forked from ThomasKaiser/sbc-bench
-
Notifications
You must be signed in to change notification settings - Fork 0
/
4Kau.txt
819 lines (692 loc) · 41 KB
/
4Kau.txt
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
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
sbc-bench v0.9.49 OrangePi 4 (Sat, 28 Oct 2023 18:11:52 +0200)
Distributor ID: Debian
Description: Armbian 23.8.3 bookworm
Release: 12
Codename: bookworm
Build system: https://github.com/armbian/build, 23.8.1, Orange Pi 4, rockchip64, rockchip64
/usr/bin/gcc (Debian 12.2.0-14) 12.2.0
Uptime: 18:11:53 up 28 min, 1 user, load average: 0.56, 0.19, 0.10, 29.4°C, 189331966
Linux 6.1.50-current-rockchip64 (orangepi4) 10/28/23 _aarch64_ (6 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
0.61 0.00 0.49 0.67 0.00 98.23
Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd
mmcblk2 7.70 286.19 62.63 0.00 484486 106021 0
zram0 0.18 0.71 0.00 0.00 1200 4 0
zram1 0.18 0.24 1.95 0.00 408 3304 0
total used free shared buff/cache available
Mem: 3.8Gi 541Mi 3.1Gi 204Mi 422Mi 3.2Gi
Swap: 1.9Gi 0B 1.9Gi
Filename Type Size Used Priority
/dev/zram0 partition 1978284 0 5
##########################################################################
Checking cpufreq OPP for cpu0-cpu3 (Cortex-A53):
Cpufreq OPP: 1512 Measured: 1509 (1509.585/1509.528/1509.453)
Cpufreq OPP: 1416 Measured: 1413 (1413.444/1413.391/1413.355)
Cpufreq OPP: 1200 Measured: 1193 (1194.092/1194.017/1192.902)
Cpufreq OPP: 1008 Measured: 1005 (1005.529/1005.529/1005.413)
Cpufreq OPP: 816 Measured: 813 (813.576/813.454/813.363)
Cpufreq OPP: 600 Measured: 597 (597.605/597.568/597.508)
Cpufreq OPP: 408 Measured: 405 (405.640/405.615/405.539)
Checking cpufreq OPP for cpu4-cpu5 (Cortex-A72):
Cpufreq OPP: 2016 Measured: 2014 (2014.620/2014.570/2014.544)
Cpufreq OPP: 1800 Measured: 1798 (1798.452/1798.362/1798.295)
Cpufreq OPP: 1608 Measured: 1606 (1606.586/1606.525/1606.305)
Cpufreq OPP: 1416 Measured: 1414 (1414.574/1414.539/1414.504)
Cpufreq OPP: 1200 Measured: 1198 (1198.540/1198.540/1198.480)
Cpufreq OPP: 1008 Measured: 1006 (1006.520/1006.507/1006.469)
Cpufreq OPP: 816 Measured: 814 (814.515/814.494/814.464)
Cpufreq OPP: 600 Measured: 598 (598.487/598.480/598.473)
Cpufreq OPP: 408 Measured: 406 (406.588/406.511/406.450)
##########################################################################
Hardware sensors:
tcpm_source_psy_4_0022-i2c-4-22
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 0.00 A (max = +0.00 A)
cpu_thermal-virtual-0
temp1: +28.1 C (crit = +100.0 C)
gpu_thermal-virtual-0
temp1: +26.2 C (crit = +95.0 C)
##########################################################################
Executing benchmark on cpu0 (Cortex-A53):
tinymembench v0.4.9-nuumio (simple benchmark for memory throughput and latency)
CFLAGS:
bandwidth test min repeats (-b): 2
bandwidth test max repeats (-B): 3
bandwidth test mem realloc (-M): no (-m for realloc)
latency test repeats (-l): 3
latency test count (-c): 1000000
==========================================================================
== Memory bandwidth tests ==
== ==
== Note 1: 1MB = 1000000 bytes ==
== Note 2: Test result is the best of repeated runs. Number of repeats ==
== is shown in brackets ==
== Note 3: Results for 'copy' tests show how many bytes can be ==
== copied per second (adding together read and writen ==
== bytes would have provided twice higher numbers) ==
== Note 4: 2-pass copy means that we are using a small temporary buffer ==
== to first fetch data into it, and only then write it to the ==
== destination (source -> L1 cache, L1 cache -> destination) ==
== Note 5: If sample standard deviation exceeds 0.1%, it is shown in ==
== brackets ==
==========================================================================
