This is an archive of the discontinued LLVM Phabricator instance.

Differential D32687

[DAGCombine] (uaddo X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry)
ClosedPublic

Authored by deadalnix on Apr 30 2017, 12:59 PM.

Details

Reviewers
jyknight
nemanjai
mkuper
spatel
RKSimon
zvi
bkramer
Commits
rG106a7eab8494: [DAGCombine] (uaddo X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry)
rL301922: [DAGCombine] (uaddo X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry)
Summary

This is a common pattern that arise when legalizing large integers operations. Only do it when Y + 1 cannot overflow as this would change the carry behavior of uaddo .

Diff Detail

Event Timeline

deadalnix created this revision.Apr 30 2017, 12:59 PM
deadalnix mentioned this in D32738: [DAGCombine] (add X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry).May 2 2017, 3:38 AM
RKSimon accepted this revision.May 2 2017, 5:51 AM

LGTM with a couple of minors

lib/CodeGen/SelectionDAG/DAGCombiner.cpp
2101

Use SDValue if possible please.

This revision is now accepted and ready to land.May 2 2017, 5:51 AM
deadalnix mentioned this in rL301919: [DAGCombine] (add X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry).May 2 2017, 6:47 AM
deadalnix updated this revision to Diff 97446.May 2 2017, 7:10 AM

Rebase and use SDValue.

Closed by commit rL301922: [DAGCombine] (uaddo X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry) (authored by deadalnix). · Explain WhyMay 2 2017, 7:29 AM
This revision was automatically updated to reflect the committed changes.

Revision Contents

PathSize
lib/
CodeGen/
SelectionDAG/
21 lines
test/
CodeGen/
X86/
22 lines
2193 lines
19 lines
166 lines

Diff 97446

lib/CodeGen/SelectionDAG/DAGCombiner.cpp

  • This file is larger than 256 KB, so syntax highlighting is disabled by default.
Show First 20 Lines • Show All 231 Lines • ▼ Show 20 Lines private:
// //
SDValue visitTokenFactor(SDNode *N); SDValue visitTokenFactor(SDNode *N);
SDValue visitMERGE_VALUES(SDNode *N); SDValue visitMERGE_VALUES(SDNode *N);
SDValue visitADD(SDNode *N); SDValue visitADD(SDNode *N);
SDValue visitADDLike(SDValue N0, SDValue N1, SDNode *LocReference); SDValue visitADDLike(SDValue N0, SDValue N1, SDNode *LocReference);
SDValue visitSUB(SDNode *N); SDValue visitSUB(SDNode *N);
SDValue visitADDC(SDNode *N); SDValue visitADDC(SDNode *N);
SDValue visitUADDO(SDNode *N); SDValue visitUADDO(SDNode *N);
SDValue visitUADDOLike(SDValue N0, SDValue N1, SDNode *N);
SDValue visitSUBC(SDNode *N); SDValue visitSUBC(SDNode *N);
SDValue visitUSUBO(SDNode *N); SDValue visitUSUBO(SDNode *N);
SDValue visitADDE(SDNode *N); SDValue visitADDE(SDNode *N);
SDValue visitADDCARRY(SDNode *N); SDValue visitADDCARRY(SDNode *N);
SDValue visitSUBE(SDNode *N); SDValue visitSUBE(SDNode *N);
SDValue visitSUBCARRY(SDNode *N); SDValue visitSUBCARRY(SDNode *N);
SDValue visitMUL(SDNode *N); SDValue visitMUL(SDNode *N);
SDValue useDivRem(SDNode *N); SDValue useDivRem(SDNode *N);
▲ Show 20 Lines • Show All 1,830 Lines • ▼ Show 20 LinesSDValue DAGCombiner::visitUADDO(SDNode *N) {
if (isNullConstant(N1)) if (isNullConstant(N1))
return CombineTo(N, N0, DAG.getConstant(0, DL, CarryVT)); return CombineTo(N, N0, DAG.getConstant(0, DL, CarryVT));
// If it cannot overflow, transform into an add. // If it cannot overflow, transform into an add.
if (DAG.computeOverflowKind(N0, N1) == SelectionDAG::OFK_Never) if (DAG.computeOverflowKind(N0, N1) == SelectionDAG::OFK_Never)
return CombineTo(N, DAG.getNode(ISD::ADD, DL, VT, N0, N1), return CombineTo(N, DAG.getNode(ISD::ADD, DL, VT, N0, N1),
DAG.getConstant(0, DL, CarryVT)); DAG.getConstant(0, DL, CarryVT));
if (SDValue Combined = visitUADDOLike(N0, N1, N))
return Combined;
if (SDValue Combined = visitUADDOLike(N1, N0, N))
return Combined;
return SDValue();
}
SDValue DAGCombiner::visitUADDOLike(SDValue N0, SDValue N1, SDNode *N) {
// (uaddo X, (addcarry Y, 0, Carry)) -> (addcarry X, Y, Carry)
// If Y + 1 cannot overflow.
if (N1.getOpcode() == ISD::ADDCARRY && isNullConstant(N1.getOperand(1))) {
SDValue Y = N1.getOperand(0);
SDValue One = DAG.getConstant(1, SDLoc(N), Y.getValueType());
RKSimonUnsubmitted
Not Done
ReplyInline Actions

Use SDValue if possible please.

RKSimon: Use SDValue if possible please.
if (DAG.computeOverflowKind(Y, One) == SelectionDAG::OFK_Never)
return DAG.getNode(ISD::ADDCARRY, SDLoc(N), N->getVTList(), N0, Y,
N1.getOperand(2));
}
return SDValue(); return SDValue();
} }
SDValue DAGCombiner::visitADDE(SDNode *N) { SDValue DAGCombiner::visitADDE(SDNode *N) {
SDValue N0 = N->getOperand(0); SDValue N0 = N->getOperand(0);
SDValue N1 = N->getOperand(1); SDValue N1 = N->getOperand(1);
SDValue CarryIn = N->getOperand(2); SDValue CarryIn = N->getOperand(2);
▲ Show 20 Lines • Show All 14,425 Lines • Show Last 20 Lines

test/CodeGen/X86/adde-carry.ll

Show First 20 Lines • Show All 80 Lines • ▼ Show 20 Lines
ret void ret void
} }
%scalar = type { [4 x i64] } %scalar = type { [4 x i64] }
define %scalar @pr31719(%scalar* nocapture readonly %this, %scalar %arg.b) { define %scalar @pr31719(%scalar* nocapture readonly %this, %scalar %arg.b) {
; CHECK-LABEL: pr31719: ; CHECK-LABEL: pr31719:
; CHECK: # BB#0: # %entry ; CHECK: # BB#0: # %entry
; CHECK-NEXT: addq (%rsi), %rdx ; CHECK-NEXT: addq 8(%rsi), %rcx
; CHECK-NEXT: sbbq %r10, %r10 ; CHECK-NEXT: sbbq %r10, %r10
; CHECK-NEXT: andl $1, %r10d ; CHECK-NEXT: andl $1, %r10d
; CHECK-NEXT: addq 8(%rsi), %rcx ; CHECK-NEXT: addq 16(%rsi), %r8
; CHECK-NEXT: sbbq %rax, %rax ; CHECK-NEXT: sbbq %rax, %rax
; CHECK-NEXT: andl $1, %eax ; CHECK-NEXT: andl $1, %eax
; CHECK-NEXT: addq %r10, %rcx
; CHECK-NEXT: adcq $0, %rax
; CHECK-NEXT: addq 16(%rsi), %r8
; CHECK-NEXT: sbbq %r10, %r10
; CHECK-NEXT: andl $1, %r10d
; CHECK-NEXT: addq 24(%rsi), %r9 ; CHECK-NEXT: addq 24(%rsi), %r9
; CHECK-NEXT: addq %rax, %r8 ; CHECK-NEXT: addq (%rsi), %rdx
; CHECK-NEXT: adcq %r10, %r9 ; CHECK-NEXT: adcq $0, %rcx
; CHECK-NEXT: adcq %r8, %r10
; CHECK-NEXT: adcq %r9, %rax
; CHECK-NEXT: movq %rdx, (%rdi) ; CHECK-NEXT: movq %rdx, (%rdi)
; CHECK-NEXT: movq %rcx, 8(%rdi) ; CHECK-NEXT: movq %rcx, 8(%rdi)
; CHECK-NEXT: movq %r8, 16(%rdi) ; CHECK-NEXT: movq %r10, 16(%rdi)
; CHECK-NEXT: movq %r9, 24(%rdi) ; CHECK-NEXT: movq %rax, 24(%rdi)
; CHECK-NEXT: movq %rdi, %rax ; CHECK-NEXT: movq %rdi, %rax
; CHECK-NEXT: retq ; CHECK-NEXT: retq
entry: entry:
%0 = extractvalue %scalar %arg.b, 0 %0 = extractvalue %scalar %arg.b, 0
%.elt = extractvalue [4 x i64] %0, 0 %.elt = extractvalue [4 x i64] %0, 0
%.elt24 = extractvalue [4 x i64] %0, 1 %.elt24 = extractvalue [4 x i64] %0, 1
%.elt26 = extractvalue [4 x i64] %0, 2 %.elt26 = extractvalue [4 x i64] %0, 2
%.elt28 = extractvalue [4 x i64] %0, 3 %.elt28 = extractvalue [4 x i64] %0, 3
Show All 38 Lines
%accumulator= type { i64, i64, i32 } %accumulator= type { i64, i64, i32 }
define void @muladd(%accumulator* nocapture %this, i64 %arg.a, i64 %arg.b) { define void @muladd(%accumulator* nocapture %this, i64 %arg.a, i64 %arg.b) {
; CHECK-LABEL: muladd: ; CHECK-LABEL: muladd:
; CHECK: # BB#0: # %entry ; CHECK: # BB#0: # %entry
; CHECK-NEXT: movq %rdx, %rax ; CHECK-NEXT: movq %rdx, %rax
; CHECK-NEXT: mulq %rsi ; CHECK-NEXT: mulq %rsi
; CHECK-NEXT: addq (%rdi), %rax ; CHECK-NEXT: addq (%rdi), %rax
; CHECK-NEXT: adcq $0, %rdx
; CHECK-NEXT: movq %rax, (%rdi) ; CHECK-NEXT: movq %rax, (%rdi)
; CHECK-NEXT: addq 8(%rdi), %rdx ; CHECK-NEXT: adcq 8(%rdi), %rdx
; CHECK-NEXT: movq %rdx, 8(%rdi) ; CHECK-NEXT: movq %rdx, 8(%rdi)
; CHECK-NEXT: adcl $0, 16(%rdi) ; CHECK-NEXT: adcl $0, 16(%rdi)
; CHECK-NEXT: retq ; CHECK-NEXT: retq
entry: entry:
%0 = zext i64 %arg.a to i128 %0 = zext i64 %arg.a to i128
%1 = zext i64 %arg.b to i128 %1 = zext i64 %arg.b to i128
%2 = mul nuw i128 %1, %0 %2 = mul nuw i128 %1, %0
%3 = getelementptr inbounds %accumulator, %accumulator* %this, i64 0, i32 0 %3 = getelementptr inbounds %accumulator, %accumulator* %this, i64 0, i32 0
Show All 21 Lines

