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
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
|
Default matrix/vector min size to enable Blas/Lapack optimization: 8
Used matrix/vector min size to enable Blas/Lapack optimization: 0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
perfMatrixMultiplication is a Catch v2.9.2 host application.
Run with -? for options
-------------------------------------------------------------------------------
Benchmark matrix-matrix multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:217
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x3) - Naive code 100 223 2.0516 ms
89 ns 88 ns 89 ns
1 ns 0 ns 3 ns
(3x3)x(3x3) - ViSP 100 103 2.06 ms
202 ns 200 ns 210 ns
17 ns 1 ns 40 ns
(6x6)x(6x6) - Naive code 100 95 2.052 ms
217 ns 216 ns 219 ns
5 ns 3 ns 9 ns
(6x6)x(6x6) - ViSP 100 114 2.0406 ms
179 ns 178 ns 185 ns
10 ns 0 ns 24 ns
(8x8)x(8x8) - Naive code 100 54 2.079 ms
388 ns 387 ns 390 ns
6 ns 0 ns 14 ns
(8x8)x(8x8) - ViSP 100 84 2.0664 ms
245 ns 244 ns 253 ns
16 ns 0 ns 38 ns
(10x10)x(10x10) - Naive code 100 28 2.0496 ms
673 ns 672 ns 678 ns
10 ns 0 ns 26 ns
(10x10)x(10x10) - ViSP 100 50 2.045 ms
385 ns 356 ns 426 ns
174 ns 135 ns 239 ns
(20x20)x(20x20) - Naive code 100 4 2.4388 ms
5.698 us 5.516 us 6.146 us
1.362 us 498 ns 2.484 us
(20x20)x(20x20) - ViSP 100 26 2.0774 ms
994 ns 950 ns 1.057 us
269 ns 209 ns 355 ns
(6x200)x(200x6) - Naive code 100 3 2.2197 ms
7.398 us 7.266 us 7.735 us
975 ns 239 ns 1.757 us
(6x200)x(200x6) - ViSP 100 32 2.0992 ms
615 ns 592 ns 669 ns
171 ns 86 ns 304 ns
(200x6)x(6x200) - Naive code 100 1 17.175 ms
165.571 us 160.191 us 173.5 us
32.984 us 24.704 us 44.521 us
(200x6)x(6x200) - ViSP 100 1 2.8158 ms
29.848 us 28.934 us 31.884 us
6.576 us 3.754 us 12.739 us
(207x119)x(119x207) - Naive code 100 1 503.475 ms
5.26692 ms 5.21554 ms 5.33146 ms
293.196 us 245.95 us 386.767 us
(207x119)x(119x207) - ViSP 100 1 24.0501 ms
241.335 us 232.314 us 253.677 us
53.573 us 41.88 us 67.573 us
(83x201)x(201x83) - Naive code 100 1 135.879 ms
1.36202 ms 1.35929 ms 1.36931 ms
21.007 us 9.253 us 44.017 us
(83x201)x(201x83) - ViSP 100 1 8.335 ms
70.298 us 67.956 us 74.124 us
14.996 us 10.355 us 21.024 us
(600x400)x(400x600) - Naive code 100 1 15.4613 s
171.254 ms 166 ms 178.327 ms
31.0091 ms 24.7589 ms 39.2442 ms
(600x400)x(400x600) - ViSP 100 1 647.363 ms
6.38457 ms 6.34948 ms 6.41008 ms
151.093 us 109.895 us 216.753 us
(400x600)x(600x400) - Naive code 100 1 10.185 s
102.595 ms 102.154 ms 103.28 ms
2.76177 ms 1.93992 ms 3.67911 ms
(400x600)x(600x400) - ViSP 100 1 424.223 ms
4.21401 ms 4.18659 ms 4.23194 ms
111.623 us 77.423 us 160.784 us
-------------------------------------------------------------------------------
Benchmark matrix-rotation matrix multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:294
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x3) - Naive code 100 162 2.0412 ms
118 ns 118 ns 119 ns
2 ns 0 ns 5 ns
(3x3)x(3x3) - ViSP 100 221 2.0332 ms
86 ns 85 ns 87 ns
2 ns 0 ns 4 ns
-------------------------------------------------------------------------------
Benchmark matrix-homogeneous matrix multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:379
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(4x4)x(4x4) - Naive code 100 135 2.052 ms
146 ns 146 ns 148 ns
3 ns 0 ns 8 ns
(4x4)x(4x4) - ViSP 100 174 2.0358 ms
112 ns 112 ns 113 ns
2 ns 0 ns 5 ns
-------------------------------------------------------------------------------
Benchmark matrix-vector multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:467
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x1) - Naive code 100 273 2.0475 ms
73 ns 71 ns 76 ns
10 ns 1 ns 19 ns
(3x3)x(3x1) - ViSP 100 169 2.0449 ms
139 ns 129 ns 153 ns
61 ns 45 ns 79 ns
(6x6)x(6x1) - Naive code 100 213 2.0448 ms
91 ns 91 ns 91 ns
1 ns 0 ns 3 ns
(6x6)x(6x1) - ViSP 100 158 2.0382 ms
123 ns 123 ns 126 ns
5 ns 1 ns 11 ns
(8x8)x(8x1) - Naive code 100 179 2.0406 ms
107 ns 107 ns 108 ns
1 ns 0 ns 4 ns
(8x8)x(8x1) - ViSP 100 157 2.041 ms
124 ns 124 ns 126 ns
4 ns 0 ns 7 ns
(10x10)x(10x1) - Naive code 100 132 2.046 ms
133 ns 133 ns 135 ns
2 ns 0 ns 5 ns
(10x10)x(10x1) - ViSP 100 136 2.04 ms
137 ns 136 ns 141 ns
10 ns 0 ns 25 ns
(20x20)x(20x1) - Naive code 100 59 2.0473 ms
311 ns 310 ns 313 ns
4 ns 0 ns 11 ns
(20x20)x(20x1) - ViSP 100 106 2.0564 ms
180 ns 178 ns 185 ns
12 ns 0 ns 29 ns
(6x200)x(200x1) - Naive code 100 23 2.1183 ms
794 ns 793 ns 795 ns
3 ns 0 ns 7 ns
(6x200)x(200x1) - ViSP 100 93 2.0646 ms
201 ns 199 ns 211 ns
20 ns 2 ns 47 ns
(200x6)x(6x1) - Naive code 100 16 2.0624 ms
1.164 us 1.161 us 1.176 us
25 ns 1 ns 59 ns
(200x6)x(6x1) - ViSP 100 25 2.1075 ms
781 ns 773 ns 822 ns
81 ns 0 ns 193 ns
(207x119)x(119x1) - Naive code 100 2 3.609 ms
16.763 us 16.421 us 17.584 us
2.573 us 1.043 us 4.592 us
(207x119)x(119x1) - ViSP 100 6 2.2386 ms
3.653 us 3.578 us 3.834 us
571 ns 275 ns 983 ns
(83x201)x(201x1) - Naive code 100 2 2.2302 ms
10.029 us 9.913 us 10.605 us
1.149 us 8 ns 2.742 us
(83x201)x(201x1) - ViSP 100 10 2.199 ms
2.092 us 2.069 us 2.195 us
211 ns 26 ns 500 ns
(600x400)x(400x1) - Naive code 100 1 22.7039 ms
215.771 us 213.856 us 220.459 us
14.681 us 7.299 us 25.746 us
(600x400)x(400x1) - ViSP 100 1 3.4367 ms
34.26 us 33.546 us 35.702 us
4.943 us 2.727 us 7.98 us
(400x600)x(600x1) - Naive code 100 1 18.0648 ms
169.05 us 166.772 us 174.014 us
16.283 us 8.605 us 28.822 us
(400x600)x(600x1) - ViSP 100 1 3.3765 ms
32.3 us 31.727 us 33.925 us
4.274 us 644 ns 8.951 us
-------------------------------------------------------------------------------
Benchmark AtA
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:548
