diff --git a/README.md b/README.md index 121ebec..2464a55 100644 --- a/README.md +++ b/README.md @@ -11,7 +11,7 @@ It has five **entirely different** usage modes: * Generate a rough CPU performance assessment for a specific SBC *in general* (under ideal conditions) * Show whether an *individual* SBC is able to perform the same and if not hopefully answering the question 'why?' * Help software developers and hardware designers to improve 'thermal performance' when using the `-t` and/or `-T` switches ([details/discussion](https://forum.armbian.com/topic/7819-sbc-bench/?do=findComment&comment=60873), [another example](https://forum.armbian.com/topic/8125-quick-review-of-nanopi-k1-plus/?do=findComment&comment=61300)) -* Plot thermal/consumption charts with `-p` to [measure efficiency of settings/devices](Efficiency_Graphing.md) +* Graph thermal/consumption charts with `-g` to [measure efficiency of settings/devices](Efficiency_Graphing.md) * Provide basic CLI monitoring functionality through the `-m` switch The monitoring now also displays some hardware information when starting: @@ -45,7 +45,7 @@ This tool therefore focuses on a controlled environment and intensive monitoring ## Execution -You need an armhf or arm64 Debian Stretch/Buster/Bullseye or Ubuntu Bionic/Focal/Jammy install. Older variants are not supported (due to distro packages being way too outdated). Then it's +You need Debian Stretch/Buster/Bullseye or Ubuntu Bionic/Focal/Jammy. Older variants are not supported (due to distro packages being way too outdated). Then it's wget https://raw.githubusercontent.com/ThomasKaiser/sbc-bench/master/sbc-bench.sh sudo /bin/bash ./sbc-bench.sh -c @@ -58,7 +58,7 @@ I chose [mhz](https://github.com/wtarreau/mhz), [tinymembench](https://github.co ### [mhz](https://github.com/wtarreau/mhz) -This tool is not a benchmark but instead calculates real CPU clockspeeds. This is helpful on platforms where cpufreq support is not available yet or we can not rely on the clockspeed values returned by the kernel. This applies to platforms where vendors are cheating (RPi, Amlogic) or where actual clockspeeds are set via jumpers while the clockspeeds available to the kernel are derived from device-tree (DT) entries. On a Clearfog Pro routerboard it will look like this for example (DT defines 666/1332 MHz while I configured 800/1600 MHz via jumper): +This tool is not a benchmark but instead measures real CPU clockspeeds. This is helpful on platforms where cpufreq support is not available yet or we can not rely on the clockspeed values returned by the kernel. This applies to platforms where vendors are cheating (RPi, Amlogic) or where actual clockspeeds are set via jumpers while the clockspeeds available to the kernel are derived from device-tree (DT) entries. On a Clearfog Pro routerboard it will look like this for example (DT defines 666/1332 MHz while I configured 800/1600 MHz via jumper): Checking cpufreq OPP: diff --git a/Results.md b/Results.md index 19f7199..af7b45a 100644 --- a/Results.md +++ b/Results.md @@ -15,6 +15,7 @@ So do **not** rely on collected numbers unless you carefully read through all th | [Akaso M8S](http://ix.io/3R3N) | 1200 MHz | 5.10 | Buster armhf | 3050 | 32050 | 32120 | 1160 | 3330 | - | | [Amazon a1.xlarge](http://ix.io/2iFY) | 2300 MHz | 4.15 | Bionic arm64 | 8610 | 458500 | 1297960 | 4280 | 14220 | - | | [AMedia X96 Max+](http://ix.io/3QOj) | 2100 MHz | 5.15 | Focal arm64 | 5270 | 197690 | 981830 | 2630 | 5150 | - | +| [Ampere A1](http://ix.io/4dsC) | 3000 MHz | 5.15 | Jammy arm64 | 16300 | 847000 | 1706150 | 11910 | 47780 | - | | [Apple M1 Pro](http://ix.io/443N) | 3030/2060 MHz | 5.18 | Gentoo 2.8 arm64 | 43800 | 620960 | 1064450 | 27110 | 71910 | 48.28 | | [BPi M2U](http://ix.io/3TKh) | 1010 Mhz | 5.16 | Buster armhf | 2230 | 15550 | 19540 | 790 | 2540 | - | | [BPi M4](http://ix.io/1Dt1) | 1400 MHz | 4.9 | Bionic arm64 | 3500 | 125430 | 651460 | 1010 | 4360 | 5.48 | @@ -154,6 +155,7 @@ So do **not** rely on collected numbers unless you carefully read through all th | [RockPro64](http://ix.io/2sZH) | 2010/1510 MHz | 5.4 | Focal arm64 | 6920 | 397740 | 1145300 | 3700 | 8430 | 11.55 | | [RockPro64](http://ix.io/2yIx) | 2010/1510 MHz | 5.8 | Bullseye arm64 | 7000 | 369820 | 1144950 | 3690 | 8360 | 11.08 | | [Star64](http://ix.io/4a3s) | 1750 MHz | 5.15 | Sid riscv64 | 4820 | 26590 | 28970 | 1170 | 1120 | - | +| [StarFive VisionFive V2](http://ix.io/4dsX) | 1250 MHz | 5.15 | Sid riscv64 | 3470 | 6530 | 5680 | 830 | 790 | - | | [SBC2D70 (SSD202D)](http://ix.io/3N1U) | 1700 MHz | 5.16 | Sid armhf | 1960 | 26240 | 33120 | 770 | 3190 | - | | [Qualcomm QRB5165](http://ix.io/49kx) | 2840/2410/1790 MHz | 4.19 | Focal arm64 | 18860 | 1125320 | 1598490 | 14470 | 23910 | 25.56 | | [T-HEAD C910 RVB-ICE](http://ix.io/41AB) | 1200 MHz | 5.10 | Sid riscv64 | 1760 | 24420 | 26930 | 3340 | 6470 | - | @@ -195,7 +197,7 @@ So do **not** rely on collected numbers unless you carefully read through all th * *memcpy* and *memset* are tinymembench measurements for memory bandwidth. On big.LITTLE SoCs numbers show big core performance * *kH/s* is a **multi threaded** cpuminer score showing the board's performance when executing NEON optimized code. To get the performance difference between big and little cores click the links in the left column * The Akaso M8S and Tronsmart MXIII Plus numbers may be representative for other Amlogic S812 devices (quad Cortex-A9 @ 1.2/1.55 GHz), Tronsmart S82 for other S802 devices (quad Cortex-A9 @ 1.6 GHz) -* The Amazon a1.xlarge numbers represent a 1st gen Graviton CPU (64-bit 'ARM Neoverse') limited to four A72 cores and 8GB memory. +* The Amazon a1.xlarge numbers represent a 1st gen Graviton CPU (64-bit 'ARM Neoverse') limited to four A72 cores and 8GB memory while the Ampere A1 numbers represent an Ampere Altra limited to four Neoverse-N1 cores. * Cubietruck and 'LeMaker Banana Pi' numbers are more or less representative for all other Allwinner A20 devices, same with Lime for Allwinner A10, Olimex Teres-I for Allwinner A64, Orange Pi "PC Plus" and "Plus 2" for Allwinner H2+/H3 and NanoPi K1 Plus, Orange Pi "PC 2" and Prime for Allwinner H5, MangoPi Mcore for Allwinner H616 (though clocked 300 MHz higher than usual). AMedia X96 Max+ numbers represent Amlogic S905X3 devices. * Honeycomb LX2 numbers (based on SolidRun's CEx7 LX2160A COM) might vary somewhat with memory configuration but are more or less representative for LX2160A in general. * Clearfog A1 and Helios4 use exactly same SoC (Armada 385) and clockspeeds and the only reason why OpenSSL numbers differ is since Helios4 numbers were made using [Marvell's CESA crypto accelerator via cryptodev](https://forum.armbian.com/topic/7763-benchmarking-cpus/?do=findComment&comment=59569) which provides nice speed improvements with larger block sizes but also some initialization overhead with tiny block sizes. Also CPU utilization is way lower so the SoC is free for other stuff while performing better at the same time. diff --git a/results/4dsC.txt b/results/4dsC.txt new file mode 100644 index 0000000..de972f2 --- /dev/null +++ b/results/4dsC.txt @@ -0,0 +1,468 @@ +sbc-bench v0.9.9 QEMU KVM Virtual Machine virt-4.2 (Tue, 18 Oct 2022 02:51:50 +0000) + +Distributor ID: Ubuntu +Description: Ubuntu 22.04.1 LTS +Release: 22.04 +Codename: jammy + +Device Info: + Manufacturer: QEMU + Product Name: KVM Virtual Machine + Version: virt-4.2 + +BIOS/UEFI: + Vendor: EFI Development Kit II / OVMF + Version: 1.5.1 + Release Date: 06/16/2021 + BIOS Revision: 0.0 + +/usr/bin/gcc (Ubuntu 11.2.0-19ubuntu1) 11.2.0 + +Uptime: 02:51:50 up 7 days, 15:11, 3 users, load average: 0.29, 0.07, 0.02, 0°C + +Linux 5.15.0-1018-oracle (instance-20220804-1721) 10/18/22 _aarch64_ (4 CPU) + +avg-cpu: %user %nice %system %iowait %steal %idle + 0.07 0.00 0.08 0.02 0.00 99.82 + +Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd +sda 1.30 3.16 24.64 0.00 2081466 16247726 0 + + total used free shared buff/cache available +Mem: 15Gi 410Mi 14Gi 4.0Mi 518Mi 15Gi +Swap: 0B 0B 0B + +########################################################################## + +Checking cpufreq OPP (Neoverse-N1): + +No cpufreq support available. Measured on cpu1: 2991 MHz (2991.650/2990.957/2990.749) + +########################################################################## + +Executing benchmark on cpu0 (Neoverse-N1): + +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 : 11393.4 MB/s (0.2%) + C copy backwards (32 byte blocks) : 11249.3 MB/s + C copy backwards (64 byte blocks) : 11269.5 MB/s (0.1%) + C copy : 11374.2 MB/s + C copy prefetched (32 bytes step) : 11581.4 MB/s + C copy prefetched (64 bytes step) : 11615.4 MB/s + C 2-pass copy : 7934.8 MB/s (0.2%) + C 2-pass copy prefetched (32 bytes step) : 8401.7 MB/s + C 2-pass copy prefetched (64 bytes step) : 8632.0 MB/s + C fill : 46840.5 MB/s + C fill (shuffle within 16 byte blocks) : 46860.3 MB/s + C fill (shuffle within 32 byte blocks) : 46946.4 MB/s + C fill (shuffle within 64 byte blocks) : 46770.4 MB/s + --- + standard memcpy : 11909.9 MB/s + standard memset : 47775.0 MB/s + --- + NEON LDP/STP copy : 11985.0 MB/s (0.1%) + NEON LDP/STP copy pldl2strm (32 bytes step) : 17090.5 MB/s (1.6%) + NEON LDP/STP copy pldl2strm (64 bytes step) : 17581.2 MB/s (1.8%) + NEON LDP/STP copy pldl1keep (32 bytes step) : 18181.7 MB/s (2.6%) + NEON LDP/STP copy pldl1keep (64 bytes step) : 18220.0 MB/s (1.1%) + NEON LD1/ST1 copy : 11998.2 MB/s + NEON STP fill : 47795.3 MB/s + NEON STNP fill : 47798.3 MB/s + ARM LDP/STP copy : 12123.0 MB/s + ARM STP fill : 47851.1 MB/s + ARM STNP fill : 47764.1 MB/s + +========================================================================== +== Framebuffer read tests. == +== == +== Many ARM devices use a part of the system memory as the framebuffer, == +== typically mapped as uncached but with write-combining enabled. == +== Writes to such framebuffers are quite fast, but reads are much == +== slower and very sensitive to the alignment and the selection of == +== CPU instructions which are used for accessing memory. == +== == +== Many x86 systems allocate the framebuffer in the GPU memory, == +== accessible for the CPU via a relatively slow PCI-E bus. Moreover, == +== PCI-E is asymmetric and handles reads a lot worse than writes. == +== == +== If uncached framebuffer reads are reasonably fast (at least 100 MB/s == +== or preferably >300 MB/s), then using the shadow framebuffer layer == +== is not necessary in Xorg DDX drivers, resulting in a nice overall == +== performance improvement. For example, the xf86-video-fbturbo DDX == +== uses this