Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- root@bpi-iot-ros-ai:~# 7za b
- 7-Zip (A) 9.20 Copyright (c) 1999-2010 Igor Pavlov 2010-11-18
- p7zip Version 9.20 (locale=en_US.UTF-8,Utf16=on,HugeFiles=on,4 CPUs)
- RAM size: 998 MB, # CPU hardware threads: 4
- RAM usage: 850 MB, # Benchmark threads: 4
- Dict Compressing | Decompressing
- Speed Usage R/U Rating | Speed Usage R/U Rating
- KB/s % MIPS MIPS | KB/s % MIPS MIPS
- 22: 1336 299 434 1300 | 36258 397 824 3271
- 23: 1314 301 444 1339 | 35334 393 823 3233
- 24: 1337 313 458 1437 | 35029 395 821 3249
- 25: 1291 317 465 1475 | 33194 383 815 3121
- ----------------------------------------------------------------
- Avr: 308 451 1388 392 821 3219
- Tot: 350 636 2303
- root@bpi-iot-ros-ai:~# sysbench --test=cpu --cpu-max-prime=20000 run --num-threads=$(grep -c '^processor' /proc/cpuinfo)
- sysbench 0.4.12: multi-threaded system evaluation benchmark
- Running the test with following options:
- Number of threads: 4
- Doing CPU performance benchmark
- Threads started!
- Done.
- Maximum prime number checked in CPU test: 20000
- Test execution summary:
- total time: 118.4670s
- total number of events: 10000
- total time taken by event execution: 473.8089
- per-request statistics:
- min: 46.72ms
- avg: 47.38ms
- max: 226.98ms
- approx. 95 percentile: 47.70ms
- Threads fairness:
- events (avg/stddev): 2500.0000/4.24
- execution time (avg/stddev): 118.4522/0.01
- root@bpi-iot-ros-ai:~# minerd --benchmark 2>&1 | grep Total
- [2017-08-05 15:01:57] Total: 2.06 khash/s
- [2017-08-05 15:02:02] Total: 2.06 khash/s
- [2017-08-05 15:02:07] Total: 2.06 khash/s
- [2017-08-05 15:02:12] Total: 2.05 khash/s
- [2017-08-05 15:02:17] Total: 2.06 khash/s
- [2017-08-05 15:02:22] Total: 2.05 khash/s
- [2017-08-05 15:02:27] Total: 2.05 khash/s
- [2017-08-05 15:02:32] Total: 2.05 khash/s
- [2017-08-05 15:02:37] Total: 2.06 khash/s
- [2017-08-05 15:02:42] Total: 2.05 khash/s
- [2017-08-05 15:02:47] Total: 2.06 khash/s
- [2017-08-05 15:02:52] Total: 2.05 khash/s
- [2017-08-05 15:02:57] Total: 2.06 khash/s
- [2017-08-05 15:03:02] Total: 2.05 khash/s
- [2017-08-05 15:03:07] Total: 2.06 khash/s
- ^C
- root@bpi-iot-ros-ai:~# /usr/local/src/tinymembench/tinymembench
- 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 : 268.7 MB/s (0.1%)
- C copy backwards (32 byte blocks) : 1024.2 MB/s
- C copy backwards (64 byte blocks) : 1029.1 MB/s (1.4%)
- C copy : 1029.3 MB/s (0.4%)
- C copy prefetched (32 bytes step) : 1112.9 MB/s (0.4%)
- C copy prefetched (64 bytes step) : 1106.1 MB/s (0.2%)
- C 2-pass copy : 787.3 MB/s (0.2%)
- C 2-pass copy prefetched (32 bytes step) : 880.5 MB/s
- C 2-pass copy prefetched (64 bytes step) : 902.0 MB/s
- C fill : 4126.0 MB/s (0.9%)
- C fill (shuffle within 16 byte blocks) : 4113.3 MB/s (0.9%)
- C fill (shuffle within 32 byte blocks) : 376.9 MB/s (3.6%)
- C fill (shuffle within 64 byte blocks) : 356.4 MB/s (1.7%)
