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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
|
dnl PowerPC-64 mpn_divrem_1 -- Divide an mpn number by an unnormalized limb.
dnl Copyright 2003, 2005 Free Software Foundation, Inc.
dnl This file is part of the GNU MP Library.
dnl The GNU MP Library is free software; you can redistribute it and/or modify
dnl it under the terms of the GNU Lesser General Public License as published
dnl by the Free Software Foundation; either version 2.1 of the License, or (at
dnl your option) any later version.
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 Lesser General Public
dnl License for more details.
dnl You should have received a copy of the GNU Lesser General Public License
dnl along with the GNU MP Library; see the file COPYING.LIB. If not, write to
dnl the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
dnl MA 02110-1301, USA.
include(`../config.m4')
C cycles/limb
C POWER3/PPC630: 40/30
C POWER4/PPC970: 40/22 (40-49 cycles for integer part, 22 stably for fraction)
C INPUT PARAMETERS
C qp = r3
C qxn = r4
C up = r5
C n = r6
C vl = r7
C This was generated by gcc, then the code was manually edited. Lots of things
C could be streamlined. It would probably be a good idea to merge the loops
C for normalized and unnormalized divisor, since the shifting stuff is done for
C free in parallel with other operations.
C FIXME: Replace divdu-based invert_limb code below!!!
ASM_START()
PROLOGUE(mpn_divrem_1)
std r27,-40(r1)
mr r27,r4
std r28,-32(r1)
mr r28,r5
std r31,-8(r1)
mr r31,r3
li r3,0
std r29,-24(r1)
std r30,-16(r1)
add. r11,r6,r27 C n = un + qxn
bne cr0,L(221)
b L(ret)
L(221): sldi r0,r11,3
rldicr. r8,r7,0,0 C test most significant vl bit
add r10,r31,r0 C qp += n
addi r31,r10,-8 C qp -= 1
beq cr0,L(salamin) C branch to 2nd code block
cmpdi cr0,r6,0
beq cr0,L(224)
C High quotient limb is 0 or 1, skip a divide step
sldi r9,r6,3
addi r31,r31,-8
add r9,r28,r9
addi r6,r6,-1
ld r3,-8(r9)
subfc r11,r7,r3
li r11,0
adde r11,r11,r11
neg r0,r11
std r11,-8(r10)
and r0,r7,r0
subf r3,r0,r3
C Start invert_limb
L(224): sldi. r0,r7,1
bne cr0,L(642)
li r9,-1 C invert value for 0x80...0
b L(274)
L(642): srdi r5,r7,32
neg r0,r7
li r11,1
sldi r11,r11,32
divdu r8,r0,r5
rldicl r30,r7,0,32
mulld r9,r8,r5
mulld r12,r8,r30
subf r10,r9,r0
mulld r10,r10,r11
cmpld cr0,r10,r12
bge cr0,L(293)
add r10,r10,r7
addi r8,r8,-1
cmpld cr0,r10,r7
blt cr0,L(293)
cmpld cr0,r10,r12
bge cr0,L(293)
addi r8,r8,-1
add r10,r10,r7
L(293): subf r10,r12,r10
li r9,1
sldi r9,r9,32
divdu r11,r10,r5
mulld r0,r11,r5
mulld r12,r11,r30
subf r0,r0,r10
mulld r0,r0,r9
cmpld cr0,r0,r12
bge cr0,L(296)
add r0,r0,r7
addi r11,r11,-1
cmpld cr0,r0,r7
blt cr0,L(296)
cmpld cr0,r0,r12
bge cr0,L(296)
addi r11,r11,-1
L(296): li r0,1
sldi r0,r0,32
mulld r0,r8,r0
or r9,r0,r11
C End invert_limb
L(274): addic. r12,r6,-1
mtctr r6
blt cr0,L(638)
sldi r0,r12,3
add r28,r28,r0
Loop1: ld r6,0(r28)
addi r28,r28,-8
mulhdu r11,r3,r9
add r11,r11,r3
mulhdu r10,r11,r7
mulld r0,r11,r7
subfc r8,r0,r6
subfe r10,r10,r3
mr r3,r8
cmpdi cr0,r10,0
beq cr0,L(332)
subfc r3,r7,r3
addme r10,r10
addi r11,r11,1
cmpdi cr0,r10,0
beq cr0,L(332)
subf r3,r7,r3
addi r11,r11,1
L(332): cmpld cr0,r3,r7
blt cr0,L(346)
subf r3,r7,r3
addi r11,r11,1
L(346): std r11,0(r31)
addi r31,r31,-8
bdnz Loop1
L(638): addic. r12,r27,-1
mtctr r27
blt cr0,L(ret)
Loop2: mulhdu r11,r3,r9
add r11,r11,r3
mulhdu r10,r11,r7
mulld r0,r11,r7
subfic r8,r0,0
subfe r10,r10,r3
mr r3,r8
cmpdi cr0,r10,0
beq cr0,L(380)
subfc r3,r7,r3
addme r10,r10
addi r11,r11,1
cmpdi cr0,r10,0
beq cr0,L(380)
subf r3,r7,r3
addi r11,r11,1
L(380): cmpld cr0,r3,r7
blt cr0,L(394)
subf r3,r7,r3
addi r11,r11,1
L(394): std r11,0(r31)
addi r31,r31,-8
bdnz Loop2
b L(ret)
L(salamin):
cmpdi cr0,r6,0
beq cr0,L(401)
sldi r9,r6,3
add r9,r28,r9
ld r5,-8(r9)
cmpld cr0,r5,r7
bge cr0,L(401)
cmpdi cr0,r11,1
std r8,-8(r10)
mr r3,r5
addi r31,r31,-8
bne cr0,L(400)
b L(ret)
L(400): addi r6,r6,-1
L(401): cntlzd r30,r7
sld r7,r7,r30
sld r3,r3,r30
C Start invert_limb
sldi. r8,r7,1
bne cr0,L(643)
li r4,-1 C invert value for 0x80...0
b L(470)
L(643): srdi r5,r7,32
neg r0,r7
li r11,1
sldi r11,r11,32
divdu r8,r0,r5
rldicl r4,r7,0,32
mulld r9,r8,r5
mulld r12,r8,r4
subf r10,r9,r0
mulld r10,r10,r11
cmpld cr0,r10,r12
bge cr0,L(489)
add r10,r10,r7
addi r8,r8,-1
cmpld cr0,r10,r7
blt cr0,L(489)
cmpld cr0,r10,r12
bge cr0,L(489)
addi r8,r8,-1
add r10,r10,r7
L(489): subf r10,r12,r10
li r9,1
sldi r9,r9,32
divdu r11,r10,r5
mulld r0,r11,r5
mulld r12,r11,r4
subf r0,r0,r10
mulld r0,r0,r9
cmpld cr0,r0,r12
bge cr0,L(492)
add r0,r0,r7
addi r11,r11,-1
cmpld cr0,r0,r7
blt cr0,L(492)
cmpld cr0,r0,r12
bge cr0,L(492)
addi r11,r11,-1
L(492): li r0,1
sldi r0,r0,32
mulld r0,r8,r0
or r4,r0,r11
C End invert_limb
L(470): cmpdi cr0,r6,0
beq cr0,L(497)
sldi r9,r6,3
subfic r11,r30,64
add r9,r28,r9
mtctr r6
addi r12,r6,-2
ld r5,-8(r9)
srd r0,r5,r11
or r3,r3,r0
bdz L(640)
mr r29,r11
sldi r0,r12,3
add r28,r28,r0
Loop3: ld r6,0(r28)
addi r28,r28,-8
mulhdu r8,r3,r4
add r8,r8,r3
mulhdu r10,r8,r7
mulld r11,r8,r7
sld r0,r5,r30
srd r9,r6,r29
or r0,r0,r9
subfc r9,r11,r0
subfe r10,r10,r3
mr r3,r9
cmpdi cr0,r10,0
beq cr0,L(529)
subfc r3,r7,r3
addme r10,r10
addi r8,r8,1
cmpdi cr0,r10,0
beq cr0,L(529)
subf r3,r7,r3
addi r8,r8,1
L(529): cmpld cr0,r3,r7
blt cr0,L(543)
subf r3,r7,r3
addi r8,r8,1
L(543): std r8,0(r31)
addi r31,r31,-8
mr r5,r6
bdnz Loop3
L(640): mulhdu r11,r3,r4
add r11,r11,r3
mulhdu r10,r11,r7
mulld r9,r11,r7
sld r0,r5,r30
subfc r8,r9,r0
subfe r10,r10,r3
mr r3,r8
cmpdi cr0,r10,0
beq cr0,L(573)
subfc r9,r7,r3
addme r10,r10
mr r3,r9
addi r11,r11,1
cmpdi cr0,r10,0
beq cr0,L(573)
subf r3,r7,r3
addi r11,r11,1
L(573): cmpld cr0,r3,r7
blt cr0,L(587)
subf r3,r7,r3
addi r11,r11,1
L(587): std r11,0(r31)
addi r31,r31,-8
L(497): addic. r12,r27,-1
mtctr r27
blt cr0,L(641)
Loop4: mulhdu r11,r3,r4
add r11,r11,r3
mulhdu r9,r11,r7
mulld r0,r11,r7
subfic r10,r0,0
subfe r9,r9,r3
mr r3,r10
cmpdi cr0,r9,0
beq cr0,L(620)
subfc r0,r7,r3
addme r9,r9
mr r3,r0
addi r11,r11,1
cmpdi cr0,r9,0
beq cr0,L(620)
subf r3,r7,r3
addi r11,r11,1
L(620): cmpld cr0,r3,r7
blt cr0,L(634)
subf r3,r7,r3
addi r11,r11,1
L(634): std r11,0(r31)
addi r31,r31,-8
bdnz Loop4
L(641): srd r3,r3,r30
L(ret): ld r27,-40(r1)
ld r28,-32(r1)
ld r29,-24(r1)
ld r30,-16(r1)
ld r31,-8(r1)
blr
EPILOGUE(mpn_divrem_1)
|
