Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- Exynos 5 Octa (Odroid-XU)
- tinymembench v0.3.9 (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 : 1408.9 MB/s
- C copy : 1458.0 MB/s
- C copy prefetched (32 bytes step) : 1632.0 MB/s
- C copy prefetched (64 bytes step) : 1632.7 MB/s
- C 2-pass copy : 1299.5 MB/s (0.6%)
- C 2-pass copy prefetched (32 bytes step) : 1555.9 MB/s (0.7%)
- C 2-pass copy prefetched (64 bytes step) : 1556.3 MB/s
- C fill : 2865.8 MB/s
- ---
- standard memcpy : 2382.8 MB/s
- standard memset : 2861.9 MB/s (0.8%)
- ---
- NEON read : 3765.0 MB/s
- NEON read prefetched (32 bytes step) : 4661.8 MB/s
- NEON read prefetched (64 bytes step) : 4673.9 MB/s
- NEON read 2 data streams : 3809.4 MB/s
- NEON read 2 data streams prefetched (32 bytes step) : 4701.5 MB/s
- NEON read 2 data streams prefetched (64 bytes step) : 4702.1 MB/s
- NEON copy : 2366.7 MB/s
- NEON copy prefetched (32 bytes step) : 2422.4 MB/s (0.7%)
- NEON copy prefetched (64 bytes step) : 2404.0 MB/s
- NEON unrolled copy : 2197.9 MB/s
- NEON unrolled copy prefetched (32 bytes step) : 2523.3 MB/s
- NEON unrolled copy prefetched (64 bytes step) : 2525.9 MB/s
- NEON copy backwards : 1449.5 MB/s
- NEON copy backwards prefetched (32 bytes step) : 1625.0 MB/s
- NEON copy backwards prefetched (64 bytes step) : 1624.6 MB/s
- NEON 2-pass copy : 1690.0 MB/s
- NEON 2-pass copy prefetched (32 bytes step) : 1793.1 MB/s (0.5%)
- NEON 2-pass copy prefetched (64 bytes step) : 1810.1 MB/s (0.5%)
- NEON unrolled 2-pass copy : 1482.5 MB/s
- NEON unrolled 2-pass copy prefetched (32 bytes step) : 1825.5 MB/s
- NEON unrolled 2-pass copy prefetched (64 bytes step) : 1851.7 MB/s
- NEON fill : 2857.2 MB/s (0.8%)
- NEON fill backwards : 2156.5 MB/s
- VFP copy : 2373.5 MB/s
- VFP 2-pass copy : 1486.1 MB/s
- ARM fill (STRD) : 2859.6 MB/s (0.4%)
- ARM fill (STM with 8 registers) : 2860.0 MB/s (0.9%)
- ARM fill (STM with 4 registers) : 2855.3 MB/s (0.2%)
- ARM copy prefetched (incr pld) : 2471.9 MB/s
- ARM copy prefetched (wrap pld) : 2422.7 MB/s
- ARM 2-pass copy prefetched (incr pld) : 1739.1 MB/s
- ARM 2-pass copy prefetched (wrap pld) : 1693.7 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 / 0.1 ns
- 32768 : 0.0 ns / 0.1 ns
- 65536 : 5.5 ns / 8.6 ns
- 131072 : 8.4 ns / 11.3 ns
- 262144 : 12.0 ns / 14.9 ns
- 524288 : 13.8 ns / 17.0 ns
- 1048576 : 14.8 ns / 18.2 ns
- 2097152 : 19.4 ns / 24.4 ns
- 4194304 : 88.2 ns / 133.9 ns
- 8388608 : 123.7 ns / 173.3 ns
- 16777216 : 141.6 ns / 190.5 ns
- 33554432 : 158.6 ns / 214.1 ns
- 67108864 : 168.8 ns / 229.5 ns
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement