_constant_insertion_v4i32:
; SSE41: # %bb.0:
-; SSE41-NEXT: movdqa {{.*#+}} xmm0 = <u,4,5,30>
+; SSE41-NEXT: movdqa {{.*#+}} xmm0 = [u,4,5,30]
; SSE41-NEXT: pinsrd $0, %edi, %xmm0
; SSE41-NEXT: retq
;
; AVX-LABEL: combine_constant_insertion_v4i32:
; AVX: # %bb.0:
-; AVX-NEXT: vmovdqa {{.*#+}} xmm0 = <u,4,5,30>
+; AVX-NEXT: vmovdqa {{.*#+}} xmm0 = [u,4,5,30]
; AVX-NEXT: vpinsrd $0, %edi, %xmm0, %xmm0
; AVX-NEXT: retq
%a0 = insertelement <4 x i32> undef, i32 %f, i32 0
@@ -2851,7 +2851,7 @@ define <4 x float> @PR30264(<4 x float> %x) {
;
; SSE41-LABEL: PR30264:
; SSE41: # %bb.0:
-; SSE41-NEXT: movaps {{.*#+}} xmm1 = <u,u,4.0E+0,1.0E+0>
+; SSE41-NEXT: movaps {{.*#+}} xmm1 = [u,u,4.0E+0,1.0E+0]
; SSE41-NEXT: insertps {{.*#+}} xmm1 = xmm0[0],zero,xmm1[2,3]
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
@@ -3298,7 +3298,7 @@ define void @PR45604(ptr %dst, ptr %src) {
; SSE41-NEXT: movdqa (%rsi), %xmm0
; SSE41-NEXT: pshufd {{.*#+}} xmm1 = xmm0[1,1,1,1]
; SSE41-NEXT: pmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
-; SSE41-NEXT: movdqa {{.*#+}} xmm2 = <u,0,11,0,u,0,11,0>
+; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [u,0,11,0,u,0,11,0]
; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3],xmm1[4],xmm2[5,6,7]
; SSE41-NEXT: pshufd {{.*#+}} xmm3 = xmm0[2,3,2,3]
; SSE41-NEXT: pmovzxwq {{.*#+}} xmm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero
@@ -3341,7 +3341,7 @@ define void @PR45604(ptr %dst, ptr %src) {
; AVX2: # %bb.0:
; AVX2-NEXT: vmovdqa (%rsi), %xmm0
; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm0[0,2,0,2]
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = <0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u]
; AVX2-NEXT: vpshufb %ymm2, %ymm1, %ymm1
; AVX2-NEXT: vpbroadcastd {{.*#+}} ymm3 = [11,0,0,0,11,0,0,0,11,0,0,0,11,0,0,0,11,0,0,0,11,0,0,0,11,0,0,0,11,0,0,0]
; AVX2-NEXT: vpblendd {{.*#+}} ymm1 = ymm1[0],ymm3[1],ymm1[2],ymm3[3],ymm1[4],ymm3[5],ymm1[6],ymm3[7]
@@ -3481,9 +3481,9 @@ define void @SpinningCube() {
; SSE2-LABEL: SpinningCube:
; SSE2: # %bb.0: # %entry
; SSE2-NEXT: movl $1065353216, (%rax) # imm = 0x3F800000
-; SSE2-NEXT: movaps {{.*#+}} xmm0 = <u,u,u,1.0E+0>
+; SSE2-NEXT: movaps {{.*#+}} xmm0 = [u,u,u,1.0E+0]
; SSE2-NEXT: movss {{.*#+}} xmm1 = [NaN,0.0E+0,0.0E+0,0.0E+0]
-; SSE2-NEXT: movapd {{.*#+}} xmm2 = <u,u,-2.0E+0,u>
+; SSE2-NEXT: movapd {{.*#+}} xmm2 = [u,u,-2.0E+0,u]
; SSE2-NEXT: movsd {{.*#+}} xmm2 = xmm1[0],xmm2[1]
; SSE2-NEXT: xorps %xmm3, %xmm3
; SSE2-NEXT: shufps {{.*#+}} xmm3 = xmm3[0,1],xmm2[2,0]
@@ -3500,9 +3500,9 @@ define void @SpinningCube() {
; SSSE3-LABEL: SpinningCube:
; SSSE3: # %bb.0: # %entry
; SSSE3-NEXT: movl $1065353216, (%rax) # imm = 0x3F800000
-; SSSE3-NEXT: movaps {{.*#+}} xmm0 = <u,u,u,1.0E+0>
+; SSSE3-NEXT: movaps {{.*#+}} xmm0 = [u,u,u,1.0E+0]
; SSSE3-NEXT: movss {{.*#+}} xmm1 = [NaN,0.0E+0,0.0E+0,0.0E+0]
-; SSSE3-NEXT: movapd {{.*#+}} xmm2 = <u,u,-2.0E+0,u>
+; SSSE3-NEXT: movapd {{.*#+}} xmm2 = [u,u,-2.0E+0,u]
; SSSE3-NEXT: movsd {{.*#+}} xmm2 = xmm1[0],xmm2[1]
; SSSE3-NEXT: xorps %xmm3, %xmm3
; SSSE3-NEXT: shufps {{.*#+}} xmm3 = xmm3[0,1],xmm2[2,0]
@@ -3519,8 +3519,8 @@ define void @SpinningCube() {
; SSE41-LABEL: SpinningCube:
; SSE41: # %bb.0: # %entry
; SSE41-NEXT: movl $1065353216, (%rax) # imm = 0x3F800000
-; SSE41-NEXT: movaps {{.*#+}} xmm0 = <u,u,u,1.0E+0>
-; SSE41-NEXT: movaps {{.*#+}} xmm1 = <0.0E+0,0.0E+0,-2.0E+0,u>
+; SSE41-NEXT: movaps {{.*#+}} xmm0 = [u,u,u,1.0E+0]
+; SSE41-NEXT: movaps {{.*#+}} xmm1 = [0.0E+0,0.0E+0,-2.0E+0,u]
; SSE41-NEXT: movss {{.*#+}} xmm2 = [NaN,0.0E+0,0.0E+0,0.0E+0]
; SSE41-NEXT: movaps %xmm1, %xmm3
; SSE41-NEXT: insertps {{.*#+}} xmm3 = xmm3[0,1,2],xmm2[0]
@@ -3539,7 +3539,7 @@ define void @SpinningCube() {
; AVX: # %bb.0: # %entry
; AVX-NEXT: movl $1065353216, (%rax) # imm = 0x3F800000
; AVX-NEXT: vbroadcastss {{.*#+}} xmm0 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
-; AVX-NEXT: vmovaps {{.*#+}} xmm1 = <0.0E+0,0.0E+0,-2.0E+0,u>
+; AVX-NEXT: vmovaps {{.*#+}} xmm1 = [0.0E+0,0.0E+0,-2.0E+0,u]
; AVX-NEXT: vmovss {{.*#+}} xmm2 = [NaN,0.0E+0,0.0E+0,0.0E+0]
; AVX-NEXT: vinsertps {{.*#+}} xmm3 = xmm1[0,1,2],xmm2[0]
; AVX-NEXT: vinsertps {{.*#+}} xmm2 = xmm0[0],xmm2[0],xmm0[2,3]
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-v1.ll b/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
index 6ef203999af6ed1..95a57d1bdf3318a 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-v1.ll
@@ -542,7 +542,7 @@ define i8 @shuf8i1_10_2_9_u_3_u_2_u(i8 %a) {
; AVX512F-NEXT: kmovw %edi, %k1
; AVX512F-NEXT: vpternlogq $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <8,2,10,u,3,u,2,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [8,2,10,u,3,u,2,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vpsllq $63, %zmm2, %zmm0
; AVX512F-NEXT: vptestmq %zmm0, %zmm0, %k0
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-v192.ll b/llvm/test/CodeGen/X86/vector-shuffle-v192.ll
index 75c26f38eb9ed7c..6941585c6a5f0fb 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-v192.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-v192.ll
@@ -11,9 +11,9 @@ define <64 x i8> @f1(ptr %p0) {
; AVX2-NEXT: vmovdqa 32(%rdi), %ymm0
; AVX2-NEXT: vmovdqa (%rdi), %xmm2
; AVX2-NEXT: vmovdqa 16(%rdi), %xmm3
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm4, %xmm3, %xmm3
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm5, %xmm2, %xmm2
; AVX2-NEXT: vpor %xmm3, %xmm2, %xmm2
; AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [1,3,7,9,13,15,0,0,0,0,0,1,5,7,11,13,1,3,7,9,13,15,0,0,0,0,0,1,5,7,11,13]
