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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
|
dnl AMD64 mpn_bdiv_q_1, mpn_pi1_bdiv_q_1 -- Hensel division by 1-limb divisor,
dnl returning quotient only.
dnl Copyright 2001, 2002, 2004-2006, 2009, 2011, 2012, 2017 Free Software
dnl 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 norm/unorm
C AMD K8,K9 10 +
C AMD K10 10 +
C AMD bull 13.7 -
C AMD pile 13.7 +
C AMD steam
C AMD excavator
C AMD bobcat 15 -
C AMD jaguar 16 -
C Intel P4 33 =
C Intel core2 13.25 =
C Intel NHM 14 =
C Intel SBR 8.5 -
C Intel IBR 8.5 -
C Intel HWL 8 =
C Intel BWL 8 =
C Intel SKL 8 =
C Intel atom 42 --
C Intel SLM 20.4 --
C VIA nano
C INPUT PARAMETERS
define(`rp', `%rdi')
define(`up', `%rsi')
define(`n', `%rdx')
define(`d', `%rcx')
define(`di', `%r8') C just mpn_pi1_bdiv_q_1
define(`ncnt', `%r9') C just mpn_pi1_bdiv_q_1
ABI_SUPPORT(DOS64)
ABI_SUPPORT(STD64)
ASM_START()
TEXT
ALIGN(16)
PROLOGUE(mpn_bdiv_q_1)
FUNC_ENTRY(4)
push %rbx
mov %rcx, %rax
xor R32(%rcx), R32(%rcx) C ncnt count
mov %rdx, %r10
bt $0, R32(%rax)
jnc L(evn) C skip bsf unless divisor is even
L(odd): mov %rax, %rbx
shr R32(%rax)
and $127, R32(%rax) C d/2, 7 bits
LEA( binvert_limb_table, %rdx)
movzbl (%rdx,%rax), R32(%rax) C inv 8 bits
mov %rbx, %r11 C d without twos
lea (%rax,%rax), R32(%rdx) C 2*inv
imul R32(%rax), R32(%rax) C inv*inv
imul R32(%rbx), R32(%rax) C inv*inv*d
sub R32(%rax), R32(%rdx) C inv = 2*inv - inv*inv*d, 16 bits
lea (%rdx,%rdx), R32(%rax) C 2*inv
imul R32(%rdx), R32(%rdx) C inv*inv
imul R32(%rbx), R32(%rdx) C inv*inv*d
sub R32(%rdx), R32(%rax) C inv = 2*inv - inv*inv*d, 32 bits
lea (%rax,%rax), %r8 C 2*inv
imul %rax, %rax C inv*inv
imul %rbx, %rax C inv*inv*d
sub %rax, %r8 C inv = 2*inv - inv*inv*d, 64 bits
jmp L(pi1)
L(evn): bsf %rax, %rcx
shr R8(%rcx), %rax
jmp L(odd)
EPILOGUE()
PROLOGUE(mpn_pi1_bdiv_q_1)
FUNC_ENTRY(4)
IFDOS(` mov 56(%rsp), %r8 ')
IFDOS(` mov 64(%rsp), %r9 ')
push %rbx
mov %rcx, %r11 C d
mov %rdx, %r10 C n
mov %r9, %rcx C ncnt
L(pi1): mov (up), %rax C up[0]
dec %r10
jz L(one)
mov 8(up), %rdx C up[1]
lea (up,%r10,8), up C up end
lea (rp,%r10,8), rp C rp end
neg %r10 C -n
shrd R8(%rcx), %rdx, %rax
xor R32(%rbx), R32(%rbx)
jmp L(ent)
ALIGN(8)
L(top):
C rax q
C rbx carry bit, 0 or 1
C rcx ncnt
C rdx
C r10 counter, limbs, negative
C r11 d
mul %r11 C carry limb in rdx
mov (up,%r10,8), %rax
mov 8(up,%r10,8), %r9
shrd R8(%rcx), %r9, %rax
nop
sub %rbx, %rax C apply carry bit
setc R8(%rbx)
sub %rdx, %rax C apply carry limb
adc $0, R32(%rbx)
L(ent): imul %r8, %rax
mov %rax, (rp,%r10,8)
inc %r10
jnz L(top)
mul %r11 C carry limb in rdx
mov (up), %rax C up high limb
shr R8(%rcx), %rax
sub %rbx, %rax C apply carry bit
sub %rdx, %rax C apply carry limb
imul %r8, %rax
mov %rax, (rp)
pop %rbx
FUNC_EXIT()
ret
L(one): shr R8(%rcx), %rax
imul %r8, %rax
mov %rax, (rp)
pop %rbx
FUNC_EXIT()
ret
EPILOGUE()
|