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
|
dnl PowerPC-32 mpn_rshift -- Shift a number right.
dnl Copyright 1995, 1998, 2000, 2002-2005 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 603e: ?
C 604e: 3.0
C 75x (G3): 3.0
C 7400,7410 (G4): 3.0
C 7445,7455 (G4+): 2.5
C 7447,7457 (G4+): 2.25
C power4/ppc970: 2.5
C power5: 2.5
C INPUT PARAMETERS
C rp r3
C up r4
C n r5
C cnt r6
ASM_START()
PROLOGUE(mpn_rshift)
cmpwi cr0, r5, 30 C more than 30 limbs?
addi r7, r3, -4 C dst-4
bgt L(BIG) C branch if more than 12 limbs
mtctr r5 C copy size into CTR
subfic r8, r6, 32
lwz r11, 0(r4) C load first s1 limb
slw r3, r11, r8 C compute function return value
bdz L(end1)
L(oop): lwzu r10, 4(r4)
srw r9, r11, r6
slw r12, r10, r8
or r9, r9, r12
stwu r9, 4(r7)
bdz L(end2)
lwzu r11, 4(r4)
srw r9, r10, r6
slw r12, r11, r8
or r9, r9, r12
stwu r9, 4(r7)
bdnz L(oop)
L(end1):
srw r0, r11, r6
stw r0, 4(r7)
blr
L(end2):
srw r0, r10, r6
stw r0, 4(r7)
blr
L(BIG):
stwu r1, -48(r1)
stmw r24, 8(r1) C save registers we are supposed to preserve
lwz r9, 0(r4)
subfic r8, r6, 32
slw r3, r9, r8 C compute function return value
srw r0, r9, r6
addi r5, r5, -1
andi. r10, r5, 3 C count for spill loop
beq L(e)
mtctr r10
lwzu r28, 4(r4)
bdz L(xe0)
L(loop0):
srw r12, r28, r6
slw r24, r28, r8
lwzu r28, 4(r4)
or r24, r0, r24
stwu r24, 4(r7)
mr r0, r12
bdnz L(loop0) C taken at most once!
L(xe0): srw r12, r28, r6
slw r24, r28, r8
or r24, r0, r24
stwu r24, 4(r7)
mr r0, r12
L(e): srwi r5, r5, 2 C count for unrolled loop
addi r5, r5, -1
mtctr r5
lwz r28, 4(r4)
lwz r29, 8(r4)
lwz r30, 12(r4)
lwzu r31, 16(r4)
L(loopU):
srw r9, r28, r6
slw r24, r28, r8
lwz r28, 4(r4)
srw r10, r29, r6
slw r25, r29, r8
lwz r29, 8(r4)
srw r11, r30, r6
slw r26, r30, r8
lwz r30, 12(r4)
srw r12, r31, r6
slw r27, r31, r8
lwzu r31, 16(r4)
or r24, r0, r24
stw r24, 4(r7)
or r25, r9, r25
stw r25, 8(r7)
or r26, r10, r26
stw r26, 12(r7)
or r27, r11, r27
stwu r27, 16(r7)
mr r0, r12
bdnz L(loopU)
srw r9, r28, r6
slw r24, r28, r8
srw r10, r29, r6
slw r25, r29, r8
srw r11, r30, r6
slw r26, r30, r8
srw r12, r31, r6
slw r27, r31, r8
or r24, r0, r24
stw r24, 4(r7)
or r25, r9, r25
stw r25, 8(r7)
or r26, r10, r26
stw r26, 12(r7)
or r27, r11, r27
stw r27, 16(r7)
stw r12, 20(r7)
lmw r24, 8(r1) C restore registers
addi r1, r1, 48
blr
EPILOGUE()
|
