Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- root@DietPi:~/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 : 1174.9 MB/s
- C copy backwards (32 byte blocks) : 1156.7 MB/s
- C copy backwards (64 byte blocks) : 2357.5 MB/s
- C copy : 1206.3 MB/s
- C copy prefetched (32 bytes step) : 1445.4 MB/s
- C copy prefetched (64 bytes step) : 1444.6 MB/s
- C 2-pass copy : 1127.6 MB/s
- C 2-pass copy prefetched (32 bytes step) : 1393.6 MB/s
- C 2-pass copy prefetched (64 bytes step) : 1392.8 MB/s
- C fill : 4911.1 MB/s (0.5%)
- C fill (shuffle within 16 byte blocks) : 1843.4 MB/s
- C fill (shuffle within 32 byte blocks) : 1842.7 MB/s
- C fill (shuffle within 64 byte blocks) : 1948.3 MB/s
- ---
- standard memcpy : 2294.8 MB/s
- standard memset : 4890.4 MB/s (1.0%)
- ---
- NEON read : 3382.3 MB/s
- NEON read prefetched (32 bytes step) : 4271.6 MB/s
- NEON read prefetched (64 bytes step) : 4284.7 MB/s
- NEON read 2 data streams : 3457.8 MB/s
- NEON read 2 data streams prefetched (32 bytes step) : 4434.3 MB/s
- NEON read 2 data streams prefetched (64 bytes step) : 4440.4 MB/s
- NEON copy : 2623.3 MB/s
- NEON copy prefetched (32 bytes step) : 2921.7 MB/s
- NEON copy prefetched (64 bytes step) : 2913.1 MB/s
- NEON unrolled copy : 2251.2 MB/s
- NEON unrolled copy prefetched (32 bytes step) : 3250.0 MB/s
- NEON unrolled copy prefetched (64 bytes step) : 3271.0 MB/s
- NEON copy backwards : 1218.2 MB/s
- NEON copy backwards prefetched (32 bytes step) : 1427.6 MB/s
- NEON copy backwards prefetched (64 bytes step) : 1427.4 MB/s
- NEON 2-pass copy : 2109.2 MB/s
- NEON 2-pass copy prefetched (32 bytes step) : 2174.2 MB/s
- NEON 2-pass copy prefetched (64 bytes step) : 2174.1 MB/s
- NEON unrolled 2-pass copy : 1398.4 MB/s
- NEON unrolled 2-pass copy prefetched (32 bytes step) : 1753.3 MB/s
- NEON unrolled 2-pass copy prefetched (64 bytes step) : 1772.1 MB/s
- NEON fill : 4885.1 MB/s (0.8%)
- NEON fill backwards : 1845.6 MB/s
- VFP copy : 2478.7 MB/s
- VFP 2-pass copy : 1336.9 MB/s
- ARM fill (STRD) : 4894.4 MB/s (0.8%)
- ARM fill (STM with 8 registers) : 4903.0 MB/s
- ARM fill (STM with 4 registers) : 4913.2 MB/s
- ARM copy prefetched (incr pld) : 3061.7 MB/s
- ARM copy prefetched (wrap pld) : 2883.5 MB/s
- ARM 2-pass copy prefetched (incr pld) : 1709.2 MB/s
- ARM 2-pass copy prefetched (wrap pld) : 1665.6 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 : 4.4 ns / 6.7 ns
- 131072 : 6.7 ns / 8.9 ns
- 262144 : 9.6 ns / 11.9 ns
- 524288 : 11.1 ns / 13.5 ns
- 1048576 : 12.0 ns / 14.5 ns
- 2097152 : 19.7 ns / 27.1 ns
- 4194304 : 95.4 ns / 143.8 ns
- 8388608 : 134.2 ns / 183.5 ns
- 16777216 : 154.1 ns / 200.2 ns
- 33554432 : 169.4 ns / 219.5 ns
- 67108864 : 179.0 ns / 232.5 ns
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement