Jump to content

gprovost

Members
  • Posts

    580
  • Joined

  • Last visited

Reputation Activity

  1. Like
    gprovost got a reaction from tkaiser in Benchmarking CPUs   
    @tkaiser Here the link of my latest benchmark of Helios4 : http://ix.io/1jCy
     
    I get the following result for OpenSSL speed
    OpenSSL results: type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes aes-128-cbc 1280.56k 5053.40k 18249.13k 52605.27k 102288.04k 109390.51k aes-128-cbc 1285.51k 5030.68k 18256.13k 53001.90k 100128.09k 109188.44k aes-192-cbc 1276.82k 4959.19k 18082.22k 51421.53k 96897.71k 103093.59k aes-192-cbc 1290.35k 4961.09k 17777.24k 51629.74k 95647.06k 102596.61k aes-256-cbc 1292.07k 5037.99k 17762.90k 50542.25k 92782.59k 98298.54k aes-256-cbc 1281.35k 5050.94k 17874.77k 49915.90k 93164.89k 98822.83k  
    In order to leverage on hw crypto engine, I had no choice but to use OpenSSL 1.1.1 lib (openssl-1.1.1-pre8) and I decided to use cryptodev-linux instead of AF_ALG since it gives me slightly better result (+5-10%).
     
    Here a bit of findings regarding OpenSSL engines implementation :
     
    As stated in the changelog
    Changes between 1.0.2h and 1.1.0 [25 Aug 2016] *) Added the AFALG engine. This is an async capable engine which is able to offload work to the Linux kernel. In this initial version it only supports AES128-CBC. The kernel must be version 4.1.0 or greater. [Catriona Lucey] So using Debian Stretch package OpenSSL 1.1.0f, or any more recent 1.1.0 version, the only cipher supported by AFALG engine was effectively AES-128-CBC
    $> openssl engine -c (dynamic) Dynamic engine loading support (afalg) AFALG engine support [AES-128-CBC]  
    Starting OpenSSL 1.1.1, even though it is not mentioned anywhere in the changelog, AES192-CBC and AES256-CBC is supported by the AFALG engine
    $> openssl engine -c (dynamic) Dynamic engine loading support (afalg) AFALG engine support [AES-128-CBC, AES-192-CBC, AES-256-CBC]  
    But one thing much more exiting about OpenSSL 1.1.1 is the following
    Changes between 1.1.0h and 1.1.1 [xx XXX xxxx] *) Add devcrypto engine. This has been implemented against cryptodev-linux, then adjusted to work on FreeBSD 8.4 as well. Enable by configuring with 'enable-devcryptoeng'. This is done by default on BSD implementations, as cryptodev.h is assumed to exist on all of them. [Richard Levitte] So now with the 1.1.1 is pretty straight forward to use cryptodev, no need to patch or configure anything in openssl, openssl will detect automatically if module cryptodev is loaded and will offload crypto operation on it if presents.
    $> openssl engine -c (devcrypto) /dev/crypto engine [DES-CBC, DES-EDE3-CBC, BF-CBC, AES-128-CBC, AES-192-CBC, AES-256-CBC, AES-128-CTR, AES-192-CTR, AES-256-CTR, AES-128-ECB, AES-192-ECB, AES-256-ECB, CAMELLIA-128-CBC, CAMELLIA-192-CBC, CAMELLIA-256-CBC, MD5, SHA1] (dynamic) Dynamic engine loading support  
    Based on those info, and making the assumption that sooner than later openssl 1.1.1 will be available in Debian Stretch (via backports most probably), i think the best approach to add openssl crypto engine support in Armbian is via the cryptodev approach. This way we can support all the ciphers now. I will look how to patch properly dpkg openssl_1.1.0f-3+deb9u2 to activate cryptodev supports. @zador.blood.stained maybe you have a different option on the topic ?
     
     
  2. Like
    gprovost got a reaction from tkaiser in Benchmarking CPUs   
    @zador.blood.stained I think there isn't any distro OpenSSL packages that is built with hardware engine support.
    Also, even if engine is installed, OpenSSL doesn't use any engine by default, you need to configure it in openssl.cnf.
    But you right about cryptsetup (dm-crypt), it uses AF_ALG by default. I was wondering why so much delta between my 'cryptsetup benchmark' and 'openssl speed' test on Helios4.
     