C copy backwards : 1693.4 MB/s (3, 4.0%)
C copy backwards (32 byte blocks) : 1700.6 MB/s (2)
C copy backwards (64 byte blocks) : 1733.0 MB/s (2)
C copy : 1779.6 MB/s (3, 0.3%)
C copy prefetched (32 bytes step) : 1336.2 MB/s (2)
C copy prefetched (64 bytes step) : 1492.3 MB/s (2)
C 2-pass copy : 1536.9 MB/s (3, 0.2%)
C 2-pass copy prefetched (32 bytes step) : 1039.8 MB/s (2)
C 2-pass copy prefetched (64 bytes step) : 1011.8 MB/s (2)
C scan 8 : 295.0 MB/s (3, 0.3%)
C scan 16 : 580.0 MB/s (3, 0.4%)
C scan 32 : 1111.5 MB/s (3, 0.1%)
C scan 64 : 1921.7 MB/s (3, 1.2%)
C fill : 8407.7 MB/s (3, 0.2%)
C fill (shuffle within 16 byte blocks) : 8411.4 MB/s (3, 0.2%)
C fill (shuffle within 32 byte blocks) : 8408.4 MB/s (3, 0.2%)
C fill (shuffle within 64 byte blocks) : 8407.8 MB/s (3, 0.4%)
---
libc memcpy copy : 1785.5 MB/s (2)
libc memchr scan : 1896.4 MB/s (3, 3.2%)
libc memset fill : 8413.2 MB/s (3, 0.2%)
---
NEON LDP/STP copy : 1827.3 MB/s (3, 0.3%)
NEON LDP/STP copy pldl2strm (32 bytes step) : 1202.5 MB/s (3, 0.5%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 1524.4 MB/s (3, 0.6%)
NEON LDP/STP copy pldl1keep (32 bytes step) : 1979.8 MB/s (3)
NEON LDP/STP copy pldl1keep (64 bytes step) : 1981.5 MB/s (2)
NEON LD1/ST1 copy : 1817.0 MB/s (3, 0.1%)
NEON LDP load : 2660.2 MB/s (2)
NEON LDNP load : 2079.5 MB/s (3, 0.3%)
NEON STP fill : 8438.7 MB/s (3, 0.2%)
NEON STNP fill : 1967.3 MB/s (3, 0.4%)
ARM LDP/STP copy : 1823.0 MB/s (3, 0.1%)
ARM LDP load : 2658.9 MB/s (2)
ARM LDNP load : 2078.7 MB/s (2)
ARM STP fill : 8439.3 MB/s (3, 0.2%)
ARM STNP fill : 1991.9 MB/s (3, 1.2%)
==========================================================================
== Framebuffer read tests. ==
== ==
== Many ARM devices use a part of the system memory as the framebuffer, ==
== typically mapped as uncached but with write-combining enabled. ==
== Writes to such framebuffers are quite fast, but reads are much ==
== slower and very sensitive to the alignment and the selection of ==
== CPU instructions which are used for accessing memory. ==
== ==
== Many x86 systems allocate the framebuffer in the GPU memory, ==
== accessible for the CPU via a relatively slow PCI-E bus. Moreover, ==
== PCI-E is asymmetric and handles reads a lot worse than writes. ==
== ==
== If uncached framebuffer reads are reasonably fast (at least 100 MB/s ==
== or preferably >300 MB/s), then using the shadow framebuffer layer ==
== is not necessary in Xorg DDX drivers, resulting in a nice overall ==
== performance improvement. For example, the xf86-video-fbturbo DDX ==
== uses this trick. ==
==========================================================================
NEON LDP/STP copy (from framebuffer) : 1852.7 MB/s (3, 0.2%)
NEON LDP/STP 2-pass copy (from framebuffer) : 1521.6 MB/s (2)
NEON LD1/ST1 copy (from framebuffer) : 1824.2 MB/s (3)
NEON LD1/ST1 2-pass copy (from framebuffer) : 1491.6 MB/s (2)
ARM LDP/STP copy (from framebuffer) : 1853.5 MB/s (2)
ARM LDP/STP 2-pass copy (from framebuffer) : 1520.7 MB/s (3, 0.1%)
==========================================================================
== Memory latency test ==
== ==
== Average time is measured for random memory accesses in the buffers ==
== of different sizes. The larger is the buffer, the more significant ==
== are relative contributions of TLB, L1/L2 cache misses and SDRAM ==
== accesses. For extremely large buffer sizes we are expecting to see ==
== page table walk with several requests to SDRAM for almost every ==
== memory access (though 64MiB is not nearly large enough to experience ==
== this effect to its fullest). ==
== ==
== Note 1: All the numbers are representing extra time, which needs to ==
== be added to L1 cache latency. The cycle timings for L1 cache ==
