A day has passed and, of course, I am curious about the performance of my little RockPro64 powered NAS-to-be. Thinking about how to benchmark this board (SPEC CPU2017?) I have come across sbc-bench, which is a well balanced collection of benchmarks covering a number of different aspects (memory, SIMD, server workloads, AES, etc.) of system performance.
sbc-bench v0.7.0 Pine64 RockPro64 (Sun, 12 Apr 2020 15:15:30 +0100)
Distributor ID: Debian
Description: Debian GNU/Linux 9.12 (stretch)
Release: 9.12
Codename: stretch
Architecture: arm64
/usr/bin/gcc (Debian 6.3.0-18+deb9u1) 6.3.0 20170516
Uptime: 15:15:30 up 11 min, 1 user, load average: 0.17, 0.27, 0.19
Linux 4.4.167-1213-rockchip-ayufan-g34ae07687fce (nas) 04/12/20 _aarch64_ (6 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
2.11 0.00 0.87 0.89 0.00 96.13
Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn
mtdblock0 0.08 0.32 0.00 216 0
mtdblock1 0.01 0.06 0.00 40 0
mtdblock2 0.03 0.11 0.00 72 0
mmcblk0 14.26 467.66 585.62 318347 398652
total used free shared buff/cache available
Mem: 3.8G 101M 3.3G 33M 390M 3.6G
Swap: 0B 0B 0B
##########################################################################
Checking cpufreq OPP for cpu0-cpu3:
Cpufreq OPP: 1416 Measured: 1412.909/1412.878/1413.157
Cpufreq OPP: 1200 Measured: 1196.620/1197.327/1196.689
Cpufreq OPP: 1008 Measured: 1004.966/1005.211/1004.417
Cpufreq OPP: 816 Measured: 812.790/812.950/812.890
Cpufreq OPP: 600 Measured: 596.623/596.798/597.256
Cpufreq OPP: 408 Measured: 405.040/404.638/404.798
Checking cpufreq OPP for cpu4-cpu5:
Cpufreq OPP: 1800 Measured: 1797.626/1797.986/1797.666
Cpufreq OPP: 1608 Measured: 1605.877/1605.857/1605.817
Cpufreq OPP: 1416 Measured: 1413.466/1413.853/1413.977
Cpufreq OPP: 1200 Measured: 1198.049/1197.535/1198.147
Cpufreq OPP: 1008 Measured: 1005.823/1005.859/1005.749
Cpufreq OPP: 816 Measured: 813.541/814.051/814.001
Cpufreq OPP: 600 Measured: 597.803/597.641/597.884
Cpufreq OPP: 408 Measured: 405.535/405.554/405.627
##########################################################################
Executing tinymembench on a little core:
tinymembench v0.4.9 (simple benchmark for memory throughput and latency)
==========================================================================
== Memory bandwidth tests ==
== ==
== Note 1: 1MB = 1000000 bytes ==
== Note 2: 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 3: 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 4: If sample standard deviation exceeds 0.1%, it is shown in ==
== brackets ==
==========================================================================
C copy backwards : 1309.1 MB/s (0.4%)
C copy backwards (32 byte blocks) : 1320.8 MB/s (0.4%)
C copy backwards (64 byte blocks) : 1326.9 MB/s (0.3%)
C copy : 1383.6 MB/s (0.5%)
C copy prefetched (32 bytes step) : 986.8 MB/s
C copy prefetched (64 bytes step) : 1116.4 MB/s
C 2-pass copy : 1218.0 MB/s
C 2-pass copy prefetched (32 bytes step) : 845.7 MB/s
C 2-pass copy prefetched (64 bytes step) : 742.0 MB/s (0.1%)
C fill : 4807.9 MB/s
C fill (shuffle within 16 byte blocks) : 4807.6 MB/s
C fill (shuffle within 32 byte blocks) : 4807.5 MB/s
C fill (shuffle within 64 byte blocks) : 4808.1 MB/s
---
standard memcpy : 1355.2 MB/s (0.2%)
standard memset : 4811.1 MB/s
---
NEON LDP/STP copy : 1387.5 MB/s
NEON LDP/STP copy pldl2strm (32 bytes step) : 925.4 MB/s (0.4%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 1153.7 MB/s (0.1%)
NEON LDP/STP copy pldl1keep (32 bytes step) : 1495.6 MB/s
NEON LDP/STP copy pldl1keep (64 bytes step) : 1496.5 MB/s (0.2%)
NEON LD1/ST1 copy : 1386.1 MB/s (0.3%)
NEON STP fill : 4810.4 MB/s
NEON STNP fill : 2127.7 MB/s (0.6%)
ARM LDP/STP copy : 1392.7 MB/s (0.4%)
ARM STP fill : 4809.4 MB/s
ARM STNP fill : 2121.4 MB/s (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) : 188.3 MB/s (0.2%)
NEON LDP/STP 2-pass copy (from framebuffer) : 174.9 MB/s
NEON LD1/ST1 copy (from framebuffer) : 46.2 MB/s
NEON LD1/ST1 2-pass copy (from framebuffer) : 44.8 MB/s
ARM LDP/STP copy (from framebuffer) : 93.7 MB/s
ARM LDP/STP 2-pass copy (from framebuffer) : 89.5 MB/s
==========================================================================
== 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
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 : 4.8 ns / 8.2 ns
131072 : 7.4 ns / 11.2 ns
262144 : 8.8 ns / 12.6 ns
524288 : 13.4 ns / 18.3 ns
1048576 : 108.9 ns / 163.7 ns
2097152 : 158.7 ns / 209.2 ns
4194304 : 186.4 ns / 229.9 ns
8388608 : 200.4 ns / 239.8 ns
16777216 : 209.3 ns / 246.6 ns
33554432 : 215.1 ns / 249.4 ns
67108864 : 218.9 ns / 251.9 ns
Executing tinymembench on a big core:
tinymembench v0.4.9 (simple benchmark for memory throughput and latency)
==========================================================================
== Memory bandwidth tests ==
== ==
== Note 1: 1MB = 1000000 bytes ==
== Note 2: 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 3: 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 4: If sample standard deviation exceeds 0.1%, it is shown in ==
== brackets ==
==========================================================================
C copy backwards : 2756.5 MB/s
C copy backwards (32 byte blocks) : 2756.1 MB/s (0.1%)
C copy backwards (64 byte blocks) : 2556.8 MB/s
C copy : 2645.0 MB/s (1.1%)
C copy prefetched (32 bytes step) : 2558.6 MB/s
C copy prefetched (64 bytes step) : 2575.6 MB/s
C 2-pass copy : 2497.4 MB/s
C 2-pass copy prefetched (32 bytes step) : 2480.3 MB/s (0.3%)
C 2-pass copy prefetched (64 bytes step) : 2476.8 MB/s
C fill : 4873.7 MB/s (0.5%)
C fill (shuffle within 16 byte blocks) : 4875.5 MB/s (0.3%)
C fill (shuffle within 32 byte blocks) : 4872.8 MB/s (0.3%)
C fill (shuffle within 64 byte blocks) : 4879.6 MB/s (0.3%)
---
standard memcpy : 2768.9 MB/s
standard memset : 4877.1 MB/s (0.6%)
---
NEON LDP/STP copy : 2764.2 MB/s
NEON LDP/STP copy pldl2strm (32 bytes step) : 2764.7 MB/s (0.4%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 2764.4 MB/s
NEON LDP/STP copy pldl1keep (32 bytes step) : 2714.8 MB/s
NEON LDP/STP copy pldl1keep (64 bytes step) : 2713.2 MB/s
NEON LD1/ST1 copy : 2765.9 MB/s
NEON STP fill : 4875.0 MB/s (0.5%)
NEON STNP fill : 4853.9 MB/s (0.8%)
ARM LDP/STP copy : 2764.4 MB/s
ARM STP fill : 4878.0 MB/s (0.5%)
ARM STNP fill : 4836.1 MB/s (2.4%)
==========================================================================
== 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) : 606.6 MB/s (0.2%)
NEON LDP/STP 2-pass copy (from framebuffer) : 553.1 MB/s
NEON LD1/ST1 copy (from framebuffer) : 662.1 MB/s
NEON LD1/ST1 2-pass copy (from framebuffer) : 606.5 MB/s (0.2%)
ARM LDP/STP copy (from framebuffer) : 441.8 MB/s (0.4%)
ARM LDP/STP 2-pass copy (from framebuffer) : 426.5 MB/s
==========================================================================
== 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