test/CodeGen/X86/mul-i1024.ll

Show First 20 Lines • Show All 4,382 Lines • ▼ Show 20 Lines
; X64-LABEL: test_1024: ; X64-LABEL: test_1024:
; X64: # BB#0: ; X64: # BB#0:
; X64-NEXT: pushq %rbp ; X64-NEXT: pushq %rbp
; X64-NEXT: pushq %r15 ; X64-NEXT: pushq %r15
; X64-NEXT: pushq %r14 ; X64-NEXT: pushq %r14
; X64-NEXT: pushq %r13 ; X64-NEXT: pushq %r13
; X64-NEXT: pushq %r12 ; X64-NEXT: pushq %r12
; X64-NEXT: pushq %rbx ; X64-NEXT: pushq %rbx
; X64-NEXT: subq $360, %rsp # imm = 0x168 ; X64-NEXT: subq $352, %rsp # imm = 0x160
; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq 48(%rdi), %r9 ; X64-NEXT: movq 48(%rdi), %r8
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq 40(%rdi), %rcx ; X64-NEXT: movq 40(%rdi), %rcx
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq 32(%rdi), %rax ; X64-NEXT: movq 32(%rdi), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdi, %r10 ; X64-NEXT: movq %rdi, %r13
; X64-NEXT: xorl %r8d, %r8d ; X64-NEXT: xorl %r9d, %r9d
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rdi, %rcx ; X64-NEXT: addq %rbx, %rcx
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: addq %r11, %rcx ; X64-NEXT: addq %r11, %rcx
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdi, %rbp ; X64-NEXT: adcq %rbx, %rbp
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: movq %rbx, %rcx
; X64-NEXT: addq %rbx, %rbp
; X64-NEXT: sbbq %rbx, %rbx ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: andl $1, %ebx ; X64-NEXT: andl $1, %ebx
; X64-NEXT: addq %rax, %rbp ; X64-NEXT: addq %rax, %rbp
; X64-NEXT: adcq %rdx, %rbx ; X64-NEXT: adcq %rdx, %rbx
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r11, %r13 ; X64-NEXT: movq %r11, %r12
; X64-NEXT: addq %rax, %r13 ; X64-NEXT: addq %rax, %r12
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rdi, %r15 ; X64-NEXT: movq %rcx, %r8
; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rdx, %rax ; X64-NEXT: adcq %rdx, %rax
; X64-NEXT: addq %rbp, %r13 ; X64-NEXT: addq %rbp, %r12
; X64-NEXT: movq %r13, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rbx, %rax ; X64-NEXT: adcq %rbx, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rsi, %r8 ; X64-NEXT: movq (%rsi), %rax
; X64-NEXT: movq (%r8), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: xorl %ebp, %ebp ; X64-NEXT: xorl %ebp, %ebp
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq 8(%r8), %rax ; X64-NEXT: movq 8(%rsi), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rbp
; X64-NEXT: xorl %r9d, %r9d ; X64-NEXT: xorl %r9d, %r9d
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rcx, %r12 ; X64-NEXT: addq %rcx, %r15
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: addq %r14, %r12 ; X64-NEXT: addq %r10, %r15
; X64-NEXT: movq %rcx, %rbx ; X64-NEXT: adcq %rcx, %rbp
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, %rdi
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %rbp, %rbx ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: andl $1, %ebx
; X64-NEXT: andl $1, %ebp ; X64-NEXT: addq %rax, %rbp
; X64-NEXT: addq %rax, %rbx ; X64-NEXT: adcq %rdx, %rbx
; X64-NEXT: adcq %rdx, %rbp ; X64-NEXT: movq 16(%rsi), %rax
; X64-NEXT: movq 16(%r8), %rax ; X64-NEXT: movq %rsi, %r14
; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r14, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r9 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r14, %r9 ; X64-NEXT: movq %r10, %rcx
; X64-NEXT: addq %rax, %r9 ; X64-NEXT: movq %rcx, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: adcq %rdx, %rax ; X64-NEXT: movq %rdi, %r9
; X64-NEXT: addq %rbx, %r9 ; X64-NEXT: adcq %rdx, %r9
; X64-NEXT: adcq %rbp, %rax ; X64-NEXT: addq %rbp, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rsi, %r10
; X64-NEXT: movq %r11, %rax ; X64-NEXT: adcq %rbx, %r9
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %r14, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: adcq %rcx, %r15 ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: adcq %rdi, %r8
; X64-NEXT: movq (%r10), %rax ; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r13, %rbp
; X64-NEXT: movq (%rbp), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: xorl %r15d, %r15d ; X64-NEXT: xorl %r8d, %r8d
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r14, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: movq %r14, %rdi
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: adcq %rcx, %rax ; X64-NEXT: adcq %rdi, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq 32(%r8), %rax ; X64-NEXT: movq 32(%r14), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r8
; X64-NEXT: xorl %r8d, %r8d ; X64-NEXT: movq %rax, %r13
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, (%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq %rbx, %rcx
; X64-NEXT: movq %rbx, %r14 ; X64-NEXT: movq %rbx, %r14
; X64-NEXT: addq %rax, %rcx ; X64-NEXT: addq %r13, %rax
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: adcq %rdx, %rax ; X64-NEXT: adcq %rdx, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %rdi, %r11 ; X64-NEXT: addq %rcx, %r11
; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdi, %r11 ; X64-NEXT: movq %rcx, %r8
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: adcq %r12, %rax ; X64-NEXT: adcq %r15, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r9, %r13 ; X64-NEXT: adcq %r10, %r12
; X64-NEXT: movq %r13, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r12, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r9, %rdi ; X64-NEXT: movq %r10, %rcx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: adcq %rbp, %rax ; X64-NEXT: adcq %r9, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbp, %r9 ; X64-NEXT: movq %r9, %r10
; X64-NEXT: movq 8(%r10), %rax ; X64-NEXT: movq 8(%rbp), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbp, %rdi
; X64-NEXT: mulq %r8
; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rsi, %r15
; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: adcq $0, %rbx
; X64-NEXT: addq %r14, %r15
; X64-NEXT: movq %rsi, %rbp
; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rbp
; X64-NEXT: addq %rbx, %rbp
; X64-NEXT: sbbq %r8, %r8
; X64-NEXT: andl $1, %r8d
; X64-NEXT: addq %rax, %rbp
; X64-NEXT: adcq %rdx, %r8
; X64-NEXT: movq 16(%r10), %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r14, %rbx
; X64-NEXT: addq %rax, %rbx
; X64-NEXT: movq %rsi, %r10
; X64-NEXT: adcq %rdx, %r10
; X64-NEXT: addq %rbp, %rbx
; X64-NEXT: adcq %r8, %r10
; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq %r14, (%rsp) # 8-byte Spill
; X64-NEXT: addq %r11, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r15, %rcx
; X64-NEXT: adcq %rcx, %r12
; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rbx, %rdi
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbx, %r8
; X64-NEXT: adcq %r10, %r9
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: movq 40(%r13), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: xorl %edx, %edx ; X64-NEXT: xorl %edx, %edx
; X64-NEXT: mulq %rdx ; X64-NEXT: mulq %rdx
; X64-NEXT: xorl %r11d, %r11d ; X64-NEXT: xorl %r9d, %r9d
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %r12
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r9 # 8-byte Reload ; X64-NEXT: addq %rsi, %r12
; X64-NEXT: addq %r9, %rsi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: addq %r14, %r12
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: adcq %rsi, %rbp
; X64-NEXT: addq %rdi, %rsi ; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r9, %rbp ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: andl $1, %ebx
; X64-NEXT: addq %rax, %rbp
; X64-NEXT: adcq %rdx, %rbx
; X64-NEXT: movq 16(%rdi), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r9
; X64-NEXT: xorl %edi, %edi
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r14, %r9
; X64-NEXT: addq %rax, %r9
; X64-NEXT: adcq %rdx, %rsi
; X64-NEXT: addq %rbp, %r9
; X64-NEXT: movq %r9, %rdx
; X64-NEXT: adcq %rbx, %rsi
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq %r14, %rsi
; X64-NEXT: movq %r14, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %r8, %rsi
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r12, %r15
; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %r15
; X64-NEXT: adcq %rax, %r10
; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq 40(%rsi), %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rdi
; X64-NEXT: xorl %r8d, %r8d
; X64-NEXT: movq %rax, %rcx
; X64-NEXT: movq (%rsp), %rdi # 8-byte Reload
; X64-NEXT: addq %rdi, %rcx
; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: addq %rbx, %rbp ; X64-NEXT: addq %r13, %rcx
; X64-NEXT: adcq %rdi, %rbp
; X64-NEXT: sbbq %rbx, %rbx ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: andl $1, %ebx ; X64-NEXT: andl $1, %ebx
; X64-NEXT: addq %rax, %rbp ; X64-NEXT: addq %rax, %rbp
; X64-NEXT: adcq %rdx, %rbx ; X64-NEXT: adcq %rdx, %rbx
; X64-NEXT: movq 48(%r13), %rax ; X64-NEXT: movq 48(%rsi), %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r11 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdi, %r15 ; X64-NEXT: movq %r13, %r8
; X64-NEXT: movq %rdi, %r11 ; X64-NEXT: addq %rax, %r8
; X64-NEXT: addq %rax, %r15 ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %r9, %rdi ; X64-NEXT: adcq %rdx, %rax
; X64-NEXT: adcq %rdx, %rdi ; X64-NEXT: addq %rbp, %r8
; X64-NEXT: addq %rbp, %r15 ; X64-NEXT: adcq %rbx, %rax
; X64-NEXT: adcq %rbx, %rdi ; X64-NEXT: movq %r13, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %r11, %r14 ; X64-NEXT: addq %r13, %r14
; X64-NEXT: movq %r14, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r14, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rsi, %rcx ; X64-NEXT: adcq %rcx, %r12
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r15, %r8 ; X64-NEXT: adcq %r8, %r15
; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rdi, %r10 ; X64-NEXT: adcq %rax, %r10
; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r14 # 8-byte Reload ; X64-NEXT: movq %rax, %rdx
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: addq %r11, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: addq %r13, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: adcq %r9, %rax ; X64-NEXT: adcq %rdi, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: addq %r11, %rax ; X64-NEXT: addq %r13, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload
; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r8
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq 56(%rax), %rsi ; X64-NEXT: movq 56(%rax), %r11
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdi, %r10 ; X64-NEXT: movq %rdi, %r9
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rbx, %rbp ; X64-NEXT: addq %rsi, %rbx
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: addq %rbp, %r8 ; X64-NEXT: addq %rbx, %r10
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: addq %rdi, %rbx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: andl $1, %ecx
; X64-NEXT: andl $1, %edi ; X64-NEXT: movq %r11, %rax
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: mulq %rdi
; X64-NEXT: mulq %rcx ; X64-NEXT: movq %rdi, %r14
; X64-NEXT: movq %rcx, %r11 ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: adcq %rdi, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r15 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r15 # 8-byte Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r12 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r12 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %r15 ; X64-NEXT: addq %rax, %r15
; X64-NEXT: adcq %rdx, %r12 ; X64-NEXT: adcq %rdx, %r12
