pen2:                                   # @pen2
# %bb.0:
	push	rbp
	push	r15
	push	r14
	push	r13
	push	r12
	push	rbx
	sub	rsp, 40
	mov	r12, rdx
	mov	r15, rsi
	mov	rbp, rdi
	mov	cl, byte ptr [rbp - 15]
	test	cl, cl
	je	.LBB0_6
# %bb.1:
	mov	al, byte ptr [r15 - 15]
	test	al, al
	je	.LBB0_6
# %bb.2:
	cmp	cl, 5
	sete	bl
	add	bl, bl
	cmp	al, 5
	sete	dl
	or	dl, bl
	cmp	dl, 3
	je	.LBB0_7
# %bb.3:
	cmp	dl, 2
	je	.LBB0_11
# %bb.4:
	cmp	dl, 1
	jne	.LBB0_14
# %bb.5:
	mov	rdi, r15
	call	mut10
	mov	r13, rax
	mov	rsi, qword ptr [r13 + 8]
	mov	rdi, rbp
	jmp	.LBB0_12
.LBB0_6:
	mov	qword ptr [rsp + 16], rbp
	mov	qword ptr [rsp + 24], r15
	shl	r12, 5
	lea	rdi, [r12 + cuvw_+1040]
	lea	rsi, [rsp + 16]
	mov	edx, 2
	call	eac
	mov	r13, rax
	jmp	.LBB0_13
.LBB0_7:
	mov	rsi, qword ptr [r15]
	shl	r12, 5
	lea	rdx, [r12 + cuvw_+1040]
	lea	rcx, [r15 + 8]
	mov	r8d, 1
	mov	rdi, rbp
	call	amd
	mov	r13, rax
	cmp	dword ptr [r15 - 12], 0
	jne	.LBB0_13
# %bb.8:
	movzx	eax, byte ptr [r15 - 15]
	mov	ecx, 2017
	bt	ecx, eax
	jae	.LBB0_10
# %bb.9:
	mov	rdi, r15
	call	mdAll
.LBB0_10:
	movzx	eax, byte ptr [r15 - 16]
	mov	rcx, qword ptr [8*rax + mx]
	mov	qword ptr [r15 - 16], rcx
	mov	qword ptr [8*rax + mx], r15
	jmp	.LBB0_13
.LBB0_11:
	mov	rdi, rbp
	call	mut10
	mov	r13, rax
	mov	rdi, qword ptr [r13 + 8]
	mov	rsi, r15
.LBB0_12:
	mov	rdx, r12
	call	pen2
	add	dword ptr [rax - 12], 1
	mov	qword ptr [r13 + 8], rax
.LBB0_13:
	mov	rax, r13
	add	rsp, 40
	pop	rbx
	pop	r12
	pop	r13
	pop	r14
	pop	r15
	pop	rbp
	vzeroupper
	ret
.LBB0_14:
	mov	ebx, ecx
	shr	bl, 6
	and	bl, 2
	mov	edx, eax
	shr	dl, 7
	or	dl, bl
	cmp	dl, 2
	mov	rsi, r15
	cmovb	rsi, rbp
	mov	rbx, qword ptr [rsi - 8]
	test	dl, dl
	jne	.LBB0_16
# %bb.15:
	mov	rsi, qword ptr [rbp - 8]
	cmp	rsi, qword ptr [r15 - 8]
	jne	.LBB0_37
.LBB0_16:
	movzx	r14d, dl
	cmp	cl, -4
	je	.LBB0_20
# %bb.17:
	cmp	cl, 4
	je	.LBB0_20
# %bb.18:
	cmp	al, 4
	je	.LBB0_20
# %bb.19:
	cmp	al, -4
	jne	.LBB0_75
.LBB0_20:
	mov	r13d, offset gd
	mov	al, 1
	mov	dword ptr [rsp + 12], eax # 4-byte Spill
.LBB0_21:
	mov	rdi, rbp
	call	r13
	mov	rdi, r15
	mov	r15, rax
	call	r13
	mov	rbp, rax
	cmp	dword ptr [r15 - 12], 0
	jne	.LBB0_23
# %bb.22:
	cmp	qword ptr [r15 - 8], rbx
	mov	r13, r15
	je	.LBB0_26
.LBB0_23:
	cmp	dword ptr [rbp - 12], 0
	jne	.LBB0_25
# %bb.24:
	cmp	qword ptr [rbp - 8], rbx
	mov	r13, rbp
	je	.LBB0_26
.LBB0_25:
	lea	rax, [rbx + 1]
	lzcnt	rax, rax
	mov	edi, 67
	sub	rdi, rax
	or	rdi, 512
	mov	rsi, rbx
	call	ma
	mov	r13, rax
.LBB0_26:
	mov	r9, rbx
	xor	eax, eax
	cmp	r14b, 3
	sete	cl
	setne	bl
	mov	edx, ebx
	shl	dl, 2
	add	dl, -2
	mov	esi, dword ptr [rsp + 12] # 4-byte Reload
	test	sil, sil
	jne	.LBB0_28
# %bb.27:
	mov	ebx, edx
	cmp	r12, 8
	jl	.LBB0_30
.LBB0_29:
	mov	ebx, edx
.LBB0_30:
	mov	byte ptr [r13 - 15], bl
	mov	al, cl
	sub	r14, rax
	movzx	eax, sil
	lea	rax, [rax + 2*rax]
	add	rax, r14
	cmp	rax, 5
	ja	.LBB0_38
# %bb.31:
	jmp	qword ptr [8*rax + .LJTI0_0]
.LBB0_32:
	cmp	r12, 10
	mov	rbx, r9
	ja	.LBB0_39
# %bb.33:
	jmp	qword ptr [8*r12 + .LJTI0_6]
.LBB0_34:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.35:
	cmp	rbx, 4
	jb	.LBB0_36
# %bb.292:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_516
# %bb.293:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_516
.LBB0_36:
	xor	eax, eax
	jmp	.LBB0_742
.LBB0_28:
	shl	bl, 3
	add	bl, -4
	cmp	r12, 8
	jl	.LBB0_30
	jmp	.LBB0_29
.LBB0_37:
	mov	edi, offset .L.str
	call	e_
	mov	r13, rax
	jmp	.LBB0_13
.LBB0_38:
	mov	edi, offset .L.str.1
	call	e_
.LBB0_39:
	cmp	r15, r13
	je	.LBB0_44
# %bb.40:
	cmp	dword ptr [r15 - 12], 0
	jne	.LBB0_44
# %bb.41:
	movzx	eax, byte ptr [r15 - 15]
	mov	ecx, 2017
	bt	ecx, eax
	jae	.LBB0_43
# %bb.42:
	mov	rdi, r15
	vzeroupper
	call	mdAll
.LBB0_43:
	movzx	eax, byte ptr [r15 - 16]
	mov	rcx, qword ptr [8*rax + mx]
	mov	qword ptr [r15 - 16], rcx
	mov	qword ptr [8*rax + mx], r15
.LBB0_44:
	cmp	rbp, r13
	je	.LBB0_49
# %bb.45:
	cmp	dword ptr [rbp - 12], 0
	jne	.LBB0_13
# %bb.46:
	movzx	eax, byte ptr [rbp - 15]
	mov	ecx, 2017
	bt	ecx, eax
	jae	.LBB0_48
# %bb.47:
	mov	rdi, rbp
	vzeroupper
	call	mdAll
.LBB0_48:
	movzx	eax, byte ptr [rbp - 16]
	mov	rcx, qword ptr [8*rax + mx]
	mov	qword ptr [rbp - 16], rcx
	mov	qword ptr [8*rax + mx], rbp
	jmp	.LBB0_13
.LBB0_49:
	mov	r13, rbp
	jmp	.LBB0_13
.LBB0_50:
	cmp	r12, 10
	mov	rbx, r9
	ja	.LBB0_39
# %bb.51:
	mov	rcx, qword ptr [rbp]
	jmp	qword ptr [8*r12 + .LJTI0_5]
.LBB0_52:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.53:
	cmp	rbx, 4
	jae	.LBB0_295
# %bb.54:
	xor	eax, eax
	jmp	.LBB0_298
.LBB0_55:
	cmp	r12, 10
	mov	rbx, r9
	ja	.LBB0_39
# %bb.56:
	mov	rcx, qword ptr [r15]
	jmp	qword ptr [8*r12 + .LJTI0_4]
.LBB0_57:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.58:
	cmp	rbx, 4
	jae	.LBB0_300
# %bb.59:
	xor	eax, eax
	jmp	.LBB0_521
.LBB0_60:
	cmp	r12, 10
	mov	rsi, r9
	ja	.LBB0_39
# %bb.61:
	jmp	qword ptr [8*r12 + .LJTI0_3]
.LBB0_62:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.63:
	cmp	rsi, 4
	jae	.LBB0_303
# %bb.64:
	xor	eax, eax
	jmp	.LBB0_525
.LBB0_65:
	cmp	r12, 10
	mov	rsi, r9
	ja	.LBB0_39
# %bb.66:
	vmovsd	xmm0, qword ptr [rbp]   # xmm0 = mem[0],zero
	jmp	qword ptr [8*r12 + .LJTI0_2]
.LBB0_67:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.68:
	cmp	rsi, 4
	jae	.LBB0_306
# %bb.69:
	xor	eax, eax
	jmp	.LBB0_309
.LBB0_70:
	cmp	r12, 10
	mov	rsi, r9
	ja	.LBB0_39
# %bb.71:
	vmovsd	xmm0, qword ptr [r15]   # xmm0 = mem[0],zero
	jmp	qword ptr [8*r12 + .LJTI0_1]
.LBB0_72:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.73:
	cmp	rsi, 4
	jae	.LBB0_311
# %bb.74:
	xor	eax, eax
	jmp	.LBB0_529
.LBB0_75:
	mov	r13d, offset gl
	mov	dword ptr [rsp + 12], 0 # 4-byte Folded Spill
	jmp	.LBB0_21
.LBB0_76:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.77:
	cmp	rbx, 4
	jae	.LBB0_314
# %bb.78:
	xor	eax, eax
	jmp	.LBB0_744
.LBB0_79:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.80:
	cmp	rbx, 4
	jae	.LBB0_318
# %bb.81:
	xor	eax, eax
	jmp	.LBB0_746
.LBB0_82:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.83:
	cmp	rbx, 4
	jae	.LBB0_322
# %bb.84:
	xor	eax, eax
	jmp	.LBB0_748
.LBB0_85:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.86:
	cmp	rbx, 4
	jae	.LBB0_326
# %bb.87:
	xor	esi, esi
	jmp	.LBB0_685
.LBB0_88:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.89:
	movabs	rcx, -9223372036854775808
	xor	esi, esi
.LBB0_90:                               # =>This Inner Loop Header: Depth=1
	mov	rdi, qword ptr [r15 + 8*rsi]
	mov	rax, qword ptr [rbp + 8*rsi]
	test	rdi, rdi
	jle	.LBB0_93
# %bb.91:                               #   in Loop: Header=BB0_90 Depth=1
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_96
# %bb.92:                               #   in Loop: Header=BB0_90 Depth=1
	cqo
	idiv	rdi
	mov	rax, rdx
	jmp	.LBB0_97
.LBB0_93:                               #   in Loop: Header=BB0_90 Depth=1
	mov	rdx, rcx
	je	.LBB0_101
# %bb.94:                               #   in Loop: Header=BB0_90 Depth=1
	neg	rdi
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_100
# %bb.95:                               #   in Loop: Header=BB0_90 Depth=1
	cqo
	idiv	rdi
	mov	rdx, rax
	jmp	.LBB0_101
.LBB0_96:                               #   in Loop: Header=BB0_90 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
	mov	eax, edx
.LBB0_97:                               #   in Loop: Header=BB0_90 Depth=1
	add	rax, rdi
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_99
# %bb.98:                               #   in Loop: Header=BB0_90 Depth=1
	cqo
	idiv	rdi
	jmp	.LBB0_101
.LBB0_99:                               #   in Loop: Header=BB0_90 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
                                        # kill: def %edx killed %edx def %rdx
	jmp	.LBB0_101
.LBB0_100:                              #   in Loop: Header=BB0_90 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
	mov	edx, eax
.LBB0_101:                              #   in Loop: Header=BB0_90 Depth=1
	mov	qword ptr [r13 + 8*rsi], rdx
	add	rsi, 1
	cmp	rbx, rsi
	jne	.LBB0_90
	jmp	.LBB0_39
.LBB0_102:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.103:
	cmp	rbx, 4
	jae	.LBB0_350
# %bb.104:
	xor	eax, eax
	jmp	.LBB0_750
.LBB0_105:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.106:
	cmp	rbx, 4
	jae	.LBB0_354
# %bb.107:
	xor	eax, eax
	jmp	.LBB0_752
.LBB0_108:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.109:
	cmp	rbx, 4
	jae	.LBB0_358
# %bb.110:
	xor	eax, eax
	jmp	.LBB0_754
.LBB0_111:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.112:
	cmp	rbx, 4
	jae	.LBB0_362
# %bb.113:
	xor	eax, eax
	jmp	.LBB0_756
.LBB0_114:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.115:
	cmp	rbx, 4
	jae	.LBB0_366
# %bb.116:
	xor	eax, eax
	jmp	.LBB0_758
.LBB0_117:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.118:
	cmp	rbx, 4
	jb	.LBB0_119
# %bb.370:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_530
# %bb.371:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_530
.LBB0_119:
	xor	eax, eax
	jmp	.LBB0_760
.LBB0_120:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.121:
	cmp	rbx, 4
	jb	.LBB0_122
# %bb.373:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_532
# %bb.374:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_532
.LBB0_122:
	xor	eax, eax
	jmp	.LBB0_762
.LBB0_123:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.124:
	cmp	rbx, 4
	jb	.LBB0_125
# %bb.376:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_534
# %bb.377:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_534
.LBB0_125:
	xor	eax, eax
	jmp	.LBB0_764
.LBB0_126:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.127:
	test	rcx, rcx
	je	.LBB0_493
# %bb.128:
	xor	esi, esi
.LBB0_129:                              # =>This Inner Loop Header: Depth=1
	mov	rax, qword ptr [r15 + 8*rsi]
	mov	rdx, rax
	or	rdx, rcx
	shr	rdx, 32
	je	.LBB0_131
# %bb.130:                              #   in Loop: Header=BB0_129 Depth=1
	cqo
	idiv	rcx
	jmp	.LBB0_132
.LBB0_131:                              #   in Loop: Header=BB0_129 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	ecx
                                        # kill: def %eax killed %eax def %rax
.LBB0_132:                              #   in Loop: Header=BB0_129 Depth=1
	mov	qword ptr [r13 + 8*rsi], rax
	add	rsi, 1
	cmp	rbx, rsi
	jne	.LBB0_129
	jmp	.LBB0_39
.LBB0_133:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.134:
	movabs	r8, -9223372036854775808
	xor	edi, edi
.LBB0_135:                              # =>This Inner Loop Header: Depth=1
	mov	rsi, qword ptr [r15 + 8*rdi]
	test	rsi, rsi
	jle	.LBB0_138
# %bb.136:                              #   in Loop: Header=BB0_135 Depth=1
	mov	rax, rcx
	or	rax, rsi
	shr	rax, 32
	je	.LBB0_141
# %bb.137:                              #   in Loop: Header=BB0_135 Depth=1
	mov	rax, rcx
	cqo
	idiv	rsi
	mov	rax, rdx
	jmp	.LBB0_142
.LBB0_138:                              #   in Loop: Header=BB0_135 Depth=1
	mov	rdx, r8
	je	.LBB0_146
# %bb.139:                              #   in Loop: Header=BB0_135 Depth=1
	neg	rsi
	mov	rax, rcx
	or	rax, rsi
	shr	rax, 32
	je	.LBB0_145
# %bb.140:                              #   in Loop: Header=BB0_135 Depth=1
	mov	rax, rcx
	cqo
	idiv	rsi
	mov	rdx, rax
	jmp	.LBB0_146
.LBB0_141:                              #   in Loop: Header=BB0_135 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	esi
	mov	eax, edx
.LBB0_142:                              #   in Loop: Header=BB0_135 Depth=1
	add	rax, rsi
	mov	rdx, rax
	or	rdx, rsi
	shr	rdx, 32
	je	.LBB0_144
# %bb.143:                              #   in Loop: Header=BB0_135 Depth=1
	cqo
	idiv	rsi
	jmp	.LBB0_146
.LBB0_144:                              #   in Loop: Header=BB0_135 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	esi
                                        # kill: def %edx killed %edx def %rdx
	jmp	.LBB0_146
.LBB0_145:                              #   in Loop: Header=BB0_135 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	esi
	mov	edx, eax
.LBB0_146:                              #   in Loop: Header=BB0_135 Depth=1
	mov	qword ptr [r13 + 8*rdi], rdx
	add	rdi, 1
	cmp	rbx, rdi
	jne	.LBB0_135
	jmp	.LBB0_39
.LBB0_147:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.148:
	cmp	rbx, 4
	jae	.LBB0_379
# %bb.149:
	xor	eax, eax
	jmp	.LBB0_539
.LBB0_150:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.151:
	cmp	rbx, 4
	jae	.LBB0_382
# %bb.152:
	xor	eax, eax
	jmp	.LBB0_543
.LBB0_153:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.154:
	cmp	rbx, 4
	jae	.LBB0_385
# %bb.155:
	xor	eax, eax
	jmp	.LBB0_547
.LBB0_156:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.157:
	cmp	rbx, 4
	jae	.LBB0_388
# %bb.158:
	xor	eax, eax
	jmp	.LBB0_551
.LBB0_159:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.160:
	cmp	rbx, 4
	jae	.LBB0_391
# %bb.161:
	xor	eax, eax
	jmp	.LBB0_555
.LBB0_162:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.163:
	cmp	rbx, 4
	jae	.LBB0_394
# %bb.164:
	xor	eax, eax
	jmp	.LBB0_559
.LBB0_165:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.166:
	cmp	rbx, 4
	jae	.LBB0_397
# %bb.167:
	xor	eax, eax
	jmp	.LBB0_563
.LBB0_168:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.169:
	cmp	rbx, 4
	jae	.LBB0_400
# %bb.170:
	xor	eax, eax
	jmp	.LBB0_567
.LBB0_171:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.172:
	movabs	rax, -9223372036854775808
	lea	rdx, [rax - 1]
	lea	r8, [rax + 1]
	test	rcx, rcx
	cmovg	r8, rdx
	je	.LBB0_495
# %bb.173:
	cmp	rbx, 4
	jae	.LBB0_507
# %bb.174:
	xor	edi, edi
	mov	rsi, qword ptr [rbp + 8*rdi]
	mov	rax, r8
	test	rsi, rsi
	jne	.LBB0_664
	jmp	.LBB0_667
.LBB0_175:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.176:
	test	rcx, rcx
	jle	.LBB0_497
# %bb.177:
	xor	esi, esi
.LBB0_178:                              # =>This Inner Loop Header: Depth=1
	mov	rax, qword ptr [rbp + 8*rsi]
	mov	rdx, rax
	or	rdx, rcx
	shr	rdx, 32
	je	.LBB0_180
# %bb.179:                              #   in Loop: Header=BB0_178 Depth=1
	cqo
	idiv	rcx
	mov	rax, rdx
	jmp	.LBB0_181
.LBB0_180:                              #   in Loop: Header=BB0_178 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	ecx
	mov	eax, edx
.LBB0_181:                              #   in Loop: Header=BB0_178 Depth=1
	add	rax, rcx
	mov	rdx, rax
	or	rdx, rcx
	shr	rdx, 32
	je	.LBB0_183
# %bb.182:                              #   in Loop: Header=BB0_178 Depth=1
	cqo
	idiv	rcx
	jmp	.LBB0_184
.LBB0_183:                              #   in Loop: Header=BB0_178 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	ecx
                                        # kill: def %edx killed %edx def %rdx
.LBB0_184:                              #   in Loop: Header=BB0_178 Depth=1
	mov	qword ptr [r13 + 8*rsi], rdx
	add	rsi, 1
	cmp	rbx, rsi
	jne	.LBB0_178
	jmp	.LBB0_39
.LBB0_185:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.186:
	cmp	rbx, 4
	jae	.LBB0_403
# %bb.187:
	xor	eax, eax
	jmp	.LBB0_571
.LBB0_188:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.189:
	cmp	rbx, 4
	jae	.LBB0_406
# %bb.190:
	xor	eax, eax
	jmp	.LBB0_575
.LBB0_191:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.192:
	cmp	rbx, 4
	jae	.LBB0_409
# %bb.193:
	xor	eax, eax
	jmp	.LBB0_579
.LBB0_194:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.195:
	cmp	rbx, 4
	jae	.LBB0_412
# %bb.196:
	xor	eax, eax
	jmp	.LBB0_583
.LBB0_197:
	test	rbx, rbx
	jle	.LBB0_39
# %bb.198:
	cmp	rbx, 4
	jae	.LBB0_415
# %bb.199:
	xor	eax, eax
	jmp	.LBB0_587
.LBB0_200:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.201:
	cmp	rsi, 4
	jae	.LBB0_418
# %bb.202:
	xor	eax, eax
	jmp	.LBB0_699
.LBB0_203:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.204:
	cmp	rsi, 4
	jae	.LBB0_423
# %bb.205:
	xor	eax, eax
	jmp	.LBB0_701
.LBB0_206:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.207:
	cmp	rsi, 4
	jae	.LBB0_428
# %bb.208:
	xor	eax, eax
	jmp	.LBB0_703
.LBB0_209:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.210:
	cmp	rsi, 4
	jae	.LBB0_433
# %bb.211:
	xor	eax, eax
	jmp	.LBB0_705
.LBB0_212:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.213:
	cmp	rsi, 4
	jae	.LBB0_438
# %bb.214:
	xor	eax, eax
	jmp	.LBB0_441
.LBB0_215:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.216:
	cmp	rsi, 4
	jae	.LBB0_443
# %bb.217:
	xor	eax, eax
	jmp	.LBB0_712
.LBB0_218:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.219:
	cmp	rsi, 4
	jae	.LBB0_448
# %bb.220:
	xor	eax, eax
	jmp	.LBB0_714
.LBB0_221:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.222:
	xor	eax, eax
.LBB0_223:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [rbp + 8*rax] # xmm0 = mem[0],zero
	xor	ecx, ecx
	vucomisd	xmm0, qword ptr [r15 + 8*rax]
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_223
	jmp	.LBB0_39
.LBB0_224:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.225:
	xor	eax, eax
.LBB0_226:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	xor	ecx, ecx
	vucomisd	xmm0, qword ptr [rbp + 8*rax]
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_226
	jmp	.LBB0_39
.LBB0_227:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.228:
	xor	eax, eax
.LBB0_229:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [rbp + 8*rax] # xmm0 = mem[0],zero
	vcmpeqsd	xmm0, xmm0, qword ptr [r15 + 8*rax]
	vmovq	rcx, xmm0
	and	ecx, 1
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_229
	jmp	.LBB0_39
.LBB0_230:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.231:
	cmp	rsi, 4
	jae	.LBB0_453
# %bb.232:
	xor	eax, eax
	jmp	.LBB0_591
.LBB0_233:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.234:
	cmp	rsi, 4
	jae	.LBB0_456
# %bb.235:
	xor	eax, eax
	jmp	.LBB0_595
.LBB0_236:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.237:
	cmp	rsi, 4
	jae	.LBB0_459
# %bb.238:
	xor	eax, eax
	jmp	.LBB0_599
.LBB0_239:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.240:
	vxorpd	xmm1, xmm1, xmm1
	vucomisd	xmm0, xmm1
	jne	.LBB0_241
	jnp	.LBB0_503
.LBB0_241:
	cmp	rsi, 4
	jae	.LBB0_510
# %bb.242:
	xor	eax, eax
	jmp	.LBB0_671
.LBB0_243:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.244:
	cmp	rsi, 4
	jae	.LBB0_462
# %bb.245:
	xor	eax, eax
	jmp	.LBB0_465
.LBB0_246:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.247:
	cmp	rsi, 4
	jae	.LBB0_467
