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 : 2084.8 MB/s (1.4%)
C copy backwards (32 byte blocks) : 2077.7 MB/s (1.0%)
C copy backwards (64 byte blocks) : 2120.8 MB/s (1.7%)
C copy : 2160.7 MB/s (2.0%)
C copy prefetched (32 bytes step) : 1567.8 MB/s (0.8%)
C copy prefetched (64 bytes step) : 1676.1 MB/s (0.1%)
C 2-pass copy : 1661.0 MB/s
C 2-pass copy prefetched (32 bytes step) : 994.4 MB/s
C 2-pass copy prefetched (64 bytes step) : 1397.5 MB/s
C fill : 5846.3 MB/s
C fill (shuffle within 16 byte blocks) : 5847.8 MB/s
C fill (shuffle within 32 byte blocks) : 5846.9 MB/s
C fill (shuffle within 64 byte blocks) : 5846.4 MB/s
---
standard memcpy : 2232.0 MB/s (0.4%)
standard memset : 5854.1 MB/s
---
NEON LDP/STP copy : 2232.4 MB/s
NEON LDP/STP copy pldl2strm (32 bytes step) : 1305.4 MB/s (0.6%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 1902.3 MB/s (0.5%)
NEON LDP/STP copy pldl1keep (32 bytes step) : 2628.7 MB/s
NEON LDP/STP copy pldl1keep (64 bytes step) : 2644.5 MB/s
NEON LD1/ST1 copy : 2183.3 MB/s (0.6%)
NEON STP fill : 5842.7 MB/s
NEON STNP fill : 5791.7 MB/s (0.1%)
ARM LDP/STP copy : 2234.7 MB/s (0.1%)
ARM STP fill : 5848.6 MB/s
ARM STNP fill : 5781.1 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, [MADV_NOHUGEPAGE]
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 : 4.1 ns / 7.5 ns
131072 : 6.4 ns / 10.7 ns
262144 : 8.2 ns / 13.5 ns
524288 : 86.2 ns / 136.7 ns
1048576 : 137.1 ns / 185.2 ns
2097152 : 164.9 ns / 202.1 ns
4194304 : 179.2 ns / 208.4 ns
8388608 : 186.1 ns / 211.3 ns
16777216 : 189.5 ns / 212.9 ns
33554432 : 191.4 ns / 214.0 ns
67108864 : 201.0 ns / 232.5 ns
block size : single random read / dual random read, [MADV_HUGEPAGE]
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 : 4.1 ns / 7.3 ns
131072 : 6.3 ns / 10.9 ns
262144 : 8.1 ns / 13.4 ns
524288 : 86.1 ns / 136.8 ns
1048576 : 137.0 ns / 185.2 ns
2097152 : 164.0 ns / 201.3 ns
4194304 : 177.8 ns / 207.1 ns
8388608 : 184.8 ns / 209.4 ns
16777216 : 188.4 ns / 210.5 ns
33554432 : 190.0 ns / 211.0 ns
67108864 : 190.9 ns / 211.1 ns