1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
|
dnl AMD64 mpn_sublshC_n -- rp[] = up[] - (vp[] << C), optimised for Core 2 and
dnl Core iN.
dnl Contributed to the GNU project by Torbjorn Granlund.
dnl Copyright 2008, 2010-2012 Free Software Foundation, Inc.
dnl This file is part of the GNU MP Library.
dnl
dnl The GNU MP Library is free software; you can redistribute it and/or modify
dnl it under the terms of either:
dnl
dnl * the GNU Lesser General Public License as published by the Free
dnl Software Foundation; either version 3 of the License, or (at your
dnl option) any later version.
dnl
dnl or
dnl
dnl * the GNU General Public License as published by the Free Software
dnl Foundation; either version 2 of the License, or (at your option) any
dnl later version.
dnl
dnl or both in parallel, as here.
dnl
dnl The GNU MP Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
dnl for more details.
dnl
dnl You should have received copies of the GNU General Public License and the
dnl GNU Lesser General Public License along with the GNU MP Library. If not,
dnl see https://www.gnu.org/licenses/.
C cycles/limb
C AMD K8,K9 4.25
C AMD K10 ?
C Intel P4 ?
C Intel core2 3
C Intel NHM 3.1
C Intel SBR 2.47
C Intel atom ?
C VIA nano ?
C INPUT PARAMETERS
define(`rp',`%rdi')
define(`up',`%rsi')
define(`vp',`%rdx')
define(`n', `%rcx')
ASM_START()
TEXT
ALIGN(8)
PROLOGUE(func)
FUNC_ENTRY(4)
push %rbx
push %r12
mov R32(%rcx), R32(%rax)
lea 24(up,n,8), up
lea 24(vp,n,8), vp
lea 24(rp,n,8), rp
neg n
xor R32(%r11), R32(%r11)
mov -24(vp,n,8), %r8 C do first limb early
shrd $RSH, %r8, %r11
and $3, R32(%rax)
je L(b0)
cmp $2, R32(%rax)
jc L(b1)
je L(b2)
L(b3): mov -16(vp,n,8), %r9
shrd $RSH, %r9, %r8
mov -8(vp,n,8), %r10
shrd $RSH, %r10, %r9
mov -24(up,n,8), %r12
ADDSUB %r11, %r12
mov %r12, -24(rp,n,8)
mov -16(up,n,8), %r12
ADCSBB %r8, %r12
mov %r12, -16(rp,n,8)
mov -8(up,n,8), %r12
ADCSBB %r9, %r12
mov %r12, -8(rp,n,8)
mov %r10, %r11
sbb R32(%rax), R32(%rax) C save cy
add $3, n
js L(top)
jmp L(end)
L(b1): mov -24(up,n,8), %r12
ADDSUB %r11, %r12
mov %r12, -24(rp,n,8)
mov %r8, %r11
sbb R32(%rax), R32(%rax) C save cy
inc n
js L(top)
jmp L(end)
L(b2): mov -16(vp,n,8), %r9
shrd $RSH, %r9, %r8
mov -24(up,n,8), %r12
ADDSUB %r11, %r12
mov %r12, -24(rp,n,8)
mov -16(up,n,8), %r12
ADCSBB %r8, %r12
mov %r12, -16(rp,n,8)
mov %r9, %r11
sbb R32(%rax), R32(%rax) C save cy
add $2, n
js L(top)
jmp L(end)
ALIGN(16)
L(top): mov -24(vp,n,8), %r8
shrd $RSH, %r8, %r11
L(b0): mov -16(vp,n,8), %r9
shrd $RSH, %r9, %r8
mov -8(vp,n,8), %r10
shrd $RSH, %r10, %r9
mov (vp,n,8), %rbx
shrd $RSH, %rbx, %r10
add R32(%rax), R32(%rax) C restore cy
mov -24(up,n,8), %r12
ADCSBB %r11, %r12
mov %r12, -24(rp,n,8)
mov -16(up,n,8), %r12
ADCSBB %r8, %r12
mov %r12, -16(rp,n,8)
mov -8(up,n,8), %r12
ADCSBB %r9, %r12
mov %r12, -8(rp,n,8)
mov (up,n,8), %r12
ADCSBB %r10, %r12
mov %r12, (rp,n,8)
mov %rbx, %r11
sbb R32(%rax), R32(%rax) C save cy
add $4, n
js L(top)
L(end): shr $RSH, %r11
pop %r12
pop %rbx
sub R32(%r11), R32(%rax)
neg R32(%rax)
FUNC_EXIT()
ret
EPILOGUE()
|