trick. == +========================================================================== + + NEON LDP/STP copy (from framebuffer) : 12930.0 MB/s + NEON LDP/STP 2-pass copy (from framebuffer) : 9932.1 MB/s + NEON LD1/ST1 copy (from framebuffer) : 12964.0 MB/s + NEON LD1/ST1 2-pass copy (from framebuffer) : 8712.5 MB/s + ARM LDP/STP copy (from framebuffer) : 13060.0 MB/s (0.2%) + ARM LDP/STP 2-pass copy (from framebuffer) : 9377.6 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, [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 : 0.0 ns / 0.0 ns + 131072 : 1.2 ns / 1.7 ns + 262144 : 2.2 ns / 2.8 ns + 524288 : 3.2 ns / 3.7 ns + 1048576 : 7.9 ns / 11.7 ns + 2097152 : 19.2 ns / 27.0 ns + 4194304 : 27.6 ns / 33.2 ns + 8388608 : 40.6 ns / 45.8 ns + 16777216 : 46.4 ns / 52.6 ns + 33554432 : 61.6 ns / 73.5 ns + 67108864 : 91.9 ns / 113.9 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 : 0.0 ns / 0.0 ns + 131072 : 1.2 ns / 1.7 ns + 262144 : 1.8 ns / 2.2 ns + 524288 : 2.1 ns / 2.4 ns + 1048576 : 2.6 ns / 2.8 ns + 2097152 : 18.1 ns / 25.9 ns + 4194304 : 25.7 ns / 31.8 ns + 8388608 : 30.4 ns / 34.6 ns + 16777216 : 33.1 ns / 35.3 ns + 33554432 : 39.8 ns / 41.2 ns + 67108864 : 70.6 ns / 90.2 ns + +########################################################################## + +Executing ramlat on cpu0 (Neoverse-N1), results in ns: + + size: 1x32 2x32 1x64 2x64 1xPTR 2xPTR 4xPTR 8xPTR + 4k: 1.338 1.337 1.337 1.337 1.337 1.338 1.337 2.544 + 8k: 1.337 1.337 1.337 1.337 1.338 1.337 1.338 2.606 + 16k: 1.337 1.337 1.337 1.337 1.338 1.337 1.337 2.605 + 32k: 1.338 1.341 1.338 1.339 1.337 1.337 1.338 2.607 + 64k: 1.340 1.339 1.341 1.340 1.340 1.340 1.340 2.612 + 128k: 4.701 4.703 4.703 4.701 4.702 5.254 6.368 10.13 + 256k: 5.349 5.501 5.353 5.496 5.349 5.374 6.523 10.14 + 512k: 5.403 5.423 5.400 5.422 5.387 5.395 6.552 10.19 + 1024k: 15.50 11.33 15.48 11.33 15.34 12.53 13.08 17.22 + 2048k: 36.35 30.97 36.22 30.88 36.37 32.04 30.88 34.87 + 4096k: 37.73 34.97 37.50 34.89 37.58 35.08 34.85 36.13 + 8192k: 58.80 47.61 52.27 47.52 52.19 47.42 48.92 51.63 + 16384k: 57.18 51.12 56.60 50.21 55.62 50.13 52.04 56.38 + +########################################################################## + +Executing benchmark twice on cluster 0 (Neoverse-N1) + +OpenSSL 3.0.2, built on 15 Mar 2022 (Library: OpenSSL 3.0.2 15 Mar 2022) +type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes +aes-128-cbc 846345.85k 1688161.66k 2166582.02k 2321589.25k 2383500.63k 2388650.67k +aes-128-cbc 847665.62k 1685302.21k 2166568.53k 2320926.38k 2381348.86k 2387192.49k +aes-192-cbc 791764.98k 1480822.76k 1834909.27k 1932311.89k 1986065.75k 1990295.55k +aes-192-cbc 789145.95k 1480497.47k 1835889.32k 1933159.42k 1987586.73k 1991764.65k +aes-256-cbc 759765.50k 1315441.11k 1590348.37k 1671956.48k 1703679.32k 1706136.92k +aes-256-cbc 761225.02k 1316275.56k 1590188.03k 1672106.67k 1703029.42k 1706169.69k + +########################################################################## + +Executing benchmark single-threaded on cpu0 (Neoverse-N1) + +7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C.UTF-8,Utf16=on,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: - - - - - - - - - + +RAM size: 15964 MB, # CPU hardware threads: 4 +RAM usage: 435 MB, # Benchmark threads: 1 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 4322 100 4205 4205 | 49260 100 4206 4206 +23: 3814 100 3887 3887 | 48582 100 4206 4205 +24: 3397 100 3653 3653 | 47864 100 4202 4202 +25: 3132 100 3577 3577 | 46516 100 4141 4140 +---------------------------------- | ------------------------------ +Avr: 100 3831 3830 | 100 4189 4188 +Tot: 100 4010 4009 + +########################################################################## + +Executing benchmark 3 times multi-threaded on CPUs 0-3 + +7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C.UTF-8,Utf16=on,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: - - - 64000000 - - - - - + +RAM size: 15964 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 16376 364 4380 15931 | 195462 397 4203 16676 +23: 15754 379 4236 16052 | 191619 396 4190 16580 +24: 14985 380 4235 16112 | 186796 396 4145 16398 +25: 13992 370 4323 15976 | 183194 397 4107 16304 +---------------------------------- | ------------------------------ +Avr: 373 4294 16018 | 396 4161 16489 +Tot: 385 4227 16254 + +7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C.UTF-8,Utf16=on,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: 64000000 - - - - - - - - + +RAM size: 15964 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 16427 364 4390 15980 | 194790 396 4198 16619 +23: 15548 374 4231 15842 | 191848 397 4183 16600 +24: 14953 378 4253 16078 | 186249 395 4143 16350 +25: 14519 383 4333 16578 | 182807 395 4115 16269 +---------------------------------- | ------------------------------ +Avr: 375 4302 16120 | 396 4160 16459 +Tot: 385 4231 16289 + +7-Zip (a) [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C.UTF-8,Utf16=on,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: - - - - 128000000 - - - - + +RAM size: 15964 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 16530 364 4418 16081 | 195270 396 4204 16660 +23: 15638 377 4224 15934 | 192115 397 4188 16623 +24: 14843 375 4257 15960 | 188178 397 4164 16519 +25: 14646 384 4358 16723 | 183086 396 4113 16294 +---------------------------------- | ------------------------------ +Avr: 375 4314 16174 | 397 4167 16524 +Tot: 386 4241 16349 + +Compression: 16018,16120,16174 +Decompression: 16489,16459,16524 +Total: 16254,16289,16349 + +########################################################################## + +Testing maximum cpufreq again, still under full load. System health now: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +03:00:56: --- 3.37 95% 2% 93% 0% 0% 0% 0°C + +Checking cpufreq OPP (Neoverse-N1): + +No cpufreq support available. Measured on cpu1: 2994 MHz (2995.120/2994.634/2994.148) + +########################################################################## + +System health while running tinymembench: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +02:51:54: --- 0.35 0% 0% 0% 0% 0% 0% -- +02:52:34: --- 0.66 25% 0% 25% 0% 0% 0% -- +02:53:14: --- 0.83 25% 0% 25% 0% 0% 0% -- +02:53:54: --- 0.91 25% 0% 25% 0% 0% 0% -- +02:54:34: --- 0.96 26% 0% 25% 0% 0% 0% -- +02:55:14: --- 0.98 25% 0% 25% 0% 0% 0% -- +02:55:54: --- 0.99 25% 0% 25% 0% 0% 0% -- +02:56:34: --- 1.00 25% 0% 25% 0% 0% 0% -- + +System health while running ramlat: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +02:56:56: --- 1.00 0% 0% 0% 0% 0% 0% -- +02:56:59: --- 1.00 25% 0% 25% 0% 0% 0% -- +02:57:02: --- 1.00 26% 0% 25% 0% 0% 0% -- +02:57:05: --- 1.00 26% 0% 25% 0% 0% 0% -- +02:57:08: --- 1.00 25% 0% 25% 0% 0% 0% -- +02:57:11: --- 1.00 25% 0% 25% 0% 0% 0% -- +02:57:14: --- 1.00 25% 0% 25% 0% 0% 0% -- +02:57:17: --- 1.00 25% 0% 25% 0% 0% 0% -- + +System health while running OpenSSL benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +02:57:18: --- 1.00 0% 0% 0% 0% 0% 0% -- +02:57:34: --- 1.08 25% 0% 25% 0% 0% 0% -- +02:57:50: --- 1.06 25% 0% 25% 0% 0% 0% -- +02:58:07: --- 1.05 25% 0% 25% 0% 0% 0% -- +02:58:23: --- 1.04 25% 0% 25% 0% 0% 0% -- +02:58:39: --- 1.03 25% 0% 25% 0% 0% 0% -- +02:58:55: --- 1.02 25% 0% 25% 0% 0% 0% -- + +System health while running 7-zip single core benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +02:59:07: --- 1.02 0% 0% 0% 0% 0% 0% -- +02:59:12: --- 1.01 25% 0% 25% 0% 0% 0% -- +02:59:17: --- 1.01 25% 0% 24% 0% 0% 0% -- +02:59:22: --- 1.01 25% 0% 25% 0% 0% 0% -- +02:59:27: --- 1.01 25% 0% 24% 0% 0% 0% -- +02:59:32: --- 1.01 25% 0% 24% 0% 0% 0% -- +02:59:37: --- 1.01 25% 0% 24% 0% 0% 0% -- +02:59:42: --- 1.01 25% 0% 24% 0% 0% 0% -- + +System health while running 7-zip multi core benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +02:59:43: --- 1.01 0% 0% 0% 0% 0% 0% -- +02:59:54: --- 1.47 97% 0% 96% 0% 0% 0% -- +03:00:05: --- 1.93 94% 1% 92% 0% 0% 0% -- +03:00:15: --- 2.46 95% 1% 93% 0% 0% 0% -- +03:00:26: --- 2.69 97% 1% 96% 0% 0% 0% -- +03:00:36: --- 2.90 93% 1% 91% 0% 0% 0% -- +03:00:46: --- 3.07 96% 0% 95% 0% 0% 0% -- +03:00:56: --- 3.37 95% 2% 93% 0% 0% 0% -- + +########################################################################## + +dmesg output while running the benchmarks: + +[659528.431159] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=92.63.197.131 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=244 ID=60574 PROTO=TCP SPT=45960 DPT=35334 WINDOW=1024 RES=0x00 SYN URGP=0 +[659554.862065] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=91.240.118.71 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=247 ID=14492 PROTO=TCP SPT=49890 DPT=7998 WINDOW=1024 RES=0x00 SYN URGP=0 +[659572.403282] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=64.225.15.53 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=242 ID=56827 PROTO=TCP SPT=61953 DPT=2095 WINDOW=1024 RES=0x00 SYN URGP=0 +[659605.840812] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=176.111.174.91 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=248 ID=64916 PROTO=TCP SPT=48488 DPT=23194 WINDOW=1024 RES=0x00 SYN URGP=0 +[659606.740348] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=87.246.7.49 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=248 ID=32394 PROTO=TCP SPT=44388 DPT=4600 WINDOW=1024 RES=0x00 SYN URGP=0 +[659633.727648] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=5.188.206.18 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=248 ID=31483 PROTO=TCP SPT=41030 DPT=6509 WINDOW=1024 RES=0x00 SYN URGP=0 +[659648.479718] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=123.160.221.60 DST=10.0.0.13 LEN=52 TOS=0x00 PREC=0x00 TTL=50 ID=39632 DF PROTO=TCP SPT=30893 DPT=8058 WINDOW=65535 RES=0x00 SYN URGP=0 +[659667.298974] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=205.210.31.18 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=253 ID=32466 PROTO=TCP SPT=53161 DPT=2121 WINDOW=1024 RES=0x00 SYN URGP=0 +[659710.950370] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=192.241.212.192 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=241 ID=54321 PROTO=TCP SPT=40907 DPT=555 WINDOW=65535 RES=0x00 SYN URGP=0 +[659710.950407] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=192.241.212.192 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=241 ID=54321 PROTO=TCP SPT=40907 DPT=555 WINDOW=65535 RES=0x00 SYN URGP=0 +[659729.995004] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=107.152.37.65 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=249 ID=14693 PROTO=TCP SPT=49486 DPT=5000 WINDOW=1024 RES=0x00 SYN URGP=0 +[659791.899456] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=194.26.29.86 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=246 ID=54871 PROTO=TCP SPT=50323 DPT=4442 WINDOW=1024 RES=0x00 SYN URGP=0 +[659791.899484] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=194.26.29.86 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=246 ID=54871 PROTO=TCP SPT=50323 DPT=4442 WINDOW=1024 RES=0x00 SYN URGP=0 +[659791.899492] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=194.26.29.86 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=246 ID=54871 PROTO=TCP SPT=50323 DPT=4442 WINDOW=1024 RES=0x00 SYN URGP=0 +[659807.155771] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=45.79.134.67 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=242 ID=48199 PROTO=TCP SPT=61000 DPT=25 WINDOW=1024 RES=0x00 SYN URGP=0 +[659836.139063] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=87.246.7.198 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=248 ID=33631 PROTO=TCP SPT=54292 DPT=13239 WINDOW=1024 RES=0x00 SYN URGP=0 +[659860.025257] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=185.167.96.138 DST=10.0.0.13 LEN=40 TOS=0x08 PREC=0x20 TTL=249 ID=54321 PROTO=TCP SPT=56083 DPT=989 WINDOW=65535 RES=0x00 SYN URGP=0 +[659875.081260] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=167.71.234.134 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=242 ID=9751 PROTO=TCP SPT=43384 DPT=5686 WINDOW=1024 RES=0x00 SYN URGP=0 +[659888.023813] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=205.210.31.6 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=253 ID=54321 PROTO=TCP SPT=54508 DPT=9443 WINDOW=65535 RES=0x00 SYN URGP=0 +[659908.186876] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=89.248.163.189 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=243 ID=31286 PROTO=TCP SPT=56455 DPT=5133 WINDOW=1024 RES=0x00 SYN URGP=0 +[659935.169804] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=205.210.31.56 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=253 ID=54321 PROTO=TCP SPT=56802 DPT=2078 WINDOW=65535 RES=0x00 SYN URGP=0 +[659954.424148] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=139.144.233.101 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=242 ID=55296 PROTO=TCP SPT=61000 DPT=448 WINDOW=1024 RES=0x00 SYN URGP=0 +[659970.863639] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=23.225.163.216 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=51 ID=0 DF PROTO=TCP SPT=60858 DPT=49152 WINDOW=1024 RES=0x00 SYN URGP=0 +[660002.104128] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=159.203.175.196 DST=10.0.0.13 LEN=48 TOS=0x00 PREC=0x00 TTL=115 ID=31930 PROTO=TCP SPT=49004 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0 +[660016.488992] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=179.106.190.223 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=234 ID=24920 PROTO=TCP SPT=45607 DPT=1433 WINDOW=1024 RES=0x00 SYN URGP=0 +[660032.396936] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=162.142.125.137 DST=10.0.0.13 LEN=44 TOS=0x00 PREC=0x00 TTL=42 ID=26073 PROTO=TCP SPT=20414 DPT=2222 WINDOW=1024 RES=0x00 SYN URGP=0 +[660066.085440] [UFW BLOCK] IN=enp0s3 OUT= MAC=02:00:17:01:47:46:00:00:17:02:37:c6:08:00 SRC=185.177.25.224 DST=10.0.0.13 LEN=40 TOS=0x00 PREC=0x00 TTL=233 ID=5358 PROTO=TCP SPT=42005 DPT=3389 WINDOW=1024 RES=0x00 SYN URGP=0 + +########################################################################## + +Linux 5.15.0-1018-oracle (instance-20220804-1721) 10/18/22 _aarch64_ (4 CPU) + +avg-cpu: %user %nice %system %iowait %steal %idle + 0.10 0.00 0.08 0.02 0.00 99.79 + +Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd +sda 1.30 3.17 24.74 0.00 2089822 16332990 0 + + total used free shared buff/cache available +Mem: 15Gi 419Mi 14Gi 4.0Mi 527Mi 15Gi +Swap: 0B 0B 0B + +CPU sysfs topology (clusters, cpufreq members, clockspeeds) + cpufreq min max + CPU cluster policy speed speed core type + 0 0 0 - - Neoverse-N1 / r3p1 + 1 0 0 - - Neoverse-N1 / r3p1 + 2 0 0 - - Neoverse-N1 / r3p1 + 3 0 0 - - Neoverse-N1 / r3p1 + +Architecture: aarch64 +CPU op-mode(s): 32-bit, 64-bit +Byte Order: Little Endian +CPU(s): 4 +On-line CPU(s) list: 0-3 +Vendor ID: ARM +Model name: Neoverse-N1 +Model: 1 +Thread(s) per core: 1 +Core(s) per cluster: 4 +Socket(s): 1 +Cluster(s): 1 +Stepping: r3p1 +BogoMIPS: 50.00 +Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm lrcpc dcpop asimddp ssbs +NUMA node(s): 1 +NUMA node0 CPU(s): 0-3 +Vulnerability Itlb multihit: Not affected +Vulnerability L1tf: Not affected +Vulnerability Mds: Not affected +Vulnerability Meltdown: Not affected +Vulnerability Mmio stale data: Not affected +Vulnerability Retbleed: Not affected +Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl +Vulnerability Spectre v1: Mitigation; __user pointer sanitization +Vulnerability Spectre v2: Mitigation; CSV2, BHB +Vulnerability Srbds: Not affected +Vulnerability Tsx async abort: Not affected + +Signature: 00NeoverseN1r3p100NeoverseN1r3p100NeoverseN1r3p100NeoverseN1r3p1 + Compiler: /usr/bin/gcc (Ubuntu 11.2.0-19ubuntu1) 11.2.0 / aarch64-linux-gnu + Userland: arm64 + Kernel: 5.15.0-1018-oracle/aarch64 (kvm) + CONFIG_HZ=250 + CONFIG_HZ_250=y + CONFIG_PREEMPT_NONE=y + CONFIG_PREEMPT_NOTIFIERS=y + +| QEMU KVM Virtual Machine virt-4.2 | ~3000 MHz | 5.15 | Ubuntu 22.04.1 LTS arm64 | 16300 | 847000 | 1706150 | 11910 | 47780 | - | \ No newline at end of file diff --git a/results/4dsX.txt b/results/4dsX.txt new file mode 100644 index 0000000..fc618af --- /dev/null +++ b/results/4dsX.txt @@ -0,0 +1,344 @@ +sbc-bench v0.9.9 StarFive VisionFive V2 (Tue, 18 Oct 2022 03:53:48 +0000) + +Distributor ID: Debian +Description: Debian GNU/Linux bookworm/sid +Release: unstable +Codename: sid + +/usr/bin/gcc (Debian 11.3.0-3) 11.3.0 + +Uptime: 03:53:51 up 20 min, 2 users, load average: 0.64, 0.29, 0.18, 0°C + +Linux 5.15.0-starfive (starfive) 10/18/22 _riscv64_ (4 CPU) + +avg-cpu: %user %nice %system %iowait %steal %idle + 3.45 0.00 0.99 1.09 0.00 94.47 + +Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd +mmcblk1 13.12 323.88 705.04 0.00 406244 884324 0 + + total used free shared buff/cache available +Mem: 3.8Gi 197Mi 3.4Gi 4.0Mi 208Mi 3.6Gi +Swap: 0B 0B 0B + +########################################################################## + +Checking cpufreq OPP (sifive,u74-mc): + +No cpufreq support available. Measured on cpu1: 1249 MHz (1249.084/1249.025/1248.995) + +########################################################################## + +Hardware sensors: + +120e0000.tmon-isa-0000 +temp1: +63.6 C + +########################################################################## + +Executing benchmark on cpu0 (sifive,u74-mc): + +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 : 824.1 MB/s + C copy backwards (32 byte blocks) : 823.2 MB/s (0.7%) + C copy backwards (64 byte blocks) : 823.2 MB/s + C copy : 830.8 MB/s (0.2%) + C copy prefetched (32 bytes step) : 829.3 MB/s + C copy prefetched (64 bytes step) : 829.0 MB/s + C 2-pass copy : 598.5 MB/s + C 2-pass copy prefetched (32 bytes step) : 598.2 MB/s + C 2-pass copy prefetched (64 bytes step) : 598.5 MB/s + C fill : 786.5 MB/s + C fill (shuffle within 16 byte blocks) : 787.1 MB/s + C fill (shuffle within 32 byte blocks) : 787.7 MB/s + C fill (shuffle within 64 byte blocks) : 787.8 MB/s + --- + standard memcpy : 827.1 MB/s (0.1%) + standard memset : 787.6 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 : 6.5 ns / 12.6 ns + 131072 : 9.7 ns / 19.1 ns + 262144 : 11.4 ns / 22.4 ns + 524288 : 17.6 ns / 34.3 ns + 1048576 : 19.8 ns / 36.8 ns + 2097152 : 32.6 ns / 60.3 ns + 4194304 : 90.2 ns / 177.0 ns + 8388608 : 129.4 ns / 256.1 ns + 16777216 : 152.2 ns / 301.6 ns + 33554432 : 166.3 ns / 330.7 ns + 67108864 : 177.3 ns / 353.7 ns + +########################################################################## + +Executing benchmark twice on cluster 0 (sifive,u74-mc) + +OpenSSL 1.1.1f, built on 31 Mar 2020 +type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes +aes-128-cbc 6559.80k 7260.93k 7516.50k 7579.99k 7602.18k 7602.18k +aes-128-cbc 6500.55k 7256.92k 7510.02k 7558.83k 7599.45k 7585.79k +aes-192-cbc 5694.41k 6252.97k 6444.89k 6492.50k 6498.99k 6498.99k +aes-192-cbc 5718.22k 6259.09k 6446.51k 6494.89k 6509.91k 6504.45k +aes-256-cbc 5001.78k 5475.18k 5634.13k 5670.91k 5685.25k 5679.79k +aes-256-cbc 5067.73k 5494.17k 5634.05k 5676.37k 5669.08k 5685.25k + +########################################################################## + +Executing benchmark single-threaded on cpu0 (sifive,u74-mc) + +7-Zip (a) 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: - - - - 128000000 256000000 512000000 1024000000 - + +RAM size: 3893 MB, # CPU hardware threads: 4 +RAM usage: 435 MB, # Benchmark threads: 1 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 798 100 777 777 | 14222 100 1215 1214 +23: 765 100 780 780 | 13976 100 1211 1210 +24: 741 100 798 797 | 13754 100 1208 1207 +25: 714 100 817 816 | 13454 100 1198 1197 +---------------------------------- | ------------------------------ +Avr: 100 793 793 | 100 1208 1207 +Tot: 100 1001 1000 + +########################################################################## + +Executing benchmark 3 times multi-threaded on CPUs 0-3 + +7-Zip (a) 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: 64000000 64000000 64000000 64000000 128000000 - 512000000 - - + +RAM size: 3893 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 2328 327 693 2265 | 54895 395 1186 4683 +23: 2284 336 693 2327 | 54663 400 1184 4730 +24: 2246 344 703 2415 | 52870 394 1177 4641 +25: 2155 343 719 2462 | 52295 399 1166 4654 +---------------------------------- | ------------------------------ +Avr: 337 702 2367 | 397 1178 4677 +Tot: 367 940 3522 + +7-Zip (a) 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: - - - - - - - - - + +RAM size: 3893 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 2162 301 698 2103 | 55152 397 1187 4705 +23: 2286 335 695 2329 | 54673 400 1184 4731 +24: 2251 344 704 2421 | 53047 396 1177 4657 +25: 1174 180 745 1341 | 52298 399 1166 4654 +---------------------------------- | ------------------------------ +Avr: 290 711 2049 | 398 1178 4687 +Tot: 344 945 3368 + +7-Zip (a) 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21 +p7zip Version 16.02 (locale=C,Utf16=off,HugeFiles=on,64 bits,4 CPUs LE) + +LE +CPU Freq: 64000000 64000000 - - 128000000 256000000 512000000 - - + +RAM size: 3893 MB, # CPU hardware threads: 4 +RAM usage: 882 MB, # Benchmark threads: 4 + + Compressing | Decompressing +Dict Speed Usage R/U Rating | Speed Usage R/U Rating + KiB/s % MIPS MIPS | KiB/s % MIPS MIPS + +22: 2263 317 694 2202 | 55561 399 1188 4740 +23: 2268 333 694 2311 | 54162 396 1183 4686 +24: 2247 343 704 2416 | 53470 399 1176 4694 +25: 2201 350 717 2514 | 52355 399 1167 4660 +---------------------------------- | ------------------------------ +Avr: 336 702 2361 | 398 1179 4695 +Tot: 367 941 3528 + +Compression: 2367,2049,2361 +Decompression: 4677,4687,4695 +Total: 3522,3368,3528 + +########################################################################## + +Testing maximum cpufreq again, still under full load. System health now: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +04:10:14: --- 3.81 96% 1% 95% 0% 0% 0% 0°C + +Checking cpufreq OPP (sifive,u74-mc): + +No cpufreq support available. Measured on cpu1: 1249 MHz (1249.113/1249.025/1248.966) + +########################################################################## + +Hardware sensors: + +120e0000.tmon-isa-0000 +temp1: +64.9 C + +########################################################################## + +System health while running tinymembench: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +03:53:55: --- 0.67 5% 0% 3% 0% 1% 0% -- +03:54:35: --- 0.88 25% 0% 25% 0% 0% 0% -- +03:55:15: --- 1.00 25% 0% 24% 0% 0% 0% -- +03:55:55: --- 1.00 25% 0% 25% 0% 0% 0% -- +03:56:35: --- 1.04 25% 0% 25% 0% 0% 0% -- +03:57:15: --- 1.08 25% 0% 25% 0% 0% 0% -- +03:57:55: --- 1.04 25% 0% 24% 0% 0% 0% -- +03:58:35: --- 1.06 25% 0% 25% 0% 0% 0% -- +03:59:15: --- 1.10 25% 0% 25% 0% 0% 0% -- + +System health while running OpenSSL benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +03:59:25: --- 1.08 9% 0% 7% 0% 0% 0% -- +03:59:41: --- 1.06 25% 0% 24% 0% 0% 0% -- +03:59:57: --- 1.05 25% 0% 25% 0% 0% 0% -- +04:00:14: --- 1.04 25% 0% 25% 0% 0% 0% -- +04:00:30: --- 1.03 25% 0% 24% 0% 0% 0% -- +04:00:46: --- 1.02 25% 0% 25% 0% 0% 0% -- +04:01:02: --- 1.01 25% 0% 25% 0% 0% 0% -- + +System health while running 7-zip single core benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +04:01:14: --- 1.01 10% 0% 9% 0% 0% 0% -- +04:01:25: --- 1.01 25% 0% 24% 0% 0% 0% -- +04:01:36: --- 1.01 25% 0% 25% 0% 0% 0% -- +04:01:47: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:01:58: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:02:09: --- 1.00 25% 0% 25% 0% 0% 0% -- +04:02:20: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:02:31: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:02:42: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:02:53: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:03:04: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:03:15: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:03:26: --- 1.00 25% 0% 24% 0% 0% 0% -- +04:03:37: --- 1.00 25% 0% 24% 0% 0% 0% -- + +System health while running 7-zip multi core benchmark: + +Time CPU n/a load %cpu %sys %usr %nice %io %irq Temp +04:03:45: --- 1.00 11% 0% 10% 0% 0% 0% -- +04:04:10: --- 1.90 89% 0% 88% 0% 0% 0% -- +04:04:33: --- 2.45 92% 0% 92% 0% 0% 0% -- +04:04:58: --- 2.98 90% 1% 89% 0% 0% 0% -- +04:05:22: --- 3.06 82% 1% 80% 0% 0% 0% -- +04:05:46: --- 3.59 95% 0% 94% 0% 0% 0% -- +04:06:11: --- 3.79 87% 0% 86% 0% 0% 0% -- +04:06:35: --- 3.92 93% 0% 92% 0% 0% 0% -- +04:07:00: --- 3.95 90% 1% 89% 0% 0% 0% -- +04:07:23: --- 3.55 52% 1% 50% 0% 0% 0% -- +04:07:46: --- 3.15 44% 0% 43% 0% 0% 0% -- +04:08:12: --- 3.27 70% 0% 69% 0% 0% 0% -- +04:08:37: --- 3.57 88% 0% 87% 0% 0% 0% -- +04:09:02: --- 3.78 94% 0% 93% 0% 0% 0% -- +04:09:27: --- 4.04 89% 1% 87% 0% 0% 0% -- +04:09:51: --- 3.82 85% 1% 83% 0% 0% 0% -- +04:10:14: --- 3.81 96% 1% 95% 0% 0% 0% -- + +########################################################################## + +Linux 5.15.0-starfive (starfive) 10/18/22 _riscv64_ (4 CPU) + +avg-cpu: %user %nice %system %iowait %steal %idle + 23.01 0.00 0.79 0.62 0.00 75.58 + +Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd +mmcblk1 7.76 183.92 398.74 0.00 412172 893592 0 + + total used free shared buff/cache available +Mem: 3.8Gi 198Mi 3.4Gi 4.0Mi 215Mi 3.6Gi +Swap: 0B 0B 0B + +CPU sysfs topology (clusters, cpufreq members, clockspeeds) + cpufreq min max + CPU cluster policy speed speed core type + 0 -1 0 - - sifive,u74-mc + 1 -1 0 - - sifive,u74-mc + 2 -1 0 - - sifive,u74-mc + 3 -1 0 - - sifive,u74-mc + +Architecture: riscv64 +Byte Order: Little Endian +CPU(s): 4 +On-line CPU(s) list: 0-3 + +SoC guess: StarFive JH7110 +DT compat: starfive,visionfive-v2 + starfive,jh7110 + Compiler: /usr/bin/gcc (Debian 11.3.0-3) 11.3.0 / riscv64-linux-gnu + Userland: riscv64 + Kernel: 5.15.0-starfive/riscv64 + CONFIG_HZ=100 + CONFIG_HZ_100=y + CONFIG_PREEMPT_COUNT=y + CONFIG_PREEMPT_NONE=y + +| StarFive VisionFive V2 | ~1250 MHz | 5.15 | Debian GNU/Linux bookworm/sid riscv64 | 3470 | 6530 | 5680 | 830 | 790 | - | \ No newline at end of file diff --git a/results/ARMv8-Crypto-Extensions.md b/results/ARMv8-Crypto-Extensions.md index b6e41b1..d5de789 100644 --- a/results/ARMv8-Crypto-Extensions.md +++ b/results/ARMv8-Crypto-Extensions.md @@ -116,6 +116,7 @@ Crawling through [sbc-bench results collection](../Results.md) comparing +30 dif | [A311D2](http://ix.io/3Wq0) | 2200 | 1252070 | 569 | | [A311D](http://ix.io/3VfL) | 2400 | 1365900 | 569 | | Neoverse-N1 | | | | +| [Ampere Altra](http://ix.io/4dsC) | 3000 | 1706150 | 569 | | [Amazon m6g.8xlarge](http://ix.io/2FrG) | 2500 | 1424770 | 570 | | Cortex-A76 | | | | | [RK3588](http://ix.io/3XzI) | 985 | 560200 | 569 | @@ -125,3 +126,5 @@ Crawling through [sbc-bench results collection](../Results.md) comparing +30 dif | [QRB5165](http://ix.io/49kx) | 2840 | 1598490 | 563 | | Cortex-A78AE | | | | | [NVIDIA Orin](http://ix.io/4ax9) | 2200 | 1242940 | 565 | + + diff --git a/results/opp-tables/jh7110-5.15.0-StarFive_VisionFive_V2.txt b/results/opp-tables/jh7110-5.15.0-StarFive_VisionFive_V2.txt new file mode 100644 index 0000000..e89b733 --- /dev/null +++ b/results/opp-tables/jh7110-5.15.0-StarFive_VisionFive_V2.txt @@ -0,0 +1,12 @@ + opp-table-0: + 375 MHz 880.0 mV + 500 MHz 880.0 mV + 625 MHz 880.0 mV + 750 MHz 880.0 mV + 875 MHz 880.0 mV + 1000 MHz 900.0 mV + 1250 MHz 950.0 mV + 1375 MHz 1000.0 mV + 1500 MHz 1100.0 mV + 1625 MHz 1100.0 mV + 1750 MHz 1200.0 mV diff --git a/results/opp-tables/meson8b-5.14.0-Endless_Computers_Endless_Mini.txt b/results/opp-tables/meson8b-5.14.0-Endless_Computers_Endless_Mini.txt new file mode 100644 index 0000000..fa8cb26 --- /dev/null +++ b/results/opp-tables/meson8b-5.14.0-Endless_Computers_Endless_Mini.txt @@ -0,0 +1,21 @@ + gpu-opp-table: + 255 MHz 1100.0 mV + 364 MHz 1100.0 mV + 425 MHz 1100.0 mV + 510 MHz 1100.0 mV + 638 MHz 1100.0 mV + + opp-table: + 96 MHz 860.0 mV + 192 MHz 860.0 mV + 312 MHz 860.0 mV + 408 MHz 860.0 mV + 504 MHz 860.0 mV + 600 MHz 860.0 mV + 720 MHz 860.0 mV + 816 MHz 900.0 mV + 1008 MHz 1140.0 mV + 1200 MHz 1140.0 mV + 1320 MHz 1140.0 mV + 1488 MHz 1140.0 mV + 1536 MHz 1140.0 mV diff --git a/results/opp-tables/rk3229-5.10.2-Generic_RK322x_Tv_Box_board.txt b/results/opp-tables/rk3229-5.10.2-Generic_RK322x_Tv_Box_board.txt new file mode 100644 index 0000000..62a802b --- /dev/null +++ b/results/opp-tables/rk3229-5.10.2-Generic_RK322x_Tv_Box_board.txt @@ -0,0 +1,11 @@ + opp-table2: + 200 MHz 1050.0 mV + 300 MHz 1050.0 mV + 500 MHz 1150.0 mV + + opp_table0: + 408 MHz 950.0 mV + 600 MHz 975.0 mV + 816 MHz 1000.0 mV + 1008 MHz 1175.0 mV + 1200 MHz 1275.0 mV diff --git a/results/opp-tables/rk3328-5.19.14-Firefly_roc-rk3328-cc.txt b/results/opp-tables/rk3328-5.19.14-Firefly_roc-rk3328-cc.txt new file mode 100644 index 0000000..2351c0f --- /dev/null +++ b/results/opp-tables/rk3328-5.19.14-Firefly_roc-rk3328-cc.txt @@ -0,0 +1,20 @@ + dmc-opp-table: + 786 MHz 1075.0 mV + 798 MHz 1075.0 mV + 840 MHz 1075.0 mV + 924 MHz 1100.0 mV + 1068 MHz 1175.0 mV + + gpu-opp-table: + 200 MHz 1050.0 mV + 300 MHz 1050.0 mV + 400 MHz 1050.0 mV + 500 MHz 1150.0 mV + + opp-table-0: + 408 MHz 950.0 mV + 600 MHz 950.0 mV + 816 MHz 1000.0 mV + 1008 MHz 1100.0 mV + 1200 MHz 1225.0 mV + 1296 MHz 1300.0 mV diff --git a/results/opp-tables/sun50iw9p1-4.9.255-Orange_Pi_Zero_2.txt b/results/opp-tables/sun50iw9p1-4.9.255-Orange_Pi_Zero_2.txt new file mode 100644 index 0000000..4f55f1a --- /dev/null +++ b/results/opp-tables/sun50iw9p1-4.9.255-Orange_Pi_Zero_2.txt @@ -0,0 +1,14 @@ + opp_l_table: + 480 MHz 820.0 mV (0003) + 480 MHz 880.0 mV (0004) + 600 MHz 820.0 mV (0003) + 600 MHz 880.0 mV (0004) + 792 MHz 860.0 mV (0003) + 792 MHz 940.0 mV (0004) + 1008 MHz 900.0 mV (0003) + 1008 MHz 1020.0 mV (0004) + 1200 MHz 960.0 mV (0003) + 1200 MHz 1100.0 mV (0004) + 1296 MHz 1100.0 mV (0002) + 1344 MHz 1120.0 mV (0004) + 1512 MHz 1100.0 mV (0001) diff --git a/results/opp-tables/sun8i-a83t-5.15.48-Banana_Pi_BPI-M3.txt b/results/opp-tables/sun8i-a83t-5.15.48-Banana_Pi_BPI-M3.txt new file mode 100644 index 0000000..6a9f500 --- /dev/null +++ b/results/opp-tables/sun8i-a83t-5.15.48-Banana_Pi_BPI-M3.txt @@ -0,0 +1,15 @@ + opp-table-cluster0: + 576 MHz 840.0 mV + 768 MHz 840.0 mV + 1152 MHz 900.0 mV + 1344 MHz 920.0 mV + 1536 MHz 1000.0 mV + 1728 MHz 1080.0 mV + + opp-table-cluster1: + 576 MHz 840.0 mV + 768 MHz 840.0 mV + 1152 MHz 900.0 mV + 1344 MHz 920.0 mV + 1536 MHz 1000.0 mV + 1728 MHz 1080.0 mV diff --git a/results/results.tar.xz b/results/results.tar.xz index df5506e..ec7af5a 100644 Binary files a/results/results.tar.xz and b/results/results.tar.xz differ diff --git a/results/validation.md b/results/validation.md index ba21409..3245696 100644 --- a/results/validation.md +++ b/results/validation.md @@ -212,3 +212,5 @@ | [4bbv](4bbv.txt) | v0.9.8 Khadas VIM1S | 0.5%/1.1% | 1.9Gi/1.1Gi | X | X | X | X | 3% | 0% | | | [4d1U](4d1U.txt) | v0.9.9 SolidRun Clearfog A1 | 0%/0% | 1.0Gi/873Mi | X | X | X | X | 1% | 0% | | | [4dd5](4dd5.txt) | v0.9.9 AAEON UP-APL03 V1.0 / Atom | 0%/0% | 3.7Gi/2.6Gi | X | X | X | X | 2% | 1% | | +| [4dsC](4dsC.txt) | v0.9.9 QEMU KVM Virtual Machine virt-4. | 0%/0% | 15Gi/15Gi | X | X | X | X | 2% | 0% | | +| [4dsX](4dsX.txt) | v0.9.9 StarFive VisionFive V2 | 0.1%/0% | 3.8Gi/3.6Gi | X | X | X | X | 1% | 0% | | diff --git a/sbc-bench.sh b/sbc-bench.sh index 5e48cdb..6529e2d 100755 --- a/sbc-bench.sh +++ b/sbc-bench.sh @@ -2157,7 +2157,7 @@ CheckClockspeedsAndSensors() { echo "Please be aware that for reasons yet unknown single-threaded or multi-" >>${ResultLog} echo "threaded workloads with less threads than available logical P cores always" >>${ResultLog} echo "end up running on P cores even if pinned to E cores (taskset/cgroups seem" >>${ResultLog} - echo "to not working in the intended way on Alder/Raptor Lake)." >>${ResultLog} + echo "both to not work in the intended way on Alder/Raptor Lake)." >>${ResultLog} fi fi # if powercapping seems to be available on Intel then add a hint