- ---
- standard memcpy : 1110.6 MB/s (0.3%)
- standard memset : 3571.6 MB/s (0.2%)
- ---
- NEON read : 1300.5 MB/s
- NEON read prefetched (32 bytes step) : 1474.6 MB/s
- NEON read prefetched (64 bytes step) : 1476.7 MB/s
- NEON read 2 data streams : 367.2 MB/s
- NEON read 2 data streams prefetched (32 bytes step) : 706.9 MB/s
- NEON read 2 data streams prefetched (64 bytes step) : 752.3 MB/s
- NEON copy : 1049.4 MB/s (0.3%)
- NEON copy prefetched (32 bytes step) : 1034.0 MB/s (0.3%)
- NEON copy prefetched (64 bytes step) : 1124.4 MB/s (0.8%)
- NEON unrolled copy : 1040.3 MB/s (0.4%)
- NEON unrolled copy prefetched (32 bytes step) : 1055.4 MB/s (0.3%)
- NEON unrolled copy prefetched (64 bytes step) : 1093.3 MB/s (0.3%)
- NEON copy backwards : 1041.2 MB/s (0.3%)
- NEON copy backwards prefetched (32 bytes step) : 1028.5 MB/s (0.5%)
- NEON copy backwards prefetched (64 bytes step) : 1088.7 MB/s (0.3%)
- NEON 2-pass copy : 884.0 MB/s (1.2%)
- NEON 2-pass copy prefetched (32 bytes step) : 952.6 MB/s
- NEON 2-pass copy prefetched (64 bytes step) : 990.1 MB/s (0.2%)
- NEON unrolled 2-pass copy : 773.3 MB/s (0.2%)
- NEON unrolled 2-pass copy prefetched (32 bytes step) : 751.4 MB/s (0.5%)
- NEON unrolled 2-pass copy prefetched (64 bytes step) : 784.5 MB/s
- NEON fill : 3908.3 MB/s
- NEON fill backwards : 3918.2 MB/s (0.5%)
- VFP copy : 1043.8 MB/s
- VFP 2-pass copy : 777.2 MB/s
- ARM fill (STRD) : 3009.9 MB/s
- ARM fill (STM with 8 registers) : 3927.3 MB/s (0.5%)
- ARM fill (STM with 4 registers) : 3570.3 MB/s
- ARM copy prefetched (incr pld) : 1073.5 MB/s (0.2%)
- ARM copy prefetched (wrap pld) : 1081.0 MB/s (2.3%)
- ARM 2-pass copy prefetched (incr pld) : 876.7 MB/s (0.3%)
- ARM 2-pass copy prefetched (wrap pld) : 829.5 MB/s
- ==========================================================================
- == 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 read (from framebuffer) : 50.1 MB/s
- NEON copy (from framebuffer) : 49.4 MB/s
- NEON 2-pass copy (from framebuffer) : 49.3 MB/s
- NEON unrolled copy (from framebuffer) : 50.0 MB/s
- NEON 2-pass unrolled copy (from framebuffer) : 49.0 MB/s
- VFP copy (from framebuffer) : 343.2 MB/s (0.2%)
- VFP 2-pass copy (from framebuffer) : 304.6 MB/s
- ARM copy (from framebuffer) : 187.5 MB/s (0.2%)
- ARM 2-pass copy (from framebuffer) : 174.5 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 / 0.1 ns
- 65536 : 5.4 ns / 9.7 ns
- 131072 : 9.4 ns / 14.0 ns
- 262144 : 11.8 ns / 17.5 ns
- 524288 : 21.1 ns / 29.2 ns
- 1048576 : 98.0 ns / 142.2 ns
- 2097152 : 135.5 ns / 179.6 ns
- 4194304 : 156.8 ns / 191.3 ns
- 8388608 : 168.2 ns / 198.7 ns
- 16777216 : 177.1 ns / 208.3 ns
- 33554432 : 186.7 ns / 222.8 ns
- 67108864 : 201.1 ns / 248.5 ns
- root@bpi-iot-ros-ai:~# cat /sys/devices/1c62000.dramfreq/devfreq/dramfreq/cur_freq
- 576000
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement