Advertisement
Guest User

OPi Zero Plus with preliminary Armbian

a guest
Nov 16th, 2017
407
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 9.08 KB | None | 0 0
  1. Orange Pi Zero Plus with preliminary Armbian Xenial arm64 image (64-bit kernel 4.13.13, max cpufreq limited to 1008 MHz and clocking DRAM at 408 MHz):
  2.  
  3. root@orangepizeroplus:~/tinymembench# ./tinymembench
  4. tinymembench v0.4.9 (simple benchmark for memory throughput and latency)
  5.  
  6. ==========================================================================
  7. == Memory bandwidth tests ==
  8. == ==
  9. == Note 1: 1MB = 1000000 bytes ==
  10. == Note 2: Results for 'copy' tests show how many bytes can be ==
  11. == copied per second (adding together read and writen ==
  12. == bytes would have provided twice higher numbers) ==
  13. == Note 3: 2-pass copy means that we are using a small temporary buffer ==
  14. == to first fetch data into it, and only then write it to the ==
  15. == destination (source -> L1 cache, L1 cache -> destination) ==
  16. == Note 4: If sample standard deviation exceeds 0.1%, it is shown in ==
  17. == brackets ==
  18. ==========================================================================
  19.  
  20. C copy backwards : 632.3 MB/s (0.3%)
  21. C copy backwards (32 byte blocks) : 636.3 MB/s
  22. C copy backwards (64 byte blocks) : 625.3 MB/s (0.8%)
  23. C copy : 630.5 MB/s (0.4%)
  24. C copy prefetched (32 bytes step) : 484.3 MB/s
  25. C copy prefetched (64 bytes step) : 521.4 MB/s
  26. C 2-pass copy : 623.2 MB/s
  27. C 2-pass copy prefetched (32 bytes step) : 451.3 MB/s (0.3%)
  28. C 2-pass copy prefetched (64 bytes step) : 256.3 MB/s
  29. C fill : 1552.3 MB/s
  30. C fill (shuffle within 16 byte blocks) : 1552.8 MB/s
  31. C fill (shuffle within 32 byte blocks) : 1553.1 MB/s
  32. C fill (shuffle within 64 byte blocks) : 1552.9 MB/s
  33. ---
  34. standard memcpy : 634.8 MB/s (0.4%)
  35. standard memset : 1553.0 MB/s
  36. ---
  37. NEON LDP/STP copy : 638.4 MB/s (0.3%)
  38. NEON LDP/STP copy pldl2strm (32 bytes step) : 477.6 MB/s (0.3%)
  39. NEON LDP/STP copy pldl2strm (64 bytes step) : 552.1 MB/s
  40. NEON LDP/STP copy pldl1keep (32 bytes step) : 659.5 MB/s
  41. NEON LDP/STP copy pldl1keep (64 bytes step) : 659.6 MB/s
  42. NEON LD1/ST1 copy : 639.6 MB/s (0.3%)
  43. NEON STP fill : 1550.8 MB/s
  44. NEON STNP fill : 1399.7 MB/s (0.2%)
  45. ARM LDP/STP copy : 638.3 MB/s (0.3%)
  46. ARM STP fill : 1551.0 MB/s
  47. ARM STNP fill : 1399.1 MB/s (0.2%)
  48.  
  49. ==========================================================================
  50. == Memory latency test ==
  51. == ==
  52. == Average time is measured for random memory accesses in the buffers ==
  53. == of different sizes. The larger is the buffer, the more significant ==
  54. == are relative contributions of TLB, L1/L2 cache misses and SDRAM ==
  55. == accesses. For extremely large buffer sizes we are expecting to see ==
  56. == page table walk with several requests to SDRAM for almost every ==
  57. == memory access (though 64MiB is not nearly large enough to experience ==
  58. == this effect to its fullest). ==
  59. == ==
  60. == Note 1: All the numbers are representing extra time, which needs to ==
  61. == be added to L1 cache latency. The cycle timings for L1 cache ==
  62. == latency can be usually found in the processor documentation. ==
  63. == Note 2: Dual random read means that we are simultaneously performing ==
  64. == two independent memory accesses at a time. In the case if ==
  65. == the memory subsystem can't handle multiple outstanding ==
  66. == requests, dual random read has the same timings as two ==
  67. == single reads performed one after another. ==
  68. ==========================================================================
  69.  
  70. block size : single random read / dual random read, [MADV_NOHUGEPAGE]
  71. 1024 : 0.0 ns / 0.0 ns
  72. 2048 : 0.0 ns / 0.0 ns
  73. 4096 : 0.0 ns / 0.0 ns
  74. 8192 : 0.0 ns / 0.0 ns
  75. 16384 : 0.0 ns / 0.0 ns
  76. 32768 : 0.0 ns / 0.0 ns
  77. 65536 : 6.8 ns / 11.4 ns
  78. 131072 : 10.4 ns / 15.9 ns
  79. 262144 : 12.3 ns / 17.7 ns
  80. 524288 : 14.6 ns / 20.5 ns
  81. 1048576 : 127.4 ns / 199.7 ns
  82. 2097152 : 188.1 ns / 261.6 ns
  83. 4194304 : 225.6 ns / 291.9 ns
  84. 8388608 : 244.8 ns / 305.7 ns
  85. 16777216 : 256.3 ns / 313.5 ns
  86. 33554432 : 263.5 ns / 318.1 ns
  87. 67108864 : 267.7 ns / 322.4 ns
  88.  
  89. block size : single random read / dual random read, [MADV_HUGEPAGE]
  90. 1024 : 0.0 ns / 0.0 ns
  91. 2048 : 0.0 ns / 0.0 ns
  92. 4096 : 0.0 ns / 0.0 ns
  93. 8192 : 0.0 ns / 0.0 ns
  94. 16384 : 0.0 ns / 0.0 ns
  95. 32768 : 0.0 ns / 0.0 ns
  96. 65536 : 6.8 ns / 11.6 ns
  97. 131072 : 10.4 ns / 16.2 ns
  98. 262144 : 12.3 ns / 18.2 ns
  99. 524288 : 14.6 ns / 21.2 ns
  100. 1048576 : 127.5 ns / 199.7 ns
  101. 2097152 : 188.1 ns / 261.5 ns
  102. 4194304 : 225.6 ns / 291.9 ns
  103. 8388608 : 244.8 ns / 305.7 ns
  104. 16777216 : 256.3 ns / 313.6 ns
  105. 33554432 : 263.5 ns / 318.2 ns
  106. 67108864 : 267.8 ns / 322.4 ns
  107.  
  108.  
  109. root@orangepizeroplus:~# /bin/bash /tmp/sysbench.sh
  110. 408: execution time (avg/stddev): 89.9097/0.00
  111. 648: execution time (avg/stddev): 56.5343/0.00
  112. 816: execution time (avg/stddev): 44.8736/0.00
  113. 912: execution time (avg/stddev): 40.1440/0.00
  114. 960: execution time (avg/stddev): 38.1315/0.00
  115. 1008: execution time (avg/stddev): 36.3298/0.00
  116. 408: execution time (avg/stddev): 22.4773/0.00
  117. 648: execution time (avg/stddev): 14.1447/0.00
  118. 816: execution time (avg/stddev): 11.2191/0.00
  119. 912: execution time (avg/stddev): 10.0360/0.00
  120. 960: execution time (avg/stddev): 9.5338/0.00
  121. 1008: execution time (avg/stddev): 9.0787/0.00
  122.  
  123.  
  124. root@orangepizeroplus:~# 7zr b
  125.  
  126. 7-Zip (A) 9.20 Copyright (c) 1999-2010 Igor Pavlov 2010-11-18
  127. p7zip Version 9.20 (locale=en_US.UTF-8,Utf16=on,HugeFiles=on,4 CPUs)
  128.  
  129. RAM size: 482 MB, # CPU hardware threads: 4
  130. RAM usage: 434 MB, # Benchmark threads: 4
  131.  
  132. Dict Compressing | Decompressing
  133. Speed Usage R/U Rating | Speed Usage R/U Rating
  134. KB/s % MIPS MIPS | KB/s % MIPS MIPS
  135.  
  136. 22: 1268 323 381 1234 | 36899 399 834 3329
  137. 23: 1255 330 388 1279 | 36255 399 830 3317
  138. 24: 1112 326 367 1195 | 35144 399 817 3260
  139. ----------------------------------------------------------------
  140. Avr: 326 378 1236 399 827 3302
  141. Tot: 363 603 2269
  142.  
  143.  
  144. for i in 128 192 256 ; do openssl speed -elapsed -evp aes-${i}-cbc ; done
  145. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
  146. aes-128-cbc 126791.45k 339021.63k 566736.38k 704634.20k 758407.17k
  147. aes-192-cbc 118403.43k 292092.78k 452832.09k 538759.85k 570266.97k
  148. aes-256-cbc 113400.10k 263222.98k 387456.43k 448689.15k 470237.18k
  149.  
  150.  
  151. root@orangepizeroplus:~/StabilityTester# ./stabilityTester.sh
  152. Testing frequency 648000
  153. Cooling down151 CPU Freq: 648000 CPU Core: 3300000
  154. Testing frequency 816000
  155. Cooling down: 54934 CPU Freq: 816000 CPU Core: 3300000
  156. Testing frequency 912000
  157. Cooling down: 53966 CPU Freq: 912000 CPU Core: 3300000
  158. Testing frequency 960000
  159. Cooling down: 53966 CPU Freq: 960000 CPU Core: 3300000
  160. Testing frequency 1008000
  161. Cooling down: 54208 CPU Freq: 960000 CPU Core: 3300000
  162.  
  163. Done testing stability:
  164. Frequency: 648000 Voltage: 3300000 Success: 1 Result: 0.0048034
  165. Frequency: 816000 Voltage: 3300000 Success: 1 Result: 0.0048034
  166. Frequency: 912000 Voltage: 3300000 Success: 1 Result: 0.0048034
  167. Frequency: 960000 Voltage: 3300000 Success: 1 Result: 0.0048034
  168. Frequency: 1008000 Voltage: 3300000 Success: 1 Result: 0.0048034
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement