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
|
; HP-PA 7100/7200 __gmpn_submul_1 -- Multiply a limb vector with a limb and
; subtract the result from a second limb vector.
; Copyright (C) 1995, 2000 Free Software Foundation, Inc.
; This file is part of the GNU MP Library.
; The GNU MP Library is free software; you can redistribute it and/or modify
; it under the terms of the GNU Lesser General Public License as published by
; the Free Software Foundation; either version 2.1 of the License, or (at your
; option) any later version.
; The GNU MP Library is distributed in the hope that it will be useful, but
; WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
; License for more details.
; You should have received a copy of the GNU Lesser General Public License
; along with the GNU MP Library; see the file COPYING.LIB. If not, write to
; the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
; MA 02111-1307, USA.
; INPUT PARAMETERS
#define res_ptr %r26
#define s1_ptr %r25
#define size %r24
#define s2_limb %r23
#define cylimb %r28
#define s0 %r19
#define s1 %r20
#define s2 %r3
#define s3 %r4
#define lo0 %r21
#define lo1 %r5
#define lo2 %r6
#define lo3 %r7
#define hi0 %r22
#define hi1 %r23 /* safe to reuse */
#define hi2 %r29
#define hi3 %r1
.code
.export __gmpn_submul_1
__gmpn_submul_1
.proc
.callinfo frame=128,no_calls
.entry
ldo 128(%r30),%r30
stws s2_limb,-16(%r30)
add %r0,%r0,cylimb ; clear cy and cylimb
addib,< -4,size,L$few_limbs
fldws -16(%r30),%fr31R
ldo -112(%r30),%r31
stw %r3,-96(%r30)
stw %r4,-92(%r30)
stw %r5,-88(%r30)
stw %r6,-84(%r30)
stw %r7,-80(%r30)
bb,>=,n s1_ptr,29,L$0
fldws,ma 4(s1_ptr),%fr4
ldws 0(res_ptr),s0
xmpyu %fr4,%fr31R,%fr5
fstds %fr5,-16(%r31)
ldws -16(%r31),cylimb
ldws -12(%r31),lo0
sub s0,lo0,s0
add s0,lo0,%r0 ; invert cy
addib,< -1,size,L$few_limbs
stws,ma s0,4(res_ptr)
; start software pipeline ----------------------------------------------------
L$0 fldds,ma 8(s1_ptr),%fr4
fldds,ma 8(s1_ptr),%fr8
xmpyu %fr4L,%fr31R,%fr5
xmpyu %fr4R,%fr31R,%fr6
xmpyu %fr8L,%fr31R,%fr9
xmpyu %fr8R,%fr31R,%fr10
fstds %fr5,-16(%r31)
fstds %fr6,-8(%r31)
fstds %fr9,0(%r31)
fstds %fr10,8(%r31)
ldws -16(%r31),hi0
ldws -12(%r31),lo0
ldws -8(%r31),hi1
ldws -4(%r31),lo1
ldws 0(%r31),hi2
ldws 4(%r31),lo2
ldws 8(%r31),hi3
ldws 12(%r31),lo3
addc lo0,cylimb,lo0
addc lo1,hi0,lo1
addc lo2,hi1,lo2
addc lo3,hi2,lo3
addib,< -4,size,L$end
addc %r0,hi3,cylimb ; propagate carry into cylimb
; main loop ------------------------------------------------------------------
L$loop fldds,ma 8(s1_ptr),%fr4
fldds,ma 8(s1_ptr),%fr8
ldws 0(res_ptr),s0
xmpyu %fr4L,%fr31R,%fr5
ldws 4(res_ptr),s1
xmpyu %fr4R,%fr31R,%fr6
ldws 8(res_ptr),s2
xmpyu %fr8L,%fr31R,%fr9
ldws 12(res_ptr),s3
xmpyu %fr8R,%fr31R,%fr10
fstds %fr5,-16(%r31)
sub s0,lo0,s0
fstds %fr6,-8(%r31)
subb s1,lo1,s1
fstds %fr9,0(%r31)
subb s2,lo2,s2
fstds %fr10,8(%r31)
subb s3,lo3,s3
subb %r0,%r0,lo0 ; these two insns ...
add lo0,lo0,%r0 ; ... just invert cy
ldws -16(%r31),hi0
ldws -12(%r31),lo0
ldws -8(%r31),hi1
ldws -4(%r31),lo1
ldws 0(%r31),hi2
ldws 4(%r31),lo2
ldws 8(%r31),hi3
ldws 12(%r31),lo3
addc lo0,cylimb,lo0
stws,ma s0,4(res_ptr)
addc lo1,hi0,lo1
stws,ma s1,4(res_ptr)
addc lo2,hi1,lo2
stws,ma s2,4(res_ptr)
addc lo3,hi2,lo3
stws,ma s3,4(res_ptr)
addib,>= -4,size,L$loop
addc %r0,hi3,cylimb ; propagate carry into cylimb
; finish software pipeline ---------------------------------------------------
L$end ldws 0(res_ptr),s0
ldws 4(res_ptr),s1
ldws 8(res_ptr),s2
ldws 12(res_ptr),s3
sub s0,lo0,s0
stws,ma s0,4(res_ptr)
subb s1,lo1,s1
stws,ma s1,4(res_ptr)
subb s2,lo2,s2
stws,ma s2,4(res_ptr)
subb s3,lo3,s3
stws,ma s3,4(res_ptr)
subb %r0,%r0,lo0 ; these two insns ...
add lo0,lo0,%r0 ; ... invert cy
; restore callee-saves registers ---------------------------------------------
ldw -96(%r30),%r3
ldw -92(%r30),%r4
ldw -88(%r30),%r5
ldw -84(%r30),%r6
ldw -80(%r30),%r7
L$few_limbs
addib,=,n 4,size,L$ret
L$loop2 fldws,ma 4(s1_ptr),%fr4
ldws 0(res_ptr),s0
xmpyu %fr4,%fr31R,%fr5
fstds %fr5,-16(%r30)
ldws -16(%r30),hi0
ldws -12(%r30),lo0
addc lo0,cylimb,lo0
addc %r0,hi0,cylimb
sub s0,lo0,s0
add s0,lo0,%r0 ; invert cy
stws,ma s0,4(res_ptr)
addib,<> -1,size,L$loop2
nop
L$ret addc %r0,cylimb,cylimb
bv 0(%r2)
ldo -128(%r30),%r30
.exit
.procend
|