Posts by sorcrosc

1) Message boards : Wish List : please remove my account..... (Message 1749)
Posted 3 Apr 2020 by sorcrosc

TN-Grid as almost all other arm enabled BOINC projects provides application for platform arm-unknown-linux-gnueabihf. This will probably work just fine with your device.
You can choose this platform while compiling BOINC or just add this configuration to your cc_config.xml:

<cc_config> <log_flags> </log_flags> <options> <alt_platform>arm-unknown-linux-gnueabihf</alt_platform> </options> </cc_config>

2) Message boards : Number crunching : Error: disk limit exceeded (Message 1122)
Posted 1 Oct 2017 by sorcrosc

Where do I find the log?

In the boinc manager under Tools -> Event log

3) Message boards : Number crunching : Error: disk limit exceeded (Message 1120)
Posted 30 Sep 2017 by sorcrosc

"Use at most 0% of total disk space"

I'm not sure this is relevant but 0% here is not fine. If you want no limit in disk usage it should be 100%

Check "max disk usage" in boinc logs. What does it say?

4) Message boards : Unix/Linux : Your 64 bit ARM and 32 bit ARM applications: Raspberry Pi 3B (Message 1108)
Posted 13 Aug 2017 by sorcrosc

Look at applications page:
http://gene.disi.unitn.it/test/apps.php
There are three applications for armhf (32bit) and one for aarch64 (64bit).

Pi 1 get the plain 32bit version only with Raspbian OS.

Pi 2 get all three 32bit applications.

Pi 3 with 32bit os get all three 32bit applications.
Pi 3 with 64bit os get aarch64 64bit application.

5) Message boards : Number crunching : Optimization (Message 1054)
Posted 16 Apr 2017 by sorcrosc

This is a test run on my FX8320 with a real vv workunit kept from the project, the same for every app. Best time of 3 runs.

-> gene_pcim_v1.03_linux64__sse2 user 36m 4s -> gene_pcim_v1.03_linux64__avx user 35m 44s -> gene_pcim_v1.03_linux64__fma user 34m 47s -> TN-Grid.linux-x86-64-sse2-v1.2 user 32m 41s -> TN-Grid.linux-x86-64-avx-v1.2 user 32m 15s -> TN-Grid.linux-x86-64-fma-v1.2 user 31m 40s

6) Message boards : Number crunching : don't get any wus on android (Message 906)
Posted 11 Feb 2017 by sorcrosc

Need a rooted phone, sorry :)
Also tried NativeBOINC but the client refuse to start. I guess it's too much old for Android 6

7) Message boards : Number crunching : File size too big error (Message 874)
Posted 7 Feb 2017 by sorcrosc

It seems a problem with vv workunits. I found one too:
http://gene.disi.unitn.it/test/result.php?resultid=6650470

8) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 861)
Posted 6 Feb 2017 by sorcrosc

See here:
http://gene.disi.unitn.it/test/forum_thread.php?id=135&postid=793

9) Message boards : Number crunching : Optimization (Message 793)
Posted 23 Jan 2017 by sorcrosc

Here you can download other applications for Linux on ARM. Same Opti v1.1 code.

armv6
armv7_vfpv3
armv8

@koschi: try the latest one on your C2 :)

10) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 776)
Posted 21 Jan 2017 by sorcrosc

I have deleted armv8 and armv7_vfpv4 applications because are now distributed by the project. You can delete your app_info.xml and restart boinc if you are using them.

armv6 remains for older devices. I think it could be provided as a base armhf application (without plan class) if valterc think it worths.

armv7_vfpv3 could be useful for those without vfpv4 or that have found it a little faster. It is tuned for hardware considered base target by Debian and maybe all armhf distros (except Raspbian), so I leave it there.

11) Message boards : Number crunching : Optimization (Message 762)
Posted 14 Jan 2017 by sorcrosc

Thanks. Pretty much to the point I was making. Looks like you're in the same boat. What CPUs actually use avx effectively? Comparative times?

Of mine only the FX have avx. I posted my test on this machine above and there was a little advantage over sse2 but I verified later that running the application on all cores the advantage gets negligible.
Anyway we no more need to care about this. Optimized applications are distributed by the project, just delete app_info.xml and restart boinc (better if you abort and report workunits first) and you will get the right app.

12) Message boards : Number crunching : Optimization (Message 756)
Posted 14 Jan 2017 by sorcrosc

Immediate failure. I tested on 12/25 and those WUs are out of the database. On most machines the avx app ran but wasn't any faster (often slower) than the sse2 app. In fact I found no machines on which the avx app was superior. The fma app was a little bit faster than the other 2 versions on my AMD X8 boxes.

Be aware you need a capable cpu to run those applications. Looking at your computers, ony AMD FX have both avx and fma, Athlon 5350 have only avx. however, they are not known to be excellent in these capabilities :)

13) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 731)
Posted 6 Jan 2017 by sorcrosc

mmm, I have very different results. It's an Hikey board (8xCortex-a53, 1.2 GHz, fp asimd evtstrm aes pmull sha1 sha2 crc32) running Debian

armv6 doesn't work here and armv8 is the best

-> pc_armv7_vfpv3 real 5m50.693s -> pc_armv7_vfpv4 real 5m57.138s -> pc_armv8-a real 5m31.885s

14) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 727)
Posted 5 Jan 2017 by sorcrosc

I have just read that AARCH64 CPUs has new NEON SIMD instructions with double precision support, so it should be possible to get additional speed boost by using them. Probably it is time to get some Odroid C2 and play with it a bit :)

I like this :)

15) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 725)
Posted 5 Jan 2017 by sorcrosc

Here are all tarballs with new application based on Daniel optimized code and app_info included:

armv6
armv7_vfpv3
armv7_vfpv4
armv8

If you want to test the fastest one then install git and clone the repo:

git clone https://github.com/sorcrosc/rpi-boinc-ap

Then cd to the bin directory and untar one by one the apps you want to test

cd rpi-boinc-ap/TN-Grid/bin tar -xzf pc_armv7_vfpv3.tgz tar -xzf pc_armv7_vfpv3.tgz tar -xzf .....

Go to the upper directory and run the test script (stop boinc computation first). It should take 5-10 minutes every app.

cd .. ./test_run.sh

Further info:
I crosscompiled all the apps with latest gcc 6.2 release from Linaro here. Fresh and ready to use in case project admin want to look in to it ;)
armv8 with aarch64-linux-gnu
armv7 with arm-linux-gnueabihf
armv6 like arm-linux-gnueabihf but I recompiled it through crosstool-ng because released binaries are configured for armv7. Maybe it doesn't worth the pain because I am the only one who still use the first old Raspberry Pi 1 here :)

16) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 717)
Posted 3 Jan 2017 by sorcrosc

I'm replacing all versions with others based on your code, Daniel. Only the armv6 is missing now.
The armv8 koschi tested is already yours and seems also about 5% faster over armv7 on my device.
I also changed the test data with the Daniel's one (ec experiments) which show more realistic results.

17) Message boards : Number crunching : Optimization (Message 670)
Posted 20 Dec 2016 by sorcrosc

Impressive! This is on my AMD FX8320:

bin/pc_avx input/tile2.txt output/output2.txt 0.05 1 2470 real 0m48.941s user 0m46.920s sys 0m0.032s diff: output/output2.txt: File o directory non esistente bin/pc_fma input/tile2.txt output/output2.txt 0.05 1 2470 real 0m47.241s user 0m45.216s sys 0m0.036s diff: output/output2.txt: File o directory non esistente bin/pc_sse2 input/tile2.txt output/output2.txt 0.05 1 2470 real 0m50.553s user 0m48.520s sys 0m0.048s diff: output/output2.txt: File o directory non esistente bin/pc_tngrid_expansion_v9_linux64 input/tile2.txt output/output2.txt 0.05 1 2470 real 1m20.760s user 1m18.744s sys 0m0.032s diff: output/output2.txt: File o directory non esistente

I'm not able to obtain a result file to compare. Is there a trick?

18) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 632)
Posted 10 Dec 2016 by sorcrosc

I added right now an armv8 application. This should work in 64 bit only os.
I can't edit my first post, download here:

armv8

19) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 624)
Posted 4 Dec 2016 by sorcrosc

I used some -m flags but they don't make much difference. See the build scripts in my repo
:
See also the Makefile, I added there the -fsigned-char flag. This is very important, without that the code don't works. I suspect this is your problem :)

20) Message boards : Number crunching : Gene application for GNU/Linux on ARM devices (Message 622)
Posted 4 Dec 2016 by sorcrosc

Thank you for testing koschi.

The project source code is available as announced here.

I am thinking to take a 64 bit board, maybe an Odroid C2 or an Hikey and I'll try to compile but I really don't know if there will be any gain cause I'm not a programmer.

Next 20