@@ -22,10 +22,10 @@ define <64 x i8> @f1(ptr %p0) {
; AVX2-NEXT: vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
; AVX2-NEXT: vpblendvb %ymm6, %ymm2, %ymm0, %ymm0
; AVX2-NEXT: vmovdqa 80(%rdi), %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13]
; AVX2-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-NEXT: vmovdqa 64(%rdi), %xmm8
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128]
; AVX2-NEXT: vpshufb %xmm9, %xmm8, %xmm8
; AVX2-NEXT: vpor %xmm2, %xmm8, %xmm2
; AVX2-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -51,17 +51,17 @@ define <64 x i8> @f1(ptr %p0) {
; AVX512F-LABEL: f1:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa 112(%rdi), %xmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovdqa 96(%rdi), %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vmovdqa 176(%rdi), %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13]
; AVX512F-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX512F-NEXT: vmovdqa 160(%rdi), %xmm5
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128]
; AVX512F-NEXT: vpshufb %xmm6, %xmm5, %xmm5
; AVX512F-NEXT: vpor %xmm2, %xmm5, %xmm2
; AVX512F-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -92,17 +92,17 @@ define <64 x i8> @f1(ptr %p0) {
; AVX512BW-LABEL: f1:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa 112(%rdi), %xmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = <128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = [128,128,128,128,128,128,3,5,9,11,15,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512BW-NEXT: vmovdqa 96(%rdi), %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [1,3,7,9,13,15,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512BW-NEXT: vmovdqa 176(%rdi), %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = <u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm4 = [u,u,u,u,u,u,128,128,128,128,128,1,5,7,11,13]
; AVX512BW-NEXT: vpshufb %xmm4, %xmm2, %xmm2
; AVX512BW-NEXT: vmovdqa 160(%rdi), %xmm5
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = <u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = [u,u,u,u,u,u,3,5,9,11,15,128,128,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm6, %xmm5, %xmm5
; AVX512BW-NEXT: vpor %xmm2, %xmm5, %xmm2
; AVX512BW-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -138,7 +138,7 @@ define <64 x i8> @f1(ptr %p0) {
; AVX512VBMI-LABEL: f1:
; AVX512VBMI: # %bb.0:
; AVX512VBMI-NEXT: vmovdqa64 (%rdi), %zmm0
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = <1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,67,69,73,75,79,81,85,87,91,93,97,99,103,105,109,111,115,117,121,123,127,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = [1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,67,69,73,75,79,81,85,87,91,93,97,99,103,105,109,111,115,117,121,123,127,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b 64(%rdi), %zmm0, %zmm1
; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,65,69,71,75,77,81,83,87,89,93,95,99,101,105,107,111,113,117,119,123,125]
; AVX512VBMI-NEXT: vpermi2b 128(%rdi), %zmm1, %zmm0
@@ -155,9 +155,9 @@ define <64 x i8> @f2(ptr %p0) {
; AVX2-NEXT: vmovdqa 32(%rdi), %ymm0
; AVX2-NEXT: vmovdqa (%rdi), %xmm2
; AVX2-NEXT: vmovdqa 16(%rdi), %xmm3
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm5, %xmm3, %xmm3
; AVX2-NEXT: vpor %xmm2, %xmm3, %xmm2
; AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [1,5,7,11,13,0,0,0,0,0,0,3,5,9,11,15,1,5,7,11,13,0,0,0,0,0,0,3,5,9,11,15]
@@ -166,10 +166,10 @@ define <64 x i8> @f2(ptr %p0) {
; AVX2-NEXT: vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
; AVX2-NEXT: vpblendvb %ymm6, %ymm2, %ymm0, %ymm0
; AVX2-NEXT: vmovdqa 80(%rdi), %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15]
; AVX2-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-NEXT: vmovdqa 64(%rdi), %xmm8
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128]
; AVX2-NEXT: vpshufb %xmm9, %xmm8, %xmm8
; AVX2-NEXT: vpor %xmm2, %xmm8, %xmm2
; AVX2-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -194,18 +194,18 @@ define <64 x i8> @f2(ptr %p0) {
; AVX512F-LABEL: f2:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa 176(%rdi), %xmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15]
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovdqa 160(%rdi), %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512F-NEXT: vmovdqa (%rdi), %xmm2
; AVX512F-NEXT: vmovdqa 16(%rdi), %xmm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm5, %xmm2, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm6, %xmm4, %xmm4
; AVX512F-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX512F-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm0
@@ -234,18 +234,18 @@ define <64 x i8> @f2(ptr %p0) {
; AVX512BW-LABEL: f2:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa 176(%rdi), %xmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,128,128,128,128,128,128,3,5,9,11,15]
; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512BW-NEXT: vmovdqa 160(%rdi), %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,1,3,7,9,13,15,128,128,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512BW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512BW-NEXT: vmovdqa (%rdi), %xmm2
; AVX512BW-NEXT: vmovdqa 16(%rdi), %xmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [1,5,7,11,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,128,128,1,3,7,9,13,15,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm6, %xmm4, %xmm4
; AVX512BW-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX512BW-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm0
@@ -275,7 +275,7 @@ define <64 x i8> @f2(ptr %p0) {
; AVX512VBMI-LABEL: f2:
; AVX512VBMI: # %bb.0:
; AVX512VBMI-NEXT: vmovdqa64 64(%rdi), %zmm0
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = <65,69,71,75,77,81,83,87,89,93,95,99,101,105,107,111,113,117,119,123,125,1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = [65,69,71,75,77,81,83,87,89,93,95,99,101,105,107,111,113,117,119,123,125,1,3,7,9,13,15,19,21,25,27,31,33,37,39,43,45,49,51,55,57,61,63,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b (%rdi), %zmm0, %zmm1
; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,67,69,73,75,79,81,85,87,91,93,97,99,103,105,109,111,115,117,121,123,127]
; AVX512VBMI-NEXT: vpermi2b 128(%rdi), %zmm1, %zmm0
@@ -291,18 +291,18 @@ define <64 x i8> @f3(ptr %p0) {
; AVX2-NEXT: vmovdqa 128(%rdi), %ymm1
; AVX2-NEXT: vmovdqa 32(%rdi), %ymm0
; AVX2-NEXT: vmovdqa 64(%rdi), %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128]
; AVX2-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX2-NEXT: vmovdqa 80(%rdi), %xmm4
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = <u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = [u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14]
; AVX2-NEXT: vpshufb %xmm5, %xmm4, %xmm4
; AVX2-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX2-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
; AVX2-NEXT: vmovdqa (%rdi), %xmm4
; AVX2-NEXT: vmovdqa 16(%rdi), %xmm6
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm7, %xmm6, %xmm6
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm8 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm8 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm8, %xmm4, %xmm4
; AVX2-NEXT: vpor %xmm6, %xmm4, %xmm4
; AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm6 = [2,4,8,10,14,0,0,0,0,0,0,2,6,8,12,14,2,4,8,10,14,0,0,0,0,0,0,2,6,8,12,14]
@@ -332,10 +332,10 @@ define <64 x i8> @f3(ptr %p0) {
; AVX512F-LABEL: f3:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa 160(%rdi), %xmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128]
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovdqa 176(%rdi), %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
@@ -346,10 +346,10 @@ define <64 x i8> @f3(ptr %p0) {
; AVX512F-NEXT: vmovdqa {{.*#+}} ymm5 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0,0,0,0,0,0,0]
; AVX512F-NEXT: vpternlogq $216, %ymm5, %ymm2, %ymm0
; AVX512F-NEXT: vmovdqa 112(%rdi), %xmm6
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm7 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm7 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm7, %xmm6, %xmm6
; AVX512F-NEXT: vmovdqa 96(%rdi), %xmm8
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm9 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm9 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm9, %xmm8, %xmm8
; AVX512F-NEXT: vpor %xmm6, %xmm8, %xmm6
; AVX512F-NEXT: vpblendw {{.*#+}} xmm2 = xmm6[0,1,2,3,4],xmm2[5,6,7]
@@ -376,18 +376,18 @@ define <64 x i8> @f3(ptr %p0) {
; AVX512BW-LABEL: f3:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa 64(%rdi), %xmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,0,4,6,10,12,128,128,128,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512BW-NEXT: vmovdqa 80(%rdi), %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,128,128,128,128,128,0,2,6,8,12,14]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512BW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512BW-NEXT: vmovdqa (%rdi), %xmm2
; AVX512BW-NEXT: vmovdqa 16(%rdi), %xmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,0,4,6,10,12,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm4, %xmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = <2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = [2,4,8,10,14,128,128,128,128,128,u,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm6, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm4, %xmm2, %xmm2
; AVX512BW-NEXT: vmovdqa 32(%rdi), %ymm4
@@ -421,7 +421,7 @@ define <64 x i8> @f3(ptr %p0) {
; AVX512VBMI-LABEL: f3:
; AVX512VBMI: # %bb.0:
; AVX512VBMI-NEXT: vmovdqa64 (%rdi), %zmm0
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = <2,4,8,10,14,16,20,22,26,28,32,34,38,40,44,46,50,52,56,58,62,64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = [2,4,8,10,14,16,20,22,26,28,32,34,38,40,44,46,50,52,56,58,62,64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b 64(%rdi), %zmm0, %zmm1
; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,64,66,70,72,76,78,82,84,88,90,94,96,100,102,106,108,112,114,118,120,124,126]
; AVX512VBMI-NEXT: vpermi2b 128(%rdi), %zmm1, %zmm0
@@ -438,9 +438,9 @@ define <64 x i8> @f4(ptr %p0) {
; AVX2-NEXT: vmovdqa 32(%rdi), %ymm0
; AVX2-NEXT: vmovdqa (%rdi), %xmm2
; AVX2-NEXT: vmovdqa 16(%rdi), %xmm3
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm4 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm4, %xmm2, %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm5 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX2-NEXT: vpshufb %xmm5, %xmm3, %xmm3
; AVX2-NEXT: vpor %xmm2, %xmm3, %xmm2
; AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm3 = [0,4,6,10,12,0,0,0,0,0,0,2,4,8,10,14,0,4,6,10,12,0,0,0,0,0,0,2,4,8,10,14]
@@ -449,10 +449,10 @@ define <64 x i8> @f4(ptr %p0) {
; AVX2-NEXT: vmovdqa {{.*#+}} xmm6 = [255,255,255,255,255,255,255,255,255,255,255,0,0,0,0,0]
; AVX2-NEXT: vpblendvb %ymm6, %ymm2, %ymm0, %ymm0
; AVX2-NEXT: vmovdqa 80(%rdi), %xmm2
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = <u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm7 = [u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14]
; AVX2-NEXT: vpshufb %xmm7, %xmm2, %xmm2
; AVX2-NEXT: vmovdqa 64(%rdi), %xmm8
-; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = <u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128>
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm9 = [u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128]
; AVX2-NEXT: vpshufb %xmm9, %xmm8, %xmm8
; AVX2-NEXT: vpor %xmm2, %xmm8, %xmm2
; AVX2-NEXT: vinserti128 $1, %xmm2, %ymm0, %ymm2
@@ -477,18 +477,18 @@ define <64 x i8> @f4(ptr %p0) {
; AVX512F-LABEL: f4:
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa 176(%rdi), %xmm0
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14]
; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512F-NEXT: vmovdqa 160(%rdi), %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128]
; AVX512F-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512F-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512F-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512F-NEXT: vmovdqa (%rdi), %xmm2
; AVX512F-NEXT: vmovdqa 16(%rdi), %xmm4
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm5 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm5, %xmm2, %xmm2
-; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX512F-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX512F-NEXT: vpshufb %xmm6, %xmm4, %xmm4
; AVX512F-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX512F-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm0
@@ -517,18 +517,18 @@ define <64 x i8> @f4(ptr %p0) {
; AVX512BW-LABEL: f4:
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa 176(%rdi), %xmm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = <u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm1 = [u,u,u,u,u,128,128,128,128,128,128,2,4,8,10,14]
; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0
; AVX512BW-NEXT: vmovdqa 160(%rdi), %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = <u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm3 = [u,u,u,u,u,0,2,6,8,12,14,128,128,128,128,128]
; AVX512BW-NEXT: vpshufb %xmm3, %xmm2, %xmm2
; AVX512BW-NEXT: vpor %xmm0, %xmm2, %xmm0
; AVX512BW-NEXT: vinserti128 $1, %xmm0, %ymm0, %ymm0
; AVX512BW-NEXT: vmovdqa (%rdi), %xmm2
; AVX512BW-NEXT: vmovdqa 16(%rdi), %xmm4
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = <0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm5 = [0,4,6,10,12,128,128,128,128,128,128,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm5, %xmm2, %xmm2
-; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = <128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm6 = [128,128,128,128,128,0,2,6,8,12,14,u,u,u,u,u]
; AVX512BW-NEXT: vpshufb %xmm6, %xmm4, %xmm4
; AVX512BW-NEXT: vpor %xmm2, %xmm4, %xmm2
; AVX512BW-NEXT: vinserti64x4 $1, %ymm0, %zmm2, %zmm0
@@ -558,7 +558,7 @@ define <64 x i8> @f4(ptr %p0) {
; AVX512VBMI-LABEL: f4:
; AVX512VBMI: # %bb.0:
; AVX512VBMI-NEXT: vmovdqa64 64(%rdi), %zmm0
-; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = <64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,0,2,6,8,12,14,18,20,24,26,30,32,36,38,42,44,48,50,54,56,60,62,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm1 = [64,68,70,74,76,80,82,86,88,92,94,98,100,104,106,110,112,116,118,122,124,0,2,6,8,12,14,18,20,24,26,30,32,36,38,42,44,48,50,54,56,60,62,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u]
; AVX512VBMI-NEXT: vpermi2b (%rdi), %zmm0, %zmm1
; AVX512VBMI-NEXT: vmovdqa64 {{.*#+}} zmm0 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,66,68,72,74,78,80,84,86,90,92,96,98,102,104,108,110,114,116,120,122,126]
; AVX512VBMI-NEXT: vpermi2b 128(%rdi), %zmm1, %zmm0
diff --git a/llvm/test/CodeGen/X86/widen_arith-2.ll b/llvm/test/CodeGen/X86/widen_arith-2.ll
index 1a39b7635766be3..a0f78db8746f45a 100644
--- a/llvm/test/CodeGen/X86/widen_arith-2.ll
+++ b/llvm/test/CodeGen/X86/widen_arith-2.ll
@@ -9,7 +9,7 @@ define void @update(ptr %dst_i, ptr %src_i, i32 %n) nounwind {
; CHECK-NEXT: subl $12, %esp
; CHECK-NEXT: movl $0, (%esp)
; CHECK-NEXT: pcmpeqd %xmm0, %xmm0
-; CHECK-NEXT: movdqa {{.*#+}} xmm1 = <4,4,4,4,4,4,4,4,u,u,u,u,u,u,u,u>
+; CHECK-NEXT: movdqa {{.*#+}} xmm1 = [4,4,4,4,4,4,4,4,u,u,u,u,u,u,u,u]
; CHECK-NEXT: .p2align 4, 0x90
; CHECK-NEXT: .LBB0_1: # %forcond
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
diff --git a/llvm/test/CodeGen/X86/widen_arith-4.ll b/llvm/test/CodeGen/X86/widen_arith-4.ll
index 9f6232a69d1ae2d..8bb716374b48226 100644
--- a/llvm/test/CodeGen/X86/widen_arith-4.ll
+++ b/llvm/test/CodeGen/X86/widen_arith-4.ll
@@ -14,8 +14,8 @@ define void @update(ptr %dst, ptr %src, i32 %n) nounwind {
; SSE2-NEXT: movq %rax, -{{[0-9]+}}(%rsp)
; SSE2-NEXT: movw $0, -{{[0-9]+}}(%rsp)
; SSE2-NEXT: movl $0, -{{[0-9]+}}(%rsp)
-; SSE2-NEXT: movdqa {{.*#+}} xmm0 = <271,271,271,271,271,u,u,u>
-; SSE2-NEXT: movdqa {{.*#+}} xmm1 = <2,4,2,2,2,u,u,u>
+; SSE2-NEXT: movdqa {{.*#+}} xmm0 = [271,271,271,271,271,u,u,u]
+; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [2,4,2,2,2,u,u,u]
; SSE2-NEXT: .p2align 4, 0x90
; SSE2-NEXT: .LBB0_1: # %forcond
; SSE2-NEXT: # =>This Inner Loop Header: Depth=1
@@ -48,7 +48,7 @@ define void @update(ptr %dst, ptr %src, i32 %n) nounwind {
; SSE41-NEXT: movq %rax, -{{[0-9]+}}(%rsp)
; SSE41-NEXT: movw $0, -{{[0-9]+}}(%rsp)
; SSE41-NEXT: movl $0, -{{[0-9]+}}(%rsp)
-; SSE41-NEXT: movdqa {{.*#+}} xmm0 = <271,271,271,271,271,u,u,u>
+; SSE41-NEXT: movdqa {{.*#+}} xmm0 = [271,271,271,271,271,u,u,u]
; SSE41-NEXT: .p2align 4, 0x90
; SSE41-NEXT: .LBB0_1: # %forcond
; SSE41-NEXT: # =>This Inner Loop Header: Depth=1
diff --git a/llvm/test/CodeGen/X86/widen_arith-5.ll b/llvm/test/CodeGen/X86/widen_arith-5.ll
index 52fcbf4e0e885ff..8f1cbd1767f8816 100644
--- a/llvm/test/CodeGen/X86/widen_arith-5.ll
+++ b/llvm/test/CodeGen/X86/widen_arith-5.ll
@@ -13,7 +13,7 @@ define void @update(ptr %dst, ptr %src, i32 %n) nounwind {
; CHECK-NEXT: movq %rax, -{{[0-9]+}}(%rsp)
; CHECK-NEXT: movl $1, -{{[0-9]+}}(%rsp)
; CHECK-NEXT: movl $0, -{{[0-9]+}}(%rsp)
-; CHECK-NEXT: movdqa {{.*#+}} xmm0 = <3,3,3,u>
+; CHECK-NEXT: movdqa {{.*#+}} xmm0 = [3,3,3,u]
; CHECK-NEXT: .p2align 4, 0x90
; CHECK-NEXT: .LBB0_1: # %forcond
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
diff --git a/llvm/test/CodeGen/X86/widen_arith-6.ll b/llvm/test/CodeGen/X86/widen_arith-6.ll
index 5fea31688e993f4..239fe25b0c4a9ad 100644
--- a/llvm/test/CodeGen/X86/widen_arith-6.ll
+++ b/llvm/test/CodeGen/X86/widen_arith-6.ll
@@ -14,7 +14,7 @@ define void @update(ptr %dst, ptr %src, i32 %n) nounwind {
; CHECK-NEXT: movl $1073741824, {{[0-9]+}}(%esp) # imm = 0x40000000
; CHECK-NEXT: movl $1065353216, {{[0-9]+}}(%esp) # imm = 0x3F800000
; CHECK-NEXT: movl $0, {{[0-9]+}}(%esp)
-; CHECK-NEXT: movaps {{.*#+}} xmm0 = <1.97604004E+3,1.97604004E+3,1.97604004E+3,u>
+; CHECK-NEXT: movaps {{.*#+}} xmm0 = [1.97604004E+3,1.97604004E+3,1.97604004E+3,u]
; CHECK-NEXT: .p2align 4, 0x90
; CHECK-NEXT: .LBB0_1: # %forcond
; CHECK-NEXT: # =>This Inner Loop Header: Depth=1
diff --git a/llvm/test/CodeGen/X86/x86-interleaved-access.ll b/llvm/test/CodeGen/X86/x86-interleaved-access.ll
index 4f20a9d5db6ac5c..742a3a0c66a2ac6 100644
--- a/llvm/test/CodeGen/X86/x86-interleaved-access.ll
+++ b/llvm/test/CodeGen/X86/x86-interleaved-access.ll
@@ -379,7 +379,7 @@ define <8 x i8> @interleaved_load_vf8_i8_stride4(ptr %ptr) nounwind {
; AVX1OR2: # %bb.0:
; AVX1OR2-NEXT: vmovdqa (%rdi), %xmm0
; AVX1OR2-NEXT: vmovdqa 16(%rdi), %xmm1
-; AVX1OR2-NEXT: vmovdqa {{.*#+}} xmm2 = <1,u,5,u,9,u,13,u,13,u,5,u,12,u,13,u>
+; AVX1OR2-NEXT: vmovdqa {{.*#+}} xmm2 = [1,u,5,u,9,u,13,u,13,u,5,u,12,u,13,u]
; AVX1OR2-NEXT: vpshufb %xmm2, %xmm1, %xmm3
; AVX1OR2-NEXT: vpshufb %xmm2, %xmm0, %xmm2
; AVX1OR2-NEXT: vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
@@ -388,7 +388,7 @@ define <8 x i8> @interleaved_load_vf8_i8_stride4(ptr %ptr) nounwind {
; AVX1OR2-NEXT: vpblendw {{.*#+}} xmm3 = xmm0[0],xmm3[1],xmm0[2],xmm3[3],xmm0[4],xmm3[5],xmm0[6],xmm3[7]
; AVX1OR2-NEXT: vpackusdw %xmm4, %xmm3, %xmm3
; AVX1OR2-NEXT: vpaddb %xmm2, %xmm3, %xmm2
-; AVX1OR2-NEXT: vmovdqa {{.*#+}} xmm3 = <3,u,7,u,11,u,15,u,7,u,15,u,6,u,7,u>
+; AVX1OR2-NEXT: vmovdqa {{.*#+}} xmm3 = [3,u,7,u,11,u,15,u,7,u,15,u,6,u,7,u]
; AVX1OR2-NEXT: vpshufb %xmm3, %xmm1, %xmm4
; AVX1OR2-NEXT: vpshufb %xmm3, %xmm0, %xmm3
; AVX1OR2-NEXT: vpunpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm4[0]
@@ -404,7 +404,7 @@ define <8 x i8> @interleaved_load_vf8_i8_stride4(ptr %ptr) nounwind {
; AVX512: # %bb.0:
; AVX512-NEXT: vmovdqu (%rdi), %ymm0
; AVX512-NEXT: vmovdqa 16(%rdi), %xmm1
-; AVX512-NEXT: vmovdqa {{.*#+}} xmm2 = <3,u,7,u,11,u,15,u,7,u,15,u,6,u,7,u>
+; AVX512-NEXT: vmovdqa {{.*#+}} xmm2 = [3,u,7,u,11,u,15,u,7,u,15,u,6,u,7,u]
; AVX512-NEXT: vpshufb %xmm2, %xmm1, %xmm3
; AVX512-NEXT: vpshufb %xmm2, %xmm0, %xmm2
; AVX512-NEXT: vpunpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm3[0]
@@ -1080,9 +1080,9 @@ define void @interleaved_store_vf64_i8_stride3(<64 x i8> %a, <64 x i8> %b, <64 x
; AVX1-NEXT: vmovdqa %ymm0, %ymm2
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm9
; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm12
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm13 = <u,u,u,u,u,128,128,128,128,128,128,6,7,8,9,10>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm13 = [u,u,u,u,u,128,128,128,128,128,128,6,7,8,9,10]
; AVX1-NEXT: vpshufb %xmm13, %xmm12, %xmm6
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm14 = <u,u,u,u,u,5,6,7,8,9,10,128,128,128,128,128>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm14 = [u,u,u,u,u,5,6,7,8,9,10,128,128,128,128,128]
; AVX1-NEXT: vextractf128 $1, %ymm3, %xmm11
; AVX1-NEXT: vpshufb %xmm14, %xmm11, %xmm7
; AVX1-NEXT: vpor %xmm6, %xmm7, %xmm0
@@ -1107,7 +1107,7 @@ define void @interleaved_store_vf64_i8_stride3(<64 x i8> %a, <64 x i8> %b, <64 x
; AVX1-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-NEXT: vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm6 # 32-byte Reload
; AVX1-NEXT: vpunpckhbw {{.*#+}} xmm1 = xmm6[8],xmm4[8],xmm6[9],xmm4[9],xmm6[10],xmm4[10],xmm6[11],xmm4[11],xmm6[12],xmm4[12],xmm6[13],xmm4[13],xmm6[14],xmm4[14],xmm6[15],xmm4[15]
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <u,u,u,u,u,4,6,8,10,12,14,7,9,11,13,15>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [u,u,u,u,u,4,6,8,10,12,14,7,9,11,13,15]
; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
; AVX1-NEXT: vmovdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-NEXT: vextractf128 $1, %ymm6, %xmm9
@@ -1290,9 +1290,9 @@ define <64 x i8> @interleaved_load_vf64_i8_stride3(ptr %ptr){
; AVX1-NEXT: vpshufb %xmm9, %xmm10, %xmm7
; AVX1-NEXT: vpshufb %xmm9, %xmm11, %xmm8
; AVX1-NEXT: vpshufb %xmm9, %xmm13, %xmm9
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm14 = <1,4,7,10,13,128,128,128,128,128,128,u,u,u,u,u>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm14 = [1,4,7,10,13,128,128,128,128,128,128,u,u,u,u,u]
; AVX1-NEXT: vpshufb %xmm14, %xmm5, %xmm5
-; AVX1-NEXT: vmovdqa {{.*#+}} xmm15 = <128,128,128,128,128,0,3,6,9,12,15,u,u,u,u,u>
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm15 = [128,128,128,128,128,0,3,6,9,12,15,u,u,u,u,u]
; AVX1-NEXT: vpshufb %xmm15, %xmm2, %xmm12
; AVX1-NEXT: vmovdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; AVX1-NEXT: vpor %xmm5, %xmm12, %xmm0
diff --git a/llvm/test/CodeGen/X86/zero_extend_vector_inreg.ll b/llvm/test/CodeGen/X86/zero_extend_vector_inreg.ll
index 8d6c1483d817f0c..023ac96181d3cc6 100644
--- a/llvm/test/CodeGen/X86/zero_extend_vector_inreg.ll
+++ b/llvm/test/CodeGen/X86/zero_extend_vector_inreg.ll
@@ -2329,7 +2329,7 @@ define void @vec256_v8i32_to_v2i128_factor4(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX2-FAST: # %bb.0:
; AVX2-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
@@ -2855,7 +2855,7 @@ define void @vec384_v48i8_to_v16i24_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa (%rdi), %ymm0
; AVX512BW-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2,u,3,3,u,4,4,u,5>
+; AVX512BW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2,u,3,3,u,4,4,u,5]
; AVX512BW-NEXT: vpermw %ymm0, %ymm1, %ymm1
; AVX512BW-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX512BW-NEXT: vpshufb {{.*#+}} xmm0 = zero,xmm0[11],zero,zero,xmm0[12],zero,zero,xmm0[13],zero,zero,xmm0[14],zero,zero,xmm0[15],zero,zero
@@ -3824,7 +3824,7 @@ define void @vec384_v24i16_to_v8i48_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX2-SLOW: # %bb.0:
; AVX2-SLOW-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-SLOW-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2>
+; AVX2-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2]
; AVX2-SLOW-NEXT: vpermd %ymm0, %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX2-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,3,3,3]
@@ -3841,7 +3841,7 @@ define void @vec384_v24i16_to_v8i48_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX2-FAST-PERLANE: # %bb.0:
; AVX2-FAST-PERLANE-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-PERLANE-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2>
+; AVX2-FAST-PERLANE-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2]
; AVX2-FAST-PERLANE-NEXT: vpermd %ymm0, %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX2-FAST-PERLANE-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[12,13],zero,zero,zero,zero,xmm0[14,15],zero,zero,zero,zero
@@ -3856,7 +3856,7 @@ define void @vec384_v24i16_to_v8i48_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX2-FAST: # %bb.0:
; AVX2-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm1, %ymm1
; AVX2-FAST-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[12,13],zero,zero,zero,zero,xmm0[14,15],zero,zero,zero,zero
@@ -3871,7 +3871,7 @@ define void @vec384_v24i16_to_v8i48_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX512F-SLOW: # %bb.0:
; AVX512F-SLOW-NEXT: vmovdqa (%rdi), %ymm0
; AVX512F-SLOW-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2>
+; AVX512F-SLOW-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2]
; AVX512F-SLOW-NEXT: vpermd %ymm0, %ymm1, %ymm1
; AVX512F-SLOW-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX512F-SLOW-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,3,3,3]
@@ -3888,7 +3888,7 @@ define void @vec384_v24i16_to_v8i48_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX512F-FAST: # %bb.0:
; AVX512F-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX512F-FAST-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,1,1,u,2,2>
+; AVX512F-FAST-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,u,1,1,u,2,2]
; AVX512F-FAST-NEXT: vpermd %ymm0, %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm1, %ymm1
; AVX512F-FAST-NEXT: vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[12,13],zero,zero,zero,zero,xmm0[14,15],zero,zero,zero,zero
@@ -4855,7 +4855,7 @@ define void @vec384_v12i32_to_v4i96_factor3(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa (%rdi), %ymm0
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <16,1,2,17,4,5,18,7,8,19,10,11,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [16,1,2,17,4,5,18,7,8,19,10,11,u,u,u,u]
; AVX512F-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
@@ -5009,7 +5009,7 @@ define void @vec384_v12i32_to_v3i128_factor4(ptr %in.vec.base.ptr, ptr %in.vec.b
; AVX2-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX2-FAST-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm1 = ymm2[0],ymm1[1,2,3],ymm2[4],ymm1[5,6,7]
; AVX2-FAST-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[8,9,10,11],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
@@ -5024,7 +5024,7 @@ define void @vec384_v12i32_to_v3i128_factor4(ptr %in.vec.base.ptr, ptr %in.vec.b
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa (%rdi), %ymm0
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = <16,1,2,3,17,5,6,7,18,9,10,11,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm1 = [16,1,2,3,17,5,6,7,18,9,10,11,u,u,u,u]
; AVX512F-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX512F-NEXT: vpermt2d %zmm0, %zmm1, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
@@ -5369,7 +5369,7 @@ define void @vec384_v6i64_to_v3i128_factor2(ptr %in.vec.base.ptr, ptr %in.vec.bi
; AVX512F-NEXT: vmovdqa (%rdi), %ymm0
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,9,1,11,2,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,9,1,11,2,13,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -7178,10 +7178,10 @@ define void @vec512_v16i32_to_v4i128_factor4(ptr %in.vec.base.ptr, ptr %in.vec.b
; AVX2-FAST-NEXT: vmovdqa (%rdi), %ymm0