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %r10, %rsi ; X64-NEXT: movq %r9, %rcx
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r10 ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r10, %rbx ; X64-NEXT: addq %r9, %rbx
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rdi, %r10 ; X64-NEXT: mulq %r14
; X64-NEXT: mulq %r11 ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq %rbp, %rcx
; X64-NEXT: addq %rbp, %rdi
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: sbbq %rbp, %rbp
; X64-NEXT: andl $1, %ebp ; X64-NEXT: andl $1, %ebp
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rsi, %rbx
; X64-NEXT: movq %rcx, %rsi ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r11 ; X64-NEXT: mulq %r14
; X64-NEXT: movq %rdx, %r13 ; X64-NEXT: movq %rdx, %r13
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %rdi, %rbx ; X64-NEXT: addq %rcx, %rsi
; X64-NEXT: adcq %rbp, %r13 ; X64-NEXT: adcq %rbp, %r13
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: addq %r9, %rbx ; X64-NEXT: addq %r8, %rsi
; X64-NEXT: adcq %r8, %r13 ; X64-NEXT: adcq %r10, %r13
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: adcq $0, %r15
; X64-NEXT: adcq $0, %r12 ; X64-NEXT: adcq $0, %r12
; X64-NEXT: movq %r10, %rbp ; X64-NEXT: movq %rdi, %rbp
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r9 # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq %rsi, %r8 ; X64-NEXT: movq %rbx, %r14
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rdi ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rdi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq 24(%rax), %r14 ; X64-NEXT: movq 24(%rax), %rcx
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rcx, %rbp
; X64-NEXT: movq %rbp, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rdi, %rax ; X64-NEXT: movq %rax, %r8
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: addq %rbx, %r8
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rdi, %rcx
; X64-NEXT: addq %rsi, %rcx ; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: andl $1, %edi
; X64-NEXT: andl $1, %esi ; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq %r8, %rax ; X64-NEXT: mulq %rbp
; X64-NEXT: mulq %r14
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rdi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r11 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r14 # 8-byte Reload
; X64-NEXT: addq %r11, %rbp ; X64-NEXT: adcq %r14, %rbp
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: addq %rax, %rbx
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq %rdx, %rbp
; X64-NEXT: addq %rax, %rbp ; X64-NEXT: addq %rsi, %r10
; X64-NEXT: adcq %rdx, %rsi ; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %rbx, %r9 ; X64-NEXT: adcq %r13, %r8
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r13, %rdi ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: addq %r15, %rbx
; X64-NEXT: addq %r15, %rbp ; X64-NEXT: adcq %r12, %rbp
; X64-NEXT: adcq %r12, %rsi ; X64-NEXT: movl $0, %r8d
; X64-NEXT: movl $0, %r10d ; X64-NEXT: adcq $0, %r8
; X64-NEXT: adcq $0, %r10 ; X64-NEXT: sbbq %r10, %r10
; X64-NEXT: sbbq %r15, %r15 ; X64-NEXT: andl $1, %r10d
; X64-NEXT: andl $1, %r15d
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq %r9, %rsi
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %r12 ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq %rax, %r13 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %r12, %rdi ; X64-NEXT: addq %r9, %rdi
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: addq %rdi, %r9 ; X64-NEXT: movq %rax, %r13
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: addq %rdi, %r13
; X64-NEXT: addq %rbx, %rcx ; X64-NEXT: adcq %rsi, %r9
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %edi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %r14, %r12 ; X64-NEXT: movq %rcx, %r12
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %r9, %rax
; X64-NEXT: adcq %rdi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: addq %r11, %rcx ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq %r14, %rsi
; X64-NEXT: addq %rax, %rcx ; X64-NEXT: addq %rax, %rcx
; X64-NEXT: adcq %rdx, %rdi ; X64-NEXT: adcq %rdx, %rsi
; X64-NEXT: addq %rbp, %r13 ; X64-NEXT: addq %rbx, %r15
; X64-NEXT: adcq %rsi, %r9 ; X64-NEXT: adcq %rbp, %r13
; X64-NEXT: adcq %r10, %rcx ; X64-NEXT: adcq %r8, %rcx
; X64-NEXT: adcq %r15, %rdi ; X64-NEXT: adcq %r10, %rsi
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: movq %r13, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r13, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rax, %r9
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq 24(%rax), %r8 ; X64-NEXT: movq 24(%rax), %rbp
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbp, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdi, %r13 ; X64-NEXT: movq %rsi, %r14
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r15 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rbx, %r15 ; X64-NEXT: addq %rbx, %r15
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rdi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rbp, %r14 ; X64-NEXT: movq %rdi, %r11
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %r8 ; X64-NEXT: addq %rax, %r8
; X64-NEXT: adcq %rdx, %r10 ; X64-NEXT: adcq %rdx, %r10
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: movq %r14, %rsi
; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rdi
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq %rsi, %r9
; X64-NEXT: mulq %r14
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: addq %rdi, %rsi ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: mulq %r11
; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r14
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: addq %rsi, %rbx ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rdi, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: andl $1, %ebx
; X64-NEXT: addq %r11, %rbx ; X64-NEXT: movq %r14, %rax
; X64-NEXT: adcq %r15, %rcx ; X64-NEXT: mulq %r11
; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %rcx, %rsi
; X64-NEXT: adcq %rbx, %r11
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: addq %r9, %rsi
; X64-NEXT: adcq %r15, %r11
; X64-NEXT: adcq $0, %r8 ; X64-NEXT: adcq $0, %r8
; X64-NEXT: adcq $0, %r10 ; X64-NEXT: adcq $0, %r10
; X64-NEXT: movq %r9, %rsi ; X64-NEXT: movq %rdi, %rbx
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %r9 ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: movq %rbp, %r14
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdi, %r13 ; X64-NEXT: movq %rdi, %r15
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %r9, %rbp ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %rdi
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r12 ; X64-NEXT: movq %r12, %r13
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: mulq %r13
; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rbp, %rax
; X64-NEXT: movq %rax, %r15 ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq %rdi, %rcx
; X64-NEXT: addq %rdi, %rsi
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: andl $1, %edi ; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r12 ; X64-NEXT: mulq %r13
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rdi, %rdx ; X64-NEXT: adcq %rdi, %rdx
; X64-NEXT: movq (%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r14 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r14 # 8-byte Reload
; X64-NEXT: addq %r14, %rsi ; X64-NEXT: addq %r14, %rbx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: addq %rax, %rbx
; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: addq %rsi, %r9
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r11, %rbp
; X64-NEXT: movq %rbp, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rbx
; X64-NEXT: adcq $0, %rcx
; X64-NEXT: addq %r8, %rbx
; X64-NEXT: adcq %r10, %rcx
; X64-NEXT: movl $0, %r12d
; X64-NEXT: adcq $0, %r12
; X64-NEXT: sbbq %r9, %r9
; X64-NEXT: andl $1, %r9d
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: addq %rax, %rsi ; X64-NEXT: mulq %r15
; X64-NEXT: adcq %rdx, %rbp ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: addq %rbx, %r11 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: adcq %rcx, %r15 ; X64-NEXT: movq %r10, %rax
; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %r8, %rdi
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: addq %r8, %rsi
; X64-NEXT: adcq %r10, %rbp
; X64-NEXT: movl $0, %r10d
; X64-NEXT: adcq $0, %r10
; X64-NEXT: sbbq %r15, %r15
; X64-NEXT: andl $1, %r15d
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r13
; X64-NEXT: movq %rdx, %r9 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: addq %rdi, %r15
; X64-NEXT: movq %r8, %rax ; X64-NEXT: adcq %rsi, %r8
; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r10, %rax
; X64-NEXT: movq %r10, %rbp
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r13
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: addq %r8, %rax
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: addq %r9, %rbx ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: movq %r10, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: addq %r14, %rsi
; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: mulq %r12 ; X64-NEXT: movq %r8, %rdi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: addq %rbx, %r9 ; X64-NEXT: adcq %rdx, %rdi
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: addq %rbx, %r11
; X64-NEXT: addq %rdi, %rcx ; X64-NEXT: adcq %rcx, %r15
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: adcq %r12, %rsi
; X64-NEXT: andl $1, %edi ; X64-NEXT: adcq %r9, %rdi
; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %r12
; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rdi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: addq %r14, %rcx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload
; X64-NEXT: movq %r12, %r13
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %rcx
; X64-NEXT: adcq %rdx, %r13
; X64-NEXT: addq %rsi, %r11
; X64-NEXT: adcq %rbp, %r9
; X64-NEXT: adcq %r10, %rcx
; X64-NEXT: adcq %r15, %r13
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r15, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq $0, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: adcq $0, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, %r14
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rcx, %r11 ; X64-NEXT: movq %rcx, %r13
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rdi, %rbp ; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rbp, %r10 ; X64-NEXT: addq %rcx, %r11
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rbx, %rsi
; X64-NEXT: addq %rbx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: andl $1, %ecx
; X64-NEXT: andl $1, %esi ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdi, %r14 ; X64-NEXT: movq %rdi, %r15
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r9 # 8-byte Reload ; X64-NEXT: movq %r10, %r9
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload
; X64-NEXT: movq %r12, %r15 ; X64-NEXT: movq %r8, %r12
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: adcq (%rsp), %r12 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %r9 ; X64-NEXT: addq %rax, %r9
; X64-NEXT: adcq %rdx, %r15 ; X64-NEXT: adcq %rdx, %r12
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %r11, %rdi ; X64-NEXT: movq %r13, %rbp
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %r8 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r11 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r13, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r8, %rbp ; X64-NEXT: addq %r8, %rbx