== latency can be usually found in the processor documentation. ==
== Note 2: Dual random read means that we are simultaneously performing ==
== two independent memory accesses at a time. In the case if ==
== the memory subsystem can't handle multiple outstanding ==
== requests, dual random read has the same timings as two ==
== single reads performed one after another. ==
==========================================================================
block size : single random read / dual random read, [MADV_NOHUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.1 ns / 0.0 ns
65536 : 4.6 ns / 7.7 ns
131072 : 7.2 ns / 10.7 ns
262144 : 8.9 ns / 11.4 ns
524288 : 16.6 ns / 25.3 ns
1048576 : 84.5 ns / 128.5 ns
2097152 : 124.0 ns / 167.3 ns
4194304 : 149.6 ns / 187.7 ns
8388608 : 163.4 ns / 197.1 ns
16777216 : 171.5 ns / 203.9 ns
33554432 : 176.7 ns / 208.9 ns
67108864 : 180.0 ns / 212.7 ns
block size : single random read / dual random read, [MADV_HUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.1 ns / 0.0 ns
65536 : 4.6 ns / 7.7 ns
131072 : 7.2 ns / 10.8 ns
262144 : 9.0 ns / 12.0 ns
524288 : 11.5 ns / 12.9 ns
1048576 : 85.6 ns / 131.5 ns
2097152 : 125.9 ns / 170.0 ns
4194304 : 146.5 ns / 183.2 ns
8388608 : 156.7 ns / 188.0 ns
16777216 : 161.6 ns / 189.8 ns
33554432 : 164.4 ns / 190.5 ns
67108864 : 165.9 ns / 190.8 ns
Executing benchmark on cpu4 (Cortex-A72):
tinymembench v0.4.9-nuumio (simple benchmark for memory throughput and latency)
CFLAGS:
bandwidth test min repeats (-b): 2
bandwidth test max repeats (-B): 3
bandwidth test mem realloc (-M): no (-m for realloc)
latency test repeats (-l): 3
latency test count (-c): 1000000
==========================================================================
== Memory bandwidth tests ==
== ==
== Note 1: 1MB = 1000000 bytes ==
== Note 2: Test result is the best of repeated runs. Number of repeats ==
== is shown in brackets ==
== Note 3: Results for 'copy' tests show how many bytes can be ==
== copied per second (adding together read and writen ==
== bytes would have provided twice higher numbers) ==
== Note 4: 2-pass copy means that we are using a small temporary buffer ==
== to first fetch data into it, and only then write it to the ==
== destination (source -> L1 cache, L1 cache -> destination) ==
== Note 5: If sample standard deviation exceeds 0.1%, it is shown in ==
== brackets ==
==========================================================================
C copy backwards : 3538.0 MB/s (3, 0.4%)
C copy backwards (32 byte blocks) : 3532.9 MB/s (2)
C copy backwards (64 byte blocks) : 3542.8 MB/s (3, 0.1%)
C copy : 3485.0 MB/s (2)
C copy prefetched (32 bytes step) : 3484.3 MB/s (3, 0.3%)
C copy prefetched (64 bytes step) : 3491.9 MB/s (3, 0.1%)
C 2-pass copy : 3079.9 MB/s (3, 0.4%)
C 2-pass copy prefetched (32 bytes step) : 3126.4 MB/s (3, 0.9%)
C 2-pass copy prefetched (64 bytes step) : 3114.9 MB/s (2)
C scan 8 : 998.5 MB/s (2)
C scan 16 : 1986.8 MB/s (2)
C scan 32 : 3909.7 MB/s (2)
C scan 64 : 6913.8 MB/s (3, 0.1%)
C fill : 8384.9 MB/s (2)
C fill (shuffle within 16 byte blocks) : 8437.0 MB/s (3, 0.1%)
C fill (shuffle within 32 byte blocks) : 8440.3 MB/s (3, 0.1%)
C fill (shuffle within 64 byte blocks) : 8389.6 MB/s (2)
---
libc memcpy copy : 3488.1 MB/s (3, 0.6%)
libc memchr scan : 6735.8 MB/s (3, 0.1%)
libc memset fill : 8425.5 MB/s (3, 0.3%)
---
NEON LDP/STP copy : 3493.5 MB/s (3, 0.9%)
NEON LDP/STP copy pldl2strm (32 bytes step) : 3490.6 MB/s (2)
NEON LDP/STP copy pldl2strm (64 bytes step) : 3491.9 MB/s (2)
NEON LDP/STP copy pldl1keep (32 bytes step) : 3491.1 MB/s (3, 0.6%)
NEON LDP/STP copy pldl1keep (64 bytes step) : 3491.0 MB/s (3, 0.1%)
NEON LD1/ST1 copy : 3501.2 MB/s (3, 0.1%)
NEON LDP load : 7286.2 MB/s (3, 0.4%)
