Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- 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 : 267.2 MB/s
- C copy backwards (32 byte blocks) : 1030.5 MB/s
- C copy backwards (64 byte blocks) : 977.2 MB/s
- C copy : 958.0 MB/s
- C copy prefetched (32 bytes step) : 875.4 MB/s
- C copy prefetched (64 bytes step) : 988.5 MB/s
- C 2-pass copy : 728.3 MB/s
- C 2-pass copy prefetched (32 bytes step) : 698.6 MB/s
- C 2-pass copy prefetched (64 bytes step) : 727.6 MB/s
- C fill : 3351.5 MB/s (0.1%)
- C fill (shuffle within 16 byte blocks) : 3337.1 MB/s
- C fill (shuffle within 32 byte blocks) : 339.8 MB/s
- C fill (shuffle within 64 byte blocks) : 345.5 MB/s
- ---
- standard memcpy : 986.4 MB/s (0.1%)
- standard memset : 2552.1 MB/s
- ---
- NEON read : 1275.5 MB/s
- NEON read prefetched (32 bytes step) : 1402.5 MB/s
- NEON read prefetched (64 bytes step) : 1451.8 MB/s
- NEON read 2 data streams : 359.6 MB/s
- NEON read 2 data streams prefetched (32 bytes step) : 686.1 MB/s
- NEON read 2 data streams prefetched (64 bytes step) : 735.2 MB/s
- NEON copy : 978.5 MB/s
- NEON copy prefetched (32 bytes step) : 1060.0 MB/s
- NEON copy prefetched (64 bytes step) : 1129.4 MB/s
- NEON unrolled copy : 1066.0 MB/s
- NEON unrolled copy prefetched (32 bytes step) : 933.3 MB/s
- NEON unrolled copy prefetched (64 bytes step) : 1059.3 MB/s
- NEON copy backwards : 1041.5 MB/s
- NEON copy backwards prefetched (32 bytes step) : 1041.2 MB/s
- NEON copy backwards prefetched (64 bytes step) : 1074.2 MB/s
- NEON 2-pass copy : 814.4 MB/s
- NEON 2-pass copy prefetched (32 bytes step) : 884.2 MB/s
- NEON 2-pass copy prefetched (64 bytes step) : 906.5 MB/s
- NEON unrolled 2-pass copy : 707.5 MB/s
- NEON unrolled 2-pass copy prefetched (32 bytes step) : 668.3 MB/s
- NEON unrolled 2-pass copy prefetched (64 bytes step) : 707.3 MB/s
- NEON fill : 3355.7 MB/s (0.2%)
- NEON fill backwards : 3335.6 MB/s
- VFP copy : 1070.8 MB/s
- VFP 2-pass copy : 719.4 MB/s
- ARM fill (STRD) : 2552.5 MB/s
- ARM fill (STM with 8 registers) : 3339.0 MB/s
- ARM fill (STM with 4 registers) : 3021.9 MB/s
- ARM copy prefetched (incr pld) : 1093.1 MB/s
- ARM copy prefetched (wrap pld) : 1066.5 MB/s
- ARM 2-pass copy prefetched (incr pld) : 791.2 MB/s
- ARM 2-pass copy prefetched (wrap pld) : 776.5 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 : 6.2 ns / 10.8 ns
- 131072 : 9.6 ns / 15.1 ns
- 262144 : 11.3 ns / 16.7 ns
- 524288 : 13.5 ns / 19.0 ns
- 1048576 : 89.2 ns / 137.4 ns
- 2097152 : 134.0 ns / 181.2 ns
- 4194304 : 157.1 ns / 197.1 ns
- 8388608 : 170.7 ns / 206.4 ns
- 16777216 : 180.4 ns / 215.2 ns
- 33554432 : 191.0 ns / 230.2 ns
- 67108864 : 204.3 ns / 254.3 ns
- root@orangepipc:/usr/local/src/tinymembench# cat /proc/cpuinfo
- processor : 0
- model name : ARMv7 Processor rev 5 (v7l)
- BogoMIPS : 48.00
- Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm
- CPU implementer : 0x41
- CPU architecture: 7
- CPU variant : 0x0
- CPU part : 0xc07
- CPU revision : 5
- processor : 1
- model name : ARMv7 Processor rev 5 (v7l)
- BogoMIPS : 48.00
- Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm
- CPU implementer : 0x41
- CPU architecture: 7
- CPU variant : 0x0
- CPU part : 0xc07
- CPU revision : 5
- processor : 2
- model name : ARMv7 Processor rev 5 (v7l)
- BogoMIPS : 48.00
- Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm
- CPU implementer : 0x41
- CPU architecture: 7
- CPU variant : 0x0
- CPU part : 0xc07
- CPU revision : 5
- processor : 3
- model name : ARMv7 Processor rev 5 (v7l)
- BogoMIPS : 48.00
- Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm
- CPU implementer : 0x41
- CPU architecture: 7
- CPU variant : 0x0
- CPU part : 0xc07
- CPU revision : 5
- Hardware : Allwinner sun8i Family
- Revision : 0000
- Serial : 0200462044000000
- root@orangepipc:/usr/local/src/tinymembench# uname -a
- Linux orangepipc 4.6.0-rc1-sunxi #10 SMP Sat Apr 2 16:19:47 CEST 2016 armv7l GNU/Linux
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