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rcx, %r8 ; X64-NEXT: movq %rcx, %rbp
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rbx, %rcx ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: andl $1, %esi
; X64-NEXT: andl $1, %ebp ; X64-NEXT: movq %r13, %rax
; X64-NEXT: movq %r11, %rax ; X64-NEXT: mulq %r15
; X64-NEXT: mulq %r14 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq %rbp, %rsi ; X64-NEXT: adcq %rsi, %r8
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq %r14, %rbx
; X64-NEXT: adcq %r10, %rsi ; X64-NEXT: adcq %r11, %r8
; X64-NEXT: adcq $0, %r9 ; X64-NEXT: adcq $0, %r9
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: adcq $0, %r12
; X64-NEXT: movq %r8, %rbp ; X64-NEXT: movq %rbp, %rsi
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r14 ; X64-NEXT: movq %rdx, %r14
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r13, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r12 ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %r14, %rcx ; X64-NEXT: addq %r14, %rcx
; X64-NEXT: adcq $0, %r12 ; X64-NEXT: adcq $0, %rdi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq 56(%rax), %rdi ; X64-NEXT: movq 56(%rax), %r15
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq %rdi, %rsi
; X64-NEXT: addq %r12, %rbp
; X64-NEXT: sbbq %rcx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r13, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r15
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: movq (%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r11 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: addq %r11, %rcx ; X64-NEXT: addq %r13, %rdi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: adcq %r12, %rdi ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: addq %rax, %rcx ; X64-NEXT: addq %rax, %rdi
; X64-NEXT: adcq %rdx, %rdi ; X64-NEXT: adcq %rdx, %rsi
; X64-NEXT: addq %rbx, %r8 ; X64-NEXT: addq %rbx, %r10
; X64-NEXT: movq %r8, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %rsi, %r10
; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r10, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %r8, %r11
; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %rdi
; X64-NEXT: addq %r9, %rcx
; X64-NEXT: adcq %r15, %rdi
; X64-NEXT: movl $0, %r8d
; X64-NEXT: adcq $0, %r8
; X64-NEXT: sbbq %r9, %r9
; X64-NEXT: andl $1, %r9d
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rdx, %r10
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload
; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %r10, %rbp
; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rbp, %r15
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: addq %rbx, %rsi ; X64-NEXT: addq %r9, %rdi
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: adcq %r12, %rsi
; X64-NEXT: andl $1, %ebp ; X64-NEXT: movl $0, %r14d
; X64-NEXT: movq %r14, %rax ; X64-NEXT: adcq $0, %r14
; X64-NEXT: mulq %r10 ; X64-NEXT: sbbq %r10, %r10
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: andl $1, %r10d
; X64-NEXT: adcq %rbp, %rdx ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: addq %r11, %r10 ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload ; X64-NEXT: mulq %rcx
; X64-NEXT: adcq %r12, %r14 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: addq %rax, %r10 ; X64-NEXT: movq %rax, %r12
; X64-NEXT: adcq %rdx, %r14 ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r9 # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq %r9, %rax
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: mulq %rcx
; X64-NEXT: adcq %rdi, %r15 ; X64-NEXT: movq %rdx, %r11
; X64-NEXT: adcq %r8, %r10 ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: adcq %r9, %r14 ; X64-NEXT: addq %r8, %rcx
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq $0, %r11
; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq %r11, %r8
; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %r15
; X64-NEXT: addq %r8, %rax
; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r15 # 8-byte Reload
; X64-NEXT: addq %r13, %r15
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r11 # 8-byte Reload
; X64-NEXT: adcq %rbp, %r11
; X64-NEXT: addq %rax, %r15
; X64-NEXT: adcq %rdx, %r11
; X64-NEXT: addq %rdi, %r12
; X64-NEXT: adcq %rsi, %rbx
; X64-NEXT: adcq %r14, %r15
; X64-NEXT: adcq %r10, %r11
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r12 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: addq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: addq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq %r13, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq $0, %rax ; X64-NEXT: adcq %rcx, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq $0, %r12
; X64-NEXT: adcq $0, %rbx
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: adcq $0, %r15
; X64-NEXT: adcq $0, %r10 ; X64-NEXT: adcq $0, %r11
; X64-NEXT: adcq $0, %r14 ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r12 # 8-byte Folded Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: movq %rbx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r15 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload
; X64-NEXT: movl $0, %eax ; X64-NEXT: movl $0, %eax
; X64-NEXT: adcq $0, %rax ; X64-NEXT: adcq $0, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movl $0, %eax ; X64-NEXT: movl $0, %eax
; X64-NEXT: adcq $0, %rax ; X64-NEXT: adcq $0, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movl $0, %eax ; X64-NEXT: movl $0, %eax
; X64-NEXT: adcq $0, %rax ; X64-NEXT: adcq $0, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: sbbq %rax, %rax ; X64-NEXT: sbbq %rax, %rax
; X64-NEXT: andl $1, %eax ; X64-NEXT: andl $1, %eax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rcx, %r9
; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, %r13 ; X64-NEXT: movq %rax, %r13
; X64-NEXT: addq %rcx, %r13 ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: addq %rbx, %rsi
; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdi, %r11 ; X64-NEXT: movq %rdi, %r14
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload ; X64-NEXT: addq %r8, %rbp
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r12 # 8-byte Folded Reload ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: addq %rax, %r12 ; X64-NEXT: mulq %r8
; X64-NEXT: adcq %rdx, %r8 ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r12
; X64-NEXT: addq %rbp, %r12
; X64-NEXT: adcq %rbx, %rcx
; X64-NEXT: sbbq %rbp, %rbp
; X64-NEXT: andl $1, %ebp
; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rbp, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r9 # 8-byte Reload
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload
; X64-NEXT: movq (%rsp), %r10 # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %r9
; X64-NEXT: adcq %rdx, %r10
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r9 ; X64-NEXT: movq %r14, %rbx
; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, (%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %r9 ; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rdi, %r9 ; X64-NEXT: movq %rdi, %r14
; X64-NEXT: mulq %r11 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: addq %rbp, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rbx, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: andl $1, %edi ; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %r11 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %r11 ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rsi ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq %rdi, %r11 ; X64-NEXT: adcq %rdi, %rbx
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq %r13, %rbp
; X64-NEXT: adcq %r13, %r11 ; X64-NEXT: adcq %r12, %rbx
; X64-NEXT: adcq $0, %r12 ; X64-NEXT: adcq $0, %r9
; X64-NEXT: adcq $0, %r8 ; X64-NEXT: movq %r9, %r12
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: adcq $0, %r10
; X64-NEXT: movq %r9, %rdi ; X64-NEXT: movq %r10, %r8
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, %r13
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq %rbp, %r9
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %r8, %rcx
; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: movq %rax, %r9
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: movq %rsi, %r10
; X64-NEXT: addq %rbp, %rbx
; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: addq %rcx, %rdi
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: mulq %rcx
; X64-NEXT: addq %rax, %rdi ; X64-NEXT: movq %rdx, %r14
; X64-NEXT: addq %rdi, %rax
; X64-NEXT: movq %rax, %rdi
; X64-NEXT: adcq %rsi, %r14
; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %rcx
; X64-NEXT: addq %r14, %rax
; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload
; X64-NEXT: addq %r14, %rsi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: adcq %r13, %rcx
; X64-NEXT: addq %rax, %rsi
; X64-NEXT: adcq %rdx, %rcx ; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: addq %rsi, %r13 ; X64-NEXT: addq %rbp, %r9
; X64-NEXT: adcq %r11, %r8 ; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: adcq %rbx, %rdi
; X64-NEXT: adcq $0, %rdi
; X64-NEXT: adcq $0, %rcx
; X64-NEXT: addq %r12, %rdi
; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: adcq $0, %rcx
; X64-NEXT: movl $0, %r12d ; X64-NEXT: addq %r12, %rsi
; X64-NEXT: adcq $0, %r12 ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r8, %rcx
; X64-NEXT: movq %rcx, %r12
; X64-NEXT: movl $0, %r10d
; X64-NEXT: adcq $0, %r10
; X64-NEXT: sbbq %r9, %r9 ; X64-NEXT: sbbq %r9, %r9
; X64-NEXT: andl $1, %r9d ; X64-NEXT: andl $1, %r9d
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rcx, %rbp ; X64-NEXT: addq %rcx, %rdi
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rbp, %rbx ; X64-NEXT: addq %rdi, %rbx
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rbp
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq %r14, %rsi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq %r13, %rcx
; X64-NEXT: addq %rax, %rsi ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: adcq %rdx, %rcx ; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: adcq %r12, %rsi ; X64-NEXT: adcq %r12, %rbx
; X64-NEXT: adcq %r10, %rsi
; X64-NEXT: adcq %r9, %rcx ; X64-NEXT: adcq %r9, %rcx
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: addq %rax, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: addq %rax, (%rsp) # 8-byte Folded Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: adcq %rax, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq %r15, {{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq %r15, {{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq %r10, %r13 ; X64-NEXT: adcq %r11, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload
; X64-NEXT: movq %r13, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r13, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r14, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: movq %rbx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r9 # 8-byte Reload
; X64-NEXT: movq 64(%rsi), %r14 ; X64-NEXT: movq 64(%r9), %r11
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %r11
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %r11
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq 72(%rsi), %rcx ; X64-NEXT: movq 72(%r9), %rcx
; X64-NEXT: movq %rsi, %r13 ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rcx, %rsi ; X64-NEXT: movq %rcx, %rsi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r8
; X64-NEXT: addq %rbx, %r10 ; X64-NEXT: addq %rbx, %r8
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rbp, %rcx