NEON LDNP load : 7347.0 MB/s (3, 0.1%)
NEON STP fill : 8407.2 MB/s (2)
NEON STNP fill : 8401.1 MB/s (3, 1.4%)
ARM LDP/STP copy : 3501.5 MB/s (3, 0.1%)
ARM LDP load : 7272.7 MB/s (2)
ARM LDNP load : 7346.1 MB/s (3, 0.4%)
ARM STP fill : 8435.7 MB/s (3, 1.5%)
ARM STNP fill : 8426.7 MB/s (3, 0.3%)
==========================================================================
== Framebuffer read tests. ==
== ==
== Many ARM devices use a part of the system memory as the framebuffer, ==
== typically mapped as uncached but with write-combining enabled. ==
== Writes to such framebuffers are quite fast, but reads are much ==
== slower and very sensitive to the alignment and the selection of ==
== CPU instructions which are used for accessing memory. ==
== ==
== Many x86 systems allocate the framebuffer in the GPU memory, ==
== accessible for the CPU via a relatively slow PCI-E bus. Moreover, ==
== PCI-E is asymmetric and handles reads a lot worse than writes. ==
== ==
== If uncached framebuffer reads are reasonably fast (at least 100 MB/s ==
== or preferably >300 MB/s), then using the shadow framebuffer layer ==
== is not necessary in Xorg DDX drivers, resulting in a nice overall ==
== performance improvement. For example, the xf86-video-fbturbo DDX ==
== uses this trick. ==
==========================================================================
NEON LDP/STP copy (from framebuffer) : 3621.2 MB/s (2)
NEON LDP/STP 2-pass copy (from framebuffer) : 3198.5 MB/s (3, 0.4%)
NEON LD1/ST1 copy (from framebuffer) : 3628.2 MB/s (3, 0.1%)
NEON LD1/ST1 2-pass copy (from framebuffer) : 3144.8 MB/s (3)
ARM LDP/STP copy (from framebuffer) : 3629.4 MB/s (3, 0.1%)
ARM LDP/STP 2-pass copy (from framebuffer) : 3192.0 MB/s (2)
==========================================================================
== Memory latency test ==
== ==
== Average time is measured for random memory accesses in the buffers ==
== of different sizes. The larger is the buffer, the more significant ==
== are relative contributions of TLB, L1/L2 cache misses and SDRAM ==
== accesses. For extremely large buffer sizes we are expecting to see ==
== page table walk with several requests to SDRAM for almost every ==
== memory access (though 64MiB is not nearly large enough to experience ==
== this effect to its fullest). ==
== ==
== Note 1: All the numbers are representing extra time, which needs to ==
== be added to L1 cache latency. The cycle timings for L1 cache ==
== latency can be usually found in the processor documentation. ==
== Note 2: Dual random read means that we are simultaneously performing ==
== two independent memory accesses at a time. In the case if ==
== the memory subsystem can't handle multiple outstanding ==
== requests, dual random read has the same timings as two ==
== single reads performed one after another. ==
==========================================================================
block size : single random read / dual random read, [MADV_NOHUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 1.1 ns / 1.8 ns
65536 : 4.1 ns / 6.4 ns
131072 : 6.2 ns / 8.6 ns
262144 : 9.1 ns / 11.4 ns
524288 : 10.8 ns / 13.1 ns
1048576 : 28.3 ns / 44.4 ns
2097152 : 97.0 ns / 146.2 ns
4194304 : 135.2 ns / 180.8 ns
8388608 : 159.3 ns / 201.5 ns
16777216 : 171.6 ns / 207.6 ns
33554432 : 178.6 ns / 215.0 ns
67108864 : 187.4 ns / 227.8 ns
block size : single random read / dual random read, [MADV_HUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.0 ns / 0.0 ns
65536 : 4.1 ns / 6.4 ns
131072 : 6.2 ns / 8.6 ns
262144 : 7.4 ns / 9.4 ns
524288 : 8.3 ns / 9.8 ns
1048576 : 14.9 ns / 20.2 ns
2097152 : 95.3 ns / 145.4 ns
4194304 : 134.0 ns / 181.1 ns
8388608 : 152.0 ns / 190.4 ns
16777216 : 161.3 ns / 194.4 ns
33554432 : 166.2 ns / 197.4 ns