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 : 4.5 ns / 7.1 ns
131072 : 6.8 ns / 9.6 ns
262144 : 9.8 ns / 12.8 ns
524288 : 11.4 ns / 14.6 ns
1048576 : 14.9 ns / 19.2 ns
2097152 : 115.1 ns / 175.6 ns
4194304 : 164.4 ns / 221.6 ns
8388608 : 193.7 ns / 245.5 ns
16777216 : 208.0 ns / 253.6 ns
33554432 : 215.3 ns / 260.0 ns
67108864 : 225.0 ns / 273.5 ns
##########################################################################
OpenSSL 1.1.0l, built on 10 Sep 2019
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes
aes-128-cbc 97709.71k 308028.16k 643292.84k 916046.17k 1052936.87k 1062300.33k
aes-128-cbc 330453.73k 764148.01k 1141503.23k 1270500.69k 1348408.66k 1348943.87k
aes-192-cbc 93020.03k 275658.75k 532274.01k 711471.45k 793711.96k 791625.73k
aes-192-cbc 233385.53k 700044.50k 973976.23k 1128532.99k 1186998.95k 1176911.87k
aes-256-cbc 87798.06k 250653.33k 462068.14k 598402.05k 655354.54k 659270.31k
aes-256-cbc 285443.29k 642637.61k 893565.70k 974888.28k 1020400.98k 1020717.74k
##########################################################################
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: 1406 1411 1412 1412 1412 1412 1412 1412 1412
RAM size: 3874 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: 680 100 662 662 | 15181 100 1295 1295
23: 655 100 668 667 | 14833 100 1284 1284
24: 629 100 677 677 | 14471 100 1270 1270
25: 600 100 686 686 | 13914 100 1238 1238
---------------------------------- | ------------------------------
Avr: 100 673 673 | 100 1272 1272
Tot: 100 973 972
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: 1792 1796 1797 1797 1797 1797 1797 1796 1797
RAM size: 3874 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: 1333 100 1297 1297 | 21445 100 1830 1829
23: 1269 100 1294 1294 | 20965 100 1815 1814
24: 1207 100 1299 1298 | 20461 100 1797 1796
25: 1155 100 1319 1319 | 19878 100 1769 1769
---------------------------------- | ------------------------------
Avr: 100 1302 1302 | 100 1803 1802
Tot: 100 1552 1552
##########################################################################
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: 1338 1794 1792 1796 1797 1797 1797 1797 1797
RAM size: 3874 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: 4190 494 825 4077 | 94585 530 1522 8066
23: 4008 500 816 4084 | 91926 529 1503 7954
24: 3965 530 805 4263 | 89389 529 1484 7846
25: 3878 553 801 4429 | 86534 529 1455 7701
---------------------------------- | ------------------------------
Avr: 519 811 4213 | 529 1491 7892
Tot: 524 1151 6053
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: 1789 1797 1797 1797 1797 1795 1797 1797 1797
RAM size: 3874 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: 3962 469 822 3855 | 94422 530 1520 8052
23: 4116 516 813 4194 | 91779 528 1505 7942
24: 4100 551 800 4409 | 88911 527 1481 7804
25: 3821 540 809 4363 | 86619 530 1454 7709
---------------------------------- | ------------------------------
Avr: 519 811 4205 | 529 1490 7877
Tot: 524 1150 6041
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: 1787 1797 1797 1796 1797 1797 1796 1797 1797
RAM size: 3874 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: 4303 489 856 4186 | 94458 529 1523 8056
23: 4184 520 820 4264 | 91930 529 1504 7955
24: 4041 541 803 4346 | 89485 528 1486 7854
25: 3949 571 790 4510 | 86428 528 1456 7692
---------------------------------- | ------------------------------
Avr: 530 817 4326 | 529 1492 7889
Tot: 529 1155 6108
Compression: 4213,4205,4326
Decompression: 7892,7877,7889
Total: 6053,6041,6108
##########################################################################
** cpuminer-multi 1.3.3 by tpruvot@github **
BTC donation address: 1FhDPLPpw18X4srecguG3MxJYe4a1JsZnd (tpruvot)
[2020-04-12 15:44:47] 6 miner threads started, using 'scrypt' algorithm.
[2020-04-12 15:44:47] CPU #5: 1.85 kH/s
[2020-04-12 15:44:47] CPU #4: 1.81 kH/s
[2020-04-12 15:44:47] CPU #0: 1.10 kH/s
[2020-04-12 15:44:47] CPU #2: 1.11 kH/s
[2020-04-12 15:44:47] CPU #1: 1.10 kH/s
[2020-04-12 15:44:47] CPU #3: 1.11 kH/s
[2020-04-12 15:44:52] Total: 8.07 kH/s
[2020-04-12 15:44:57] CPU #4: 1.85 kH/s
[2020-04-12 15:44:57] CPU #5: 1.85 kH/s
[2020-04-12 15:44:57] Total: 8.13 kH/s
[2020-04-12 15:44:57] CPU #0: 1.12 kH/s
[2020-04-12 15:44:57] CPU #1: 1.11 kH/s
[2020-04-12 15:44:57] CPU #3: 1.11 kH/s
[2020-04-12 15:44:57] CPU #2: 1.11 kH/s
[2020-04-12 15:45:02] Total: 7.94 kH/s
[2020-04-12 15:45:07] CPU #2: 1.10 kH/s
[2020-04-12 15:45:07] CPU #3: 1.10 kH/s
[2020-04-12 15:45:07] CPU #1: 1.07 kH/s
[2020-04-12 15:45:08] CPU #0: 1.07 kH/s
[2020-04-12 15:45:08] CPU #5: 1.61 kH/s
[2020-04-12 15:45:08] Total: 7.72 kH/s
[2020-04-12 15:45:08] CPU #4: 1.58 kH/s
[2020-04-12 15:45:11] Total: 8.04 kH/s
[2020-04-12 15:45:12] Total: 8.14 kH/s
[2020-04-12 15:45:17] CPU #1: 1.11 kH/s
[2020-04-12 15:45:17] CPU #2: 1.11 kH/s
[2020-04-12 15:45:17] CPU #3: 1.11 kH/s
[2020-04-12 15:45:17] CPU #0: 1.11 kH/s
[2020-04-12 15:45:17] CPU #4: 1.85 kH/s
[2020-04-12 15:45:17] CPU #5: 1.85 kH/s
[2020-04-12 15:45:17] Total: 8.14 kH/s
[2020-04-12 15:45:22] Total: 8.13 kH/s
[2020-04-12 15:45:27] CPU #1: 1.11 kH/s
[2020-04-12 15:45:27] CPU #3: 1.11 kH/s
[2020-04-12 15:45:27] CPU #2: 1.11 kH/s
[2020-04-12 15:45:27] CPU #0: 1.11 kH/s
[2020-04-12 15:45:27] CPU #4: 1.85 kH/s
[2020-04-12 15:45:27] CPU #5: 1.85 kH/s
[2020-04-12 15:45:27] Total: 8.14 kH/s
[2020-04-12 15:45:32] Total: 8.13 kH/s
[2020-04-12 15:45:37] CPU #1: 1.11 kH/s
[2020-04-12 15:45:37] CPU #3: 1.11 kH/s
[2020-04-12 15:45:37] CPU #2: 1.11 kH/s
[2020-04-12 15:45:37] CPU #0: 1.11 kH/s
[2020-04-12 15:45:37] CPU #4: 1.85 kH/s
[2020-04-12 15:45:37] CPU #5: 1.85 kH/s
[2020-04-12 15:45:37] Total: 8.14 kH/s
[2020-04-12 15:45:42] Total: 8.14 kH/s
[2020-04-12 15:45:47] CPU #1: 1.11 kH/s
[2020-04-12 15:45:47] CPU #2: 1.11 kH/s
[2020-04-12 15:45:47] CPU #3: 1.11 kH/s
[2020-04-12 15:45:47] CPU #0: 1.12 kH/s
[2020-04-12 15:45:47] CPU #4: 1.85 kH/s
[2020-04-12 15:45:47] CPU #5: 1.85 kH/s
[2020-04-12 15:45:47] Total: 8.14 kH/s
[2020-04-12 15:45:53] CPU #5: 1.78 kH/s
[2020-04-12 15:45:53] Total: 8.00 kH/s
[2020-04-12 15:45:56] Total: 8.06 kH/s
[2020-04-12 15:45:57] CPU #1: 1.11 kH/s
[2020-04-12 15:45:57] CPU #2: 1.11 kH/s
[2020-04-12 15:45:57] CPU #3: 1.11 kH/s
[2020-04-12 15:45:57] CPU #0: 1.11 kH/s
[2020-04-12 15:45:57] CPU #4: 1.84 kH/s
[2020-04-12 15:45:57] Total: 8.13 kH/s
[2020-04-12 15:46:02] CPU #5: 1.84 kH/s
[2020-04-12 15:46:02] Total: 8.12 kH/s
[2020-04-12 15:46:07] CPU #2: 1.11 kH/s
[2020-04-12 15:46:07] CPU #1: 1.11 kH/s
[2020-04-12 15:46:07] CPU #3: 1.11 kH/s
[2020-04-12 15:46:07] CPU #0: 1.11 kH/s
[2020-04-12 15:46:07] CPU #4: 1.85 kH/s
[2020-04-12 15:46:07] Total: 8.14 kH/s
[2020-04-12 15:46:12] CPU #5: 1.85 kH/s
[2020-04-12 15:46:12] Total: 8.14 kH/s
[2020-04-12 15:46:17] CPU #2: 1.11 kH/s
[2020-04-12 15:46:17] CPU #1: 1.11 kH/s
[2020-04-12 15:46:17] CPU #0: 1.11 kH/s
[2020-04-12 15:46:17] CPU #3: 1.11 kH/s