; X64-NEXT: addq %rdi, %rcx ; X64-NEXT: sbbq %rbp, %rbp
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: andl $1, %ebp
; X64-NEXT: andl $1, %edi ; X64-NEXT: movq %r10, %rax
; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rsi, %r8 ; X64-NEXT: movq %rsi, %r10
; X64-NEXT: movq %r10, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rcx, %rbp ; X64-NEXT: addq %rcx, %rdi
; X64-NEXT: adcq %rdi, %rsi ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: xorl %ecx, %ecx ; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: movq %rdx, %r11 ; X64-NEXT: movq %rdx, %r14
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r15 # 8-byte Reload
; X64-NEXT: addq %rbx, %r15
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r12 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r12 # 8-byte Reload
; X64-NEXT: adcq %r11, %r12 ; X64-NEXT: addq %rbx, %r12
; X64-NEXT: addq %rbp, %r15 ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r15 # 8-byte Reload
; X64-NEXT: adcq %rsi, %r12 ; X64-NEXT: adcq %r14, %r15
; X64-NEXT: addq %rdi, %r12
; X64-NEXT: adcq %rsi, %r15
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %r14, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, %rsi
; X64-NEXT: mulq %r14 ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %r9 ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %r9, %rbp ; X64-NEXT: addq %r11, %rdi
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rcx, %r14 ; X64-NEXT: movq %rcx, %r11
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: mulq %r10
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rdi, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq %rdi, %r9 ; X64-NEXT: movq %rbp, %rdi
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r10
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: adcq %r13, %r14
; X64-NEXT: addq %rax, %rbx ; X64-NEXT: addq %rax, %rbx
; X64-NEXT: adcq %rdx, %r11 ; X64-NEXT: adcq %rdx, %r14
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq %r10, %r11 ; X64-NEXT: adcq %r8, %r14
; X64-NEXT: adcq $0, %r15
; X64-NEXT: adcq $0, %r12 ; X64-NEXT: adcq $0, %r12
; X64-NEXT: movq 80(%r13), %rbp ; X64-NEXT: adcq $0, %r15
; X64-NEXT: movq %r14, %rsi ; X64-NEXT: movq %r9, %rbp
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq 80(%rbp), %r8
; X64-NEXT: mulq %rbp ; X64-NEXT: movq %r11, %rax
; X64-NEXT: movq %rdx, %r8 ; X64-NEXT: movq %r11, %r9
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %r10 ; X64-NEXT: movq %rdx, %r10
; X64-NEXT: movq %rax, %r11
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %r8, %rcx ; X64-NEXT: addq %r10, %rcx
; X64-NEXT: adcq $0, %r10 ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq 88(%r13), %r13 ; X64-NEXT: movq 88(%rbp), %r10
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %rcx, %r8 ; X64-NEXT: addq %rcx, %r9
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq %rsi, %rbp
; X64-NEXT: addq %r10, %rdi
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %r10 ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rdi, %rcx ; X64-NEXT: addq %rbp, %rcx
; X64-NEXT: adcq %rsi, %r10 ; X64-NEXT: adcq %rsi, %rdi
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: xorl %edx, %edx ; X64-NEXT: xorl %edx, %edx
; X64-NEXT: mulq %rdx ; X64-NEXT: mulq %rdx
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: addq %rdi, %rsi ; X64-NEXT: movq %r13, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: adcq %rdx, %rax ; X64-NEXT: adcq %rdx, %rax
; X64-NEXT: movq %rdx, %r13
; X64-NEXT: addq %rcx, %rsi ; X64-NEXT: addq %rcx, %rsi
; X64-NEXT: adcq %r10, %rax ; X64-NEXT: adcq %rdi, %rax
; X64-NEXT: addq %rbx, %r14 ; X64-NEXT: addq %rbx, %r11
; X64-NEXT: movq %r14, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r11, %r8 ; X64-NEXT: adcq %r14, %r9
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: adcq $0, %rax ; X64-NEXT: adcq $0, %rax
; X64-NEXT: addq %r15, %rsi ; X64-NEXT: addq %r12, %rsi
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, %r14
; X64-NEXT: adcq %r12, %rax ; X64-NEXT: adcq %r15, %rax
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: movl $0, %r15d ; X64-NEXT: movl $0, %r12d
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: adcq $0, %r12
; X64-NEXT: sbbq %r12, %r12 ; X64-NEXT: sbbq %r11, %r11
; X64-NEXT: andl $1, %r12d ; X64-NEXT: andl $1, %r11d
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %r8 ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r11 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r8, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %r10
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: addq %rdi, %rsi ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: adcq %r13, %rcx
; X64-NEXT: addq %rax, %rsi ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: adcq %rdx, %rcx ; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload ; X64-NEXT: addq %r14, %r15
; X64-NEXT: movq %r14, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r10, %rbx ; X64-NEXT: adcq %r9, %rdi
; X64-NEXT: movq %rbx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r15, %rsi ; X64-NEXT: adcq %r12, %rsi
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r12, %rcx ; X64-NEXT: adcq %r11, %rcx
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: imulq %rax, %r13 ; X64-NEXT: imulq %rax, %r10
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %r13, %rdx ; X64-NEXT: addq %r10, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: imulq %rdi, %rbp ; X64-NEXT: imulq %r10, %r8
; X64-NEXT: addq %rdx, %rbp ; X64-NEXT: addq %rdx, %r8
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r12 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: imulq %r12, %rsi ; X64-NEXT: imulq %rbx, %rsi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rsi, %rdx ; X64-NEXT: addq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: imulq %rcx, %rax ; X64-NEXT: imulq %rbp, %rax
; X64-NEXT: addq %rdx, %rax ; X64-NEXT: addq %rdx, %rax
; X64-NEXT: addq %r9, %r10 ; X64-NEXT: addq %r9, %r11
; X64-NEXT: adcq %rbp, %rax ; X64-NEXT: adcq %r8, %rax
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r14
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: movq %rcx, %rbp ; X64-NEXT: mulq %rdi
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %r12, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: movq %rbx, %r8
; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rcx, %rbx ; X64-NEXT: addq %rcx, %rbx
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: movq %r10, %rbp
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: addq %rbx, %r14 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: addq %rbx, %r15
; X64-NEXT: addq %rsi, %rbp ; X64-NEXT: adcq %rsi, %rdi
; X64-NEXT: sbbq %rcx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r12, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %r12 ; X64-NEXT: movq %rdx, %r12
; X64-NEXT: movq %rax, %r8 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %rbp, %r8 ; X64-NEXT: addq %rdi, %r9
; X64-NEXT: adcq %rcx, %r12 ; X64-NEXT: adcq %rcx, %r12
; X64-NEXT: addq %r10, %r8 ; X64-NEXT: addq %r11, %r9
; X64-NEXT: adcq %r9, %r12 ; X64-NEXT: adcq %r14, %r12
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdx # 8-byte Reload
; X64-NEXT: movq 120(%rdx), %rcx ; X64-NEXT: movq 120(%rdx), %rcx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r9 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: imulq %r9, %rcx ; X64-NEXT: imulq %r10, %rcx
; X64-NEXT: movq 112(%rdx), %rsi ; X64-NEXT: movq 112(%rdx), %rsi
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rax, %r15 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rcx, %rdx ; X64-NEXT: addq %rcx, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: imulq %r10, %rsi ; X64-NEXT: imulq %r8, %rsi
; X64-NEXT: addq %rdx, %rsi ; X64-NEXT: addq %rdx, %rsi
; X64-NEXT: movq 96(%rdi), %rbp ; X64-NEXT: movq 96(%rbp), %rdi
; X64-NEXT: movq 104(%rdi), %rbx ; X64-NEXT: movq 104(%rbp), %rbx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: imulq %rbx, %rcx ; X64-NEXT: imulq %rbx, %rcx
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rax, %r13 ; X64-NEXT: movq %rax, %r13
; X64-NEXT: addq %rcx, %rdx ; X64-NEXT: addq %rcx, %rdx
; X64-NEXT: imulq %rbp, %r11 ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: addq %rdx, %r11 ; X64-NEXT: imulq %rdi, %rax
; X64-NEXT: addq %r15, %r13 ; X64-NEXT: addq %rdx, %rax
; X64-NEXT: adcq %rsi, %r11 ; X64-NEXT: addq %r11, %r13
; X64-NEXT: movq %r11, %r15 ; X64-NEXT: adcq %rsi, %rax
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: adcq $0, %rdi
; X64-NEXT: movq %rbp, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %r14
; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq $0, %rsi
; X64-NEXT: addq %rdi, %rsi
; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %r10
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %rcx, %rsi
; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rsi, %rdi
; X64-NEXT: adcq %rbp, %rcx
; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: addq %r13, %rax ; X64-NEXT: addq %r13, %rax
; X64-NEXT: adcq %r15, %rdx ; X64-NEXT: adcq %r11, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload
; X64-NEXT: adcq %r14, %rbp ; X64-NEXT: adcq %r15, %rdi
; X64-NEXT: adcq %r8, %rax ; X64-NEXT: adcq %r9, %rax
; X64-NEXT: adcq %r12, %rdx ; X64-NEXT: adcq %r12, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r14, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: movq %rbp, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq 80(%rsi), %r9 ; X64-NEXT: movq 80(%rsi), %rcx
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq 88(%rsi), %r8 ; X64-NEXT: movq 88(%rsi), %r10
; X64-NEXT: movq %rsi, %r11 ; X64-NEXT: movq %rsi, %r12
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r10, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdi, %rbx ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: addq %r9, %rbx
; X64-NEXT: addq %rcx, %rdi ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %rcx, %r9
; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r15 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r15 # 8-byte Reload
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: addq %rdi, %r14 ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: addq %rsi, %rbp
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: andl $1, %edi ; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rbp, %rcx ; X64-NEXT: addq %rsi, %rbx
; X64-NEXT: adcq %rdi, %rsi ; X64-NEXT: adcq %rdi, %rbp
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %r9, %rax
; X64-NEXT: xorl %r13d, %r13d ; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: mulq %r13 ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %r10, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r9 # 8-byte Reload
; X64-NEXT: addq %r9, %r10
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: adcq %r13, %r11
; X64-NEXT: addq %rbx, %r10
; X64-NEXT: adcq %rbp, %r11
; X64-NEXT: movq %r12, %rcx
; X64-NEXT: movq 64(%rcx), %rdi
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq 72(%rcx), %r14
; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rsi, %rbx
; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: movq (%rsp), %r10 # 8-byte Reload ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: addq %r10, %r12 ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rsi, %rbp
; X64-NEXT: adcq %rcx, %rbx
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r8 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r9 # 8-byte Reload ; X64-NEXT: movq %rax, %r15
; X64-NEXT: adcq %r9, %r8 ; X64-NEXT: addq %r15, %r9
; X64-NEXT: addq %rcx, %r12 ; X64-NEXT: movq %r13, %rax
; X64-NEXT: adcq %rsi, %r8 ; X64-NEXT: adcq %r8, %rax
; X64-NEXT: movq %r11, %rsi ; X64-NEXT: addq %rbp, %r9
; X64-NEXT: movq 64(%rsi), %r11 ; X64-NEXT: adcq %rbx, %rax
; X64-NEXT: movq %r11, %rax ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r9 # 8-byte Folded Reload
; X64-NEXT: movq %rbx, %rdi ; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rdi ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, %r13
; X64-NEXT: adcq $0, %r10
; X64-NEXT: adcq $0, %r11
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq 72(%rsi), %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: movq %r14, %r12
; X64-NEXT: movq %r12, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rsi, %r14
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rbp ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbp, %rax
; X64-NEXT: movq %rax, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq %rdi, %rsi
; X64-NEXT: movq %r11, %rdi
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r13 ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq %rax, %r13
; X64-NEXT: addq %r13, %r10
; X64-NEXT: adcq %r11, %r9
; X64-NEXT: addq %rbp, %r10
; X64-NEXT: adcq %rsi, %r9
; X64-NEXT: addq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload
; X64-NEXT: movq %r10, (%rsp) # 8-byte Spill
; X64-NEXT: adcq %r14, %r9
; X64-NEXT: movq %r9, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %r12
; X64-NEXT: adcq $0, %r8
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r9 ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq %rax, %r15
; X64-NEXT: movq %rbx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %r9, %rsi
; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rsi, %rdi ; X64-NEXT: addq %rbp, %rdi
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %r9
; X64-NEXT: addq %rbp, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %r12, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %rcx
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: movq %rcx, %rbp
; X64-NEXT: addq %r9, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r13 # 8-byte Folded Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload ; X64-NEXT: addq %r12, %r15
; X64-NEXT: adcq %r14, %r11 ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r8 # 8-byte Folded Reload
; X64-NEXT: addq %rax, %r13 ; X64-NEXT: addq %rax, %r15
; X64-NEXT: adcq %rdx, %r11 ; X64-NEXT: adcq %rdx, %r8
; X64-NEXT: addq (%rsp), %r15 # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: movq %r15, (%rsp) # 8-byte Spill ; X64-NEXT: movq %rbx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq %r13, %rdi
; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %r13 ; X64-NEXT: adcq $0, %r15
; X64-NEXT: adcq $0, %r11
; X64-NEXT: addq %r12, %r13
; X64-NEXT: adcq %r8, %r11
; X64-NEXT: movl $0, %r8d
; X64-NEXT: adcq $0, %r8 ; X64-NEXT: adcq $0, %r8
; X64-NEXT: sbbq %r9, %r9 ; X64-NEXT: addq %r10, %r15
; X64-NEXT: andl $1, %r9d ; X64-NEXT: adcq %r11, %r8
; X64-NEXT: movl $0, %r9d
; X64-NEXT: adcq $0, %r9
; X64-NEXT: sbbq %r13, %r13
; X64-NEXT: andl $1, %r13d
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r15 # 8-byte Reload ; X64-NEXT: movq %r14, %rsi
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %r12 ; X64-NEXT: movq %rdx, %r14
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rsi, %r11
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r12, %rbp ; X64-NEXT: addq %r14, %rbx
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: movq %rdi, %r14
; X64-NEXT: mulq %rbp
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq %r12, %rsi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: adcq %r14, %rcx ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: addq %rax, %rsi ; X64-NEXT: addq %rax, %rsi
; X64-NEXT: adcq %rdx, %rcx ; X64-NEXT: adcq %rdx, %rcx
; X64-NEXT: addq %r13, %rdi ; X64-NEXT: addq %r15, %r10
; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r10, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r11, %rbp ; X64-NEXT: adcq %r8, %rbx
; X64-NEXT: movq %rbp, {{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbx, {{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r8, %rsi ; X64-NEXT: adcq %r9, %rsi
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r9, %rcx ; X64-NEXT: adcq %r13, %rcx
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq 96(%rbp), %rcx ; X64-NEXT: movq 96(%rsi), %rcx
; X64-NEXT: imulq %rcx, %r10 ; X64-NEXT: imulq %rcx, %rbp
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r15 ; X64-NEXT: movq %r11, %rdi
; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %r10, %rdx ; X64-NEXT: addq %rbp, %rdx
; X64-NEXT: movq 104(%rbp), %r8 ; X64-NEXT: movq 104(%rsi), %r8
; X64-NEXT: imulq %r8, %r15 ; X64-NEXT: imulq %r8, %rdi
; X64-NEXT: addq %rdx, %r15 ; X64-NEXT: addq %rdx, %rdi
; X64-NEXT: movq 112(%rbp), %rax ; X64-NEXT: movq %rdi, %r10
; X64-NEXT: movq %rbp, %rdi ; X64-NEXT: movq 112(%rsi), %rax
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rsi, %rbp
; X64-NEXT: movq %rax, %rdi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: imulq %rbx, %rsi ; X64-NEXT: imulq %rbx, %rdi
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rax, %r13 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rsi, %rdx ; X64-NEXT: addq %rdi, %rdx
; X64-NEXT: movq 120(%rdi), %rdi ; X64-NEXT: movq 120(%rbp), %rdi
; X64-NEXT: imulq %rbp, %rdi ; X64-NEXT: imulq %rsi, %rdi
; X64-NEXT: addq %rdx, %rdi ; X64-NEXT: addq %rdx, %rdi
; X64-NEXT: addq %r9, %r13 ; X64-NEXT: addq %r9, %r11
; X64-NEXT: adcq %r15, %rdi ; X64-NEXT: adcq %r10, %rdi
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq %rbp, %r9 ; X64-NEXT: movq %rsi, %r10
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: movq %rbx, %r9
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rbp, %rsi ; X64-NEXT: addq %rsi, %rbx
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq $0, %rcx
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rsi, %r12 ; X64-NEXT: addq %rbx, %r15
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq %rcx, %rsi
; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: sbbq %rcx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %r8 ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r8
; X64-NEXT: addq %rbp, %r9 ; X64-NEXT: addq %rsi, %r8
; X64-NEXT: adcq %rcx, %r8 ; X64-NEXT: adcq %rcx, %rbx
; X64-NEXT: addq %r13, %r9 ; X64-NEXT: addq %r11, %r8
; X64-NEXT: adcq %rdi, %r8 ; X64-NEXT: adcq %rdi, %rbx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: imulq %rbx, %rsi ; X64-NEXT: imulq %rax, %rsi
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rax, %r9
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rsi, %rdx ; X64-NEXT: addq %rsi, %rdx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r11 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %r12 # 8-byte Reload
; X64-NEXT: imulq %r11, %rcx ; X64-NEXT: imulq %r12, %rcx
; X64-NEXT: addq %rdx, %rcx ; X64-NEXT: addq %rdx, %rcx
; X64-NEXT: movq %rcx, %rsi ; X64-NEXT: movq %rcx, %rbp
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %r14 # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: imulq %r14, %rcx ; X64-NEXT: imulq %rsi, %rcx
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rax, %r13 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: addq %rcx, %rdx ; X64-NEXT: addq %rcx, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq %r14, %r13
; X64-NEXT: imulq %rbp, %rax ; X64-NEXT: imulq %rdi, %r13
; X64-NEXT: addq %rdx, %rax ; X64-NEXT: addq %rdx, %r13
; X64-NEXT: addq %r10, %r13 ; X64-NEXT: addq %r11, %r10
; X64-NEXT: adcq %rsi, %rax ; X64-NEXT: adcq %rbp, %r13
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rdi, %r11
; X64-NEXT: movq %rbp, %r10 ; X64-NEXT: mulq %r9
; X64-NEXT: mulq %rbx ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %rsi, %r14
; X64-NEXT: mulq %rbx ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: addq %rsi, %rcx ; X64-NEXT: addq %rbp, %rcx
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq $0, %r9
; X64-NEXT: movq %r10, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %r11 ; X64-NEXT: movq %r12, %rbp
; X64-NEXT: movq %rdx, %r15 ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: addq %rcx, %r10 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: addq %rcx, %r11
; X64-NEXT: addq %rdi, %r15 ; X64-NEXT: adcq %r9, %rsi
; X64-NEXT: sbbq %rcx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r11 ; X64-NEXT: mulq %rbp
; X64-NEXT: addq %r15, %rax ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: addq %r13, %rax ; X64-NEXT: addq %r10, %rax
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq %r13, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: adcq %r12, %r10 ; X64-NEXT: adcq %r15, %r11
; X64-NEXT: adcq %r9, %rax ; X64-NEXT: adcq %r8, %rax
; X64-NEXT: adcq %r8, %rdx ; X64-NEXT: adcq %rbx, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: addq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq (%rsp), %rbx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbx # 8-byte Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: addq {{[0-9]+}}(%rsp), %rcx # 8-byte Folded Reload ; X64-NEXT: addq (%rsp), %rcx # 8-byte Folded Reload
; X64-NEXT: movq %rcx, %r8 ; X64-NEXT: movq %rcx, %r8
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: movq %rdi, %r9 ; X64-NEXT: movq %rsi, %r9
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rbx # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rbp # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %r11 # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq {{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, (%rcx) ; X64-NEXT: movq %rsi, (%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 8(%rcx) ; X64-NEXT: movq %rsi, 8(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 16(%rcx) ; X64-NEXT: movq %rsi, 16(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 24(%rcx) ; X64-NEXT: movq %rsi, 24(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 32(%rcx) ; X64-NEXT: movq %rsi, 32(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 40(%rcx) ; X64-NEXT: movq %rsi, 40(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 48(%rcx) ; X64-NEXT: movq %rsi, 48(%rcx)
; X64-NEXT: movq {{[0-9]+}}(%rsp), %rdi # 8-byte Reload ; X64-NEXT: movq {{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq %rdi, 56(%rcx) ; X64-NEXT: movq %rsi, 56(%rcx)
; X64-NEXT: movq %r8, 64(%rcx) ; X64-NEXT: movq %r8, 64(%rcx)
; X64-NEXT: movq %r9, 72(%rcx) ; X64-NEXT: movq %r9, 72(%rcx)
; X64-NEXT: movq %rbx, 80(%rcx) ; X64-NEXT: movq %rbx, 80(%rcx)
; X64-NEXT: movq %rsi, 88(%rcx) ; X64-NEXT: movq %rbp, 88(%rcx)
; X64-NEXT: movq %rbp, 96(%rcx) ; X64-NEXT: movq %rdi, 96(%rcx)
; X64-NEXT: movq %r10, 104(%rcx) ; X64-NEXT: movq %r11, 104(%rcx)
; X64-NEXT: movq %rax, 112(%rcx) ; X64-NEXT: movq %rax, 112(%rcx)
; X64-NEXT: movq %rdx, 120(%rcx) ; X64-NEXT: movq %rdx, 120(%rcx)
; X64-NEXT: addq $360, %rsp # imm = 0x168 ; X64-NEXT: addq $352, %rsp # imm = 0x160
; X64-NEXT: popq %rbx ; X64-NEXT: popq %rbx
; X64-NEXT: popq %r12 ; X64-NEXT: popq %r12
; X64-NEXT: popq %r13 ; X64-NEXT: popq %r13
; X64-NEXT: popq %r14 ; X64-NEXT: popq %r14
; X64-NEXT: popq %r15 ; X64-NEXT: popq %r15
; X64-NEXT: popq %rbp ; X64-NEXT: popq %rbp
; X64-NEXT: retq ; X64-NEXT: retq
%av = load i1024, i1024* %a %av = load i1024, i1024* %a
%bv = load i1024, i1024* %b %bv = load i1024, i1024* %b
%r = mul i1024 %av, %bv %r = mul i1024 %av, %bv
store i1024 %r, i1024* %out store i1024 %r, i1024* %out
ret void ret void
} }

test/CodeGen/X86/mul-i256.ll

Show First 20 Lines • Show All 213 Lines • ▼ Show 20 Lines
; X64-NEXT: .Lcfi4: ; X64-NEXT: .Lcfi4:
; X64-NEXT: .cfi_offset %rbx, -40 ; X64-NEXT: .cfi_offset %rbx, -40
; X64-NEXT: .Lcfi5: ; X64-NEXT: .Lcfi5:
; X64-NEXT: .cfi_offset %r12, -32 ; X64-NEXT: .cfi_offset %r12, -32
; X64-NEXT: .Lcfi6: ; X64-NEXT: .Lcfi6:
; X64-NEXT: .cfi_offset %r14, -24 ; X64-NEXT: .cfi_offset %r14, -24
; X64-NEXT: .Lcfi7: ; X64-NEXT: .Lcfi7:
; X64-NEXT: .cfi_offset %r15, -16 ; X64-NEXT: .cfi_offset %r15, -16
; X64-NEXT: movq %rdx, %r10 ; X64-NEXT: movq %rdx, %r9
; X64-NEXT: movq (%rdi), %r14 ; X64-NEXT: movq (%rdi), %r14
; X64-NEXT: movq 8(%rdi), %r8 ; X64-NEXT: movq 8(%rdi), %r8
; X64-NEXT: movq 16(%rdi), %rcx ; X64-NEXT: movq 16(%rdi), %rcx
; X64-NEXT: movq 16(%rsi), %rbx ; X64-NEXT: movq 16(%rsi), %rbx
; X64-NEXT: movq (%rsi), %r12 ; X64-NEXT: movq (%rsi), %r12
; X64-NEXT: movq 8(%rsi), %r15 ; X64-NEXT: movq 8(%rsi), %r15
; X64-NEXT: movq 24(%rdi), %rdi ; X64-NEXT: movq 24(%rdi), %rdi
; X64-NEXT: imulq %r12, %rdi ; X64-NEXT: imulq %r12, %rdi
; X64-NEXT: movq %r12, %rax ; X64-NEXT: movq %r12, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: addq %rdi, %rdx ; X64-NEXT: addq %rdi, %rdx
; X64-NEXT: imulq %r15, %rcx ; X64-NEXT: imulq %r15, %rcx
; X64-NEXT: addq %rdx, %rcx ; X64-NEXT: addq %rdx, %rcx
; X64-NEXT: movq %rbx, %rdi ; X64-NEXT: movq %rbx, %rdi
; X64-NEXT: imulq %r8, %rdi ; X64-NEXT: imulq %r8, %rdi
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %r14 ; X64-NEXT: mulq %r14
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %rdi, %rdx ; X64-NEXT: addq %rdi, %rdx
; X64-NEXT: movq 24(%rsi), %rbx ; X64-NEXT: movq 24(%rsi), %rbx
; X64-NEXT: imulq %r14, %rbx ; X64-NEXT: imulq %r14, %rbx
; X64-NEXT: addq %rdx, %rbx ; X64-NEXT: addq %rdx, %rbx
; X64-NEXT: addq %r9, %r11 ; X64-NEXT: addq %r10, %r11
; X64-NEXT: adcq %rcx, %rbx ; X64-NEXT: adcq %rcx, %rbx
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r12 ; X64-NEXT: mulq %r12
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %r12 ; X64-NEXT: mulq %r12
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rsi, %rdi ; X64-NEXT: addq %rsi, %rdi
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq $0, %rcx
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r15
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %r14 ; X64-NEXT: movq %rax, %r14
; X64-NEXT: addq %rdi, %r14 ; X64-NEXT: addq %rdi, %r14
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq %rcx, %rsi
; X64-NEXT: addq %rcx, %rsi
; X64-NEXT: sbbq %rcx, %rcx ; X64-NEXT: sbbq %rcx, %rcx
; X64-NEXT: andl $1, %ecx ; X64-NEXT: andl $1, %ecx
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: mulq %r15 ; X64-NEXT: mulq %r15
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: adcq %rcx, %rdx ; X64-NEXT: adcq %rcx, %rdx
; X64-NEXT: addq %r11, %rax ; X64-NEXT: addq %r11, %rax
; X64-NEXT: adcq %rbx, %rdx ; X64-NEXT: adcq %rbx, %rdx
; X64-NEXT: movq %r9, (%r10) ; X64-NEXT: movq %r10, (%r9)
; X64-NEXT: movq %r14, 8(%r10) ; X64-NEXT: movq %r14, 8(%r9)
; X64-NEXT: movq %rax, 16(%r10) ; X64-NEXT: movq %rax, 16(%r9)
; X64-NEXT: movq %rdx, 24(%r10) ; X64-NEXT: movq %rdx, 24(%r9)
; X64-NEXT: popq %rbx ; X64-NEXT: popq %rbx
; X64-NEXT: popq %r12 ; X64-NEXT: popq %r12
; X64-NEXT: popq %r14 ; X64-NEXT: popq %r14
; X64-NEXT: popq %r15 ; X64-NEXT: popq %r15
; X64-NEXT: retq ; X64-NEXT: retq
entry: entry:
%av = load i256, i256* %a %av = load i256, i256* %a
%bv = load i256, i256* %b %bv = load i256, i256* %b
%r = mul i256 %av, %bv %r = mul i256 %av, %bv
store i256 %r, i256* %out store i256 %r, i256* %out
ret void ret void
} }
attributes #0 = { norecurse nounwind uwtable "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" } attributes #0 = { norecurse nounwind uwtable "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" }