67108864 : 171.0 ns / 198.2 ns
##########################################################################
Executing ramlat on cpu0 (Cortex-A53), results in ns:
size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR
4k: 2.693 2.661 1.998 1.992 1.989 1.988 2.736 5.558
8k: 2.652 2.651 1.989 1.989 1.993 1.988 2.734 5.552
16k: 2.651 2.657 1.990 1.989 1.989 1.989 2.742 5.552
32k: 2.660 2.657 1.993 1.997 1.993 1.994 2.747 5.567
64k: 16.69 16.99 14.93 16.31 14.76 16.41 23.34 44.18
128k: 17.91 18.46 17.85 18.17 17.87 18.22 25.24 50.39
256k: 17.91 18.41 18.17 18.49 18.19 18.47 25.72 51.42
512k: 42.16 54.17 41.45 53.16 42.43 67.29 84.44 154.5
1024k: 160.1 161.8 157.0 160.3 157.8 160.4 216.6 406.0
2048k: 174.2 175.5 173.0 173.9 173.0 174.0 221.6 423.4
4096k: 182.2 183.4 182.0 182.9 182.1 182.6 224.3 428.8
8192k: 182.4 183.8 181.5 183.3 181.6 183.2 227.1 473.7
16384k: 194.2 197.2 191.2 195.2 184.1 187.2 233.3 446.6
32768k: 188.6 192.6 188.8 192.8 188.8 193.4 236.6 455.8
65536k: 189.1 191.5 189.0 191.9 188.5 193.8 235.7 465.8
131072k: 189.4 194.6 190.5 191.8 189.1 217.1 235.2 443.1
Executing ramlat on cpu4 (Cortex-A72), results in ns:
size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR
4k: 2.485 2.485 2.482 2.482 1.986 1.987 1.988 3.972
8k: 2.482 2.482 2.482 2.485 1.986 1.986 1.994 3.972
16k: 2.485 2.483 2.483 2.483 1.986 1.988 2.797 3.972
32k: 6.480 6.903 6.476 6.909 6.213 6.398 9.834 19.02
64k: 9.932 10.38 9.930 10.38 9.439 11.96 22.60 44.01
128k: 10.45 10.74 10.63 10.69 9.930 11.97 23.01 46.08
256k: 14.88 14.91 14.89 14.91 14.38 15.19 23.58 46.13
512k: 15.87 15.01 14.98 15.13 15.10 15.10 23.20 46.19
1024k: 92.67 73.77 80.26 72.44 90.43 84.91 85.87 117.1
2048k: 146.4 137.4 151.2 139.1 146.5 148.9 154.3 191.4
4096k: 164.3 166.6 171.1 167.6 170.0 167.4 170.2 217.7
8192k: 188.2 188.4 189.1 188.1 186.8 183.2 189.2 223.9
16384k: 190.4 191.5 191.1 191.6 190.4 192.3 195.7 237.9
32768k: 190.7 193.5 191.4 192.8 189.7 197.2 204.0 235.8
65536k: 206.5 203.3 198.5 203.5 198.0 207.6 217.6 240.1
131072k: 210.9 207.2 204.5 206.4 202.6 212.1 220.9 243.8
##########################################################################
Executing benchmark on each cluster individually
OpenSSL 3.0.11, built on 19 Sep 2023 (Library: OpenSSL 3.0.11 19 Sep 2023)
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes
aes-128-cbc 109018.92k 338432.58k 697436.67k 989102.76k 1125340.50k 1136448.85k (Cortex-A53)
aes-128-cbc 342756.83k 815776.02k 1252253.78k 1426930.35k 1500261.03k 1510380.89k (Cortex-A72)
aes-192-cbc 104873.79k 306752.51k 578010.28k 767753.90k 848196.95k 854447.45k (Cortex-A53)
aes-192-cbc 318997.87k 748793.15k 1078684.59k 1265191.94k 1323199.15k 1331669.67k (Cortex-A72)
aes-256-cbc 102188.35k 284982.36k 505413.63k 643340.97k 699766.10k 703780.18k (Cortex-A53)
aes-256-cbc 312775.00k 692917.01k 987584.26k 1093514.58k 1141557.93k 1145842.35k (Cortex-A72)
##########################################################################
Executing benchmark single-threaded on cpu0 (Cortex-A53)
7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: 64000000 64000000 64000000 - - - - - -
RAM size: 3863 MB, # CPU hardware threads: 6
RAM usage: 435 MB, # Benchmark threads: 1
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 882 100 859 859 | 16606 100 1419 1418
23: 833 100 850 849 | 16303 100 1412 1411
24: 793 100 854 853 | 15959 100 1402 1401
25: 748 100 854 854 | 15545 100 1384 1384
---------------------------------- | ------------------------------
Avr: 100 854 854 | 100 1404 1403
Tot: 100 1129 1129
Executing benchmark single-threaded on cpu4 (Cortex-A72)
7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: - - - - - - - - -
RAM size: 3863 MB, # CPU hardware threads: 6
RAM usage: 435 MB, # Benchmark threads: 1