[2020-04-12 15:46:17] CPU #4: 1.83 kH/s
[2020-04-12 15:46:17] Total: 8.12 kH/s
[2020-04-12 15:46:22] CPU #5: 1.84 kH/s
[2020-04-12 15:46:22] Total: 8.10 kH/s
[2020-04-12 15:46:27] CPU #2: 1.11 kH/s
[2020-04-12 15:46:27] CPU #1: 1.11 kH/s
[2020-04-12 15:46:27] CPU #3: 1.11 kH/s
[2020-04-12 15:46:27] CPU #0: 1.11 kH/s
[2020-04-12 15:46:27] CPU #4: 1.85 kH/s
[2020-04-12 15:46:27] Total: 8.14 kH/s
[2020-04-12 15:46:32] CPU #5: 1.85 kH/s
[2020-04-12 15:46:32] Total: 8.13 kH/s
[2020-04-12 15:46:37] CPU #2: 1.11 kH/s
[2020-04-12 15:46:37] CPU #1: 1.11 kH/s
[2020-04-12 15:46:37] CPU #0: 1.11 kH/s
[2020-04-12 15:46:37] CPU #3: 1.11 kH/s
[2020-04-12 15:46:37] CPU #4: 1.83 kH/s
[2020-04-12 15:46:37] Total: 8.12 kH/s
[2020-04-12 15:46:42] CPU #5: 1.85 kH/s
[2020-04-12 15:46:42] Total: 8.14 kH/s
[2020-04-12 15:46:47] CPU #2: 1.11 kH/s
[2020-04-12 15:46:47] CPU #1: 1.11 kH/s
[2020-04-12 15:46:47] CPU #0: 1.12 kH/s
[2020-04-12 15:46:47] CPU #3: 1.11 kH/s
[2020-04-12 15:46:47] CPU #4: 1.85 kH/s
[2020-04-12 15:46:47] Total: 8.14 kH/s
[2020-04-12 15:46:53] CPU #5: 1.81 kH/s
[2020-04-12 15:46:53] Total: 8.10 kH/s
[2020-04-12 15:46:56] Total: 8.13 kH/s
[2020-04-12 15:46:57] CPU #2: 1.11 kH/s
[2020-04-12 15:46:57] CPU #1: 1.11 kH/s
[2020-04-12 15:46:57] CPU #0: 1.11 kH/s
[2020-04-12 15:46:57] CPU #3: 1.11 kH/s
[2020-04-12 15:46:57] CPU #4: 1.84 kH/s
[2020-04-12 15:46:57] Total: 8.13 kH/s
[2020-04-12 15:47:03] CPU #5: 1.82 kH/s
[2020-04-12 15:47:03] Total: 8.10 kH/s
[2020-04-12 15:47:06] Total: 8.12 kH/s
[2020-04-12 15:47:07] CPU #2: 1.11 kH/s
[2020-04-12 15:47:07] CPU #0: 1.12 kH/s
[2020-04-12 15:47:07] CPU #1: 1.11 kH/s
[2020-04-12 15:47:07] CPU #3: 1.11 kH/s
[2020-04-12 15:47:07] CPU #4: 1.85 kH/s
[2020-04-12 15:47:07] Total: 8.14 kH/s
[2020-04-12 15:47:12] CPU #5: 1.84 kH/s
[2020-04-12 15:47:12] Total: 8.14 kH/s
[2020-04-12 15:47:17] CPU #2: 1.11 kH/s
[2020-04-12 15:47:17] CPU #0: 1.12 kH/s
[2020-04-12 15:47:17] CPU #1: 1.11 kH/s
[2020-04-12 15:47:17] CPU #3: 1.11 kH/s
[2020-04-12 15:47:17] CPU #4: 1.85 kH/s
[2020-04-12 15:47:17] Total: 8.14 kH/s
[2020-04-12 15:47:23] CPU #5: 1.79 kH/s
[2020-04-12 15:47:23] Total: 8.07 kH/s
[2020-04-12 15:47:26] Total: 8.13 kH/s
[2020-04-12 15:47:27] CPU #2: 1.11 kH/s
[2020-04-12 15:47:27] CPU #1: 1.11 kH/s
[2020-04-12 15:47:27] CPU #0: 1.11 kH/s
[2020-04-12 15:47:27] CPU #3: 1.11 kH/s
[2020-04-12 15:47:27] CPU #4: 1.85 kH/s
[2020-04-12 15:47:27] Total: 8.14 kH/s
[2020-04-12 15:47:33] CPU #5: 1.84 kH/s
[2020-04-12 15:47:33] Total: 8.13 kH/s
[2020-04-12 15:47:36] Total: 8.14 kH/s
[2020-04-12 15:47:37] CPU #2: 1.11 kH/s
[2020-04-12 15:47:37] CPU #1: 1.11 kH/s
[2020-04-12 15:47:37] CPU #0: 1.11 kH/s
[2020-04-12 15:47:37] CPU #3: 1.11 kH/s
[2020-04-12 15:47:37] CPU #4: 1.85 kH/s
[2020-04-12 15:47:37] Total: 8.14 kH/s
[2020-04-12 15:47:43] CPU #5: 1.84 kH/s
[2020-04-12 15:47:43] Total: 8.12 kH/s
[2020-04-12 15:47:46] Total: 8.13 kH/s
[2020-04-12 15:47:47] CPU #2: 1.11 kH/s
[2020-04-12 15:47:47] CPU #0: 1.11 kH/s
[2020-04-12 15:47:47] CPU #1: 1.11 kH/s
[2020-04-12 15:47:47] CPU #3: 1.11 kH/s
[2020-04-12 15:47:47] CPU #4: 1.84 kH/s
[2020-04-12 15:47:47] Total: 8.14 kH/s
[2020-04-12 15:47:53] CPU #5: 1.82 kH/s
[2020-04-12 15:47:53] Total: 8.10 kH/s
[2020-04-12 15:47:57] Total: 8.13 kH/s
[2020-04-12 15:47:57] CPU #2: 1.11 kH/s
[2020-04-12 15:47:57] CPU #0: 1.12 kH/s
[2020-04-12 15:47:57] CPU #1: 1.11 kH/s
[2020-04-12 15:47:57] CPU #3: 1.11 kH/s