# %bb.248:
	xor	eax, eax
	jmp	.LBB0_603
.LBB0_249:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.250:
	cmp	rsi, 4
	jae	.LBB0_470
# %bb.251:
	xor	eax, eax
	jmp	.LBB0_607
.LBB0_252:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.253:
	xor	eax, eax
.LBB0_254:                              # =>This Inner Loop Header: Depth=1
	xor	ecx, ecx
	vucomisd	xmm0, qword ptr [r15 + 8*rax]
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_254
	jmp	.LBB0_39
.LBB0_255:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.256:
	xor	eax, eax
.LBB0_257:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [r15 + 8*rax] # xmm1 = mem[0],zero
	xor	ecx, ecx
	vucomisd	xmm1, xmm0
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_257
	jmp	.LBB0_39
.LBB0_258:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.259:
	xor	eax, eax
.LBB0_260:                              # =>This Inner Loop Header: Depth=1
	vcmpeqsd	xmm1, xmm0, qword ptr [r15 + 8*rax]
	vmovq	rcx, xmm1
	and	ecx, 1
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_260
	jmp	.LBB0_39
.LBB0_261:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.262:
	cmp	rsi, 4
	jae	.LBB0_473
# %bb.263:
	xor	eax, eax
	jmp	.LBB0_611
.LBB0_264:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.265:
	cmp	rsi, 4
	jae	.LBB0_476
# %bb.266:
	xor	eax, eax
	jmp	.LBB0_615
.LBB0_267:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.268:
	cmp	rsi, 4
	jae	.LBB0_479
# %bb.269:
	xor	eax, eax
	jmp	.LBB0_619
.LBB0_270:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.271:
	vxorpd	xmm1, xmm1, xmm1
	xor	eax, eax
	vucomisd	xmm0, xmm1
	seta	cl
	jne	.LBB0_272
	jnp	.LBB0_505
.LBB0_272:
	mov	al, cl
	vmovsd	xmm1, qword ptr [8*rax + .LCPI0_1] # xmm1 = mem[0],zero
	cmp	rsi, 4
	jb	.LBB0_273
# %bb.513:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_672
# %bb.514:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_672
.LBB0_273:
	xor	eax, eax
.LBB0_675:
	vxorpd	xmm2, xmm2, xmm2
.LBB0_676:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm4, qword ptr [rbp + 8*rax] # xmm4 = mem[0],zero
	vucomisd	xmm4, xmm2
	vmovapd	xmm3, xmm1
	jne	.LBB0_677
	jnp	.LBB0_678
.LBB0_677:                              #   in Loop: Header=BB0_676 Depth=1
	vdivsd	xmm3, xmm0, xmm4
.LBB0_678:                              #   in Loop: Header=BB0_676 Depth=1
	vmovsd	qword ptr [r13 + 8*rax], xmm3
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_676
	jmp	.LBB0_39
.LBB0_274:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.275:
	cmp	rsi, 4
	jae	.LBB0_482
# %bb.276:
	xor	eax, eax
	jmp	.LBB0_485
.LBB0_277:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.278:
	cmp	rsi, 4
	jae	.LBB0_487
# %bb.279:
	xor	eax, eax
	jmp	.LBB0_623
.LBB0_280:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.281:
	cmp	rsi, 4
	jae	.LBB0_490
# %bb.282:
	xor	eax, eax
	jmp	.LBB0_627
.LBB0_283:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.284:
	xor	eax, eax
.LBB0_285:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [rbp + 8*rax] # xmm1 = mem[0],zero
	xor	ecx, ecx
	vucomisd	xmm1, xmm0
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_285
	jmp	.LBB0_39
.LBB0_286:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.287:
	xor	eax, eax
.LBB0_288:                              # =>This Inner Loop Header: Depth=1
	xor	ecx, ecx
	vucomisd	xmm0, qword ptr [rbp + 8*rax]
	seta	cl
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_288
	jmp	.LBB0_39
.LBB0_289:
	test	rsi, rsi
	jle	.LBB0_39
# %bb.290:
	xor	eax, eax
.LBB0_291:                              # =>This Inner Loop Header: Depth=1
	vcmpeqsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovq	rcx, xmm1
	and	ecx, 1
	shl	rcx, 5
	mov	rcx, qword ptr [rcx + cb_+16]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_291
	jmp	.LBB0_39
.LBB0_295:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	mov	rdx, rax
	mov	rsi, r13
.LBB0_296:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymmword ptr [rsi], ymm0
	add	rsi, 32
	add	rdx, -4
	jne	.LBB0_296
# %bb.297:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_298:
	lea	rdx, [8*rax]
	add	rdx, r13
	sub	rbx, rax
.LBB0_299:                              # =>This Inner Loop Header: Depth=1
	mov	qword ptr [rdx], rcx
	add	rdx, 8
	add	rbx, -1
	jne	.LBB0_299
	jmp	.LBB0_39
.LBB0_300:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_518
# %bb.301:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_518
# %bb.302:
	xor	eax, eax
	jmp	.LBB0_521
.LBB0_303:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_522
# %bb.304:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_522
# %bb.305:
	xor	eax, eax
	jmp	.LBB0_525
.LBB0_306:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	mov	rcx, rax
	mov	rdx, r13
.LBB0_307:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymmword ptr [rdx], ymm1
	add	rdx, 32
	add	rcx, -4
	jne	.LBB0_307
# %bb.308:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_309:
	lea	rcx, [8*rax]
	add	rcx, r13
	sub	rsi, rax
.LBB0_310:                              # =>This Inner Loop Header: Depth=1
	vmovsd	qword ptr [rcx], xmm0
	add	rcx, 8
	add	rsi, -1
	jne	.LBB0_310
	jmp	.LBB0_39
.LBB0_311:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_526
# %bb.312:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_526
# %bb.313:
	xor	eax, eax
	jmp	.LBB0_529
.LBB0_314:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_681
# %bb.315:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_744
# %bb.316:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_317:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [rbp + 8*rcx]
	vpaddq	ymm0, ymm0, ymmword ptr [r15 + 8*rcx]
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_317
	jmp	.LBB0_743
.LBB0_318:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_682
# %bb.319:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_746
# %bb.320:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_321:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vpsubq	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_321
	jmp	.LBB0_745
.LBB0_322:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_683
# %bb.323:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_748
# %bb.324:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_325:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rcx]
	vpsrlq	ymm2, ymm1, 32
	vpmuludq	ymm2, ymm2, ymm0
	vpsrlq	ymm3, ymm0, 32
	vpmuludq	ymm3, ymm1, ymm3
	vpaddq	ymm2, ymm3, ymm2
	vpsllq	ymm2, ymm2, 32
	vpmuludq	ymm0, ymm1, ymm0
	vpaddq	ymm0, ymm0, ymm2
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_325
	jmp	.LBB0_747
.LBB0_326:
	lea	rcx, [8*rbx]
	add	rcx, r13
	lea	rax, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rax
	setb	dil
	cmp	r15, rcx
	setb	bl
	cmp	r13, rdx
	setb	al
	cmp	rbp, rcx
	setb	cl
	xor	esi, esi
	test	dil, bl
	jne	.LBB0_684
# %bb.327:
	and	al, cl
	mov	rbx, r9
	jne	.LBB0_685
# %bb.328:
	mov	rsi, rbx
	and	rsi, -4
	xor	ecx, ecx
	vpxor	xmm0, xmm0, xmm0
	vpcmpeqd	ymm9, ymm9, ymm9
	vbroadcastsd	ymm11, qword ptr [rip + .LCPI0_5] # ymm11 = [9223372036854775807,9223372036854775807,9223372036854775807,9223372036854775807]
	vbroadcastsd	ymm3, qword ptr [rip + .LCPI0_6] # ymm3 = [9223372036854775809,9223372036854775809,9223372036854775809,9223372036854775809]
	vbroadcastsd	ymm4, qword ptr [rip + .LCPI0_4] # ymm4 = [9223372036854775808,9223372036854775808,9223372036854775808,9223372036854775808]
.LBB0_329:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm5, ymmword ptr [r15 + 8*rcx]
	vmovdqu	ymm7, ymmword ptr [rbp + 8*rcx]
	vpcmpeqq	ymm6, ymm7, ymm0
	vpxor	ymm8, ymm6, ymm9
	vextracti128	xmm1, ymm8, 1
	vpackssdw	xmm1, xmm8, xmm1
	vpextrb	eax, xmm8, 0
                                        # implicit-def: %ymm8
	test	al, 1
	je	.LBB0_330
# %bb.333:                              #   in Loop: Header=BB0_329 Depth=1
	vmovq	rax, xmm5
	vmovq	rdi, xmm7
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_335
# %bb.334:                              #   in Loop: Header=BB0_329 Depth=1
	cqo
	idiv	rdi
	vmovq	xmm8, rax
	vpextrb	eax, xmm1, 4
	test	al, 1
	jne	.LBB0_336
.LBB0_331:                              #   in Loop: Header=BB0_329 Depth=1
	vpextrb	eax, xmm1, 8
	test	al, 1
	je	.LBB0_332
.LBB0_340:                              #   in Loop: Header=BB0_329 Depth=1
	vextracti128	xmm2, ymm5, 1
	vmovq	rax, xmm2
	vextracti128	xmm2, ymm7, 1
	vmovq	rdi, xmm2
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_342
# %bb.341:                              #   in Loop: Header=BB0_329 Depth=1
	cqo
	idiv	rdi
	jmp	.LBB0_343
.LBB0_330:                              #   in Loop: Header=BB0_329 Depth=1
	vpextrb	eax, xmm1, 4
	test	al, 1
	je	.LBB0_331
.LBB0_336:                              #   in Loop: Header=BB0_329 Depth=1
	vpextrq	rax, xmm5, 1
	vpextrq	rdi, xmm7, 1
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_338
# %bb.337:                              #   in Loop: Header=BB0_329 Depth=1
	cqo
	idiv	rdi
	jmp	.LBB0_339
.LBB0_335:                              #   in Loop: Header=BB0_329 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
                                        # kill: def %eax killed %eax def %rax
	vmovq	xmm8, rax
	vpextrb	eax, xmm1, 4
	test	al, 1
	je	.LBB0_331
	jmp	.LBB0_336
.LBB0_342:                              #   in Loop: Header=BB0_329 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
                                        # kill: def %eax killed %eax def %rax
.LBB0_343:                              #   in Loop: Header=BB0_329 Depth=1
	vextracti128	xmm2, ymm8, 1
	vpinsrq	xmm2, xmm2, rax, 0
	vinserti128	ymm8, ymm8, xmm2, 1
	vpextrb	eax, xmm1, 12
	test	al, 1
	je	.LBB0_348
	jmp	.LBB0_344
.LBB0_338:                              #   in Loop: Header=BB0_329 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
                                        # kill: def %eax killed %eax def %rax
.LBB0_339:                              #   in Loop: Header=BB0_329 Depth=1
	vpinsrq	xmm10, xmm8, rax, 1
	vpblendd	ymm8, ymm8, ymm10, 15 # ymm8 = ymm10[0,1,2,3],ymm8[4,5,6,7]
	vpextrb	eax, xmm1, 8
	test	al, 1
	jne	.LBB0_340
.LBB0_332:                              #   in Loop: Header=BB0_329 Depth=1
	vpextrb	eax, xmm1, 12
	test	al, 1
	je	.LBB0_348
.LBB0_344:                              #   in Loop: Header=BB0_329 Depth=1
	vextracti128	xmm1, ymm5, 1
	vpextrq	rax, xmm1, 1
	vextracti128	xmm1, ymm7, 1
	vpextrq	rdi, xmm1, 1
	mov	rdx, rax
	or	rdx, rdi
	shr	rdx, 32
	je	.LBB0_346
# %bb.345:                              #   in Loop: Header=BB0_329 Depth=1
	cqo
	idiv	rdi
	jmp	.LBB0_347
.LBB0_346:                              #   in Loop: Header=BB0_329 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	edi
                                        # kill: def %eax killed %eax def %rax
.LBB0_347:                              #   in Loop: Header=BB0_329 Depth=1
	vextracti128	xmm1, ymm8, 1
	vpinsrq	xmm1, xmm1, rax, 1
	vinserti128	ymm8, ymm8, xmm1, 1
.LBB0_348:                              #   in Loop: Header=BB0_329 Depth=1
	vpor	ymm1, ymm5, ymm7
	vpcmpeqq	ymm2, ymm5, ymm0
	vpcmpgtq	ymm5, ymm5, ymm0
	vblendvpd	ymm5, ymm3, ymm11, ymm5
	vpandn	ymm2, ymm2, ymm6
	vblendvpd	ymm2, ymm8, ymm5, ymm2
	vpcmpeqq	ymm1, ymm1, ymm0
	vblendvpd	ymm1, ymm2, ymm4, ymm1
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm1
	add	rcx, 4
	cmp	rsi, rcx
	jne	.LBB0_329
# %bb.349:
	cmp	rbx, rsi
	je	.LBB0_39
	jmp	.LBB0_685
.LBB0_350:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_693
# %bb.351:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_750
# %bb.352:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_353:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rcx]
	vpcmpgtq	ymm2, ymm1, ymm0
	vblendvpd	ymm0, ymm1, ymm0, ymm2
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_353
	jmp	.LBB0_749
.LBB0_354:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_694
# %bb.355:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_752
# %bb.356:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_357:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rcx]
	vpcmpgtq	ymm2, ymm0, ymm1
	vblendvpd	ymm0, ymm1, ymm0, ymm2
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_357
	jmp	.LBB0_751
.LBB0_358:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_695
# %bb.359:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_754
# %bb.360:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_361:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [rbp + 8*rcx]
	vpcmpgtq	ymm0, ymm0, ymmword ptr [r15 + 8*rcx]
	vpsrlq	ymm0, ymm0, 63
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_361
	jmp	.LBB0_753
.LBB0_362:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_696
# %bb.363:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_756
# %bb.364:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_365:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vpcmpgtq	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vpsrlq	ymm0, ymm0, 63
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_365
	jmp	.LBB0_755
.LBB0_366:
	lea	rax, [8*rbx]
	add	rax, r13
	lea	rcx, [r15 + 8*rbx]
	lea	rdx, [8*rbx]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_697
# %bb.367:
	and	cl, dl
	mov	rbx, r9
	jne	.LBB0_758
# %bb.368:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_369:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm0, ymmword ptr [r15 + 8*rcx]
	vpcmpeqq	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vpsrlq	ymm0, ymm0, 63
	vmovdqu	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_369
	jmp	.LBB0_757
.LBB0_379:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_536
# %bb.380:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_536
# %bb.381:
	xor	eax, eax
	jmp	.LBB0_539
.LBB0_382:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_540
# %bb.383:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_540
# %bb.384:
	xor	eax, eax
	jmp	.LBB0_543
.LBB0_385:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_544
# %bb.386:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_544
# %bb.387:
	xor	eax, eax
	jmp	.LBB0_547
.LBB0_388:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_548
# %bb.389:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_548
# %bb.390:
	xor	eax, eax
	jmp	.LBB0_551
.LBB0_391:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_552
# %bb.392:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_552
# %bb.393:
	xor	eax, eax
	jmp	.LBB0_555
.LBB0_394:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_556
# %bb.395:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_556
# %bb.396:
	xor	eax, eax
	jmp	.LBB0_559
.LBB0_397:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_560
# %bb.398:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_560
# %bb.399:
	xor	eax, eax
	jmp	.LBB0_563
.LBB0_400:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_564
# %bb.401:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_564
# %bb.402:
	xor	eax, eax
	jmp	.LBB0_567
.LBB0_403:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_568
# %bb.404:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_568
# %bb.405:
	xor	eax, eax
	jmp	.LBB0_571
.LBB0_406:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_572
# %bb.407:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_572
# %bb.408:
	xor	eax, eax
	jmp	.LBB0_575
.LBB0_409:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_576
# %bb.410:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_576
# %bb.411:
	xor	eax, eax
	jmp	.LBB0_579
.LBB0_412:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_580
# %bb.413:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_580
# %bb.414:
	xor	eax, eax
	jmp	.LBB0_583
.LBB0_415:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_584
# %bb.416:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_584
# %bb.417:
	xor	eax, eax
	jmp	.LBB0_587
.LBB0_418:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_698
# %bb.419:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_699
# %bb.420:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_421:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [r15 + 8*rcx]
	vaddpd	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_421
# %bb.422:
	cmp	rsi, rax
	jne	.LBB0_699
	jmp	.LBB0_39
.LBB0_423:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_700
# %bb.424:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_701
# %bb.425:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_426:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [r15 + 8*rcx]
	vsubpd	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_426
# %bb.427:
	cmp	rsi, rax
	jne	.LBB0_701
	jmp	.LBB0_39
.LBB0_428:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_702
# %bb.429:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_703
# %bb.430:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_431:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [r15 + 8*rcx]
	vmulpd	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_431
# %bb.432:
	cmp	rsi, rax
	jne	.LBB0_703
	jmp	.LBB0_39
.LBB0_433:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_704
# %bb.434:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_705
# %bb.435:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
	vxorpd	xmm0, xmm0, xmm0
	vbroadcastsd	ymm1, qword ptr [rip + .LCPI0_2] # ymm1 = [+Inf,+Inf,+Inf,+Inf]
	vbroadcastsd	ymm2, qword ptr [rip + .LCPI0_3] # ymm2 = [-Inf,-Inf,-Inf,-Inf]
	vbroadcastsd	ymm3, qword ptr [rip + .LCPI0_0] # ymm3 = [NaN,NaN,NaN,NaN]
.LBB0_436:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm4, ymmword ptr [r15 + 8*rcx]
	vmovupd	ymm5, ymmword ptr [rbp + 8*rcx]
	vcmpeqpd	ymm6, ymm5, ymm0
	vcmpeqpd	ymm7, ymm4, ymm0
	vcmpltpd	ymm8, ymm0, ymm4
	vblendvpd	ymm8, ymm2, ymm1, ymm8
	vcmpneqpd	ymm9, ymm4, ymm0
	vdivpd	ymm4, ymm4, ymm5
	vandpd	ymm5, ymm9, ymm6
	vblendvpd	ymm4, ymm4, ymm8, ymm5
	vandpd	ymm5, ymm6, ymm7
	vblendvpd	ymm4, ymm4, ymm3, ymm5
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm4
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_436
# %bb.437:
	cmp	rsi, rax
	je	.LBB0_39
	jmp	.LBB0_705
.LBB0_438:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm0, qword ptr [rip + .LCPI0_0] # ymm0 = [NaN,NaN,NaN,NaN]
	mov	rcx, rax
	mov	rdx, r13
.LBB0_439:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymmword ptr [rdx], ymm0
	add	rdx, 32
	add	rcx, -4
	jne	.LBB0_439
# %bb.440:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_441:
	lea	rcx, [8*rax]
	add	rcx, r13
	sub	rsi, rax
	movabs	rax, 9221120237041090560
.LBB0_442:                              # =>This Inner Loop Header: Depth=1
	mov	qword ptr [rcx], rax
	add	rcx, 8
	add	rsi, -1
	jne	.LBB0_442
	jmp	.LBB0_39
.LBB0_443:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_711
# %bb.444:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_712
# %bb.445:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_446:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [r15 + 8*rcx]
	vminpd	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_446
# %bb.447:
	cmp	rsi, rax
	jne	.LBB0_712
	jmp	.LBB0_39
.LBB0_448:
	lea	rax, [8*rsi]
	add	rax, r13
	lea	rcx, [r15 + 8*rsi]
	lea	rdx, [8*rsi]
	add	rdx, rbp
	cmp	r13, rcx
	setb	sil
	cmp	r15, rax
	setb	bl
	cmp	r13, rdx
	setb	cl
	cmp	rbp, rax
	setb	dl
	xor	eax, eax
	test	sil, bl
	jne	.LBB0_713
# %bb.449:
	and	cl, dl
	mov	rsi, r9
	jne	.LBB0_714
# %bb.450:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_451:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [r15 + 8*rcx]
	vmaxpd	ymm0, ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_451
# %bb.452:
	cmp	rsi, rax
	jne	.LBB0_714
	jmp	.LBB0_39
.LBB0_453:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_588
# %bb.454:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_588
# %bb.455:
	xor	eax, eax
	jmp	.LBB0_591
.LBB0_456:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_592
# %bb.457:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_592
# %bb.458:
	xor	eax, eax
	jmp	.LBB0_595
.LBB0_459:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_596
# %bb.460:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_596
# %bb.461:
	xor	eax, eax
	jmp	.LBB0_599
.LBB0_462:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm0, qword ptr [rip + .LCPI0_0] # ymm0 = [NaN,NaN,NaN,NaN]
	mov	rcx, rax
	mov	rdx, r13
.LBB0_463:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymmword ptr [rdx], ymm0
	add	rdx, 32
	add	rcx, -4
	jne	.LBB0_463
# %bb.464:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_465:
	lea	rcx, [8*rax]
	add	rcx, r13
	sub	rsi, rax
	movabs	rax, 9221120237041090560
.LBB0_466:                              # =>This Inner Loop Header: Depth=1
	mov	qword ptr [rcx], rax
	add	rcx, 8
	add	rsi, -1
	jne	.LBB0_466
	jmp	.LBB0_39
.LBB0_467:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_600
# %bb.468:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_600
# %bb.469:
	xor	eax, eax
	jmp	.LBB0_603
.LBB0_470:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_604
# %bb.471:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_604
# %bb.472:
	xor	eax, eax
	jmp	.LBB0_607
.LBB0_473:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_608
# %bb.474:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_608
# %bb.475:
	xor	eax, eax
	jmp	.LBB0_611
.LBB0_476:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_612
# %bb.477:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_612
# %bb.478:
	xor	eax, eax
	jmp	.LBB0_615
.LBB0_479:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_616
# %bb.480:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_616
# %bb.481:
	xor	eax, eax
	jmp	.LBB0_619
.LBB0_482:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm0, qword ptr [rip + .LCPI0_0] # ymm0 = [NaN,NaN,NaN,NaN]
	mov	rcx, rax
	mov	rdx, r13
.LBB0_483:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymmword ptr [rdx], ymm0
	add	rdx, 32
	add	rcx, -4
	jne	.LBB0_483
# %bb.484:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_485:
	lea	rcx, [8*rax]
	add	rcx, r13
	sub	rsi, rax
	movabs	rax, 9221120237041090560
.LBB0_486:                              # =>This Inner Loop Header: Depth=1
	mov	qword ptr [rcx], rax
	add	rcx, 8
	add	rsi, -1
	jne	.LBB0_486
	jmp	.LBB0_39
.LBB0_487:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_620
# %bb.488:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_620
# %bb.489:
	xor	eax, eax
	jmp	.LBB0_623
.LBB0_490:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_624
# %bb.491:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_624
# %bb.492:
	xor	eax, eax
	jmp	.LBB0_627
.LBB0_493:
	cmp	rbx, 4
	jb	.LBB0_494
# %bb.628:
	lea	rax, [r15 + 8*rbx]
	cmp	r13, rax
	jae	.LBB0_715
# %bb.629:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_715
.LBB0_494:
	xor	eax, eax
.LBB0_718:
	movabs	rcx, -9223372036854775808
	lea	rdx, [rcx - 1]
	lea	rsi, [rcx + 1]
.LBB0_719:                              # =>This Inner Loop Header: Depth=1
	cmp	qword ptr [r15 + 8*rax], 0
	mov	rdi, rsi
	cmovg	rdi, rdx
	cmove	rdi, rcx
	mov	qword ptr [r13 + 8*rax], rdi
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_719
	jmp	.LBB0_39
.LBB0_495:
	cmp	rbx, 4
	jae	.LBB0_631
# %bb.496:
	xor	ecx, ecx
	jmp	.LBB0_723
.LBB0_497:
	test	rcx, rcx
	je	.LBB0_679
# %bb.498:
	neg	rcx
	xor	esi, esi
.LBB0_499:                              # =>This Inner Loop Header: Depth=1
	mov	rax, qword ptr [rbp + 8*rsi]
	mov	rdx, rax
	or	rdx, rcx
	shr	rdx, 32
	je	.LBB0_501
# %bb.500:                              #   in Loop: Header=BB0_499 Depth=1
	cqo
	idiv	rcx
	jmp	.LBB0_502
.LBB0_501:                              #   in Loop: Header=BB0_499 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	ecx
                                        # kill: def %eax killed %eax def %rax
.LBB0_502:                              #   in Loop: Header=BB0_499 Depth=1
	mov	qword ptr [r13 + 8*rsi], rax
	add	rsi, 1
	cmp	rbx, rsi
	jne	.LBB0_499
	jmp	.LBB0_39
.LBB0_503:
	cmp	rsi, 4
	jb	.LBB0_504
# %bb.634:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_724
# %bb.635:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_724
.LBB0_504:
	xor	eax, eax
.LBB0_727:
	vxorpd	xmm0, xmm0, xmm0
	vmovsd	xmm1, qword ptr [rip + .LCPI0_0] # xmm1 = mem[0],zero
.LBB0_728:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm2, qword ptr [r15 + 8*rax] # xmm2 = mem[0],zero
	xor	ecx, ecx
	vucomisd	xmm2, xmm0
	seta	cl
	vmovsd	xmm3, qword ptr [8*rcx + .LCPI0_1] # xmm3 = mem[0],zero
	vcmpneqsd	xmm2, xmm2, xmm0
	vandpd	xmm3, xmm2, xmm3
	vandnpd	xmm2, xmm2, xmm1
	vorpd	xmm2, xmm2, xmm3
	vmovsd	qword ptr [r13 + 8*rax], xmm2
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_728
	jmp	.LBB0_39
.LBB0_505:
	cmp	rsi, 4
	jb	.LBB0_506
# %bb.637:
	lea	rax, [8*rsi]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_729
# %bb.638:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_729
.LBB0_506:
	xor	eax, eax
.LBB0_732:
	vmovsd	xmm1, qword ptr [rip + .LCPI0_0] # xmm1 = mem[0],zero
	vxorpd	xmm2, xmm2, xmm2
.LBB0_733:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm4, qword ptr [rbp + 8*rax] # xmm4 = mem[0],zero
	vucomisd	xmm4, xmm2
	vmovapd	xmm3, xmm1
	jne	.LBB0_734
	jnp	.LBB0_735
.LBB0_734:                              #   in Loop: Header=BB0_733 Depth=1
	vdivsd	xmm3, xmm0, xmm4
.LBB0_735:                              #   in Loop: Header=BB0_733 Depth=1
	vmovsd	qword ptr [r13 + 8*rax], xmm3
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_733
	jmp	.LBB0_39
.LBB0_507:
	lea	rax, [8*rbx]
	add	rax, rbp
	cmp	r13, rax
	jae	.LBB0_640
# %bb.508:
	lea	rax, [8*rbx]
	add	rax, r13
	cmp	rbp, rax
	jae	.LBB0_640
# %bb.509:
	xor	edi, edi
	mov	rsi, qword ptr [rbp + 8*rdi]
	mov	rax, r8
	test	rsi, rsi
	jne	.LBB0_664
	jmp	.LBB0_667
.LBB0_510:
	lea	rax, [r15 + 8*rsi]
	cmp	r13, rax
	jae	.LBB0_668
# %bb.511:
	lea	rax, [8*rsi]
	add	rax, r13
	cmp	r15, rax
	jae	.LBB0_668
# %bb.512:
	xor	eax, eax
	jmp	.LBB0_671
.LBB0_516:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_517:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_517
	jmp	.LBB0_741
.LBB0_530:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_531:                              # =>This Inner Loop Header: Depth=1
	vpaddq	ymm1, ymm0, ymmword ptr [r15 + 8*rdx]
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_531
	jmp	.LBB0_759
.LBB0_532:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_533:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [r15 + 8*rdx]
	vpsubq	ymm1, ymm1, ymm0
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_533
	jmp	.LBB0_761
.LBB0_534:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
	vpsrlq	ymm1, ymm0, 32
.LBB0_535:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm2, ymmword ptr [r15 + 8*rdx]
	vpmuludq	ymm3, ymm2, ymm1
	vpsrlq	ymm4, ymm2, 32
	vpmuludq	ymm4, ymm4, ymm0
	vpaddq	ymm3, ymm3, ymm4
	vpsllq	ymm3, ymm3, 32
	vpmuludq	ymm2, ymm2, ymm0
	vpaddq	ymm2, ymm2, ymm3
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm2
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_535
	jmp	.LBB0_763
.LBB0_518:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
.LBB0_519:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_519
# %bb.520:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_521:                              # =>This Inner Loop Header: Depth=1
	mov	rcx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_521
	jmp	.LBB0_39
.LBB0_522:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_523:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_523
# %bb.524:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_525:                              # =>This Inner Loop Header: Depth=1
	mov	rcx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_525
	jmp	.LBB0_39
.LBB0_526:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
.LBB0_527:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm0, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm0
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_527
# %bb.528:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_529:                              # =>This Inner Loop Header: Depth=1
	mov	rcx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_529
	jmp	.LBB0_39
.LBB0_536:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_537:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [r15 + 8*rdx]
	vpcmpgtq	ymm2, ymm0, ymm1
	vblendvpd	ymm1, ymm0, ymm1, ymm2
	vmovupd	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_537
# %bb.538:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_539:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [r15 + 8*rax]
	cmp	rdx, rcx
	cmovg	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_539
	jmp	.LBB0_39
.LBB0_540:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_541:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [r15 + 8*rdx]
	vpcmpgtq	ymm2, ymm1, ymm0
	vblendvpd	ymm1, ymm0, ymm1, ymm2
	vmovupd	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_541
# %bb.542:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_543:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [r15 + 8*rax]
	cmp	rdx, rcx
	cmovl	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_543
	jmp	.LBB0_39
.LBB0_544:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_545:                              # =>This Inner Loop Header: Depth=1
	vpcmpgtq	ymm1, ymm0, ymmword ptr [r15 + 8*rdx]
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_545
# %bb.546:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_547:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	qword ptr [r15 + 8*rax], rcx
	setl	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_547
	jmp	.LBB0_39
.LBB0_548:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_549:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [r15 + 8*rdx]
	vpcmpgtq	ymm1, ymm1, ymm0
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_549
# %bb.550:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_551:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	qword ptr [r15 + 8*rax], rcx
	setg	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_551
	jmp	.LBB0_39
.LBB0_552:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_553:                              # =>This Inner Loop Header: Depth=1
	vpcmpeqq	ymm1, ymm0, ymmword ptr [r15 + 8*rdx]
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_553
# %bb.554:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_555:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	qword ptr [r15 + 8*rax], rcx
	sete	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_555
	jmp	.LBB0_39
.LBB0_556:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_557:                              # =>This Inner Loop Header: Depth=1
	vpaddq	ymm1, ymm0, ymmword ptr [rbp + 8*rdx]
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_557
# %bb.558:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_559:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [rbp + 8*rax]
	add	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_559
	jmp	.LBB0_39
.LBB0_560:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_561:                              # =>This Inner Loop Header: Depth=1
	vpsubq	ymm1, ymm0, ymmword ptr [rbp + 8*rdx]
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_561
# %bb.562:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_563:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, rcx
	sub	rdx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_563
	jmp	.LBB0_39
.LBB0_564:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
	vpsrlq	ymm1, ymm0, 32
.LBB0_565:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm2, ymmword ptr [rbp + 8*rdx]
	vpmuludq	ymm3, ymm2, ymm1
	vpsrlq	ymm4, ymm2, 32
	vpmuludq	ymm4, ymm4, ymm0
	vpaddq	ymm3, ymm3, ymm4
	vpsllq	ymm3, ymm3, 32
	vpmuludq	ymm2, ymm2, ymm0
	vpaddq	ymm2, ymm2, ymm3
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm2
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_565
# %bb.566:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_567:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [rbp + 8*rax]
	imul	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_567
	jmp	.LBB0_39
.LBB0_568:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_569:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rdx]
	vpcmpgtq	ymm2, ymm1, ymm0
	vblendvpd	ymm1, ymm1, ymm0, ymm2
	vmovupd	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_569
# %bb.570:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_571:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [rbp + 8*rax]
	cmp	rcx, rdx
	cmovle	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_571
	jmp	.LBB0_39
.LBB0_572:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_573:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rdx]
	vpcmpgtq	ymm2, ymm0, ymm1
	vblendvpd	ymm1, ymm1, ymm0, ymm2
	vmovupd	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_573
# %bb.574:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_575:                              # =>This Inner Loop Header: Depth=1
	mov	rdx, qword ptr [rbp + 8*rax]
	cmp	rcx, rdx
	cmovge	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_575
	jmp	.LBB0_39
.LBB0_576:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_577:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm1, ymmword ptr [rbp + 8*rdx]
	vpcmpgtq	ymm1, ymm1, ymm0
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_577
# %bb.578:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_579:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	setl	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_579
	jmp	.LBB0_39
.LBB0_580:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_581:                              # =>This Inner Loop Header: Depth=1
	vpcmpgtq	ymm1, ymm0, ymmword ptr [rbp + 8*rdx]
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_581
# %bb.582:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_583:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	setg	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_583
	jmp	.LBB0_39
.LBB0_584:
	mov	rax, rbx
	and	rax, -4
	vmovq	xmm0, rcx
	vpbroadcastq	ymm0, xmm0
	xor	edx, edx
.LBB0_585:                              # =>This Inner Loop Header: Depth=1
	vpcmpeqq	ymm1, ymm0, ymmword ptr [rbp + 8*rdx]
	vpsrlq	ymm1, ymm1, 63
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm1
	add	rdx, 4
	cmp	rax, rdx
	jne	.LBB0_585
# %bb.586:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_587:                              # =>This Inner Loop Header: Depth=1
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	sete	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	cmp	rbx, rax
	jne	.LBB0_587
	jmp	.LBB0_39
.LBB0_588:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_589:                              # =>This Inner Loop Header: Depth=1
	vaddpd	ymm2, ymm1, ymmword ptr [r15 + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_589
# %bb.590:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_591:                              # =>This Inner Loop Header: Depth=1
	vaddsd	xmm1, xmm0, qword ptr [r15 + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_591
	jmp	.LBB0_39
.LBB0_592:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_593:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm2, ymmword ptr [r15 + 8*rcx]
	vsubpd	ymm2, ymm2, ymm1
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_593
# %bb.594:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_595:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [r15 + 8*rax] # xmm1 = mem[0],zero
	vsubsd	xmm1, xmm1, xmm0
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_595
	jmp	.LBB0_39
.LBB0_596:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_597:                              # =>This Inner Loop Header: Depth=1
	vmulpd	ymm2, ymm1, ymmword ptr [r15 + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_597
# %bb.598:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_599:                              # =>This Inner Loop Header: Depth=1
	vmulsd	xmm1, xmm0, qword ptr [r15 + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_599
	jmp	.LBB0_39
.LBB0_600:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_601:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm2, ymmword ptr [r15 + 8*rcx]
	vminpd	ymm2, ymm2, ymm1
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_601
# %bb.602:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_603:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [r15 + 8*rax] # xmm1 = mem[0],zero
	vminsd	xmm1, xmm1, xmm0
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_603
	jmp	.LBB0_39
.LBB0_604:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_605:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm2, ymmword ptr [r15 + 8*rcx]
	vmaxpd	ymm2, ymm2, ymm1
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_605
# %bb.606:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_607:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [r15 + 8*rax] # xmm1 = mem[0],zero
	vmaxsd	xmm1, xmm1, xmm0
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_607
	jmp	.LBB0_39
.LBB0_608:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_609:                              # =>This Inner Loop Header: Depth=1
	vaddpd	ymm2, ymm1, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_609
# %bb.610:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_611:                              # =>This Inner Loop Header: Depth=1
	vaddsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_611
	jmp	.LBB0_39
.LBB0_612:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_613:                              # =>This Inner Loop Header: Depth=1
	vsubpd	ymm2, ymm1, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_613
# %bb.614:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_615:                              # =>This Inner Loop Header: Depth=1
	vsubsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_615
	jmp	.LBB0_39
.LBB0_616:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_617:                              # =>This Inner Loop Header: Depth=1
	vmulpd	ymm2, ymm1, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_617
# %bb.618:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_619:                              # =>This Inner Loop Header: Depth=1
	vmulsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_619
	jmp	.LBB0_39
.LBB0_620:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_621:                              # =>This Inner Loop Header: Depth=1
	vminpd	ymm2, ymm1, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_621
# %bb.622:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_623:                              # =>This Inner Loop Header: Depth=1
	vminsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_623
	jmp	.LBB0_39
.LBB0_624:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_625:                              # =>This Inner Loop Header: Depth=1
	vmaxpd	ymm2, ymm1, ymmword ptr [rbp + 8*rcx]
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_625
# %bb.626:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_627:                              # =>This Inner Loop Header: Depth=1
	vmaxsd	xmm1, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_627
	jmp	.LBB0_39
.LBB0_631:
	lea	rcx, [8*rbx]
	add	rcx, rbp
	cmp	r13, rcx
	jae	.LBB0_720
# %bb.632:
	lea	rcx, [8*rbx]
	add	rcx, r13
	cmp	rbp, rcx
	jae	.LBB0_720
# %bb.633:
	xor	ecx, ecx
	jmp	.LBB0_723
.LBB0_672:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm2, xmm0
	vbroadcastsd	ymm3, xmm1
	xor	ecx, ecx
	vxorpd	xmm4, xmm4, xmm4
.LBB0_673:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm5, ymmword ptr [rbp + 8*rcx]
	vcmpneqpd	ymm6, ymm5, ymm4
	vdivpd	ymm5, ymm2, ymm5
	vblendvpd	ymm5, ymm3, ymm5, ymm6
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm5
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_673
# %bb.674:
	cmp	rsi, rax
	je	.LBB0_39
	jmp	.LBB0_675
.LBB0_640:
	mov	rdi, rbx
	and	rdi, -4
	vmovq	xmm0, r8
	vpbroadcastq	ymm0, xmm0
	xor	esi, esi
	vpxor	xmm1, xmm1, xmm1
	vpcmpeqd	ymm2, ymm2, ymm2
.LBB0_641:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm4, ymmword ptr [rbp + 8*rsi]
	vpcmpeqq	ymm3, ymm4, ymm1
	vpxor	ymm5, ymm3, ymm2
	vextracti128	xmm6, ymm5, 1
	vpackssdw	xmm6, xmm5, xmm6
	vpextrb	eax, xmm5, 0
                                        # implicit-def: %ymm5
	test	al, 1
	je	.LBB0_642
# %bb.645:                              #   in Loop: Header=BB0_641 Depth=1
	vmovq	rbx, xmm4
	mov	rax, rcx
	or	rax, rbx
	shr	rax, 32
	je	.LBB0_647
# %bb.646:                              #   in Loop: Header=BB0_641 Depth=1
	mov	rax, rcx
	cqo
	idiv	rbx
	jmp	.LBB0_648
.LBB0_642:                              #   in Loop: Header=BB0_641 Depth=1
	vpextrb	eax, xmm6, 4
	test	al, 1
	je	.LBB0_643
.LBB0_649:                              #   in Loop: Header=BB0_641 Depth=1
	vpextrq	rbx, xmm4, 1
	mov	rax, rcx
	or	rax, rbx
	shr	rax, 32
	je	.LBB0_651
# %bb.650:                              #   in Loop: Header=BB0_641 Depth=1
	mov	rax, rcx
	cqo
	idiv	rbx
	jmp	.LBB0_652
.LBB0_647:                              #   in Loop: Header=BB0_641 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	ebx
                                        # kill: def %eax killed %eax def %rax
.LBB0_648:                              #   in Loop: Header=BB0_641 Depth=1
	vmovq	xmm5, rax
	mov	rbx, r9
	vpextrb	eax, xmm6, 4
	test	al, 1
	jne	.LBB0_649
.LBB0_643:                              #   in Loop: Header=BB0_641 Depth=1
	vpextrb	eax, xmm6, 8
	test	al, 1
	je	.LBB0_644
.LBB0_653:                              #   in Loop: Header=BB0_641 Depth=1
	vextracti128	xmm7, ymm4, 1
	vmovq	rbx, xmm7
	mov	rax, rcx
	or	rax, rbx
	shr	rax, 32
	je	.LBB0_655
# %bb.654:                              #   in Loop: Header=BB0_641 Depth=1
	mov	rax, rcx
	cqo
	idiv	rbx
	jmp	.LBB0_656
.LBB0_651:                              #   in Loop: Header=BB0_641 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	ebx
                                        # kill: def %eax killed %eax def %rax
.LBB0_652:                              #   in Loop: Header=BB0_641 Depth=1
	vpinsrq	xmm7, xmm5, rax, 1
	vpblendd	ymm5, ymm5, ymm7, 15 # ymm5 = ymm7[0,1,2,3],ymm5[4,5,6,7]
	mov	rbx, r9
	vpextrb	eax, xmm6, 8
	test	al, 1
	jne	.LBB0_653
.LBB0_644:                              #   in Loop: Header=BB0_641 Depth=1
	vpextrb	eax, xmm6, 12
	test	al, 1
	jne	.LBB0_657
	jmp	.LBB0_661
.LBB0_655:                              #   in Loop: Header=BB0_641 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	ebx
                                        # kill: def %eax killed %eax def %rax
.LBB0_656:                              #   in Loop: Header=BB0_641 Depth=1
	vextracti128	xmm7, ymm5, 1
	vpinsrq	xmm7, xmm7, rax, 0
	vinserti128	ymm5, ymm5, xmm7, 1
	mov	rbx, r9
	vpextrb	eax, xmm6, 12
	test	al, 1
	je	.LBB0_661
.LBB0_657:                              #   in Loop: Header=BB0_641 Depth=1
	vextracti128	xmm4, ymm4, 1
	vpextrq	rbx, xmm4, 1
	mov	rax, rcx
	or	rax, rbx
	shr	rax, 32
	je	.LBB0_659
# %bb.658:                              #   in Loop: Header=BB0_641 Depth=1
	mov	rax, rcx
	cqo
	idiv	rbx
	jmp	.LBB0_660
.LBB0_659:                              #   in Loop: Header=BB0_641 Depth=1
	xor	edx, edx
	mov	eax, ecx
	div	ebx
                                        # kill: def %eax killed %eax def %rax
.LBB0_660:                              #   in Loop: Header=BB0_641 Depth=1
	vextracti128	xmm4, ymm5, 1
	vpinsrq	xmm4, xmm4, rax, 1
	vinserti128	ymm5, ymm5, xmm4, 1
	mov	rbx, r9
.LBB0_661:                              #   in Loop: Header=BB0_641 Depth=1
	vblendvpd	ymm3, ymm5, ymm0, ymm3
	vmovupd	ymmword ptr [r13 + 8*rsi], ymm3
	add	rsi, 4
	cmp	rdi, rsi
	jne	.LBB0_641
# %bb.662:
	cmp	rbx, rdi
	je	.LBB0_39
.LBB0_663:
	mov	rsi, qword ptr [rbp + 8*rdi]
	mov	rax, r8
	test	rsi, rsi
	je	.LBB0_667
.LBB0_664:
	mov	rax, rcx
	or	rax, rsi
	shr	rax, 32
	je	.LBB0_666
# %bb.665:
	mov	rax, rcx
	cqo
	idiv	rsi
	jmp	.LBB0_667
.LBB0_666:
	xor	edx, edx
	mov	eax, ecx
	div	esi
                                        # kill: def %eax killed %eax def %rax
.LBB0_667:
	mov	qword ptr [r13 + 8*rdi], rax
	add	rdi, 1
	cmp	rbx, rdi
	jne	.LBB0_663
	jmp	.LBB0_39
.LBB0_668:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
.LBB0_669:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm2, ymmword ptr [r15 + 8*rcx]
	vdivpd	ymm2, ymm2, ymm1
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm2
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_669
# %bb.670:
	cmp	rsi, rax
	je	.LBB0_39
.LBB0_671:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm1, qword ptr [r15 + 8*rax] # xmm1 = mem[0],zero
	vdivsd	xmm1, xmm1, xmm0
	vmovsd	qword ptr [r13 + 8*rax], xmm1
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_671
	jmp	.LBB0_39
.LBB0_679:
	cmp	rbx, 4
	jae	.LBB0_736
# %bb.680:
	xor	eax, eax
	jmp	.LBB0_739
.LBB0_681:
	mov	rbx, r9
	jmp	.LBB0_744
.LBB0_682:
	mov	rbx, r9
	jmp	.LBB0_746
.LBB0_683:
	mov	rbx, r9
	jmp	.LBB0_748
.LBB0_684:
	mov	rbx, r9
.LBB0_685:
	movabs	r8, -9223372036854775808
	lea	r9, [r8 - 1]
	lea	rdi, [r8 + 1]
.LBB0_686:                              # =>This Inner Loop Header: Depth=1
	mov	rax, qword ptr [r15 + 8*rsi]
	mov	rcx, qword ptr [rbp + 8*rsi]
	test	rcx, rcx
	je	.LBB0_689
# %bb.687:                              #   in Loop: Header=BB0_686 Depth=1
	mov	rdx, rax
	or	rdx, rcx
	shr	rdx, 32
	je	.LBB0_691
# %bb.688:                              #   in Loop: Header=BB0_686 Depth=1
	cqo
	idiv	rcx
	mov	rcx, rax
	jmp	.LBB0_692
.LBB0_689:                              #   in Loop: Header=BB0_686 Depth=1
	mov	rcx, r8
	test	rax, rax
	je	.LBB0_692
# %bb.690:                              #   in Loop: Header=BB0_686 Depth=1
	mov	rcx, rdi
	cmovg	rcx, r9
	jmp	.LBB0_692
.LBB0_691:                              #   in Loop: Header=BB0_686 Depth=1
	xor	edx, edx
                                        # kill: def %eax killed %eax killed %rax
	div	ecx
	mov	ecx, eax
.LBB0_692:                              #   in Loop: Header=BB0_686 Depth=1
	mov	qword ptr [r13 + 8*rsi], rcx
	add	rsi, 1
	cmp	rbx, rsi
	jne	.LBB0_686
	jmp	.LBB0_39
.LBB0_693:
	mov	rbx, r9
	jmp	.LBB0_750
.LBB0_694:
	mov	rbx, r9
	jmp	.LBB0_752
.LBB0_695:
	mov	rbx, r9
	jmp	.LBB0_754
.LBB0_696:
	mov	rbx, r9
	jmp	.LBB0_756
.LBB0_697:
	mov	rbx, r9
	jmp	.LBB0_758
.LBB0_698:
	mov	rsi, r9
.LBB0_699:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	vaddsd	xmm0, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm0
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_699
	jmp	.LBB0_39
.LBB0_700:
	mov	rsi, r9
.LBB0_701:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	vsubsd	xmm0, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm0
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_701
	jmp	.LBB0_39
.LBB0_702:
	mov	rsi, r9
.LBB0_703:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	vmulsd	xmm0, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm0
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_703
	jmp	.LBB0_39
.LBB0_704:
	mov	rsi, r9
.LBB0_705:
	vxorpd	xmm0, xmm0, xmm0
	vmovsd	xmm1, qword ptr [rip + .LCPI0_0] # xmm1 = mem[0],zero
.LBB0_706:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm2, qword ptr [r15 + 8*rax] # xmm2 = mem[0],zero
	vmovsd	xmm3, qword ptr [rbp + 8*rax] # xmm3 = mem[0],zero
	vucomisd	xmm3, xmm0
	jne	.LBB0_707
	jnp	.LBB0_708
.LBB0_707:                              #   in Loop: Header=BB0_706 Depth=1
	vdivsd	xmm2, xmm2, xmm3
	jmp	.LBB0_710
.LBB0_708:                              #   in Loop: Header=BB0_706 Depth=1
	vucomisd	xmm2, xmm0
	vmovapd	xmm2, xmm1
	jne	.LBB0_709
	jnp	.LBB0_710
.LBB0_709:                              #   in Loop: Header=BB0_706 Depth=1
	seta	cl
	movzx	ecx, cl
	vmovsd	xmm2, qword ptr [8*rcx + .LCPI0_1] # xmm2 = mem[0],zero
.LBB0_710:                              #   in Loop: Header=BB0_706 Depth=1
	vmovsd	qword ptr [r13 + 8*rax], xmm2
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_706
	jmp	.LBB0_39
.LBB0_711:
	mov	rsi, r9
.LBB0_712:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	vminsd	xmm0, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm0
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_712
	jmp	.LBB0_39
.LBB0_713:
	mov	rsi, r9
.LBB0_714:                              # =>This Inner Loop Header: Depth=1
	vmovsd	xmm0, qword ptr [r15 + 8*rax] # xmm0 = mem[0],zero
	vmaxsd	xmm0, xmm0, qword ptr [rbp + 8*rax]
	vmovsd	qword ptr [r13 + 8*rax], xmm0
	add	rax, 1
	cmp	rsi, rax
	jne	.LBB0_714
	jmp	.LBB0_39
.LBB0_715:
	mov	rax, rbx
	and	rax, -4
	xor	ecx, ecx
	vpxor	xmm0, xmm0, xmm0
	vbroadcastsd	ymm1, qword ptr [rip + .LCPI0_5] # ymm1 = [9223372036854775807,9223372036854775807,9223372036854775807,9223372036854775807]
	vbroadcastsd	ymm2, qword ptr [rip + .LCPI0_6] # ymm2 = [9223372036854775809,9223372036854775809,9223372036854775809,9223372036854775809]
	vbroadcastsd	ymm3, qword ptr [rip + .LCPI0_4] # ymm3 = [9223372036854775808,9223372036854775808,9223372036854775808,9223372036854775808]
.LBB0_716:                              # =>This Inner Loop Header: Depth=1
	vmovdqu	ymm4, ymmword ptr [r15 + 8*rcx]
	vpcmpeqq	ymm5, ymm4, ymm0
	vpcmpgtq	ymm4, ymm4, ymm0
	vblendvpd	ymm4, ymm2, ymm1, ymm4
	vblendvpd	ymm4, ymm4, ymm3, ymm5
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm4
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_716
# %bb.717:
	mov	rbx, r9
	cmp	rbx, rax
	je	.LBB0_39
	jmp	.LBB0_718
.LBB0_724:
	mov	rax, rsi
	and	rax, -4
	xor	ecx, ecx
	vxorpd	xmm0, xmm0, xmm0
	vbroadcastsd	ymm1, qword ptr [rip + .LCPI0_2] # ymm1 = [+Inf,+Inf,+Inf,+Inf]
	vbroadcastsd	ymm2, qword ptr [rip + .LCPI0_3] # ymm2 = [-Inf,-Inf,-Inf,-Inf]
	vbroadcastsd	ymm3, qword ptr [rip + .LCPI0_0] # ymm3 = [NaN,NaN,NaN,NaN]
.LBB0_725:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm4, ymmword ptr [r15 + 8*rcx]
	vcmpneqpd	ymm5, ymm4, ymm0
	vcmpltpd	ymm4, ymm0, ymm4
	vblendvpd	ymm4, ymm2, ymm1, ymm4
	vblendvpd	ymm4, ymm3, ymm4, ymm5
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm4
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_725
# %bb.726:
	mov	rsi, r9
	cmp	rsi, rax
	je	.LBB0_39
	jmp	.LBB0_727
.LBB0_729:
	mov	rax, rsi
	and	rax, -4
	vbroadcastsd	ymm1, xmm0
	xor	ecx, ecx
	vxorpd	xmm2, xmm2, xmm2
	vbroadcastsd	ymm3, qword ptr [rip + .LCPI0_0] # ymm3 = [NaN,NaN,NaN,NaN]
.LBB0_730:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymm4, ymmword ptr [rbp + 8*rcx]
	vcmpneqpd	ymm5, ymm4, ymm2
	vdivpd	ymm4, ymm1, ymm4
	vblendvpd	ymm4, ymm3, ymm4, ymm5
	vmovupd	ymmword ptr [r13 + 8*rcx], ymm4
	add	rcx, 4
	cmp	rax, rcx
	jne	.LBB0_730
# %bb.731:
	mov	rsi, r9
	cmp	rsi, rax
	je	.LBB0_39
	jmp	.LBB0_732
.LBB0_720:
	mov	rcx, rbx
	and	rcx, -4
	xor	edx, edx
	vpxor	xmm0, xmm0, xmm0
	vpbroadcastq	ymm1, qword ptr [rip + .LCPI0_4] # ymm1 = [9223372036854775808,9223372036854775808,9223372036854775808,9223372036854775808]
.LBB0_721:                              # =>This Inner Loop Header: Depth=1
	vpcmpeqq	ymm2, ymm0, ymmword ptr [rbp + 8*rdx]
	vpand	ymm2, ymm2, ymm1
	vmovdqu	ymmword ptr [r13 + 8*rdx], ymm2
	add	rdx, 4
	cmp	rcx, rdx
	jne	.LBB0_721
# %bb.722:
	mov	rbx, r9
	cmp	rbx, rcx
	je	.LBB0_39
.LBB0_723:                              # =>This Inner Loop Header: Depth=1
	cmp	qword ptr [rbp + 8*rcx], 0
	mov	edx, 0
	cmove	rdx, rax
	mov	qword ptr [r13 + 8*rcx], rdx
	add	rcx, 1
	cmp	rbx, rcx
	jne	.LBB0_723
	jmp	.LBB0_39
.LBB0_736:
	mov	rax, rbx
	and	rax, -4
	vbroadcastsd	ymm0, qword ptr [rip + .LCPI0_4] # ymm0 = [9223372036854775808,9223372036854775808,9223372036854775808,9223372036854775808]
	mov	rcx, rax
	mov	rdx, r13
.LBB0_737:                              # =>This Inner Loop Header: Depth=1
	vmovupd	ymmword ptr [rdx], ymm0
	add	rdx, 32
	add	rcx, -4
	jne	.LBB0_737
# %bb.738:
	mov	rbx, r9
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_739:
	movabs	rcx, -9223372036854775808
	sub	rbx, rax
	lea	rax, [8*rax]
	add	rax, r13
.LBB0_740:                              # =>This Inner Loop Header: Depth=1
	mov	qword ptr [rax], rcx
	add	rax, 8
	add	rbx, -1
	jne	.LBB0_740
	jmp	.LBB0_39
.LBB0_742:
	mov	rcx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
.LBB0_741:
	cmp	rbx, rax
	jne	.LBB0_742
	jmp	.LBB0_39
.LBB0_760:
	mov	rdx, qword ptr [r15 + 8*rax]
	add	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
.LBB0_759:
	cmp	rbx, rax
	jne	.LBB0_760
	jmp	.LBB0_39
.LBB0_762:
	mov	rdx, qword ptr [r15 + 8*rax]
	sub	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
.LBB0_761:
	cmp	rbx, rax
	jne	.LBB0_762
	jmp	.LBB0_39
.LBB0_764:
	mov	rdx, qword ptr [r15 + 8*rax]
	imul	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
.LBB0_763:
	cmp	rbx, rax
	jne	.LBB0_764
	jmp	.LBB0_39
.LBB0_743:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_744:
	mov	rcx, qword ptr [rbp + 8*rax]
	add	rcx, qword ptr [r15 + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	jmp	.LBB0_743
.LBB0_745:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_746:
	mov	rcx, qword ptr [r15 + 8*rax]
	sub	rcx, qword ptr [rbp + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	jmp	.LBB0_745
.LBB0_747:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_748:
	mov	rcx, qword ptr [rbp + 8*rax]
	imul	rcx, qword ptr [r15 + 8*rax]
	mov	qword ptr [r13 + 8*rax], rcx
	add	rax, 1
	jmp	.LBB0_747
.LBB0_749:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_750:
	mov	rcx, qword ptr [r15 + 8*rax]
	mov	rdx, qword ptr [rbp + 8*rax]
	cmp	rcx, rdx
	cmovle	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	jmp	.LBB0_749
.LBB0_751:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_752:
	mov	rcx, qword ptr [r15 + 8*rax]
	mov	rdx, qword ptr [rbp + 8*rax]
	cmp	rcx, rdx
	cmovge	rdx, rcx
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	jmp	.LBB0_751
.LBB0_753:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_754:
	mov	rcx, qword ptr [r15 + 8*rax]
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	setl	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	jmp	.LBB0_753
.LBB0_755:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_756:
	mov	rcx, qword ptr [r15 + 8*rax]
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	setg	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	jmp	.LBB0_755
.LBB0_757:
	cmp	rbx, rax
	je	.LBB0_39
.LBB0_758:
	mov	rcx, qword ptr [r15 + 8*rax]
	xor	edx, edx
	cmp	rcx, qword ptr [rbp + 8*rax]
	sete	dl
	mov	qword ptr [r13 + 8*rax], rdx
	add	rax, 1
	jmp	.LBB0_757
.Lfunc_end0:
	.size	pen2, .Lfunc_end0-pen2