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 1712 100 1667 1666 | 25352 100 2165 2165
23: 1610 100 1641 1640 | 24886 100 2154 2154
24: 1523 100 1639 1638 | 24375 100 2140 2140
25: 1407 100 1607 1607 | 23743 100 2114 2113
---------------------------------- | ------------------------------
Avr: 100 1638 1638 | 100 2143 2143
Tot: 100 1891 1891
##########################################################################
Executing benchmark 3 times multi-threaded on CPUs 0-5
7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: 64000000 - - - - - - - -
RAM size: 3863 MB, # CPU hardware threads: 6
RAM usage: 1323 MB, # Benchmark threads: 6
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 5281 537 956 5137 | 100208 515 1659 8546
23: 4980 521 973 5074 | 98233 516 1647 8500
24: 4924 556 952 5295 | 95789 516 1630 8408
25: 4755 587 924 5430 | 93332 517 1608 8306
---------------------------------- | ------------------------------
Avr: 551 951 5234 | 516 1636 8440
Tot: 533 1294 6837
7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: - - - - - - - 1024000000 -
RAM size: 3863 MB, # CPU hardware threads: 6
RAM usage: 1323 MB, # Benchmark threads: 6
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 5594 564 964 5443 | 100662 517 1660 8585
23: 5310 570 949 5411 | 98403 520 1638 8515
24: 4849 550 948 5214 | 95812 516 1630 8410
25: 4613 558 944 5268 | 93141 516 1606 8289
---------------------------------- | ------------------------------
Avr: 561 951 5334 | 517 1634 8450
Tot: 539 1292 6892
7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,6 CPUs LE)
LE
CPU Freq: - - - - - - - - -
RAM size: 3863 MB, # CPU hardware threads: 6
RAM usage: 1323 MB, # Benchmark threads: 6
Compressing | Decompressing
Dict Speed Usage R/U Rating | Speed Usage R/U Rating
KiB/s % MIPS MIPS | KiB/s % MIPS MIPS
22: 5576 562 966 5424 | 99864 516 1652 8516
23: 5262 566 948 5362 | 97670 516 1636 8451
24: 5037 577 939 5416 | 95899 518 1624 8417
25: 4656 563 944 5316 | 93059 516 1605 8282
---------------------------------- | ------------------------------
Avr: 567 949 5380 | 517 1629 8417
Tot: 542 1289 6898
Compression: 5234,5334,5380
Decompression: 8440,8450,8417
Total: 6837,6892,6898
##########################################################################
Testing maximum cpufreq again, still under full load. System health now:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:27:00: 2016/1512MHz 5.86 90% 1% 88% 0% 0% 0% 63.8°C
Checking cpufreq OPP for cpu0-cpu3 (Cortex-A53):
Cpufreq OPP: 1512 Measured: 1509 (1509.523/1509.447/1509.390)
Checking cpufreq OPP for cpu4-cpu5 (Cortex-A72):
Cpufreq OPP: 2016 Measured: 2014 (2014.605/2014.555/2014.555)
##########################################################################
Hardware sensors:
tcpm_source_psy_4_0022-i2c-4-22
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 0.00 A (max = +0.00 A)
cpu_thermal-virtual-0
temp1: +45.6 C (crit = +100.0 C)
gpu_thermal-virtual-0
temp1: +40.6 C (crit = +95.0 C)
##########################################################################
Thermal source: /sys/class/hwmon/hwmon0/ (cpu_thermal)
System health while running tinymembench:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:13:07: 2016/1512MHz 0.99 2% 0% 1% 0% 0% 0% 30.6°C
18:13:27: 2016/1512MHz 0.99 16% 0% 16% 0% 0% 0% 30.6°C
18:13:47: 2016/1512MHz 0.99 16% 0% 16% 0% 0% 0% 31.7°C
18:14:08: 2016/1512MHz 1.00 17% 0% 16% 0% 0% 0% 31.7°C
18:14:28: 2016/1512MHz 1.00 16% 0% 16% 0% 0% 0% 30.6°C
18:14:48: 2016/1512MHz 1.00 16% 0% 16% 0% 0% 0% 42.2°C
18:15:08: 2016/1512MHz 1.00 17% 0% 16% 0% 0% 0% 42.2°C
18:15:28: 2016/1512MHz 1.00 17% 0% 16% 0% 0% 0% 43.9°C
18:15:49: 2016/1512MHz 1.00 17% 0% 16% 0% 0% 0% 40.6°C
System health while running ramlat:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:15:57: 2016/1512MHz 1.16 3% 0% 2% 0% 0% 0% 37.5°C
18:16:03: 2016/1512MHz 1.15 16% 0% 16% 0% 0% 0% 33.3°C
18:16:09: 2016/1512MHz 1.13 16% 0% 16% 0% 0% 0% 33.3°C
18:16:15: 2016/1512MHz 1.12 16% 0% 16% 0% 0% 0% 32.8°C
18:16:21: 2016/1512MHz 1.11 17% 0% 17% 0% 0% 0% 40.6°C
18:16:27: 2016/1512MHz 1.17 21% 2% 17% 0% 1% 0% 32.2°C
18:16:33: 2016/1512MHz 1.16 17% 0% 16% 0% 0% 0% 32.2°C
18:16:39: 2016/1512MHz 1.14 17% 0% 16% 0% 0% 0% 35.6°C
18:16:45: 2016/1512MHz 1.13 17% 0% 16% 0% 0% 0% 35.6°C
18:16:51: 2016/1512MHz 1.12 17% 0% 16% 0% 0% 0% 34.4°C
18:16:58: 2016/1512MHz 1.10 17% 0% 16% 0% 0% 0% 35.0°C
18:17:04: 2016/1512MHz 1.09 17% 0% 16% 0% 0% 0% 35.0°C
System health while running OpenSSL benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:17:08: 2016/1512MHz 1.09 4% 0% 3% 0% 0% 0% 34.4°C
18:17:24: 2016/1512MHz 1.07 17% 0% 16% 0% 0% 0% 32.2°C
18:17:41: 2016/1512MHz 1.05 17% 0% 16% 0% 0% 0% 45.0°C
18:17:57: 2016/1512MHz 1.04 17% 0% 16% 0% 0% 0% 33.3°C
18:18:13: 2016/1512MHz 1.03 17% 0% 16% 0% 0% 0% 46.9°C
18:18:29: 2016/1512MHz 1.02 16% 0% 16% 0% 0% 0% 33.9°C
18:18:45: 2016/1512MHz 1.02 17% 0% 16% 0% 0% 0% 46.9°C
System health while running 7-zip single core benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:18:57: 2016/1512MHz 1.17 4% 0% 3% 0% 0% 0% 38.8°C
18:19:08: 2016/1512MHz 1.14 17% 0% 16% 0% 0% 0% 33.3°C
18:19:19: 2016/1512MHz 1.12 17% 0% 16% 0% 0% 0% 32.2°C
18:19:30: 2016/1512MHz 1.10 17% 0% 16% 0% 0% 0% 32.2°C
18:19:41: 2016/1512MHz 1.08 17% 0% 16% 0% 0% 0% 31.1°C
18:19:52: 2016/1512MHz 1.06 17% 0% 16% 0% 0% 0% 31.7°C
18:20:03: 2016/1512MHz 1.05 17% 0% 16% 0% 0% 0% 31.1°C
18:20:14: 2016/1512MHz 1.05 17% 0% 16% 0% 0% 0% 31.1°C
18:20:26: 2016/1512MHz 1.11 17% 0% 16% 0% 0% 0% 30.6°C
18:20:37: 2016/1512MHz 1.09 17% 0% 16% 0% 0% 0% 30.6°C
18:20:48: 2016/1512MHz 1.07 17% 0% 16% 0% 0% 0% 30.6°C
18:20:59: 2016/1512MHz 1.06 17% 0% 16% 0% 0% 0% 30.6°C
18:21:10: 2016/1512MHz 1.05 17% 0% 16% 0% 0% 0% 30.6°C
18:21:21: 2016/1512MHz 1.04 17% 0% 16% 0% 0% 0% 41.7°C
18:21:32: 2016/1512MHz 1.03 17% 0% 16% 0% 0% 0% 40.0°C
18:21:43: 2016/1512MHz 1.03 17% 0% 16% 0% 0% 0% 41.1°C
18:21:54: 2016/1512MHz 1.02 17% 0% 16% 0% 0% 0% 43.3°C
18:22:06: 2016/1512MHz 1.02 17% 0% 16% 0% 0% 0% 41.7°C
18:22:17: 2016/1512MHz 1.02 17% 0% 16% 0% 0% 0% 41.1°C
18:22:28: 2016/1512MHz 1.08 17% 0% 16% 0% 0% 0% 44.4°C
System health while running 7-zip multi core benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
18:22:31: 2016/1512MHz 1.08 6% 0% 5% 0% 0% 0% 47.5°C
18:22:45: 2016/1512MHz 1.52 95% 1% 94% 0% 0% 0% 55.6°C
18:23:00: 2016/1512MHz 2.60 82% 1% 80% 0% 0% 0% 56.7°C
18:23:11: 2016/1512MHz 3.02 84% 1% 83% 0% 0% 0% 55.6°C
18:23:25: 2016/1512MHz 3.22 94% 1% 92% 0% 0% 0% 59.4°C
18:23:35: 2016/1512MHz 4.00 79% 1% 77% 0% 0% 0% 56.7°C
18:23:46: 2016/1512MHz 4.38 99% 2% 97% 0% 0% 0% 53.3°C
18:23:57: 2016/1512MHz 4.70 98% 1% 96% 0% 0% 0% 61.2°C
18:24:08: 2016/1512MHz 5.45 79% 1% 78% 0% 0% 0% 57.2°C
18:24:18: 2016/1512MHz 5.45 92% 0% 91% 0% 0% 0% 62.5°C
18:24:33: 2016/1512MHz 5.76 86% 1% 85% 0% 0% 0% 62.5°C
18:24:43: 2016/1512MHz 5.66 85% 1% 83% 0% 0% 0% 59.4°C
18:24:55: 2016/1512MHz 5.49 93% 1% 92% 0% 0% 0% 63.1°C
18:25:06: 2016/1512MHz 5.66 76% 1% 73% 0% 0% 0% 59.4°C
18:25:17: 2016/1512MHz 5.94 99% 2% 97% 0% 0% 0% 55.6°C
18:25:29: 2016/1512MHz 6.17 89% 1% 87% 0% 0% 0% 62.5°C
18:25:39: 2016/1512MHz 5.77 83% 1% 81% 0% 0% 0% 57.2°C
18:25:49: 2016/1512MHz 5.80 92% 0% 92% 0% 0% 0% 63.8°C
18:26:04: 2016/1512MHz 5.91 86% 1% 85% 0% 0% 0% 64.4°C
18:26:14: 2016/1512MHz 5.84 84% 1% 82% 0% 0% 0% 60.0°C
18:26:26: 2016/1512MHz 5.94 97% 1% 96% 0% 0% 0% 65.0°C
18:26:37: 2016/1512MHz 6.04 75% 2% 73% 0% 0% 0% 60.0°C
18:26:47: 2016/1512MHz 6.11 99% 2% 97% 0% 0% 0% 56.1°C
18:27:00: 2016/1512MHz 5.86 90% 1% 88% 0% 0% 0% 63.8°C
##########################################################################
Linux 6.1.50-current-rockchip64 (orangepi4) 10/28/23 _aarch64_ (6 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
13.60 0.01 0.61 0.45 0.00 85.33
Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd
mmcblk2 5.26 192.52 55.15 0.00 503042 144109 0
zram0 0.12 0.46 0.00 0.00 1200 4 0
zram1 0.14 0.18 1.33 0.00 476 3480 0
total used free shared buff/cache available
Mem: 3.8Gi 546Mi 3.1Gi 204Mi 440Mi 3.2Gi
Swap: 1.9Gi 0B 1.9Gi
Filename Type Size Used Priority
/dev/zram0 partition 1978284 0 5
CPU sysfs topology (clusters, cpufreq members, clockspeeds)
cpufreq min max
CPU cluster policy speed speed core type
0 0 0 408 1512 Cortex-A53 / r0p4
1 0 0 408 1512 Cortex-A53 / r0p4
2 0 0 408 1512 Cortex-A53 / r0p4
3 0 0 408 1512 Cortex-A53 / r0p4
4 0 4 408 2016 Cortex-A72 / r0p2
5 0 4 408 2016 Cortex-A72 / r0p2
Architecture: aarch64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 6
On-line CPU(s) list: 0-5
Vendor ID: ARM
Model name: Cortex-A53
Model: 4
Thread(s) per core: 1
Core(s) per socket: 4
Socket(s): 1
Stepping: r0p4
CPU(s) scaling MHz: 100%
CPU max MHz: 1512.0000
CPU min MHz: 408.0000
BogoMIPS: 48.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 cpuid
Model name: Cortex-A72
Model: 2
Thread(s) per core: 1
Core(s) per socket: 2
Socket(s): 1
Stepping: r0p2
CPU(s) scaling MHz: 100%
CPU max MHz: 2016.0000
CPU min MHz: 408.0000
BogoMIPS: 48.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 cpuid
NUMA node(s): 1
NUMA node0 CPU(s): 0-5
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Vulnerable
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Vulnerable
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
SoC guess: Rockchip RK3399
DT compat: xunlong,orangepi-4
rockchip,rk3399
Compiler: /usr/bin/gcc (Debian 12.2.0-14) 12.2.0 / aarch64-linux-gnu
Userland: arm64
Kernel: 6.1.50-current-rockchip64/aarch64
CONFIG_HZ=250
CONFIG_HZ_250=y
CONFIG_PREEMPTION=y
CONFIG_PREEMPT=y
CONFIG_PREEMPT_BUILD=y
CONFIG_PREEMPT_COUNT=y
CONFIG_PREEMPT_NOTIFIERS=y
CONFIG_PREEMPT_RCU=y
##########################################################################
Kernel 6.1.50 is not latest 6.1.60 LTS that was released on 2023-10-25.
See https://endoflife.date/linux for details. Perhaps some kernel bugs have
been fixed in the meantime and maybe vulnerabilities as well.
##########################################################################
vdd_center: 900 mV (1350 mV max)
vdd_cpu_b: 1300 mV (1500 mV max)
vdd_cpu_l: 1200 mV (1350 mV max)
vdd_gpu: 825 mV (1500 mV max)
opp-table-0:
408 MHz 825.0 mV
600 MHz 825.0 mV
816 MHz 850.0 mV
1008 MHz 925.0 mV
1200 MHz 1000.0 mV
1416 MHz 1125.0 mV
1512 MHz 1200.0 mV
opp-table-1:
408 MHz 825.0 mV
600 MHz 825.0 mV
816 MHz 825.0 mV
1008 MHz 875.0 mV
1200 MHz 950.0 mV
1416 MHz 1025.0 mV
1608 MHz 1100.0 mV
1800 MHz 1200.0 mV
2016 MHz 1300.0 mV
opp-table-2:
200 MHz 825.0 mV
297 MHz 825.0 mV
400 MHz 825.0 mV
500 MHz 875.0 mV
600 MHz 925.0 mV
800 MHz 1100.0 mV
##########################################################################
Results validation:
* Measured clockspeed not lower than advertised max CPU clockspeed
* No swapping
* Background activity (%system) OK
* Too much other background activity: 0% avg, 5% max
* No throttling
Status of performance related governors found below /sys (w/o cpufreq):
* ff9a0000.gpu: simple_ondemand / 200 MHz (powersave performance simple_ondemand / 200 297 400 500 600 800)
Status of performance related policies found below /sys:
* /sys/module/pcie_aspm/parameters/policy: default [performance] powersave powersupersave
| OrangePi 4 | 2016/1512 MHz | 6.1 | Armbian 23.8.3 bookworm arm64 | 6880 | 1891 | 1145840 | 3490 | 8430 | - |