; AVX2-FAST-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX2-FAST-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = <0,u,u,u,1,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm2 = [0,u,u,u,1,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm2, %ymm2
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm2 = ymm2[0],ymm1[1,2,3],ymm2[4],ymm1[5,6,7]
-; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = <2,u,u,u,3,u,u,u>
+; AVX2-FAST-NEXT: vmovdqa {{.*#+}} ymm3 = [2,u,u,u,3,u,u,u]
; AVX2-FAST-NEXT: vpermd %ymm0, %ymm3, %ymm0
; AVX2-FAST-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4],ymm1[5,6,7]
; AVX2-FAST-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
diff --git a/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast.ll b/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast.ll
index a7b1fa24ce17d6a..b4c79aff5cef1f2 100644
--- a/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast.ll
+++ b/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast.ll
@@ -1053,7 +1053,7 @@ define void @vec256_i8_widen_to_i16_factor2_broadcast_to_v16i16_factor16(ptr %in
; SSE42-NEXT: paddb 48(%rsi), %xmm2
; SSE42-NEXT: paddb (%rsi), %xmm0
; SSE42-NEXT: paddb 32(%rsi), %xmm1
-; SSE42-NEXT: movdqa {{.*#+}} xmm3 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSE42-NEXT: movdqa {{.*#+}} xmm3 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pshufb %xmm3, %xmm1
; SSE42-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE42-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
@@ -4736,7 +4736,7 @@ define void @vec384_i32_widen_to_i64_factor2_broadcast_to_v6i64_factor6(ptr %in.
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -4754,7 +4754,7 @@ define void @vec384_i32_widen_to_i64_factor2_broadcast_to_v6i64_factor6(ptr %in.
; AVX512DQ-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512DQ-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -4941,7 +4941,7 @@ define void @vec384_i32_widen_to_i96_factor3_broadcast_to_v4i96_factor4(ptr %in.
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -4959,7 +4959,7 @@ define void @vec384_i32_widen_to_i96_factor3_broadcast_to_v4i96_factor4(ptr %in.
; AVX512DQ-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512DQ-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -5099,7 +5099,7 @@ define void @vec384_i32_widen_to_i128_factor4_broadcast_to_v3i128_factor3(ptr %i
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -5117,7 +5117,7 @@ define void @vec384_i32_widen_to_i128_factor4_broadcast_to_v3i128_factor3(ptr %i
; AVX512DQ-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512DQ-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -5374,7 +5374,7 @@ define void @vec384_i64_widen_to_i128_factor2_broadcast_to_v3i128_factor3(ptr %i
; AVX512F-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512F-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,0,11,0,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,0,11,0,13,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
@@ -5392,7 +5392,7 @@ define void @vec384_i64_widen_to_i128_factor2_broadcast_to_v3i128_factor3(ptr %i
; AVX512DQ-NEXT: vpaddb (%rsi), %ymm0, %ymm0
; AVX512DQ-NEXT: vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,0,11,0,13,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,0,11,0,13,u,u]
; AVX512DQ-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rdx), %ymm0, %ymm0
diff --git a/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast_from_memory.ll b/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast_from_memory.ll
index b88e2921484d9b1..23f02e9245eedad 100644
--- a/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast_from_memory.ll
+++ b/llvm/test/CodeGen/X86/zero_extend_vector_inreg_of_broadcast_from_memory.ll
@@ -875,7 +875,7 @@ define void @vec256_i8_widen_to_i16_factor2_broadcast_to_v16i16_factor16(ptr %in
; SSE42-NEXT: movdqa (%rdi), %xmm0
; SSE42-NEXT: movdqa 32(%rdi), %xmm1
; SSE42-NEXT: movdqa 48(%rdi), %xmm2
-; SSE42-NEXT: movdqa {{.*#+}} xmm3 = <1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u>
+; SSE42-NEXT: movdqa {{.*#+}} xmm3 = [1,3,5,7,9,11,13,15,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pshufb %xmm3, %xmm1
; SSE42-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
; SSE42-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
@@ -2968,7 +2968,7 @@ define void @vec384_i16_widen_to_i32_factor2_broadcast_to_v12i32_factor12(ptr %i
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,25,0,27,0,29,0,31,0,41,0,43,0,45,0,47,0,49,0,51,0,53,0,55,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,25,0,27,0,29,0,31,0,41,0,43,0,45,0,47,0,49,0,51,0,53,0,55,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -3123,7 +3123,7 @@ define void @vec384_i16_widen_to_i48_factor3_broadcast_to_v8i48_factor8(ptr %in.
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,25,26,0,28,29,0,31,40,0,42,43,0,45,46,0,48,49,0,51,52,0,54,55,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,25,26,0,28,29,0,31,40,0,42,43,0,45,46,0,48,49,0,51,52,0,54,55,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -3307,7 +3307,7 @@ define void @vec384_i16_widen_to_i64_factor4_broadcast_to_v6i64_factor6(ptr %in.
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,25,26,27,0,29,30,31,0,41,42,43,0,45,46,47,0,49,50,51,0,53,54,55,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,25,26,27,0,29,30,31,0,41,42,43,0,45,46,47,0,49,50,51,0,53,54,55,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpermi2w %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -3465,7 +3465,7 @@ define void @vec384_i16_widen_to_i96_factor6_broadcast_to_v4i96_factor4(ptr %in.