    I just did a test by compiling openssl-1.1.1-pre8 with the AF_ALG (... enable-engine enable-afalgeng ...) and here are the benchmark result on Helios4 :
     
    $> openssl speed -evp aes-xxx-cbc -engine afalg -elapsed
    type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes aes-128-cbc 745.71k 3018.47k 11270.23k 36220.25k 90355.03k 101094.74k aes-256-cbc 739.49k 2964.93k 11085.23k 34178.05k 82597.21k 90461.53k  
    The difference is quite interesting, with AF_ALG it performs much better on bigger block size, but poorly on very small block size.
     
    $> openssl speed -evp aes-xxx-cbc -elapsed
    type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes aes-128-cbc 44795.07k 55274.84k 59076.27k 59920.04k 59719.68k 59353.77k aes-256-cbc 34264.93k 40524.42k 42168.92k 42496.68k 42535.59k 42500.10k  
    System : Linux helios4 4.14.57-mvebu #2 SMP Tue Jul 24 08:29:55 UTC 2018 armv7l GNU/Linux
     
    Supposedly you can even have better perf with cryptodev, but I think Crypto API AF_ALK is more elegant and easier to setup.
     
    Once I have a cleaner install of this AF_ALK (or cryptodev), I will run sbc_bench and send you ( @tkaiser ) the output.
  3. Like
    gprovost reacted to Igor in Helios4 Second Campaign is ON !   
    And updated images with many bugfixes.
     
     
  4. Like
    gprovost got a reaction from iNTiGOD in Helios4 Support   
    @iNTiGOD which build / image version are running ? 
    Maybe you can refer to the following blog post to check if the files on your system show to correct settings.
  5. Like
    gprovost got a reaction from chwe in Helios4 Support   
    @nemo19 Ok no need to look any further, you found the issue. There should be a thermal pad between the CPU and the Heatsink as shown below. Without the thermal pad no proper heat transfer can happen, therefore the CPU might have reached above Maximum Junction Temperature (115C) resulting by it to get unstable and crash. I'm really sorry about this missing thermal pad, this should definitively not have happened, I will report / complain to the company that handled the board assembly for us.
     
    FYI the thermal pad dimension we are using is 20x20x1mm.
     
    Please provide me by private message your complete shipping address. I will send you this missing thermal pad. In the meantime you can try using thermal paste, even though the gap between CPU and Heatsink is a bit too big for thermal paste.
     

  6. Like
    gprovost reacted to nemo19 in Helios4 Support   
    I received and mounted the thermal pad from SolidRun on Tuesday. My Helios4 now runs cooler and quieter under maximum load than when idling before. It's been running for 1 day 16 hours at 70% to 100% cpu load hashing files, staying below 80C and without crashing.
     
    If it keeps going like this I'll be very happy. Thank you for the great board and the great service!
  7. Like
    gprovost got a reaction from chwe in Helios4 Support   
    @bigheart Is the serial issue you described happens all the time after over night running ?
     
    The helios4 board has an onboard FTDI USB-to-UART bridge (FT230X), this way you don't need to have a RS232 port on your computer or to buy a USB-to-Serial converter.
    The FT230X is actually power supplied by the USB port of your computer that you connect to the Heliso4 micro-USB console port, so even if the Helios4 board is not powered-on the FTDI USB-to-UART bridge will still come to live and should be detected by your computer.
     