[2020-04-12 15:47:57] CPU #4: 1.85 kH/s
[2020-04-12 15:48:02] CPU #5: 1.84 kH/s
[2020-04-12 15:48:02] Total: 8.13 kH/s
[2020-04-12 15:48:07] Total: 8.11 kH/s
[2020-04-12 15:48:07] CPU #2: 1.11 kH/s
[2020-04-12 15:48:07] CPU #0: 1.11 kH/s
[2020-04-12 15:48:07] CPU #1: 1.10 kH/s
[2020-04-12 15:48:07] CPU #3: 1.11 kH/s
[2020-04-12 15:48:07] CPU #4: 1.84 kH/s
[2020-04-12 15:48:12] CPU #5: 1.85 kH/s
[2020-04-12 15:48:12] Total: 8.13 kH/s
[2020-04-12 15:48:17] Total: 8.14 kH/s
[2020-04-12 15:48:17] CPU #2: 1.11 kH/s
[2020-04-12 15:48:17] CPU #0: 1.12 kH/s
[2020-04-12 15:48:17] CPU #1: 1.11 kH/s
[2020-04-12 15:48:17] CPU #3: 1.11 kH/s
[2020-04-12 15:48:17] CPU #4: 1.85 kH/s
[2020-04-12 15:48:22] CPU #5: 1.85 kH/s
[2020-04-12 15:48:22] Total: 8.14 kH/s
[2020-04-12 15:48:27] Total: 8.08 kH/s
[2020-04-12 15:48:27] CPU #2: 1.11 kH/s
[2020-04-12 15:48:27] CPU #0: 1.11 kH/s
[2020-04-12 15:48:27] CPU #1: 1.11 kH/s
[2020-04-12 15:48:27] CPU #3: 1.11 kH/s
[2020-04-12 15:48:27] CPU #4: 1.84 kH/s
[2020-04-12 15:48:32] CPU #5: 1.85 kH/s
[2020-04-12 15:48:32] Total: 8.13 kH/s
[2020-04-12 15:48:37] Total: 8.13 kH/s
[2020-04-12 15:48:37] CPU #2: 1.11 kH/s
[2020-04-12 15:48:37] CPU #0: 1.11 kH/s
[2020-04-12 15:48:37] CPU #3: 1.11 kH/s
[2020-04-12 15:48:37] CPU #1: 1.11 kH/s
[2020-04-12 15:48:37] CPU #4: 1.85 kH/s
[2020-04-12 15:48:42] CPU #5: 1.84 kH/s
[2020-04-12 15:48:42] Total: 8.14 kH/s
[2020-04-12 15:48:47] Total: 8.14 kH/s
[2020-04-12 15:48:47] CPU #2: 1.11 kH/s
[2020-04-12 15:48:47] CPU #0: 1.11 kH/s
[2020-04-12 15:48:47] CPU #3: 1.11 kH/s
[2020-04-12 15:48:47] CPU #1: 1.10 kH/s
[2020-04-12 15:48:47] CPU #4: 1.84 kH/s
[2020-04-12 15:48:52] CPU #5: 1.81 kH/s
[2020-04-12 15:48:52] Total: 8.09 kH/s
[2020-04-12 15:48:57] Total: 8.13 kH/s
[2020-04-12 15:48:57] CPU #2: 1.11 kH/s
[2020-04-12 15:48:57] CPU #0: 1.12 kH/s
[2020-04-12 15:48:57] CPU #3: 1.11 kH/s
[2020-04-12 15:48:57] CPU #1: 1.11 kH/s
[2020-04-12 15:48:57] CPU #4: 1.85 kH/s
[2020-04-12 15:49:02] CPU #5: 1.84 kH/s
[2020-04-12 15:49:02] Total: 8.13 kH/s
[2020-04-12 15:49:07] Total: 8.13 kH/s
[2020-04-12 15:49:07] CPU #2: 1.11 kH/s
[2020-04-12 15:49:07] CPU #0: 1.11 kH/s
[2020-04-12 15:49:07] CPU #3: 1.11 kH/s
[2020-04-12 15:49:07] CPU #1: 1.11 kH/s
[2020-04-12 15:49:07] CPU #4: 1.85 kH/s
[2020-04-12 15:49:12] CPU #5: 1.83 kH/s
[2020-04-12 15:49:12] Total: 8.13 kH/s
[2020-04-12 15:49:17] Total: 8.13 kH/s
[2020-04-12 15:49:17] CPU #2: 1.11 kH/s
[2020-04-12 15:49:17] CPU #0: 1.11 kH/s
[2020-04-12 15:49:17] CPU #3: 1.11 kH/s
[2020-04-12 15:49:17] CPU #1: 1.11 kH/s
[2020-04-12 15:49:17] CPU #4: 1.85 kH/s
[2020-04-12 15:49:22] CPU #5: 1.84 kH/s
[2020-04-12 15:49:22] Total: 8.14 kH/s
[2020-04-12 15:49:27] Total: 8.10 kH/s
[2020-04-12 15:49:27] CPU #2: 1.11 kH/s
[2020-04-12 15:49:27] CPU #0: 1.11 kH/s
[2020-04-12 15:49:27] CPU #3: 1.11 kH/s
[2020-04-12 15:49:27] CPU #1: 1.11 kH/s
[2020-04-12 15:49:27] CPU #4: 1.85 kH/s
[2020-04-12 15:49:32] CPU #5: 1.83 kH/s
[2020-04-12 15:49:32] Total: 8.13 kH/s
[2020-04-12 15:49:37] Total: 8.12 kH/s
[2020-04-12 15:49:37] CPU #0: 1.11 kH/s
[2020-04-12 15:49:37] CPU #2: 1.11 kH/s
[2020-04-12 15:49:37] CPU #3: 1.11 kH/s
[2020-04-12 15:49:37] CPU #1: 1.11 kH/s
[2020-04-12 15:49:37] CPU #4: 1.85 kH/s
[2020-04-12 15:49:42] CPU #5: 1.84 kH/s
[2020-04-12 15:49:42] Total: 8.14 kH/s
[2020-04-12 15:49:47] Total: 8.14 kH/s
Total Scores: 8.14,8.13,8.12,8.11,8.10,8.09,8.08,8.07,8.06,8.04,8.00,7.94,7.72
##########################################################################
Testing clockspeeds again. System health now:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:49:32: 1800/1416MHz 6.08 100% 0% 99% 0% 0% 0% 57.8°C
Checking cpufreq OPP for cpu0-cpu3:
Cpufreq OPP: 1416 Measured: 1412.739/1412.894/1412.353
Cpufreq OPP: 1200 Measured: 1196.759/1196.786/1196.925
Cpufreq OPP: 1008 Measured: 1004.930/1004.942/1005.199
Cpufreq OPP: 816 Measured: 809.100/808.388/809.734
Cpufreq OPP: 600 Measured: 593.243/590.389/594.230
Cpufreq OPP: 408 Measured: 403.255/401.933/402.952
Checking cpufreq OPP for cpu4-cpu5:
Cpufreq OPP: 1800 Measured: 934.319/946.800/2060.959
Cpufreq OPP: 1608 Measured: 1602.728/1601.734/1597.869
Cpufreq OPP: 1416 Measured: 1410.779/1409.978/1411.535
Cpufreq OPP: 1200 Measured: 935.258/768.315/1196.537
Cpufreq OPP: 1008 Measured: 1002.638/484.971/1004.551
Cpufreq OPP: 816 Measured: 813.280/401.960/563.550
Cpufreq OPP: 600 Measured: 597.533/597.431/597.000
Cpufreq OPP: 408 Measured: 403.819/130.844/130.873
##########################################################################
System health while running tinymembench:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:15:40: 1800/1416MHz 0.29 4% 0% 2% 0% 0% 0% 37.5°C
15:17:40: 1800/1416MHz 0.97 16% 0% 16% 0% 0% 0% 38.8°C
15:19:40: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 38.8°C
15:21:40: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 43.9°C
15:23:40: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 43.9°C
15:25:41: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 42.8°C
15:27:41: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 41.7°C
System health while running OpenSSL benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:28:51: 1800/1416MHz 1.00 10% 0% 10% 0% 0% 0% 43.3°C
15:29:01: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 41.7°C
15:29:11: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 43.9°C
15:29:21: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 43.9°C
15:29:31: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 41.7°C
15:29:41: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 41.7°C
15:29:51: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 45.0°C
15:30:01: 1800/1416MHz 1.00 17% 0% 17% 0% 0% 0% 49.4°C
15:30:11: 1800/1416MHz 1.00 20% 0% 19% 0% 0% 0% 42.8°C
15:30:21: 1800/1416MHz 1.00 17% 0% 16% 0% 0% 0% 44.4°C
15:30:32: 1800/1416MHz 1.00 16% 0% 16% 0% 0% 0% 46.2°C
System health while running 7-zip single core benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:30:39: 1800/1416MHz 1.00 11% 0% 10% 0% 0% 0% 46.9°C
15:31:39: 1800/1416MHz 2.88 16% 0% 16% 0% 0% 0% 40.6°C
15:32:39: 1800/1416MHz 4.00 16% 0% 16% 0% 0% 0% 40.6°C
15:33:39: 1800/1416MHz 3.91 16% 0% 16% 0% 0% 0% 40.0°C
15:34:39: 1800/1416MHz 4.14 16% 0% 16% 0% 0% 0% 40.0°C
15:35:39: 1800/1416MHz 3.69 16% 0% 16% 0% 0% 0% 40.6°C
15:36:39: 1800/1416MHz 3.54 16% 0% 16% 0% 0% 0% 40.6°C
15:37:39: 1800/1416MHz 4.67 16% 0% 16% 0% 0% 0% 42.8°C
15:38:40: 1800/1416MHz 4.32 16% 0% 16% 0% 0% 0% 45.0°C
15:39:40: 1800/1416MHz 4.49 17% 0% 16% 0% 0% 0% 43.9°C
15:40:40: 1800/1416MHz 4.16 17% 0% 16% 0% 0% 0% 45.6°C
System health while running 7-zip multi core benchmark:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:41:13: 1800/1416MHz 4.81 12% 0% 12% 0% 0% 0% 43.9°C
15:41:34: 1800/1416MHz 5.20 78% 1% 77% 0% 0% 0% 54.4°C
15:41:54: 1800/1416MHz 5.26 80% 1% 79% 0% 0% 0% 50.6°C
15:42:15: 1800/1416MHz 5.52 91% 1% 89% 0% 0% 0% 55.6°C
15:42:35: 1800/1416MHz 5.35 79% 1% 78% 0% 0% 0% 57.8°C
15:42:56: 1800/1416MHz 5.45 84% 1% 82% 0% 0% 0% 48.1°C
15:43:17: 1800/1416MHz 5.68 84% 1% 82% 0% 0% 0% 47.5°C
15:43:37: 1800/1416MHz 5.56 88% 1% 86% 0% 0% 0% 48.1°C
15:43:57: 1800/1416MHz 5.57 81% 1% 80% 0% 0% 0% 50.6°C
15:44:17: 1800/1416MHz 5.02 80% 1% 78% 0% 0% 0% 45.6°C
15:44:38: 1800/1416MHz 5.51 96% 2% 94% 0% 0% 0% 48.8°C
System health while running cpuminer:
Time big.LITTLE load %cpu %sys %usr %nice %io %irq Temp
15:44:48: 1800/1416MHz 5.42 19% 0% 18% 0% 0% 0% 48.1°C
15:45:10: 1800/1416MHz 5.59 99% 0% 99% 0% 0% 0% 54.4°C
15:45:31: 1800/1416MHz 5.71 100% 0% 99% 0% 0% 0% 50.6°C
15:45:53: 1800/1416MHz 5.87 100% 0% 99% 0% 0% 0% 48.8°C
15:46:15: 1800/1416MHz 5.90 100% 0% 99% 0% 0% 0% 50.0°C
15:46:37: 1800/1416MHz 5.93 100% 0% 99% 0% 0% 0% 51.7°C
15:46:59: 1800/1416MHz 6.01 100% 0% 99% 0% 0% 0% 56.1°C
15:47:21: 1800/1416MHz 6.38 100% 0% 99% 0% 0% 0% 57.8°C
15:47:42: 1800/1416MHz 6.27 100% 0% 99% 0% 0% 0% 51.7°C
15:48:04: 1800/1416MHz 6.24 100% 0% 99% 0% 0% 0% 48.8°C
15:48:26: 1800/1416MHz 6.17 100% 0% 99% 0% 0% 0% 49.4°C
15:48:48: 1800/1416MHz 6.17 100% 0% 99% 0% 0% 0% 49.4°C
15:49:10: 1800/1416MHz 6.12 100% 0% 99% 0% 0% 0% 53.3°C
15:49:32: 1800/1416MHz 6.08 100% 0% 99% 0% 0% 0% 57.8°C
##########################################################################
Linux 4.4.167-1213-rockchip-ayufan-g34ae07687fce (nas) 04/12/20 _aarch64_ (6 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
27.53 0.00 0.51 0.22 0.00 71.73
Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn
mtdblock0 0.02 0.08 0.00 216 0
mtdblock1 0.00 0.01 0.00 40 0
mtdblock2 0.01 0.03 0.00 72 0
mmcblk0 3.86 117.39 147.82 322667 406296
total used free shared buff/cache available
Mem: 3.8G 159M 3.2G 34M 399M 3.6G
Swap: 0B 0B 0B
Architecture: aarch64
Byte Order: Little Endian
CPU(s): 6
On-line CPU(s) list: 0-5
Thread(s) per core: 1
Core(s) per socket: 3
Socket(s): 2
Model: 4
CPU max MHz: 1416.0000
CPU min MHz: 408.0000
BogoMIPS: 48.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32
Of course, I am also interested in network performance since this thing is supposed to drive a NAS in the near future where data throughput matters. So, I have been turning to iperf3 as a tool for measuring network throughput between Linux servers. Using the RockPro64 as a server and a Raspberry Pi 4 on the same 1Gbs Ethernet network I obtain these results. Around 111 MBytes/sec bandwidth looks right!
root@nas:~# iperf3 -s -f M ----------------------------------------------------------- Server listening on 5201 ----------------------------------------------------------- Accepted connection from 192.168.1.212, port 57634 [ 5] local 192.168.1.31 port 5201 connected to 192.168.1.212 port 57636 [ ID] Interval Transfer Bandwidth [ 5] 0.00-1.00 sec 108 MBytes 108 MBytes/sec [ 5] 1.00-2.00 sec 112 MBytes 112 MBytes/sec [ 5] 2.00-3.00 sec 112 MBytes 112 MBytes/sec [ 5] 3.00-4.00 sec 112 MBytes 112 MBytes/sec [ 5] 4.00-5.00 sec 112 MBytes 112 MBytes/sec [ 5] 5.00-6.00 sec 111 MBytes 111 MBytes/sec [ 5] 6.00-7.00 sec 112 MBytes 112 MBytes/sec [ 5] 7.00-8.00 sec 112 MBytes 112 MBytes/sec [ 5] 8.00-9.00 sec 110 MBytes 110 MBytes/sec [ 5] 9.00-10.00 sec 109 MBytes 109 MBytes/sec [ 5] 10.00-10.03 sec 3.24 MBytes 109 MBytes/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth [ 5] 0.00-10.03 sec 0.00 Bytes 0.00 MBytes/sec sender [ 5] 0.00-10.03 sec 1.09 GBytes 111 MBytes/sec receiver