test/CodeGen/X86/mul-i512.ll

Show First 20 Lines • Show All 905 Lines • ▼ Show 20 Lines
; X64-NEXT: pushq %rbp ; X64-NEXT: pushq %rbp
; X64-NEXT: pushq %r15 ; X64-NEXT: pushq %r15
; X64-NEXT: pushq %r14 ; X64-NEXT: pushq %r14
; X64-NEXT: pushq %r13 ; X64-NEXT: pushq %r13
; X64-NEXT: pushq %r12 ; X64-NEXT: pushq %r12
; X64-NEXT: pushq %rbx ; X64-NEXT: pushq %rbx
; X64-NEXT: pushq %rax ; X64-NEXT: pushq %rax
; X64-NEXT: movq %rdx, (%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, (%rsp) # 8-byte Spill
; X64-NEXT: movq 24(%rdi), %rbp ; X64-NEXT: movq 24(%rdi), %r11
; X64-NEXT: movq 16(%rdi), %r11 ; X64-NEXT: movq 16(%rdi), %r14
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq (%rsi), %rdx ; X64-NEXT: movq (%rsi), %rdx
; X64-NEXT: movq 8(%rsi), %r8 ; X64-NEXT: movq 8(%rsi), %rbp
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %r10 ; X64-NEXT: movq %rdx, %r8
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: movq %rbp, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rsi, %r9 ; X64-NEXT: movq %rsi, %r10
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %r10, %rsi ; X64-NEXT: addq %r8, %rsi
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq %r11, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r14, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %rsi, %r12 ; X64-NEXT: addq %rsi, %r9
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rbx, %rcx
; X64-NEXT: addq %rbx, %rcx
; X64-NEXT: sbbq %rbx, %rbx ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: andl $1, %ebx ; X64-NEXT: andl $1, %ebx
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rbp, %r8
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rbp ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq %rbx, %rsi ; X64-NEXT: adcq %rbx, %rsi
; X64-NEXT: xorl %ecx, %ecx ; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: movq %r9, %rbx ; X64-NEXT: movq %r10, %rbx
; X64-NEXT: movq %rbx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rbx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rbx, %rax ; X64-NEXT: movq %rbx, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r13 ; X64-NEXT: movq %rdx, %r13
; X64-NEXT: movq %rax, %r10 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: movq %r11, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, %r9 ; X64-NEXT: movq %rdx, %r12
; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r12, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %r15 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r15, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: addq %r10, %r15 ; X64-NEXT: addq %r10, %r15
; X64-NEXT: adcq %r13, %r9 ; X64-NEXT: adcq %r13, %r12
; X64-NEXT: addq %rbp, %r15 ; X64-NEXT: addq %rbp, %r15
; X64-NEXT: adcq %rsi, %r9 ; X64-NEXT: adcq %rsi, %r12
; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq (%rdi), %r14 ; X64-NEXT: movq (%rdi), %r14
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: mulq %rbx ; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rdx, %r11 ; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq 8(%rdi), %rcx ; X64-NEXT: movq 8(%rdi), %rcx
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %rbx ; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rdx, %rbp ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %r11, %rsi ; X64-NEXT: addq %r11, %rsi
; X64-NEXT: adcq $0, %rbp ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq %r14, %rax ; X64-NEXT: movq %r14, %rax
; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r8, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: addq %rsi, %rax ; X64-NEXT: addq %rsi, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq %rbp, %rbx
; X64-NEXT: addq %rbp, %rbx
; X64-NEXT: sbbq %rdi, %rdi ; X64-NEXT: sbbq %rdi, %rdi
; X64-NEXT: andl $1, %edi ; X64-NEXT: andl $1, %edi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rbx, %rbp ; X64-NEXT: addq %rbx, %rbp
; X64-NEXT: adcq %rdi, %rsi ; X64-NEXT: adcq %rdi, %rsi
; X64-NEXT: movq %r14, %rcx ; X64-NEXT: movq %r14, %rcx
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: xorl %edx, %edx ; X64-NEXT: xorl %edx, %edx
; X64-NEXT: mulq %rdx ; X64-NEXT: mulq %rdx
; X64-NEXT: movq %rdx, %r14 ; X64-NEXT: movq %rdx, %r14
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r11
; X64-NEXT: addq %r11, %r10 ; X64-NEXT: addq %r11, %r10
; X64-NEXT: adcq %r14, %r13 ; X64-NEXT: adcq %r14, %r13
; X64-NEXT: addq %rbp, %r10 ; X64-NEXT: addq %rbp, %r10
; X64-NEXT: adcq %rsi, %r13 ; X64-NEXT: adcq %rsi, %r13
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %r10 # 8-byte Folded Reload
; X64-NEXT: adcq %r12, %r13 ; X64-NEXT: adcq %r9, %r13
; X64-NEXT: adcq $0, %r15 ; X64-NEXT: adcq $0, %r15
; X64-NEXT: adcq $0, %r9 ; X64-NEXT: adcq $0, %r12
; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: movq 16(%rsi), %r8 ; X64-NEXT: movq 16(%rsi), %r8
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rcx, %r9 ; X64-NEXT: movq %rcx, %r9
; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %r9, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %rdi, %rbp ; X64-NEXT: addq %rdi, %rbx
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rbp
; X64-NEXT: movq 24(%rsi), %rdi ; X64-NEXT: movq 24(%rsi), %rdi
; X64-NEXT: movq %r9, %rax ; X64-NEXT: movq %r9, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: addq %rbp, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq %rbp, %rsi
; X64-NEXT: addq %rbx, %rsi
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: sbbq %rbp, %rbp
; X64-NEXT: andl $1, %ebp ; X64-NEXT: andl $1, %ebp
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %rsi, %r9 ; X64-NEXT: addq %rsi, %r9
; X64-NEXT: adcq %rbp, %rbx ; X64-NEXT: adcq %rbp, %rbx
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r8, %rax
; X64-NEXT: xorl %ecx, %ecx ; X64-NEXT: xorl %ecx, %ecx
; X64-NEXT: mulq %rcx ; X64-NEXT: mulq %rcx
; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rdx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rbp, %r11 ; X64-NEXT: addq %rbp, %r11
; X64-NEXT: adcq %rdx, %r14 ; X64-NEXT: adcq %rdx, %r14
; X64-NEXT: addq %r9, %r11 ; X64-NEXT: addq %r9, %r11
; X64-NEXT: adcq %rbx, %r14 ; X64-NEXT: adcq %rbx, %r14
; X64-NEXT: addq %r10, %r12 ; X64-NEXT: addq %r10, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: movq %r12, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r13, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill ; X64-NEXT: adcq %r13, -{{[0-9]+}}(%rsp) # 8-byte Folded Spill
; X64-NEXT: adcq $0, %r11 ; X64-NEXT: adcq $0, %r11
; X64-NEXT: adcq $0, %r14 ; X64-NEXT: adcq $0, %r14
; X64-NEXT: addq %r15, %r11 ; X64-NEXT: addq %r15, %r11
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %r14 # 8-byte Folded Reload ; X64-NEXT: adcq %r12, %r14
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq $0, %rcx
; X64-NEXT: movq %rcx, %r13 ; X64-NEXT: movq %rcx, %r13
; X64-NEXT: sbbq %r9, %r9 ; X64-NEXT: sbbq %r9, %r9
; X64-NEXT: andl $1, %r9d ; X64-NEXT: andl $1, %r9d
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %r15 ; X64-NEXT: movq %rdx, %r15
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, %r12
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: movq %r10, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: addq %r15, %rbx ; X64-NEXT: addq %r15, %rbx
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: addq %rbx, %rax ; X64-NEXT: addq %rbx, %rax
; X64-NEXT: movq %rax, %rbx ; X64-NEXT: movq %rax, %rbx
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rsi, %rcx
; X64-NEXT: addq %rsi, %rcx
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %r10, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rsi, %rdx ; X64-NEXT: adcq %rsi, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rsi # 8-byte Reload
; X64-NEXT: addq %rbp, %rsi ; X64-NEXT: addq %rbp, %rsi
Show All 9 Lines
; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rsi, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: adcq %r9, %rcx ; X64-NEXT: adcq %r9, %rcx
; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rcx, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq 32(%rcx), %rsi ; X64-NEXT: movq 32(%rcx), %rsi
; X64-NEXT: imulq %rsi, %rdi ; X64-NEXT: imulq %rsi, %rdi
; X64-NEXT: movq %rsi, %rax ; X64-NEXT: movq %rsi, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r10
; X64-NEXT: addq %rdi, %rdx ; X64-NEXT: addq %rdi, %rdx
; X64-NEXT: movq 40(%rcx), %r9 ; X64-NEXT: movq 40(%rcx), %r9
; X64-NEXT: imulq %r9, %r8 ; X64-NEXT: imulq %r9, %r8
; X64-NEXT: addq %rdx, %r8 ; X64-NEXT: addq %rdx, %r8
; X64-NEXT: movq 48(%rcx), %rax ; X64-NEXT: movq 48(%rcx), %rax
; X64-NEXT: movq %rcx, %rbx ; X64-NEXT: movq %rcx, %rbx
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r11 # 8-byte Reload
; X64-NEXT: imulq %r11, %rdi
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rcx # 8-byte Reload
; X64-NEXT: imulq %rcx, %rdi ; X64-NEXT: mulq %rcx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: mulq %rbp
; X64-NEXT: movq %rax, %r12 ; X64-NEXT: movq %rax, %r12
; X64-NEXT: addq %rdi, %rdx ; X64-NEXT: addq %rdi, %rdx
; X64-NEXT: movq 56(%rbx), %rbx ; X64-NEXT: movq 56(%rbx), %rbx
; X64-NEXT: imulq %rbp, %rbx ; X64-NEXT: imulq %rcx, %rbx
; X64-NEXT: addq %rdx, %rbx ; X64-NEXT: addq %rdx, %rbx
; X64-NEXT: addq %r11, %r12 ; X64-NEXT: addq %r10, %r12
; X64-NEXT: adcq %r8, %rbx ; X64-NEXT: adcq %r8, %rbx
; X64-NEXT: movq %rbp, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: movq %rbp, %r8
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill ; X64-NEXT: movq %rax, -{{[0-9]+}}(%rsp) # 8-byte Spill
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %rsi ; X64-NEXT: mulq %rsi
; X64-NEXT: movq %rdx, %rsi ; X64-NEXT: movq %rdx, %rsi
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rdi, %rbp ; X64-NEXT: addq %rdi, %rbp
; X64-NEXT: adcq $0, %rsi ; X64-NEXT: adcq $0, %rsi
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %rcx, %rax
; X64-NEXT: mulq %r9 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %rdi ; X64-NEXT: movq %rdx, %rdi
; X64-NEXT: movq %rax, %r11 ; X64-NEXT: movq %rax, %r15
; X64-NEXT: addq %rbp, %r11 ; X64-NEXT: addq %rbp, %r15
; X64-NEXT: adcq $0, %rdi ; X64-NEXT: adcq %rsi, %rdi
; X64-NEXT: addq %rsi, %rdi
; X64-NEXT: sbbq %rsi, %rsi ; X64-NEXT: sbbq %rsi, %rsi
; X64-NEXT: andl $1, %esi ; X64-NEXT: andl $1, %esi
; X64-NEXT: movq %rcx, %rax ; X64-NEXT: movq %r11, %rax
; X64-NEXT: mulq %r9 ; X64-NEXT: mulq %r9
; X64-NEXT: movq %rdx, %r14 ; X64-NEXT: movq %rdx, %r11
; X64-NEXT: movq %rax, %r15 ; X64-NEXT: movq %rax, %r14
; X64-NEXT: addq %rdi, %r15 ; X64-NEXT: addq %rdi, %r14
; X64-NEXT: adcq %rsi, %r14 ; X64-NEXT: adcq %rsi, %r11
; X64-NEXT: addq %r12, %r15 ; X64-NEXT: addq %r12, %r14
; X64-NEXT: adcq %rbx, %r14 ; X64-NEXT: adcq %rbx, %r11
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdx # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rdx # 8-byte Reload
; X64-NEXT: movq 56(%rdx), %rcx ; X64-NEXT: movq 56(%rdx), %rcx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload
; X64-NEXT: imulq %r8, %rcx ; X64-NEXT: imulq %r10, %rcx
; X64-NEXT: movq 48(%rdx), %rbp ; X64-NEXT: movq 48(%rdx), %rbx
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbp
; X64-NEXT: movq %r8, %rax ; X64-NEXT: movq %r10, %rax
; X64-NEXT: mulq %rbp ; X64-NEXT: mulq %rbx
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: addq %rcx, %rdx ; X64-NEXT: addq %rcx, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r10 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r8 # 8-byte Reload
; X64-NEXT: imulq %r10, %rbp ; X64-NEXT: imulq %r8, %rbx
; X64-NEXT: addq %rdx, %rbp ; X64-NEXT: addq %rdx, %rbx
; X64-NEXT: movq 32(%rbx), %rdi ; X64-NEXT: movq 32(%rbp), %rdi
; X64-NEXT: movq 40(%rbx), %r12 ; X64-NEXT: movq 40(%rbp), %r12
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rax # 8-byte Reload
; X64-NEXT: movq %rax, %rcx ; X64-NEXT: movq %rax, %rcx
; X64-NEXT: imulq %r12, %rcx ; X64-NEXT: imulq %r12, %rcx
; X64-NEXT: mulq %rdi ; X64-NEXT: mulq %rdi
; X64-NEXT: movq %rax, %r9 ; X64-NEXT: movq %rax, %r9
; X64-NEXT: addq %rcx, %rdx ; X64-NEXT: addq %rcx, %rdx
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r13 # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %r13 # 8-byte Reload
; X64-NEXT: imulq %rdi, %r13 ; X64-NEXT: imulq %rdi, %r13
; X64-NEXT: addq %rdx, %r13 ; X64-NEXT: addq %rdx, %r13
; X64-NEXT: addq %rsi, %r9 ; X64-NEXT: addq %rsi, %r9
; X64-NEXT: adcq %rbp, %r13 ; X64-NEXT: adcq %rbx, %r13
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rsi ; X64-NEXT: movq %rax, %rsi
; X64-NEXT: movq %r12, %rax ; X64-NEXT: movq %r12, %rax
; X64-NEXT: mulq %r8 ; X64-NEXT: mulq %r10
; X64-NEXT: movq %rdx, %rbx ; X64-NEXT: movq %rdx, %rbx
; X64-NEXT: movq %rax, %rbp ; X64-NEXT: movq %rax, %rbp
; X64-NEXT: addq %rcx, %rbp ; X64-NEXT: addq %rcx, %rbp
; X64-NEXT: adcq $0, %rbx ; X64-NEXT: adcq $0, %rbx
; X64-NEXT: movq %rdi, %rax ; X64-NEXT: movq %rdi, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %r8
; X64-NEXT: movq %rdx, %rcx ; X64-NEXT: movq %rdx, %rcx
; X64-NEXT: movq %rax, %rdi ; X64-NEXT: movq %rax, %rdi
; X64-NEXT: addq %rbp, %rdi ; X64-NEXT: addq %rbp, %rdi
; X64-NEXT: adcq $0, %rcx ; X64-NEXT: adcq %rbx, %rcx
; X64-NEXT: addq %rbx, %rcx ; X64-NEXT: sbbq %rbx, %rbx
; X64-NEXT: sbbq %rbp, %rbp ; X64-NEXT: andl $1, %ebx
; X64-NEXT: andl $1, %ebp
; X64-NEXT: movq %r12, %rax ; X64-NEXT: movq %r12, %rax
; X64-NEXT: mulq %r10 ; X64-NEXT: mulq %r8
; X64-NEXT: addq %rcx, %rax ; X64-NEXT: addq %rcx, %rax
; X64-NEXT: adcq %rbp, %rdx ; X64-NEXT: adcq %rbx, %rdx
; X64-NEXT: addq %r9, %rax ; X64-NEXT: addq %r9, %rax
; X64-NEXT: adcq %r13, %rdx ; X64-NEXT: adcq %r13, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq %r11, %rdi ; X64-NEXT: adcq %r15, %rdi
; X64-NEXT: adcq %r15, %rax ; X64-NEXT: adcq %r14, %rax
; X64-NEXT: adcq %r14, %rdx ; X64-NEXT: adcq %r11, %rdx
; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload ; X64-NEXT: addq -{{[0-9]+}}(%rsp), %rsi # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdi # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rax # 8-byte Folded Reload
; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload ; X64-NEXT: adcq -{{[0-9]+}}(%rsp), %rdx # 8-byte Folded Reload
; X64-NEXT: movq (%rsp), %rcx # 8-byte Reload ; X64-NEXT: movq (%rsp), %rcx # 8-byte Reload
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
; X64-NEXT: movq %rbp, (%rcx) ; X64-NEXT: movq %rbp, (%rcx)
; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload ; X64-NEXT: movq -{{[0-9]+}}(%rsp), %rbp # 8-byte Reload
Show All 23 Lines