    Assuming your computer is a Linux machine, as soon as you connect the Helios4 console port you should see the following in your kernel messages
    $> dmesg -w [...] [684438.411938] usb 1-2: new full-speed USB device number 18 using xhci_hcd [684438.547622] usb 1-2: New USB device found, idVendor=0403, idProduct=6015 [684438.547631] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [684438.547636] usb 1-2: Product: FT230X Basic UART [684438.547639] usb 1-2: Manufacturer: FTDI [684438.547643] usb 1-2: SerialNumber: DN00KDNN [684438.551627] ftdi_sio 1-2:1.0: FTDI USB Serial Device converter detected [684438.551703] usb 1-2: Detected FT-X [684438.552290] usb 1-2: FTDI USB Serial Device converter now attached to ttyUSB0  
    If you unplug you should see the following USB disconnect message
    [684398.900050] usb 1-1: USB disconnect, device number 17 [684398.900432] ftdi_sio ttyUSB0: FTDI USB Serial Device converter now disconnected from ttyUSB0 [684398.900463] ftdi_sio 1-1:1.0: device disconnected  
    So when the serial issue you described happens again, you should try to unplug / plug the console port and see if you see the above messages appearing in you kernel messages.
     
    If the expected kernel messages appear but you still can't access the shell console of the Linux running on your Helios4, then the issue is somewhere else than the onboard FTDI USB-to-UART bridge (unless a very unlikely issue with the Dual-Buffer IC between the FTDI chip ans the SoM) . In this case you should post the full dmesg ouput of your Helios4 when the issue appear (use https://pastebin.com/ to share the output to us).
  8. Like
    gprovost reacted to tkaiser in Support of Helios4 - Intro   
    Good news: @zador.blood.stained imported (cherry-picked) vendor provided software support for Helios4 into our build system recently, I let on my host build an OMV image and it seems it's working out of the box: https://github.com/armbian/build/pull/812#issuecomment-342006038 (though some minor issues present we can focus now on).
  9. Like
    gprovost got a reaction from tkaiser in Support of Helios4 - Intro   
    Here some fresh benchmarks with Helios4 configured as RAID6, then as RAID5 with 4x HDDs Western Digital RED 2TB.
     
    Note : This is also a 1GB RAM version clocked at 1.6GHz.
     
    RAID 6
    md0 : active raid6 sdd1[3] sdc1[2] sdb1[1] sda1[0]
          39028736 blocks super 1.2 level 6, 128k chunk, algorithm 2 [4/4] [UUUU]
     
    With buffer cache
    iozone -e -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    random random kB reclen write rewrite read reread read write 102400 4 112166 182202 878938 897663 720385 46844 102400 16 126067 193562 907704 1068469 1001711 98934 102400 512 118358 182862 985050 1006993 1001779 170459 102400 1024 117394 176268 875566 888008 889065 169862 102400 16384 114508 169885 818170 817925 827517 172793  
    With DIRECT IO
    iozone -e -a -I -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    random random kB reclen write rewrite read reread read write 102400 4 5830 5826 34179 37669 2839 602 102400 16 12445 19215 66829 71497 10074 2198 102400 512 106759 116574 199069 281008 61111 124175 102400 1024 129763 105815 173525 237719 96948 139462 102400 16384 177544 171003 284953 315916 286591 164939  
    RAID5
    md0 : active raid5 sdd1[3] sdc1[2] sdb1[1] sda1[0]
          58543104 blocks super 1.2 level 5, 128k chunk, algorithm 2 [4/4] [UUUU]
     
    With buffer cache
    iozone -e -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    random random kB reclen write rewrite read reread read write 102400 4 115183 170676 904434 895645 722963 51570 102400 16 119682 172369 1063885 1084722 1002759 116063 102400 512 123952 201012 989143 992372 1001279 148636 102400 1024 116065 197683 874443 885162 886703 198021 102400 16384 129151 199118 830783 845355 838656 180389  
    With DIRECT IO
    iozone -e -a -I -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    random random kB reclen write rewrite read reread read write 102400 4 9552 10938 36978 39222 28551 530 102400 16 11185 20621 64135 77590 33906 2159 102400 512 40443 51534 96495 533923 536836 43272 102400 1024 84312 78308 122916 486366 440706 83712 102400 16384 196534 207664 598961 597055 619819 206123  
     
    On small block operation, my setup shows CPU utilization 50% and memory usage 50%.
     
    I know there is no obvious way to say if the system offload on the hardware XOR engine. Check your kernel message you should see the kernel has detected 2 XOR engine.
    [    0.737452] mv_xor f1060800.xor: Marvell shared XOR driver
    [    0.776401] mv_xor f1060800.xor: Marvell XOR (Descriptor Mode): ( xor cpy intr pq )
    [    0.776570] mv_xor f1060900.xor: Marvell shared XOR driver
    [    0.816397] mv_xor f1060900.xor: Marvell XOR (Descriptor Mode): ( xor cpy intr pq )
    In the current armbian mvebu kernel config, mv_xor driver is part of the kernel, not a module.
     
    I'm still trying to figure out a more concrete way to confirm that the system offload on hardware engine, but based on benchmark it does.
     
     

     
  10. Like
    gprovost got a reaction from lanefu in Support of Helios4 - Intro   
    Hi Guys,
     
    That's nice to see the effervescence around Helios4. We have a pretty good start for the first 48hours, we are very optimistic.
     
    Thanks @tkaiser for being so proactive and already answering most people questions. I saw you were very active on CNX ;-) I have so many question to reply through KS, it's hard to catch-up.
     
    I take note of all your feedback. Regarding the HDD vibration absorbers and the filter screen, it's something I had already in mind but haven't found yet the right solution. Take note that the thickness of the acrylic plate is much ticker than usual PC metal case, so standard absorber won't fit. If I can find the right solution we might put these as part of the kit by default.
     
    Haha i see someone really wants its ECC RAM, but yes agree It would be a great stretch goal.
    We were also planning in the stretch goal list to include a OLED i2c screen.
     
    But well, first we need to see how the campaign progress in the next few days before dreaming too big.
     
    Cheers.
  11. Like
    gprovost got a reaction from devman in Support of Helios4 - Intro   
    Awesome Helios4 is successfully funded !!!
     
    We still accept orders till the end of the campaign : July 19, 2017
     
    https://shop.kobol.io/
     
    Thanks for the support.
  12. Like
    gprovost got a reaction from wsian in Support of Helios4 - Intro   
    Awesome Helios4 is successfully funded !!!
     
    We still accept orders till the end of the campaign : July 19, 2017
     
    https://shop.kobol.io/
     
    Thanks for the support.
  13. Like
    gprovost got a reaction from TonyMac32 in Support of Helios4 - Intro   
    Awesome Helios4 is successfully funded !!!
     
    We still accept orders till the end of the campaign : July 19, 2017
     
    https://shop.kobol.io/
     
    Thanks for the support.
  14. Like
    gprovost got a reaction from tkaiser in Support of Helios4 - Intro   
    Awesome Helios4 is successfully funded !!!
     
    We still accept orders till the end of the campaign : July 19, 2017
     
    https://shop.kobol.io/
     
    Thanks for the support.
  15. Like
    gprovost got a reaction from zador.blood.stained in Support of Helios4 - Intro   
    Awesome Helios4 is successfully funded !!!
     
    We still accept orders till the end of the campaign : July 19, 2017
     
    https://shop.kobol.io/
     
    Thanks for the support.
  16. Like
    gprovost got a reaction from Igor in Support of Helios4 - Intro   
    @tkaiser The discount code (KSBACKER10) is valid till July 3rd.
     
    Like you, we were surprised to see such positive response to the ECC model, but it looks like the majority of our backers understood the added value and how it was differentiating even more the Helios4 from proprietary entry level NAS. So thank you to have suggested (many times) this upgrade ;-)
     
    Regarding 'getting more attention', it isn't that trivial unfortunately... we think that once the first Helios4 batch is received and used by our backers it will improve our exposure.
  17. Like
    gprovost got a reaction from TonyMac32 in Support of Helios4 - Intro   
    Hey guys,
     
    Awesome news, our Kickstarter campaign is finally live. Be amount the first backers and enjoy our early bird deals :

    Our kickstarter page :  Helios4
     
    For some specs : Helios4_Specifications.pdf
     


    Basic Kit - 1GB (USD$125)
    (Save 16% off the $149 Retail - Limited to 50 units)

    Full Kit - 1GB (USD$139)
    (Save 18% off the $169 Retail - Limited to 130 units)

    Full Kit - 2GB (USD$165)
    (Save 11% off the $185 Retail - Limited to 70 units)
     
    Our kickstarter page :  Helios4
     
    Cheers ;-)
     
