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
159
|
dnl AMD64 mpn_add_n, mpn_sub_n optimised for bobcat.
dnl Copyright 2003-2005, 2007, 2008, 2010-2013 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/.
include(`../config.m4')
C cycles/limb
C AMD K8,K9 1.77
C AMD K10 1.76\1.82
C AMD bd1 1.67\2.12
C AMD bd2 1.62\1.82
C AMD bd3
C AMD bd4 1.55\2.2
C AMD zen
C AMD bt1 2.54
C AMD bt2 2
C Intel P4 11
C Intel PNR 4.76
C Intel NHM 5.27
C Intel SBR 2
C Intel IBR 1.94
C Intel HWL 1.63
C Intel BWL 1.51
C Intel SKL 1.51
C Intel atom 3.56
C Intel SLM 4
C VIA nano
C The loop of this code is the result of running a code generation and
C optimization tool suite written by David Harvey and Torbjorn Granlund.
C INPUT PARAMETERS
define(`rp', `%rdi') C rcx
define(`up', `%rsi') C rdx
define(`vp', `%rdx') C r8
define(`n', `%rcx') C r9
define(`cy', `%r8') C rsp+40 (mpn_add_nc and mpn_sub_nc)
ifdef(`OPERATION_add_n', `
define(ADCSBB, adc)
define(func, mpn_add_n)
define(func_nc, mpn_add_nc)')
ifdef(`OPERATION_sub_n', `
define(ADCSBB, sbb)
define(func, mpn_sub_n)
define(func_nc, mpn_sub_nc)')
MULFUNC_PROLOGUE(mpn_add_n mpn_add_nc mpn_sub_n mpn_sub_nc)
ABI_SUPPORT(DOS64)
ABI_SUPPORT(STD64)
ASM_START()
TEXT
ALIGN(16)
PROLOGUE(func)
FUNC_ENTRY(4)
xor %r8, %r8
L(ent): test $1, R8(n)
jnz L(bx1)
L(bx0): test $2, R8(n)
jnz L(b10)
L(b00): shr $2, n
neg %r8
mov $3, R32(%rax)
mov (up), %r10
mov 8(up), %r11
jmp L(lo0)
L(b10): shr $2, n
neg %r8
mov $1, R32(%rax)
mov (up), %r8
mov 8(up), %r9
jrcxz L(cj2)
jmp L(top)
L(bx1): test $2, R8(n)
jnz L(b11)
L(b01): shr $2, n
neg %r8
mov $0, R32(%rax)
mov (up), %r9
jrcxz L(cj1)
mov 8(up), %r10
jmp L(lo1)
ALIGN(8)
L(b11): inc n
shr $2, n
neg %r8
mov $2, R32(%rax)
mov (up), %r11
jmp L(lo3)
ALIGN(4)
L(top): mov 8(up,%rax,8), %r10
ADCSBB -8(vp,%rax,8), %r8
mov %r8, -8(rp,%rax,8)
L(lo1): mov 16(up,%rax,8), %r11
ADCSBB (vp,%rax,8), %r9
lea 4(%rax), %rax
mov %r9, -32(rp,%rax,8)
L(lo0): ADCSBB -24(vp,%rax,8), %r10
mov %r10, -24(rp,%rax,8)
L(lo3): ADCSBB -16(vp,%rax,8), %r11
dec n
mov -8(up,%rax,8), %r8
mov %r11, -16(rp,%rax,8)
L(lo2): mov (up,%rax,8), %r9
jnz L(top)
L(cj2): ADCSBB -8(vp,%rax,8), %r8
mov %r8, -8(rp,%rax,8)
L(cj1): ADCSBB (vp,%rax,8), %r9
mov %r9, (rp,%rax,8)
mov $0, R32(%rax)
adc $0, R32(%rax)
FUNC_EXIT()
ret
EPILOGUE()
ALIGN(16)
PROLOGUE(func_nc)
FUNC_ENTRY(4)
IFDOS(` mov 56(%rsp), %r8 ')
jmp L(ent)
EPILOGUE()
|
