pbies

Members
  • Content count

    33
  • Joined

  • Last visited

About pbies

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. pbies

    RK3399 Orange Pi

    If you want to move from default 4.4.103 kernel version to 4.4.143 it is possible just by replacing the "Image" file in "boot" folder (boot partition which is FAT when you mount it). No need to change the .conf or .dtb files. Tested. The instruction is provided here: http://opensource.rock-chips.com/wiki_Rockchip_Kernel Which takes only 4 commands: git clone https://github.com/rockchip-linux/kernel.git cd kernel make ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- rockchip_linux_defconfig make ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- -j4 And you have "Image" file. Surely some packages would be needed - all is on the Internet if you search for a specific error messages.
  2. pbies

    RK3399 Orange Pi

    Benchmarks 2018-08-18: $ stress-ng --cpu 1 --cpu-method matrixprod --metrics-brief -t 60 stress-ng: info: [1932] dispatching hogs: 1 cpu stress-ng: info: [1932] cache allocate: using built-in defaults as unable to determine cache details stress-ng: info: [1932] cache allocate: default cache size: 2048K stress-ng: info: [1932] successful run completed in 60.23s (1 min, 0.23 secs) stress-ng: info: [1932] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s stress-ng: info: [1932] (secs) (secs) (secs) (real time) (usr+sys time) stress-ng: info: [1932] cpu 258 60.22 60.22 0.00 4.28 4.28 $ stress-ng --cpu 6 --cpu-method matrixprod --metrics-brief -t 60 stress-ng: info: [1935] dispatching hogs: 6 cpu stress-ng: info: [1935] cache allocate: using built-in defaults as unable to determine cache details stress-ng: info: [1935] cache allocate: default cache size: 2048K stress-ng: info: [1935] successful run completed in 60.14s (1 min, 0.14 secs) stress-ng: info: [1935] stressor bogo ops real time usr time sys time bogo ops/s bogo ops/s stress-ng: info: [1935] (secs) (secs) (secs) (real time) (usr+sys time) stress-ng: info: [1935] cpu 882 60.10 359.17 0.02 14.68 2.46 $ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=1 --max-time=60 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 1 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 20000 Test execution summary: total time: 14.2567s total number of events: 10000 total time taken by event execution: 14.2539 per-request statistics: min: 1.42ms avg: 1.43ms max: 6.39ms approx. 95 percentile: 1.42ms Threads fairness: events (avg/stddev): 10000.0000/0.00 execution time (avg/stddev): 14.2539/0.00 $ sysbench --test=cpu --cpu-max-prime=20000 --num-threads=6 --max-time=60 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 6 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 20000 Test execution summary: total time: 3.5062s total number of events: 10000 total time taken by event execution: 20.9935 per-request statistics: min: 1.42ms avg: 2.10ms max: 19.40ms approx. 95 percentile: 2.62ms Threads fairness: events (avg/stddev): 1666.6667/498.81 execution time (avg/stddev): 3.4989/0.00
  3. pbies

    RK3399 Orange Pi

    Orange Pi RK3399 eMMC read speed is: 14,6GiB 0:01:13 [ 202MiB/s] [===...===>] 100% It varies from 170 MB/s to over 216 MB/s. Command used: pv /dev/mmcblk1 > /dev/null Write speed in filesystem is about 48 MB/s.
  4. pbies

    RK3399 Orange Pi

    In 7-zip benchmark Orange Pi RK3399 gives about 5330 MIPS total, when Raspberry Pi 3 B+ gives 2833 MIPS. Surely these are not very accurate benchmarks as there throttling occurs.
  5. pbies

    RK3399 Orange Pi

    Are these files the same as in manufacturers files (all OSes), or they are especially written for Armbian?
  6. pbies

    RK3399 Orange Pi

    Steps: - Download: "OrangePi-RK3399_Linux4.4_V1.0_2018_0110.tar.gz" "OrangePi-rk3399_Ubuntu16.04-Server_20171227.img" - Start Linux on PC - Unpack .tar.gz via "tar xvf OrangePi-RK3399_Linux4.4_V1.0_2018_0110.tar.gz" - it will create a subfolder "rk-linux", no need to create any folders - In folder "rk-linux" create folder "out" - Put in "out" folder the Ubuntu .img file and rename it to "system.img" - Short connect the two pins and connect board via USB-C to USB 2.0 in your PC - you may need to repeat this step if the following step gives error - Run "./build/flash_tool.sh -c rk3399 -p system -i out/system.img" from "rk-linux" folder Done!
  7. pbies

