![]() |
|
|
#1596 |
|
"Mr. Meeseeks"
Jan 2012
California, USA
23·271 Posts |
mfakto has often been a pain to use with multiple gpu's... it is somehow connected with the output of clinfo(don't exactly know how)... something like -d 10/11 etc might work.
Honestly, mfakto needs a lot of changes - it would be nice if the system gpuowl uses to detect gpu's were implemented here - although that requires code changes and it may be completely different from what mfakto needs and not work... also automatic kernel tuning and selection is needed, which would remove the need of users testing it themselves and adding to the code... |
|
|
|
|
|
#1597 |
|
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
123628 Posts |
Needs the right parameters. I run mfakto as follows on multi-gpu systems:
Code:
set gpuappinstance=mfakto-RX550 :set gpuappinstance=mfakto-RX480 mfakto -d 02 >>mfakto.txt echo At %date% %time% %computername% %gpuappinstance% exited >>\\wdmycloud\ken\sources\mersennes\appwatch\%computername%-%gpuappinstance%.txt set gpuappinstance= goto exit :all the following put mfakto activity on gpu 0: : 00 01 1 : this puts it on gpu 1: 02 : this puts it on gpu 2 (the third) 03 :exit Last fiddled with by kriesel on 2020-03-31 at 05:58 |
|
|
|
|
|
#1598 |
|
Mar 2017
Germany, Wolfsburg
3 Posts |
Hello,
I should post this: Unknown GPU name, assuming GCN. Please post the device name "gfx1012 (Advanced Micro Devices, Inc.)" (error message when doing the self test) |
|
|
|
|
|
#1599 |
|
"AMD YES!"
Jan 2020
Bellevue, WA
2·52 Posts |
Same thing with gfx902, which is Vega 8 iGPU, but the error message does no harm
|
|
|
|
|
|
#1600 |
|
Sep 2011
Germany
22×32×79 Posts |
We are running into an issue with the mfakto_Kernels.elf file. (v0.15pre6)
http://srbase.my-firewall.org/sr5/fo...27&postid=6025 I have tried this on my 5500XT and its working, on the 5700XT not. Any ideas how to fix this? Some more errors seen like this: Loading binary kernel file mfakto_Kernels.elf Cannot use binary kernel: its build options (-I. -DVECTOR_SIZE=2 -DGCN -O3 -DMORE_CLASSES -DCL_GPU_SIEVE) are different than the current build options (-I. -DVECTOR_SIZE=2 -DVLIW5 -O3 -DMORE_CLASSES -DCL_GPU_SIEVE). Rebuilding kernels. Compiling kernels. Last fiddled with by rebirther on 2020-04-04 at 21:09 |
|
|
|
|
|
#1601 | |
|
If I May
"Chris Halsall"
Sep 2002
Barbados
230028 Posts |
Quote:
I don't have any OpenCL kit -- I'm an Nvidia guy. And I don't know of any cloud providers which offer access to AMD cards by way of instances. |
|
|
|
|
|
|
#1602 |
|
"James Heinrich"
May 2004
ex-Northern Ontario
3×5×227 Posts |
I have an RX 480, but no insight, since it works fine here (as it does on his 5500XT).
|
|
|
|
|
|
#1603 |
|
"Mr. Meeseeks"
Jan 2012
California, USA
41708 Posts |
Going to try to add "support"(bandaid fix is the best term) for newer GPU's... Could I have the output from --perftest for any relatively new GPU's? (Radeon VII, any vega(RX Vega, RX 5xxx etc.
|
|
|
|
|
|
#1604 | |
|
Bemusing Prompter
"Danny"
Dec 2002
California
2·5·239 Posts |
Quote:
|
|
|
|
|
|
|
#1605 |
|
Bemusing Prompter
"Danny"
Dec 2002
California
239010 Posts |
Hey everyone, mfakto 0.15pre7 is now available on GitHub. Here are the changes:
Please note that I wasn't able to test the latest code on the Mac Pro at my office due to the COVID-19 pandemic. Feel free to slap me with a trout if I broke anything. Last fiddled with by ixfd64 on 2020-05-05 at 23:05 |
|
|
|
|
|
#1606 |
|
"James Heinrich"
May 2004
ex-Northern Ontario
D4D16 Posts |
|
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| gpuOwL: an OpenCL program for Mersenne primality testing | preda | GpuOwl | 2718 | 2021-07-06 18:30 |
| mfaktc: a CUDA program for Mersenne prefactoring | TheJudger | GPU Computing | 3497 | 2021-06-05 12:27 |
| LL with OpenCL | msft | GPU Computing | 433 | 2019-06-23 21:11 |
| OpenCL for FPGAs | TObject | GPU Computing | 2 | 2013-10-12 21:09 |
| Program to TF Mersenne numbers with more than 1 sextillion digits? | Stargate38 | Factoring | 24 | 2011-11-03 00:34 |