PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise

Advanced search

Message boards : Generalized Fermat Prime Search : High CPU Usage for GPU task

Author Message
jimjam
Send message
Joined: 19 Dec 16
Posts: 3
ID: 476882
Credit: 31,165,182
RAC: 0
PPS Sieve Sapphire: Earned 20,000,000 credits (28,373,707)GFN Ruby: Earned 2,000,000 credits (2,791,475)
Message 106379 - Posted: 25 Mar 2017 | 3:54:25 UTC
Last modified: 25 Mar 2017 | 3:54:49 UTC

Genefer WUs seem to take about 90% of a core on my computer. I see there have been some similar posts in the past but I'm not sure if they're relevant.

Here's a particularly stark example:
http://www.primegrid.com/workunit.php?wuid=519284471

The graphics card is a 1080Ti, and I'm running Linux.

Any easy fix? I've tried restarting, but it didn't help.

Yves Gallot
Volunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 675
ID: 164101
Credit: 305,042,960
RAC: 0
GFN Double Silver: Earned 200,000,000 credits (305,042,960)
Message 106383 - Posted: 25 Mar 2017 | 9:22:11 UTC - in response to Message 106379.
Last modified: 25 Mar 2017 | 9:32:04 UTC

Genefer WUs seem to take about 90% of a core on my computer. I see there have been some similar posts in the past but I'm not sure if they're relevant.

Here's a particularly stark example:
http://www.primegrid.com/workunit.php?wuid=519284471

The graphics card is a 1080Ti, and I'm running Linux.

Any easy fix? I've tried restarting, but it didn't help.


Your device is 'Graphics Device' rather than 'GeForce GTX 1080 Ti'

Running on platform 'NVIDIA CUDA', device 'Graphics Device', version 'OpenCL 1.2 CUDA' and driver '378.13'.

On Windows, GeForce GTX 1080 Ti is supported by driver >= 378.78.
I think that it will be fixed on Linux when a new driver is available.

But it's very fast!
On a GTX 1080, time @ n=21 is about 44,000 sec and your GTX 1080 Ti takes 32,000 sec.

jimjam
Send message
Joined: 19 Dec 16
Posts: 3
ID: 476882
Credit: 31,165,182
RAC: 0
PPS Sieve Sapphire: Earned 20,000,000 credits (28,373,707)GFN Ruby: Earned 2,000,000 credits (2,791,475)
Message 106386 - Posted: 25 Mar 2017 | 9:43:56 UTC - in response to Message 106383.

I'm not so much concerned with the display bug as with the CPU usage, unless they're related..?

Yves Gallot
Volunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 675
ID: 164101
Credit: 305,042,960
RAC: 0
GFN Double Silver: Earned 200,000,000 credits (305,042,960)
Message 106387 - Posted: 25 Mar 2017 | 10:08:51 UTC - in response to Message 106386.

I'm not so much concerned with the display bug as with the CPU usage, unless they're related..?

I think that they are related.
If the GPU is not recognized, the CPU may spin on a status and wait for work completion rather than using a low-latency interrupt.

jimjam
Send message
Joined: 19 Dec 16
Posts: 3
ID: 476882
Credit: 31,165,182
RAC: 0
PPS Sieve Sapphire: Earned 20,000,000 credits (28,373,707)GFN Ruby: Earned 2,000,000 credits (2,791,475)
Message 106409 - Posted: 26 Mar 2017 | 4:37:08 UTC - in response to Message 106387.

Interesting, thanks! I will stay tuned for a driver update, then.

Profile Luigi R.Project donor
Avatar
Send message
Joined: 11 Feb 14
Posts: 173
ID: 297455
Credit: 56,579,172
RAC: 131
Discovered 1 mega primeFound 1 prime in the 2020 Tour de PrimesFound 2 primes in the 2021 Tour de PrimesFound 1 mega prime in the 2021 Tour de PrimesFound 1 prime in the 2021 Tour de Primes Mountain Stage321 LLR Silver: Earned 100,000 credits (124,678)Cullen LLR Silver: Earned 100,000 credits (138,607)ESP LLR Silver: Earned 100,000 credits (167,391)Generalized Cullen/Woodall LLR Silver: Earned 100,000 credits (150,362)PPS LLR Turquoise: Earned 5,000,000 credits (5,740,696)PSP LLR Silver: Earned 100,000 credits (208,396)SoB LLR Amethyst: Earned 1,000,000 credits (1,343,981)SR5 LLR Amethyst: Earned 1,000,000 credits (1,150,519)SGS LLR Silver: Earned 100,000 credits (142,451)TRP LLR Amethyst: Earned 1,000,000 credits (1,397,758)Woodall LLR Silver: Earned 100,000 credits (120,798)321 Sieve (suspended) Silver: Earned 100,000 credits (100,021)Generalized Cullen/Woodall Sieve (suspended) Silver: Earned 100,000 credits (115,766)PPS Sieve Sapphire: Earned 20,000,000 credits (24,753,253)TRP Sieve (suspended) Silver: Earned 100,000 credits (108,608)AP 26/27 Turquoise: Earned 5,000,000 credits (5,033,535)GFN Jade: Earned 10,000,000 credits (10,045,338)WW Turquoise: Earned 5,000,000 credits (5,076,000)PSA Gold: Earned 500,000 credits (661,014)
Message 111898 - Posted: 25 Nov 2017 | 20:26:16 UTC
Last modified: 25 Nov 2017 | 20:29:11 UTC

Hello, how do you explain this?

Running on platform 'NVIDIA CUDA', device 'GeForce GTX 750 Ti', vendor 'NVIDIA Corporation', version 'OpenCL 1.2 CUDA' and driver '384.98'.

Link: https://www.primegrid.com/result.php?resultid=858183803

This issue has not been solved by a driver update for years.

My device is 'GeForce GTX 750 Ti', not 'Graphics Device'.


It is not my priority to run LLR and GFN tests simultaneously. Maybe for someone it is.

Profile composite
Volunteer tester
Send message
Joined: 16 Feb 10
Posts: 850
ID: 55391
Credit: 812,725,514
RAC: 387,460
Discovered 2 mega primesFound 1 prime in the 2018 Tour de Primes321 LLR Turquoise: Earned 5,000,000 credits (5,477,467)Cullen LLR Gold: Earned 500,000 credits (776,297)ESP LLR Ruby: Earned 2,000,000 credits (3,433,680)Generalized Cullen/Woodall LLR Ruby: Earned 2,000,000 credits (2,443,837)PPS LLR Sapphire: Earned 20,000,000 credits (27,997,486)PSP LLR Turquoise: Earned 5,000,000 credits (6,587,988)SoB LLR Sapphire: Earned 20,000,000 credits (38,274,664)SR5 LLR Turquoise: Earned 5,000,000 credits (6,110,877)SGS LLR Ruby: Earned 2,000,000 credits (3,486,325)TRP LLR Turquoise: Earned 5,000,000 credits (7,070,795)Woodall LLR Amethyst: Earned 1,000,000 credits (1,693,614)321 Sieve (suspended) Emerald: Earned 50,000,000 credits (50,256,050)Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,571,178)Generalized Cullen/Woodall Sieve (suspended) Emerald: Earned 50,000,000 credits (50,009,610)PPS Sieve Double Silver: Earned 200,000,000 credits (409,037,761)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Jade: Earned 10,000,000 credits (10,165,888)TRP Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,071,454)AP 26/27 Turquoise: Earned 5,000,000 credits (6,798,063)GFN Emerald: Earned 50,000,000 credits (53,771,465)WW Gold: Earned 500,000 credits (932,000)PSA Double Bronze: Earned 100,000,000 credits (102,762,384)
Message 111901 - Posted: 25 Nov 2017 | 21:35:09 UTC - in response to Message 111898.
Last modified: 25 Nov 2017 | 21:36:52 UTC

Hello, how do you explain this?
Running on platform 'NVIDIA CUDA', device 'GeForce GTX 750 Ti', vendor 'NVIDIA Corporation', version 'OpenCL 1.2 CUDA' and driver '384.98'.
Link: https://www.primegrid.com/result.php?resultid=858183803
This issue has not been solved by a driver update for years.
My device is 'GeForce GTX 750 Ti', not 'Graphics Device'.
It is not my priority to run LLR and GFN tests simultaneously. Maybe for someone it is.

I too am seeing this problem with the GeneferOCL app on Linux.
This is not a driver problem, this is a regression in the GeneferOCL app.
What happened to GeneferOCL between 3.3.0 and 3.3.2-7?

The decision to discontinue GeneferCUDA was announced with a claim that the CPU usage issue with OCL was fixed.
Quoting the announcement (with my ephasis added):
We've now started formal testing of GeneferOCL 3.3.0. Once 3.3.0 is live, we plan on removing the GeneferCUDA apps.

As far as we know, in all situations, GeneferOCL is faster than GeneferCUDA. The only reason for using GeneferCUDA was that GeneferCUDA wouldn't use a full CPU core. GeneferOCL would use a full core, under some circumstances. Since GeneferOCL 3.3.0 has fixed the problem with CPU hogging, there appears to be no reason to use GeneferCUDA anymore.

If you have a reason why we should keep GeneferCUDA, now is the time to speak up. Once we turn it off, it will be too late.

Yves Gallot
Volunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 675
ID: 164101
Credit: 305,042,960
RAC: 0
GFN Double Silver: Earned 200,000,000 credits (305,042,960)
Message 111914 - Posted: 26 Nov 2017 | 10:12:10 UTC - in response to Message 111901.

I too am seeing this problem with the GeneferOCL app on Linux.
This is not a driver problem, this is a regression in the GeneferOCL app.
What happened to GeneferOCL between 3.3.0 and 3.3.2-7?


The interface with the GPU of GeneferOCL 3.3.0 and 3.3.2 are identical... a regression???
Was CPU usage of GeneferOCL 3.3 tested on Linux?

You can try this
https://www.reddit.com/r/EtherMining/comments/600lpx/high_cpu_load_with_genoil/

Profile composite
Volunteer tester
Send message
Joined: 16 Feb 10
Posts: 850
ID: 55391
Credit: 812,725,514
RAC: 387,460
Discovered 2 mega primesFound 1 prime in the 2018 Tour de Primes321 LLR Turquoise: Earned 5,000,000 credits (5,477,467)Cullen LLR Gold: Earned 500,000 credits (776,297)ESP LLR Ruby: Earned 2,000,000 credits (3,433,680)Generalized Cullen/Woodall LLR Ruby: Earned 2,000,000 credits (2,443,837)PPS LLR Sapphire: Earned 20,000,000 credits (27,997,486)PSP LLR Turquoise: Earned 5,000,000 credits (6,587,988)SoB LLR Sapphire: Earned 20,000,000 credits (38,274,664)SR5 LLR Turquoise: Earned 5,000,000 credits (6,110,877)SGS LLR Ruby: Earned 2,000,000 credits (3,486,325)TRP LLR Turquoise: Earned 5,000,000 credits (7,070,795)Woodall LLR Amethyst: Earned 1,000,000 credits (1,693,614)321 Sieve (suspended) Emerald: Earned 50,000,000 credits (50,256,050)Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,571,178)Generalized Cullen/Woodall Sieve (suspended) Emerald: Earned 50,000,000 credits (50,009,610)PPS Sieve Double Silver: Earned 200,000,000 credits (409,037,761)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Jade: Earned 10,000,000 credits (10,165,888)TRP Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,071,454)AP 26/27 Turquoise: Earned 5,000,000 credits (6,798,063)GFN Emerald: Earned 50,000,000 credits (53,771,465)WW Gold: Earned 500,000 credits (932,000)PSA Double Bronze: Earned 100,000,000 credits (102,762,384)
Message 111916 - Posted: 26 Nov 2017 | 13:11:27 UTC - in response to Message 111914.
Last modified: 26 Nov 2017 | 13:15:05 UTC

I too am seeing this problem with the GeneferOCL app on Linux.
This is not a driver problem, this is a regression in the GeneferOCL app.
What happened to GeneferOCL between 3.3.0 and 3.3.2-7?


The interface with the GPU of GeneferOCL 3.3.0 and 3.3.2 are identical... a regression???
Was CPU usage of GeneferOCL 3.3 tested on Linux?

You can try this
https://www.reddit.com/r/EtherMining/comments/600lpx/high_cpu_load_with_genoil/

GeneferOCL was thoroughly tested on all platforms for numerical accuracy.
I understand that one representative test per platform was enough to register a pass.

The mining discussion talks about 100% core loading with CUDA.

However CUDA PPS Sieve is using only 17% to 20% of a CPU core.

My CPU doesn't have integrated graphics, so there's no hope of trying that solution here.

EDIT:
GeneferOCL would use a full core under some circumstances.
That wasn't elaborated. Maybe that was with no integrated graphics?

Message boards : Generalized Fermat Prime Search : High CPU Usage for GPU task

[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2021 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.55, 0.58, 0.56
Generated 6 Dec 2021 | 6:27:09 UTC