;
; AVX512BW-LABEL: vec384_i16_widen_to_i96_factor6_broadcast_to_v4i96_factor4:
; AVX512BW: # %bb.0:
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <32,57,58,59,60,61,32,63,8,9,10,11,32,13,14,15,16,17,32,19,20,21,22,23,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [32,57,58,59,60,61,32,63,8,9,10,11,32,13,14,15,16,17,32,19,20,21,22,23,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: vpermt2w (%rdi), %zmm0, %zmm1
; AVX512BW-NEXT: vpaddb (%rsi), %zmm1, %zmm0
@@ -3580,7 +3580,7 @@ define void @vec384_i16_widen_to_i128_factor8_broadcast_to_v3i128_factor3(ptr %i
;
; AVX512BW-LABEL: vec384_i16_widen_to_i128_factor8_broadcast_to_v3i128_factor3:
; AVX512BW: # %bb.0:
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <32,57,58,59,60,61,62,63,32,9,10,11,12,13,14,15,32,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [32,57,58,59,60,61,62,63,32,9,10,11,12,13,14,15,32,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: vpermt2w (%rdi), %zmm0, %zmm1
; AVX512BW-NEXT: vpaddb (%rsi), %zmm1, %zmm0
@@ -3693,7 +3693,7 @@ define void @vec384_i16_widen_to_i192_factor12_broadcast_to_v2i192_factor2(ptr %
;
; AVX512BW-LABEL: vec384_i16_widen_to_i192_factor12_broadcast_to_v2i192_factor2:
; AVX512BW: # %bb.0:
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <32,57,58,59,60,61,62,63,8,9,10,11,32,13,14,15,16,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [32,57,58,59,60,61,62,63,8,9,10,11,32,13,14,15,16,17,18,19,20,21,22,23,u,u,u,u,u,u,u,u]
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: vpermt2w (%rdi), %zmm0, %zmm1
; AVX512BW-NEXT: vpaddb (%rsi), %zmm1, %zmm0
@@ -3822,7 +3822,7 @@ define void @vec384_i32_widen_to_i64_factor2_broadcast_to_v6i64_factor6(ptr %in.
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -3836,7 +3836,7 @@ define void @vec384_i32_widen_to_i64_factor2_broadcast_to_v6i64_factor6(ptr %in.
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -3850,7 +3850,7 @@ define void @vec384_i32_widen_to_i64_factor2_broadcast_to_v6i64_factor6(ptr %in.
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,0,15,0,21,0,23,0,25,0,27,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -3978,7 +3978,7 @@ define void @vec384_i32_widen_to_i96_factor3_broadcast_to_v4i96_factor4(ptr %in.
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -3992,7 +3992,7 @@ define void @vec384_i32_widen_to_i96_factor3_broadcast_to_v4i96_factor4(ptr %in.
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -4006,7 +4006,7 @@ define void @vec384_i32_widen_to_i96_factor3_broadcast_to_v4i96_factor4(ptr %in.
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,0,20,21,0,23,24,0,26,27,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -4092,7 +4092,7 @@ define void @vec384_i32_widen_to_i128_factor4_broadcast_to_v3i128_factor3(ptr %i
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u]
; AVX512F-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -4106,7 +4106,7 @@ define void @vec384_i32_widen_to_i128_factor4_broadcast_to_v3i128_factor3(ptr %i
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u]
; AVX512DQ-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -4120,7 +4120,7 @@ define void @vec384_i32_widen_to_i128_factor4_broadcast_to_v3i128_factor3(ptr %i
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,13,14,15,0,21,22,23,0,25,26,27,u,u,u,u]
; AVX512BW-NEXT: vpermi2d %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -4230,7 +4230,7 @@ define void @vec384_i32_widen_to_i192_factor6_broadcast_to_v2i192_factor2(ptr %i
;
; AVX512BW-LABEL: vec384_i32_widen_to_i192_factor6_broadcast_to_v2i192_factor2:
; AVX512BW: # %bb.0:
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = <16,29,30,31,4,5,16,7,8,9,10,11,u,u,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm0 = [16,29,30,31,4,5,16,7,8,9,10,11,u,u,u,u]
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX512BW-NEXT: vpermt2d (%rdi), %zmm0, %zmm1
; AVX512BW-NEXT: vpaddb (%rsi), %zmm1, %zmm0
@@ -4315,7 +4315,7 @@ define void @vec384_i64_widen_to_i128_factor2_broadcast_to_v3i128_factor3(ptr %i
; AVX512F: # %bb.0:
; AVX512F-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512F-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,0,11,0,13,u,u>
+; AVX512F-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,0,11,0,13,u,u]
; AVX512F-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512F-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512F-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -4329,7 +4329,7 @@ define void @vec384_i64_widen_to_i128_factor2_broadcast_to_v3i128_factor3(ptr %i
; AVX512DQ: # %bb.0:
; AVX512DQ-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512DQ-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,0,11,0,13,u,u>
+; AVX512DQ-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,0,11,0,13,u,u]
; AVX512DQ-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512DQ-NEXT: vextracti64x4 $1, %zmm2, %ymm0
; AVX512DQ-NEXT: vpaddb 32(%rsi), %ymm0, %ymm0
@@ -4343,7 +4343,7 @@ define void @vec384_i64_widen_to_i128_factor2_broadcast_to_v3i128_factor3(ptr %i
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,0,11,0,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,0,11,0,13,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -4453,7 +4453,7 @@ define void @vec384_i64_widen_to_i192_factor3_broadcast_to_v2i192_factor2(ptr %i
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,7,10,0,12,13,u,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,7,10,0,12,13,u,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpaddb (%rsi), %zmm2, %zmm0
; AVX512BW-NEXT: vmovdqa64 %zmm0, (%rdx)
@@ -4919,7 +4919,7 @@ define void @vec512_i8_widen_to_i256_factor32_broadcast_to_v2i256_factor2(ptr %i
; AVX512BW: # %bb.0:
; AVX512BW-NEXT: vmovdqa64 (%rdi), %zmm0
; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1
-; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = <0,u,10,u,0,u,14,u>
+; AVX512BW-NEXT: vmovdqa64 {{.*#+}} zmm2 = [0,u,10,u,0,u,14,u]
; AVX512BW-NEXT: vpermi2q %zmm1, %zmm0, %zmm2
; AVX512BW-NEXT: vpandq {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %zmm2, %zmm0
; AVX512BW-NEXT: vpaddb (%rsi), %zmm0, %zmm0
More information about the llvm-commits
mailing list