  18. Like
    gprovost got a reaction from manuti in Support of Helios4 - Intro   
    Hey guys,
     
    Awesome news, our Kickstarter campaign is finally live. Be amount the first backers and enjoy our early bird deals :

    Our kickstarter page :  Helios4
     
    For some specs : Helios4_Specifications.pdf
     


    Basic Kit - 1GB (USD$125)
    (Save 16% off the $149 Retail - Limited to 50 units)

    Full Kit - 1GB (USD$139)
    (Save 18% off the $169 Retail - Limited to 130 units)

    Full Kit - 2GB (USD$165)
    (Save 11% off the $185 Retail - Limited to 70 units)
     
    Our kickstarter page :  Helios4
     
    Cheers ;-)
     
  19. Like
    gprovost got a reaction from Igor in Support of Helios4 - Intro   
    Hey guys,
     
    Awesome news, our Kickstarter campaign is finally live. Be amount the first backers and enjoy our early bird deals :

    Our kickstarter page :  Helios4
     
    For some specs : Helios4_Specifications.pdf
     


    Basic Kit - 1GB (USD$125)
    (Save 16% off the $149 Retail - Limited to 50 units)

    Full Kit - 1GB (USD$139)
    (Save 18% off the $169 Retail - Limited to 130 units)

    Full Kit - 2GB (USD$165)
    (Save 11% off the $185 Retail - Limited to 70 units)
     
    Our kickstarter page :  Helios4
     
    Cheers ;-)
     
  20. Like
    gprovost got a reaction from zador.blood.stained in Support of Helios4 - Intro   
    Hey guys,
     
    Awesome news, our Kickstarter campaign is finally live. Be amount the first backers and enjoy our early bird deals :

    Our kickstarter page :  Helios4
     
    For some specs : Helios4_Specifications.pdf
     


    Basic Kit - 1GB (USD$125)
    (Save 16% off the $149 Retail - Limited to 50 units)

    Full Kit - 1GB (USD$139)
    (Save 18% off the $169 Retail - Limited to 130 units)

    Full Kit - 2GB (USD$165)
    (Save 11% off the $185 Retail - Limited to 70 units)
     
    Our kickstarter page :  Helios4
     
    Cheers ;-)
     
  21. Like
    gprovost got a reaction from TheLinuxBug in Support of Helios4 - Intro   
    @Igor, thanks I will look up for the ongoing change related to the clearfog.

    That's super great to read that you guys think it's a cool board. I had the strong feeling that I wasn't the only one looking for a board that's finally address the need to build a proper NAS on a fully open source ARM board. Coz yes it's the intention, if the campaign is a success, the whole hardware design will be released as open source (Open Hardware).
     
    Haha, well believe it or not, I'm not doing it for the money...If it was really only $$$ that drives this project, I don't think it will actually worth it. My goal is to make it a product that the community likes and adopt...it will be my ultimate satisfaction. But ok deal, if it the campaign succeed, you will get your box of beers ;-)

    Regarding the price point, I hope people will first understand that this a high performance SoC... it can't come as dirt cheap as some Pi-like boards. And that obviously I won't be manufacturing tens of thousands units. My target is 1000 units.

    As briefly explained, I will principally offer it as a full kit (AC/DC adapter 12V 8A, casing, complete cabling, fans and a little oled screen). My assumption is that if you set it up a NAS, you don't want a messy setup with HDDs hanging around. You want something neat and clean...and sweet to look at Anyway there will also be the barebone option.

    The price of the full kit will be in the range of 170 - 200 USD.

    I'm fully aware of the competition (Synology, WD, etc...) and I can confidently say that apple to apple specs comparison is way cheaper.

    Here as some pictures of what it looks like with the casing. (note : this is the casing without the oled screen).
     




     
     
    No. It's non-ECC memory. The SoM supports ECC though, but it will bump the price quite a bit. However it could be a stretch goal or an additional option. I don't wish to open here a discussion on ECC memory in the context of NAS, there are already plenty of years old ongoing threads on [non-]ECC memory for RAID or ZFS. While Helios-4 board is clearly not targeting mission critical application, the likeliness of corrupted data due to bit rot + existing "mitigation" feature (beside doing backups), doesn't in my opinion worth the premium of ECC memory for the scope of this board.
    Any opinion are much welcome ;-)

    Cheers.

    Gauthier.
  22. Like
    gprovost got a reaction from tkaiser in Support of Helios4 - Intro   
    @Igor, thanks I will look up for the ongoing change related to the clearfog.

    That's super great to read that you guys think it's a cool board. I had the strong feeling that I wasn't the only one looking for a board that's finally address the need to build a proper NAS on a fully open source ARM board. Coz yes it's the intention, if the campaign is a success, the whole hardware design will be released as open source (Open Hardware).
     
    Haha, well believe it or not, I'm not doing it for the money...If it was really only $$$ that drives this project, I don't think it will actually worth it. My goal is to make it a product that the community likes and adopt...it will be my ultimate satisfaction. But ok deal, if it the campaign succeed, you will get your box of beers ;-)

    Regarding the price point, I hope people will first understand that this a high performance SoC... it can't come as dirt cheap as some Pi-like boards. And that obviously I won't be manufacturing tens of thousands units. My target is 1000 units.

    As briefly explained, I will principally offer it as a full kit (AC/DC adapter 12V 8A, casing, complete cabling, fans and a little oled screen). My assumption is that if you set it up a NAS, you don't want a messy setup with HDDs hanging around. You want something neat and clean...and sweet to look at Anyway there will also be the barebone option.

    The price of the full kit will be in the range of 170 - 200 USD.

    I'm fully aware of the competition (Synology, WD, etc...) and I can confidently say that apple to apple specs comparison is way cheaper.

    Here as some pictures of what it looks like with the casing. (note : this is the casing without the oled screen).
     




     
     
    No. It's non-ECC memory. The SoM supports ECC though, but it will bump the price quite a bit. However it could be a stretch goal or an additional option. I don't wish to open here a discussion on ECC memory in the context of NAS, there are already plenty of years old ongoing threads on [non-]ECC memory for RAID or ZFS. While Helios-4 board is clearly not targeting mission critical application, the likeliness of corrupted data due to bit rot + existing "mitigation" feature (beside doing backups), doesn't in my opinion worth the premium of ECC memory for the scope of this board.
    Any opinion are much welcome ;-)

    Cheers.

    Gauthier.
  23. Like
    gprovost got a reaction from gnasch in Support of Helios4 - Intro   
    Hi Igor, Hi Armbian team,
     
    I wanted to give a short teaser of an upcoming board I have been developing. The board will be launched through a crowd-funding campaign in couple of weeks.
     
    Codename : Helios-4
     

     
    I'm sure the heatsink will give you some hint The board has been build around the Marvell Armada 388 SoC / SoM from SolidRun. So part of the design is inspired from the Clearfog dev board. Actually the SolidRun helped me on the PCB layout to make this little guy reality.
     
    Here the main specs :
    - dual core A9 @1.6 GHz (max 2Ghz) - 1GB DDR3 (2GB optional) - 4x SATA III - 2x USB 3.0 - 1x Gigabit Ethernet - Micro SD card - Built-in HDD power circuitry for 4 disks  - 2x Fan PWM headers - I2C, GPIO and extension pins - Micro USB to Serial - Dimension 100mm x 100mm   So clearly I have started this project with one main goal in mind : to finally bring to market the 1st ARM based board which has been designed specially for NAS purpose. I was frustrated to not find any affordable ARM board which has more than 1 SATA port.   Optionally the board will come along with a full kit, acrylic casing fitting 4x HDD (3.5'), 2x FAN, oled screen, cabling and power adaptor. Once assembled, it's the smallest 4x 3.5' disk NAS that you could find our there ;-) If people interested, I will post some pic.   Anyhow I stop my pitch here ;-)   Having used Armbian images and the build framework / toolchain for quite some times, it was logical for this project that the baseline image for Helios4 would be done via the Armbian project.   I recently started to dig into your repo to see how to contribute and squeeze my board in. I might need some guidance later, so I hope I can seek a bit of support from the team. I'm also super keen to donate aboard if it can help ;-)   Hope it was the right place to post this.   Cheers.
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines