1)
Message boards :
Sixtrack Application :
Computation errors on Android (31 (0x0000001F) Unknown error code)
(Message 40432)
Posted 13 Nov 2019 by James Molson Post: Been getting these compute errors immediately on my smartphone: https://lhcathome.cern.ch/lhcathome/result.php?resultid=251147453 Any ideas? What you are seeing is this: https://android-developers.googleblog.com/2017/07/seccomp-filter-in-android-o.html The current "android" executable is just a big statically linked linux executable, built on a linux machine. The newer versions of android do not like this. We put a filter in on the android version to stop this executable being sent to the newer versions of android, but it does not seem to work due to a bug in boinc itself, which I hope will be fixed at some point. Perhaps Nils or Laurence can comment on that last bit since it is out of my hands :) Eventually I will make a real android executable if I get the time - the annoying part is making a fortran compiler that will work with the android libraries. |
2)
Message boards :
Number crunching :
LHC on android
(Message 40090)
Posted 8 Oct 2019 by James Molson Post: Okay, I have one Android-6-Handy that asks for work since weeks, but never got any task: https://lhcathome.cern.ch/lhcathome/show_host_detail.php?hostid=10603561 The arm/android executable is aarch64, your host is armv7, which is 32bit arm, thus you will not get sent anything. |
3)
Message boards :
Number crunching :
Is LHC@Home exporting statistics?
(Message 39483)
Posted 2 Aug 2019 by James Molson Post: Is it working now? |
4)
Message boards :
Number crunching :
Run Boinc in Router
(Message 36306)
Posted 7 Aug 2018 by James Molson Post: If you can get me some info on the system/processor, then sure, I could build a sixtrack version for them. |
5)
Message boards :
News :
test of SixTrack 5.00.00
(Message 35876)
Posted 12 Jul 2018 by James Molson Post: Nice! It should work - I will set up my pi again and build it in the next few days. |
6)
Message boards :
Sixtrack Application :
11 (0x0000000B) Unknown error code
(Message 35743)
Posted 2 Jul 2018 by James Molson Post: Hi, I made the executable you are using. It was built on the latest version of OSX. I see what you are running is quite old now. The executable is not fully statically linked, and has to pull in some libraries from the running system, so I think there is a miss-match here, and this is why you are getting the segfault. In principle sixtrack should run perfectly fine on your computer. Would it be possible to build an executable yourself and use the anonymous platform mechanism? I can give you a guide on how to do this if you are willing to do so. |
7)
Questions and Answers :
Wish list :
Lhc@Home open source?
(Message 35077)
Posted 23 Apr 2018 by James Molson Post: For sixtrack at least: https://github.com/SixTrack/SixTrack |
8)
Message boards :
Sixtrack Application :
AVX Sixtrack version
(Message 34295)
Posted 5 Feb 2018 by James Molson Post: HI, any benefit for AVX-512? Of course, but currently I do not have a machine to test it with. For the next release I want to provide an AVX-512 executable. |
9)
Message boards :
Sixtrack Application :
AVX Sixtrack version
(Message 33892)
Posted 17 Jan 2018 by James Molson Post: Hello all, The issue preventing the AVX executable from being distributed has finally been found and resolved. If your computer has AVX support, you should now be sent the AVX executable. This may take a short while to stabilise. If you think you should be sent the AVX executable and this is not happening, please comment below. |
10)
Message boards :
Number crunching :
Android x86 app
(Message 33802)
Posted 12 Jan 2018 by James Molson Post: Hi. I have an ASUS ZenFone Zoom smartphone (Intel Z3580 CPU quad-core 2.3GHz) but at the moment I can only help SETI research. Can you tell me when you will add an app for x86 Android devices? We can run on android devices, see sixtracktest here: https://lhcathome.cern.ch/lhcathome/apps.php Currently there are two issues: 1: The ram usage of sixtrack is relatively high - almost all internal arrays are allocated with a fixed size at compile time, and set to a "worst case" capacity. A lot of android devices just do not have sufficient spare ram. Work is in progress to re-structure the internals to reduce the memory usage and only allocate what is needed. 2: The security model in the current android release has restricted the usage of some syscalls, and we seem to use some of the restricted ones. This needs to be changed. Once both of these issues have been resolved, we can push the android version out to production. |
11)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 32576)
Posted 30 Sep 2017 by James Molson Post: Hi, the minimum is 2.6.32: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.32 That kernel has not been supported for about 7 years (2.6.31). |
12)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 32181)
Posted 1 Sep 2017 by James Molson Post: I made a short test with my tablet but all tasks failed: My Pixel C. The no CPU type is a BOINC bug with newer android versions. The crash is due to android 8 (I think). See: https://android-developers.googleblog.com/2017/07/seccomp-filter-in-android-o.html would it be possible to get a logcat to see which syscall is getting blocked? |
13)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 32178)
Posted 1 Sep 2017 by James Molson Post: The 32bit arm exe (i.e. arm v6 and v7) has not been pushed out yet. The 64bit one, (arm v8) is up and running now. Alessio should fix it soon :) Your host is arm v7. |
14)
Message boards :
Sixtrack Application :
More Work
(Message 32168)
Posted 1 Sep 2017 by James Molson Post: There are sixtracktest tasks available currently. Of current interest right now would be running these tasks on any machine that isn't x86 windows or linux if possible. This would be arm32/64 or ppc64 (on linux) cpus, android devices, and x86 bsd including osx. |
15)
Message boards :
Number crunching :
Can't set Project Preferences ?
(Message 32031)
Posted 21 Aug 2017 by James Molson Post: which url are you trying to access the preferences with? |
16)
Message boards :
Number crunching :
No GPU work?
(Message 32013)
Posted 20 Aug 2017 by James Molson Post: There is not any GPU for this project. BOINC just requests it? Sixtrack runs on android, i.e. 32/64 bit arm/linux, and each BSD variant. NetBSD and OpenBSD needs a little tweaking though :) We also have ppc64/linux, but I don't think there is an icon for that. If there is a fortran compiler for a platform, we can build for it. |
17)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 31986)
Posted 17 Aug 2017 by James Molson Post: Building a 32bit version for armv7 would need a little modification to the build system, but should be simple to do. It is in my todo list. Seems good :) fort.6 contains stdout which is a better place to look for errors, but that seems to have exited okay. |
18)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 31964)
Posted 16 Aug 2017 by James Molson Post: Building a 32bit version for armv7 would need a little modification to the build system, but should be simple to do. It is in my todo list. Try seeing if this will run on your system: https://cernbox.cern.ch/index.php/s/epIgB6FZL6jhxJl |
19)
Message boards :
Sixtrack Application :
SIXTRACKTEST
(Message 31907)
Posted 10 Aug 2017 by James Molson Post: I have seen that there are "apps" for arm now as well. I have a banana pi (similar to raspberry pi) which runs linux (and maybe could also run android) and i could test these new ones if this could/would help you. The current arm binaries are 64 bit (aarch64) armv8. Building a 32bit version for armv7 would need a little modification to the build system, but should be simple to do. It is in my todo list. The sources for sixtrack live on github, so if you want to you can have a go yourself. https://github.com/SixTrack/SixTrack |
20)
Message boards :
Sixtrack Application :
AVX/AVX2 support
(Message 31526)
Posted 20 Jul 2017 by James Molson Post: yes |
©2024 CERN