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
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
|
dnl x86-64 mpn_div_qr_2u_pi1
dnl -- Divide an mpn number by an unnormalized 2-limb number,
dnl using a single-limb inverse and shifting the dividend on the fly.
dnl Copyright 2007, 2008, 2010, 2011 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 c/l
C INPUT PARAMETERS
define(`qp', `%rdi')
define(`rp', `%rsi')
define(`up_param', `%rdx')
define(`un_param', `%rcx') dnl %rcx needed for shift count
define(`d1', `%r8')
define(`d0', `%r9')
define(`shift_param', `FRAME+8(%rsp)')
define(`di_param', `FRAME+16(%rsp)')
define(`di', `%r10')
define(`up', `%r11')
define(`un', `%rbp')
define(`u2', `%rbx')
define(`u1', `%r12')
define(`u0', `%rsi') dnl Same as rp, which is saved and restored.
define(`t1', `%r13')
define(`t0', `%r14')
define(`md1', `%r15')
ASM_START()
TEXT
ALIGN(16)
deflit(`FRAME', 0)
PROLOGUE(mpn_div_qr_2u_pi1)
mov di_param, di
mov up_param, up
push %r15
push %r14
push %r13
push %r12
push %rbx
push %rbp
push rp
deflit(`FRAME', 56)
lea -2(un_param), un
mov d1, md1
neg md1
C int parameter, 32 bits only
movl shift_param, R32(%rcx)
C FIXME: Different code for SHLD_SLOW
xor R32(u2), R32(u2)
mov 8(up, un, 8), u1
shld %cl, u1, u2
C Remains to read (up, un, 8) and shift u1, u0
C udiv_qr_3by2 (qh,u2,u1,u2,u1,n0, d1,d0,di)
mov di, %rax
mul u2
mov (up, un, 8), u0
shld %cl, u0, u1
mov u1, t0
add %rax, t0 C q0 in t0
adc u2, %rdx
mov %rdx, t1 C q in t1
imul md1, %rdx
mov d0, %rax
lea (%rdx, u1), u2
mul t1
mov u0, u1
shl %cl, u1
sub d0, u1
sbb d1, u2
sub %rax, u1
sbb %rdx, u2
xor R32(%rax), R32(%rax)
xor R32(%rdx), R32(%rdx)
cmp t0, u2
cmovnc d0, %rax
cmovnc d1, %rdx
adc $0, t1
nop
add %rax, u1
adc %rdx, u2
cmp d1, u2
jae L(fix_qh)
L(bck_qh):
push t1 C push qh on stack
jmp L(next)
ALIGN(16)
L(loop):
C udiv_qr_3by2 (q,u2,u1,u2,u1,n0, d1,d0,di)
C Based on the optimized divrem_2.asm code.
mov di, %rax
mul u2
mov (up, un, 8), u0
xor R32(t1), R32(t1)
shld %cl, u0, t1
or t1, u1
mov u1, t0
add %rax, t0 C q0 in t0
adc u2, %rdx
mov %rdx, t1 C q in t1
imul md1, %rdx
mov d0, %rax
lea (%rdx, u1), u2
mul t1
mov u0, u1
shl %cl, u1
sub d0, u1
sbb d1, u2
sub %rax, u1
sbb %rdx, u2
xor R32(%rax), R32(%rax)
xor R32(%rdx), R32(%rdx)
cmp t0, u2
cmovnc d0, %rax
cmovnc d1, %rdx
adc $0, t1
nop
add %rax, u1
adc %rdx, u2
cmp d1, u2
jae L(fix)
L(bck):
mov t1, (qp, un, 8)
L(next):
sub $1, un
jnc L(loop)
L(end):
C qh on stack
pop %rax
pop rp
shrd %cl, u2, u1
shr %cl, u2
mov u2, 8(rp)
mov u1, (rp)
pop %rbp
pop %rbx
pop %r12
pop %r13
pop %r14
pop %r15
ret
L(fix): C Unlikely update. u2 >= d1
seta %dl
cmp d0, u1
setae %al
orb %dl, %al C "orb" form to placate Sun tools
je L(bck)
inc t1
sub d0, u1
sbb d1, u2
jmp L(bck)
C Duplicated, just jumping back to a different address.
L(fix_qh): C Unlikely update. u2 >= d1
seta %dl
cmp d0, u1
setae %al
orb %dl, %al C "orb" form to placate Sun tools
je L(bck_qh)
inc t1
sub d0, u1
sbb d1, u2
jmp L(bck_qh)
EPILOGUE()
|
