A GPU version of Enigma software

Message boards : Number crunching : A GPU version of Enigma software

To post messages, you must log in.

Previous · 1 . . . 8 · 9 · 10 · 11

AuthorMessage
Aurel

Send message
Joined: 26 Sep 12
Posts: 18
Credit: 921,616
RAC: 1
Message 13559 - Posted: 2 Dec 2017, 10:38:43 UTC - in response to Message 13558.  

"Run test applications" in Your preferences. It helped in my case.


Activated it, atm "project has no tasks available".
ID: 13559 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
tito

Send message
Joined: 3 Sep 07
Posts: 6
Credit: 123,852,677
RAC: 21,942
Message 13561 - Posted: 4 Dec 2017, 18:57:22 UTC - in response to Message 13559.  
Last modified: 4 Dec 2017, 18:59:03 UTC

Hmmm.
My settings "ticked":
Use NVidia GPU
Run test applications

Run only the selected applications
Enigma 0.76: no
Enigma AV: no
Enigma GPU: yes
Enigma GPU - Test version: yes

After I checked that fields WU started to flow.

Edit: maybe You have app_config that cause error?
ID: 13561 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Aurel

Send message
Joined: 26 Sep 12
Posts: 18
Credit: 921,616
RAC: 1
Message 13565 - Posted: 6 Dec 2017, 12:15:50 UTC - in response to Message 13561.  
Last modified: 6 Dec 2017, 12:35:12 UTC

So, now I´m getting tasks!

But...
wrapper [26118 765 +dev#]: starting (GPU #0)
13:07:07 (33288): wrapper: running enigma_cuda.exe (-d 0 -R -o results.txt 04trigr.cur 04unigr.cur 04ciphertext)
CUDA error: C:/Proj/Other/Enigma/enigma-cuda/enigma-cuda-lib/cuda_code.cu:784:  CUDA driver version is insufficient for CUDA runtime versionInitialization failed: CUDA error: C:/Proj/Other/Enigma/enigma-cuda/enigma-cuda-lib/cuda_code.cu:784:  CUDA driver version is insufficient for CUDA runtime version
app exit status: 0x2


I´m updating my driver, which is now 1 year old... :)
They should work then, I hope.
Anyway, never had any problem on Einstein or BURP with GPU work. :/

EDIT: Seems to work. :)
ID: 13565 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
matszpk
Project tester
Volunteer developer
Volunteer tester

Send message
Joined: 20 May 11
Posts: 9
Credit: 42,566,517
RAC: 0
Message 13566 - Posted: 8 Dec 2017, 20:16:10 UTC

Few weeks ago, I published new version of enigma-opencl that supports RX VEGA under Mesa3D drivers. This version is working under drivers without full OpenCL support (even if clinfo doesn't works under RX VEGA) thanks embedded all in an assembly code.
New version in: https://github.com/matszpk/enigma-cuda/releases/tag/gcnopt0.2r1-3-test
ID: 13566 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
europe64

Send message
Joined: 9 Dec 15
Posts: 3
Credit: 100,265,550
RAC: 797
Message 13575 - Posted: 17 Dec 2017, 17:00:15 UTC

hi,
can we have some wu´s for amd gpu´s please ?
ID: 13575 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Bill Michael

Send message
Joined: 22 Sep 15
Posts: 8
Credit: 16,937,145
RAC: 6,457
Message 13578 - Posted: 19 Dec 2017, 18:25:48 UTC - in response to Message 13575.  

hi,
can we have some wu´s for amd gpu´s please ?


And for CPUs? Nothing from Enigma AV in a while...
ID: 13578 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Christophe Daulie

Send message
Joined: 29 Aug 15
Posts: 1
Credit: 323,992
RAC: 51
Message 13584 - Posted: 31 Dec 2017, 9:00:44 UTC

GPU WU's always get an error message when calculating starts.
They always get aborted.

How comes ?
ID: 13584 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Helix Von Smelix

Send message
Joined: 31 Oct 07
Posts: 36
Credit: 1,198,682,898
RAC: 2
Message 13585 - Posted: 31 Dec 2017, 10:30:16 UTC - in response to Message 13584.  

GPU WU's always get an error message when calculating starts.
They always get aborted.

How comes ?

Nvidia??
ID: 13585 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
matszpk
Project tester
Volunteer developer
Volunteer tester

Send message
Joined: 20 May 11
Posts: 9
Credit: 42,566,517
RAC: 0
Message 13589 - Posted: 12 Jan 2018, 16:49:09 UTC

New version enigma-opencl (Linux and Windows) and enigma-cuda for the Linux. I built Linux versions by using older GCC 4.8 compiler and these versions are comformant with older system C++ libraries (can works on more and older Linux systems with C++ library from GCC 4.8 compiler). New binaries are here: https://github.com/matszpk/enigma-cuda/releases/tag/gcnopt0.2r1-4-test.
If you encountered troubles with running enigma-opencl or enigma-cuda for Linux you can download these binaries and make an app_info.xml.
I hope that these versions will be deployed soon (I notifying administrator about it).
ID: 13589 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Yavanius
Avatar

Send message
Joined: 13 May 15
Posts: 15
Credit: 435,693
RAC: 0
Message 13590 - Posted: 13 Jan 2018, 18:37:10 UTC - in response to Message 13589.  

Open-GL, is that available for Intel HD GPU??

I have a dual GPU chip system which I can take advantage of both the Intel and nVidia.
ID: 13590 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Variable

Send message
Joined: 5 Jan 18
Posts: 4
Credit: 1,292,330,417
RAC: 3,041,273
Message 13596 - Posted: 16 Jan 2018, 19:14:46 UTC - in response to Message 13589.  

How does one use these test versions? I've never downloaded apps for Boinc manually, just instead gotten whatever the project sends to my machine. Is there a 'manual' for this somewhere?
ID: 13596 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
ikeke1

Send message
Joined: 26 Sep 17
Posts: 6
Credit: 129,705,722
RAC: 0
Message 13597 - Posted: 16 Jan 2018, 20:19:06 UTC - in response to Message 13596.  

How does one use these test versions? I've never downloaded apps for Boinc manually, just instead gotten whatever the project sends to my machine. Is there a 'manual' for this somewhere?


Yes, how does one use these?
ID: 13597 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
tolwyn

Send message
Joined: 14 Oct 12
Posts: 4
Credit: 3,505,320
RAC: 0
Message 13600 - Posted: 20 Jan 2018, 14:25:13 UTC

Go to the project settings and tick the box that you want to use test applications.
ID: 13600 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile nexiagsi16v

Send message
Joined: 31 Jan 15
Posts: 8
Credit: 1,189,104
RAC: 5
Message 13643 - Posted: 9 Jul 2018, 7:51:07 UTC

Are the GPU-WUs still Beta-Versions?
ID: 13643 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Amauri

Send message
Joined: 28 Feb 18
Posts: 2
Credit: 7,941,306
RAC: 10
Message 13675 - Posted: 19 Oct 2018, 23:44:01 UTC - in response to Message 13589.  

New version enigma-opencl (Linux and Windows) and enigma-cuda for the Linux. I built Linux versions by using older GCC 4.8 compiler and these versions are comformant with older system C++ libraries (can works on more and older Linux systems with C++ library from GCC 4.8 compiler). New binaries are here: https://github.com/matszpk/enigma-cuda/releases/tag/gcnopt0.2r1-4-test.
If you encountered troubles with running enigma-opencl or enigma-cuda for Linux you can download these binaries and make an app_info.xml.
I hope that these versions will be deployed soon (I notifying administrator about it).


Does anybody have a working app_info.xml for this, please?
ID: 13675 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
bluestang

Send message
Joined: 23 May 18
Posts: 5
Credit: 165,438,162
RAC: 1,134,214
Message 13685 - Posted: 19 Nov 2018, 17:58:56 UTC

I had to reset GPU performance stats listed here GPU speed because the script did not sum points correctly when the app runtime was very short and the app used less than 100s of CPU time. Mostly AMD cards running OpenCL apps were affected and I finally found the bug when I noticed RX Vega at the bottom with speed index of 0.


Is there still an issue with the Speed Index table?

I ran my Sapphire Vega 64 for a day or 2 about 4-5 weeks ago to test out what it can do and was expecting it to show/move the Vega 64 up the list quite a bit.
ID: 13685 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
mikey
Avatar

Send message
Joined: 13 Sep 16
Posts: 4
Credit: 3,484,206
RAC: 0
Message 13692 - Posted: 26 Dec 2018, 23:31:37 UTC - in response to Message 5859.  
Last modified: 26 Dec 2018, 23:32:41 UTC

For me the GPU app still only works on one GPU, allowing two WU's to run only runs them both on the same GPU which leaves any other GPU's doing nothing. For me at least nothing has changed :(

For sh#%s & giggles I even tried to changing the command to -d 1 in the job_160.xml file and another time tried pausing the WU and changing it in job.xml file in the slot directory but with the same result. As far as boinc is aware all is working as it should be with a WU running on each device but in reality there both being crunched by the same GPU.


Add a 2nd copy of Boinc in a totally different set of folders from the defualt if that's where the existing copy is installed and use the exclude command to exclude gpu0 in the existing boinc and gpu1 in the 2nd copy of Boinc, then you can mange each gpu independently.
ID: 13692 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 . . . 8 · 9 · 10 · 11

Message boards : Number crunching : A GPU version of Enigma software




Copyright © 2019 TJM