    RK3399 Orange Pi

    AT LAST! I've started the RK3399 with Ubuntu Server 16.04.3 LTS, kernel 4.4.103. Exactly flashed under Linux the provided Ubuntu image from Mega hosting. You need the Linux4.4...tar.gz file to flash the image. The mk-image.sh script is making recursive rootfs.img inside system.img - so it won't work! You can check that by 7-zip on Windows. Image compiled like it is said in manual won't work as rootfs is in this image recursively. Via Windows AndroidTool you can flash only Android image, other images, whatever they are, won't be taken. /proc/cpuinfo sees only 6 CPUs (0-5). Serial number for CPU is only zeros. You can flash the image only on Linux, using USB 2.0 port - not 3.0! Linux flash_tool.sh is in .tar.gz - you need to uncompress the whole file (binaries which are needed are elsewhere in that archive). Reboot from logged in root to login prompt takes 16 seconds. Login "root", password "orangepi" - I suspect root can't login first time. Other user - login "orangepi", password "orangepi", then "su", there is no "sudo". There are NO: ifconfig, ping, trace*, mc, pv, links, arp, rarp. There are: ifup, ifdown, ifquery, wpa_passphrase, wpa_supplicant. There is no network configuration - any! Only some files that are nearly empty - not configured. "ifquery" gives only "lo" and "eth0". Now I need to configure Ethernet and/or Wi-Fi on that board - all help appreciated! If you need to know anything - just ask!
  8. pbies

    RK3399 Orange Pi

    Well, after two weeks fight with Orange Pi RK3399 I am trying to return it to seller on AliExpress. I flashed Linux images to the board few times, but the board does not boot. Many users on manufacturer's forum are complaining that they can't do anything with the board. The only system which is working in Android, which is useless in many cases. Maybe this board is powerful, but the software and support given are ridiculous. Board has few months and there should be stable Linux provided, not only Android (even without Google Play shop). If you want to lose money - buying this board is the right direction.
  9. pbies

    RK3399 Orange Pi

    I have it in my hand. I can give you feedback about this board. Due to eMMC it is nearly not accessible from outside. Android already installed in eMMC is useless. Keyboard and mouse gives nothing with this board. Need to make some acrobatics to work this board out. First install a normal Linux system. By USB-C cable connected to my PC I can't do anything with this board, because it receives not enough power. Dedicated PSU gives possibility to boot to Android, but no other sensible port to program the eMMC.
  10. pbies

    Benchmarking CPUs

    @chwe I always try to simplify things when there is such possibility, so I'll be short on what you written: it is a specific usage and the CPU core is the most important part. That's why I need reliable benchmark of CPU, NOT RAM, NOT IO, NOT disk. And yes, cache is part of CPU - good for my application. But as you can see the rumours are different for each app. I didn't selected the last and only bench for this moment. I'm gathering knowledge and not happy enough with the current proposals and results. Also in multi-threaded and multi-core benchmarks, I think, you will not find the same results even on the same machine in the same circumstances. What I am trying to do is to compare x86/x64 with ARM in some kind of general for those two architectures bench. And also I don't belive that I am only one on the planet that is searching for such app. It should exist. Even if it doesn't exist - one should be able to write it. Linpack seems to be the right choice, but I need to drill the topic more.
  11. pbies

    Benchmarking CPUs

    I imagine such benchmark as: - operating only on registers = add, subtract, multiply, divide and other math instructions (ALU) - using many of ARM instructions, but limited to ARM, and also be reliable on x86/x64 CPUs - doing number of operations in specific (selectable) time - only operating on RAM if there is need to check the clock for the above
  12. pbies

    Benchmarking CPUs

    @NicoD I know the computer hardware and software for 30 years now, so no need to explain me anything. This is also the time I was a developer. Lost tracking the apps just 2 years ago as they got multiplied by thousands. That's why I am asking for CPU benchmark. I know that RAM will be used to start/stop the benchmark, as OS is needed and so on, but still all the operations can be done only on CPU, with small help of clock which can be just checked for time passed along with the test. So there is no need to use RAM for CPU benchmark that much. And that's the bench I was looking for. I understand that you didn't found such a bench because I don't belive that such doesn't exist. And yes, I know, that OS will interrupt the benchmark just by interrupts and IO and will be not exact, but it is fine for me - still no need to use RAM for CPU benchmark.
  13. pbies

    Benchmarking CPUs

    @tkaiser Really, I am not interested in your negative way of thinking. Please leave me alone out of your ideas. @NicoD You didn't understood what I said: I need benchmark for CPU, not RAM. And to benchmark CPU you don't need tons of RAM as for 7z. You understand now direction in which I want to go? I don't need to test tons of RAM, I agree that nothing without it would exist, but still operations on registers is enough to test CPU, and NOT on RAM.
  14. pbies

    Benchmarking CPUs

    @NicoD still these rely on RAM speed, compression. I would go for Phoronix Test Suite and stress-ng. Thanks.
  15. pbies

    Benchmarking CPUs

    @NicoD SBCs aren't CPU alone, but I need to test only CPU. 7z is dependant on RAM mainly. This way I can use MemTest and get the same level-reliable results. So what kind of software do you propose then? Do you have any software that tests only CPU and isn't memory-dependant? And please don't tell me that it doesn't exist - any such software can operate only on registers inside CPU.