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3) - Naive code 100 219 2.0367 ms
81 ns 81 ns 82 ns
1 ns 0 ns 2 ns
(3x3) - ViSP 100 87 2.0445 ms
196 ns 194 ns 202 ns
13 ns 0 ns 32 ns
(6x6) - Naive code 100 111 2.0535 ms
166 ns 165 ns 167 ns
3 ns 0 ns 7 ns
(6x6) - ViSP 100 85 2.0485 ms
223 ns 220 ns 233 ns
21 ns 0 ns 50 ns
(8x8) - Naive code 100 67 2.0435 ms
279 ns 279 ns 282 ns
5 ns 1 ns 12 ns
(8x8) - ViSP 100 77 2.0482 ms
252 ns 245 ns 266 ns
47 ns 26 ns 77 ns
(10x10) - Naive code 100 40 2.056 ms
464 ns 463 ns 467 ns
7 ns 0 ns 17 ns
(10x10) - ViSP 100 63 2.0538 ms
300 ns 298 ns 311 ns
22 ns 2 ns 53 ns
(20x20) - Naive code 100 7 2.3093 ms
3.044 us 3.041 us 3.059 us
30 ns 3 ns 71 ns
(20x20) - ViSP 100 23 2.0999 ms
815 ns 806 ns 859 ns
87 ns 2 ns 208 ns
(6x200) - Naive code 100 1 10.4797 ms
97.175 us 96.383 us 98.895 us
5.682 us 2.742 us 9.619 us
(6x200) - ViSP 100 1 2.6835 ms
25.388 us 24.922 us 27.178 us
4.216 us 750 ns 9.909 us
(200x6) - Naive code 100 5 2.0935 ms
4.184 us 4.177 us 4.201 us
47 ns 4 ns 86 ns
(200x6) - ViSP 100 27 2.0871 ms
690 ns 672 ns 733 ns
140 ns 69 ns 229 ns
(207x119) - Naive code 100 1 144.588 ms
1.45281 ms 1.44668 ms 1.46487 ms
42.491 us 24.651 us 67.98 us
(207x119) - ViSP 100 1 14.9142 ms
131.627 us 130.423 us 134.817 us
8.986 us 2.199 us 16.924 us
(83x201) - Naive code 100 1 152.614 ms
1.49636 ms 1.49332 ms 1.50178 ms
20.07 us 12.992 us 29.432 us
(83x201) - ViSP 100 1 16.69 ms
144.707 us 143.3 us 147.362 us
9.528 us 5.805 us 16.117 us
(600x400) - Naive code 100 1 5.72741 s
57.3315 ms 57.1886 ms 57.538 ms
870.225 us 658.677 us 1.14161 ms
(600x400) - ViSP 100 1 435.461 ms
4.30212 ms 4.27026 ms 4.31975 ms
117.438 us 75.304 us 184.055 us
(400x600) - Naive code 100 1 7.82654 s
78.6463 ms 78.4704 ms 78.8933 ms
1.05031 ms 816.214 us 1.43298 ms
(400x600) - ViSP 100 1 660.716 ms
6.53438 ms 6.49527 ms 6.57244 ms
195.587 us 120.116 us 296.007 us
-------------------------------------------------------------------------------
Benchmark AAt
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:619
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3) - Naive code 100 217 2.0398 ms
83 ns 83 ns 84 ns
2 ns 0 ns 4 ns
(3x3) - ViSP 100 91 2.0566 ms
211 ns 205 ns 223 ns
38 ns 17 ns 68 ns
(6x6) - Naive code 100 110 2.046 ms
193 ns 182 ns 211 ns
72 ns 51 ns 102 ns
(6x6) - ViSP 100 79 2.054 ms
267 ns 260 ns 281 ns
45 ns 25 ns 75 ns
(8x8) - Naive code 100 63 2.0412 ms
337 ns 319 ns 366 ns
113 ns 78 ns 165 ns
(8x8) - ViSP 100 67 2.0636 ms
279 ns 268 ns 302 ns
77 ns 46 ns 143 ns
(10x10) - Naive code 100 41 2.0828 ms
459 ns 454 ns 479 ns
48 ns 16 ns 109 ns
(10x10) - ViSP 100 46 2.0838 ms
363 ns 339 ns 410 ns
163 ns 96 ns 264 ns
(20x20) - Naive code 100 7 2.2001 ms
4.88 us 4.665 us 5.071 us
1.032 us 914 ns 1.14 us
(20x20) - ViSP 100 22 2.1208 ms
875 ns 845 ns 951 ns
240 ns 116 ns 404 ns
(6x200) - Naive code 100 5 2.1265 ms
4.142 us 4.042 us 4.384 us
758 ns 371 ns 1.295 us
(6x200) - ViSP 100 20 2.08 ms
947 ns 913 ns 1.026 us
259 ns 120 ns 434 ns
(200x6) - Naive code 100 1 9.9795 ms
100.619 us 96.876 us 105.807 us
22.349 us 17.575 us 29.378 us
(200x6) - ViSP 100 1 2.872 ms
26.514 us 25.036 us 32.238 us
12.726 us 3.037 us 29.115 us
-------------------------------------------------------------------------------
Benchmark matrix-velocity twist multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:690
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(6x6)x(6x6) - Naive code 100 70 2.065 ms
262 ns 257 ns 275 ns
35 ns 6 ns 64 ns
(6x6)x(6x6) - ViSP 100 106 2.0458 ms
172 ns 169 ns 179 ns
16 ns 0 ns 37 ns
(20x6)x(6x6) - Naive code 100 29 2.0851 ms
729 ns 696 ns 774 ns
196 ns 151 ns 248 ns
(20x6)x(6x6) - ViSP 100 60 2.058 ms
328 ns 313 ns 358 ns
103 ns 61 ns 160 ns
(207x6)x(6x6) - Naive code 100 4 2.3788 ms
5.941 us 5.705 us 6.301 us
1.462 us 1.09 us 2.053 us
(207x6)x(6x6) - ViSP 100 9 2.1888 ms
2.209 us 2.158 us 2.343 us
381 ns 117 ns 744 ns
(600x6)x(6x6) - Naive code 100 2 3.3366 ms
14.785 us 14.589 us 15.278 us
1.464 us 516 ns 2.755 us
(600x6)x(6x6) - ViSP 100 4 2.6204 ms
6.292 us 6.222 us 6.489 us
552 ns 179 ns 1.162 us
(1201x6)x(6x6) - Naive code 100 1 3.1746 ms
28.902 us 28.787 us 29.142 us
815 ns 485 ns 1.563 us
(1201x6)x(6x6) - ViSP 100 2 2.491 ms
12.312 us 11.953 us 13.049 us
2.508 us 1.43 us 4.146 us
-------------------------------------------------------------------------------
Benchmark matrix-force twist multiplication
-------------------------------------------------------------------------------
/home/fspindle/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:775
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(6x6)x(6x6) - Naive code 100 74 2.0646 ms
253 ns 253 ns 256 ns
6 ns 1 ns 15 ns
(6x6)x(6x6) - ViSP 100 114 2.0406 ms
166 ns 166 ns 168 ns
4 ns 0 ns 11 ns
(20x6)x(6x6) - Naive code 100 32 2.0608 ms
624 ns 621 ns 631 ns
20 ns 5 ns 36 ns
(20x6)x(6x6) - ViSP 100 66 2.0592 ms
292 ns 290 ns 298 ns
13 ns 2 ns 31 ns
(207x6)x(6x6) - Naive code 100 4 2.3 ms
5.46 us 5.28 us 5.871 us
1.32 us 729 ns 2.362 us
(207x6)x(6x6) - ViSP 100 10 2.263 ms
2.145 us 2.14 us 2.164 us
40 ns 2 ns 95 ns
(600x6)x(6x6) - Naive code 100 2 3.23 ms
14.517 us 14.484 us 14.633 us
266 ns 24 ns 602 ns
(600x6)x(6x6) - ViSP 100 4 2.5576 ms
5.914 us 5.881 us 6.063 us
299 ns 7 ns 706 ns
(1201x6)x(6x6) - Naive code 100 1 3.1943 ms
30.849 us 29.807 us 32.58 us
6.719 us 4.79 us 9.907 us
(1201x6)x(6x6) - ViSP 100 2 2.477 ms
11.978 us 11.768 us 12.505 us
1.594 us 629 ns 2.894 us
===============================================================================
All tests passed (60 assertions in 8 test cases)
|
