Linksys MR7350 (Qualcomm Atheros IPQ60XX) - ssvb/tinymembench GitHub Wiki
Device: Linksys MR7350 1.0, OpenWRT SNAPSHOT
CONFIG_EXTRA_OPTIMIZATION="-pipe -O3 -mcpu=cortex-a73+crypto+aes+sha2 -mtune=cortex-a73"
# uname -a
Linux MR7350 6.1.82 #0 SMP Fri Mar 29 00:16:13 2024 aarch64 GNU/Linux
# mhz
count=600545 us50=19894 us250=100266 diff=80372 cpu_MHz=1494.413
# cat /proc/cpuinfo
...
processor : 3
BogoMIPS : 48.00
Features : fp asimd evtstrm aes pmull sha1 sha2 crc32 cpuid
CPU implementer : 0x51
CPU architecture: 8
CPU variant : 0xa
CPU part : 0x801
CPU revision : 4
# 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 : 1353.2 MB/s (0.4%)
C copy backwards (32 byte blocks) : 1351.5 MB/s (0.7%)
C copy backwards (64 byte blocks) : 1342.5 MB/s (0.9%)
C copy : 1342.8 MB/s (1.4%)
C copy prefetched (32 bytes step) : 1064.2 MB/s (1.4%)
C copy prefetched (64 bytes step) : 1161.9 MB/s (4.3%)
C 2-pass copy : 1296.1 MB/s (1.7%)
C 2-pass copy prefetched (32 bytes step) : 737.9 MB/s (0.7%)
C 2-pass copy prefetched (64 bytes step) : 750.2 MB/s (0.6%)
C fill : 3488.3 MB/s (1.5%)
C fill (shuffle within 16 byte blocks) : 3477.7 MB/s (2.5%)
C fill (shuffle within 32 byte blocks) : 3481.6 MB/s (0.7%)
C fill (shuffle within 64 byte blocks) : 3490.8 MB/s (0.8%)
---
standard memcpy : 1342.0 MB/s (1.7%)
standard memset : 3496.1 MB/s (0.7%)
---
NEON LDP/STP copy : 1351.4 MB/s (1.7%)
NEON LDP/STP copy pldl2strm (32 bytes step) : 1039.6 MB/s (1.1%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 1190.3 MB/s (1.2%)
NEON LDP/STP copy pldl1keep (32 bytes step) : 1448.6 MB/s (0.8%)
NEON LDP/STP copy pldl1keep (64 bytes step) : 1441.8 MB/s (0.9%)
NEON LD1/ST1 copy : 1346.0 MB/s (0.6%)
NEON STP fill : 3518.6 MB/s (0.6%)
NEON STNP fill : 2724.5 MB/s (10.1%)
ARM LDP/STP copy : 1353.4 MB/s (2.0%)
ARM STP fill : 3518.0 MB/s (0.9%)
ARM STNP fill : 2661.9 MB/s (3.2%)
==========================================================================
== 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 / 0.1 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.0 ns / 0.0 ns
65536 : 4.6 ns / 7.3 ns
131072 : 7.1 ns / 10.1 ns
262144 : 8.3 ns / 11.3 ns
524288 : 10.8 ns / 14.6 ns
1048576 : 79.1 ns / 120.6 ns
2097152 : 113.8 ns / 152.8 ns
4194304 : 136.3 ns / 169.9 ns
8388608 : 148.2 ns / 177.1 ns
16777216 : 155.0 ns / 182.6 ns
33554432 : 159.2 ns / 186.3 ns
67108864 : 162.4 ns / 189.5 ns