Folding@home-Team Linux Mint-76140

[TeaSwigger]

Like some other folks I'm sure, I have an AMD APU and am obliged to use the newer radeon drivers, which don't support using the gpu side for folding at present. How would we know if/when we may try setting up gpu folding again?

Right now I'm trying a little zotac ci320 mini-pc. Nice little box. To me, it's still surprising to have something that would outperform many full tower PCs of a decade ago in the palm of one's hand. It's fanless so I've been cautious trying it for folding, but found it (just) handles folding (medium, 3 of 4 cores) ok without overheating or unduly affecting it for regular light use. It's not a strong folder but at least it doesn't take much power.

Heatwave here in Blighty, having to cope with ambient of 30 deg c. My f@h setup is in the garden shed and it's baking. And it doesn't get much cooler on a night time.

Hope it's eased up for you.

It's 41c out where I am, will be 30-40 'till late autumn. Fortunately the a/c is keeping it reasonable indoors. Not a thing would go in a garden shed unless it's on the way to dinner.

[LazyDev]

I was considering stopping folding as my water temp was getting too high. It hit just over 50.C the other day, with 60.C been the absolute limit. I've just about got the temperature under control.

I have an old Radeon 7870 that I've had for the past 4 years. Anyone know the PPD? I don't think it'll be worth it to wire it up.

[Citizen229]

The 7870 is a GCN1.0 card. You will not be able to fold it on Linux. The new AMDGPU-pro drivers will not work with it. And the FGLRX drivers from Mint17 do not have open CL2 for folding. Nothing AMd will fold unless it is a RX series card.

If you were on Mint 17.3 you ould use the FGLRX drivers to use the AMD card as you daily driver card. This way your video streaming does not hinder you Nvidia card. The open source only drivers in Mint 18 are kinda pita with screen tearing.

[Citizen229]

Actually that card may come in handy for the july6-7 FAH shutdown. Technically for the shutdown one could use a second GPU to have 10-12 gpu tasks downloaded before the shutdown. Then when the servers come back up a massive upload dump to turn in work. The AMd card will not fold, but it gets recognized enough to DL a WU. You could plug in that AMD card and add its slot to get a WU then delete the slot. The WU will stay as it will wait for the other gpu. The same can be done for cpu's as well. Your 4 core 8 thread cpu can be set up as 6 threads and 1 threads. the 8th thread being system reserved for Nvidia drivers. Use the single thread to DL a small pile of WU's for the downtime. then delete the single thread and let you 6 threads chew the wu up during the downtime.

these WU will lose bonus points for time during the 2 days, but its better than no points.

I have my r9 280x(7970) sitting here, I may do this.

[900i]

Well done team, It's official now that we have gone into the top 300,


But on a sad note, my original gtx960 has died and I have replaced it with a spare. I think I just pushed it too hard over the last week or two and the temps finally got to it.

[Citizen229]

Been lucky so far this year, again. Have only had a few days so far get to 90F. I put a fan in the window last night, and as of 10am est it is 65F in the house. I hope it stays with this pattern till mid July like last year.

I missed the start of the new page Helped a friend get a few head to the knockerhouse, then we grilled meat and drank beer.

[tinca]

900i,

under 300 for the first time, never thought the day would come. It is some achievement.

sorry to hear about your GPU. The weather here has cooled down a lot in the past day, and best of all there is a good breeze.

best regards keith

[powerwagon75]

Hardware failures....yeah you got to love them. Been on and off folding since Friday trying to diagnose a case of the slow cpu freq's (stuck at 800mhz). I was getting pretty annoyed with myself for all the time it I was spending on it, but, I think I learned more about acpi and intel cpu frequency control, c-states, latency, etc.. than I thought I would ever need to. Boiled down to something bad on the motherboard. Seems to be an issue with the MSI Z170 boards, where something goes wrong in the slow-boot/thermal protection where outside thermal sensors can send send erroneous data, causing the cpu to be throttled, to try and control the heat of the vrm's and such. Knew it wasn't likely a true thermal issue, as woke it up from sleep, where the windows had been open all night--cpu was only reporting 17C at start up, so nothing else should have been too warm yet. The work-around solution was to disable the "intel_thermal protection" in bios. Made a few other changes along the troubleshooting path, so it will fold with full cpu freq now. Proper solution is to rma the board, but think I am out of that window now, plus had to do that before and it was a two week turn-around. I "stressed" it with Full Power folding on gpu and cpu for a while, and temps (for the cpu) stayed in the high 70C's. [cpu temps are in the mid-30Cs, with the "busy core" around low 50Cs with just gpu folding.] I think I am going to rig up a couple of small fans very soon, that can be positioned to cool the vrm's, as they do run pretty darn hot to the touch.

i7-6700K: 4.07Ghz, Vcore @ 1.189v
MSI Z170 Gaming M7 motherboard

[LazyDev]

Been trying to get my IGP working, so I could use that for my desktop, and use my GPU for sole Folding. No luck at all. Seems Linux Mint doesn't like multi-GPU's.

I've tried:
Bumblebee
Nvidia Prime
Drivers 375

It's plug and play with Windows... Normally, the setup would only take 5 minutes. I need proper steps and guides.

Edit:
I have Nvidia prime set to the 1060, but when I plug a monitor into the IGP, I just get 5 dots in the center of the screen, nothing else. It's as if Linux is refusing to use 2 GPU's simultaneously.

Edit #2:
Found this: viewtopic.php?f=59&t=236026&p=1256065#p1256065
This is exactly what I want to achieve.

[LazyDev]

Code: Select all

17:07:41:WU01:FS01:0x21:Project: 11402 (Run 13, Clone 33, Gen 145)
17:07:41:WU01:FS01:0x21:Unit: 0x000000bc8ca304f255ed4ee4c5e9469a
17:07:41:WU01:FS01:0x21:CPU: 0x00000000000000000000000000000000
17:07:41:WU01:FS01:0x21:Machine: 1
17:07:41:WU01:FS01:0x21:Reading tar file core.xml
17:07:41:WU01:FS01:0x21:Reading tar file system.xml
17:07:41:WU01:FS01:0x21:Reading tar file integrator.xml
17:07:41:WU01:FS01:0x21:Reading tar file state.xml
17:07:41:WU01:FS01:0x21:Digital signatures verified
17:07:41:WU01:FS01:0x21:Folding@home GPU Core21 Folding@home Core
17:07:41:WU01:FS01:0x21:Version 0.0.18

17:07:41:WU01:FS01:0x21:ERROR:126: Bad platformId size. <--------------------

17:07:41:WU01:FS01:0x21:Saving result file logfile_01.txt
17:07:41:WU01:FS01:0x21:Saving result file log.txt
17:07:41:WU01:FS01:0x21:Folding@home Core Shutdown: BAD_WORK_UNIT
17:07:42:WARNING:WU01:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72)
17:07:42:WU01:FS01:Sending unit results: id:01 state:SEND error:FAULTY project:11402 run:13 clone:33 gen:145 core:0x21 unit:0x000000bc8ca304f255ed4ee4c5e9469a
17:07:42:WU01:FS01:Uploading 1.97KiB to 140.163.4.242

Code: Select all

user@user-MS-7851 ~/Downloads $ optirun nvidia-smi
Sun Jul  2 17:55:17 2017       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 381.22                 Driver Version: 381.22                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 106...  Off  | 0000:01:00.0     Off |                  N/A |
| 40%   24C    P0    28W / 120W |      6MiB /  6072MiB |      2%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0      7033    G   /usr/lib/xorg/Xorg                               6MiB |
+-----------------------------------------------------------------------------+

Maybe F@H doesn't like Nvidia Optimus?

[LazyDev]

Quick update:

I reset my config as it seems F@H doesn't like Nvidia Optimus, in fact, it refuses to use the GPU.

On later observations, I noticed that my GPU is spending alot of time on projects based on Cancer, and my CPU is running alot of unspecified projects.

[Citizen229]

Looks like we have a user who has returned. Welcome back Karlie!

[Citizen229]

Reminder! FAH will be down july 6th and 7th.

https://foldingforum.org/viewtopic.php?f=24&t=30020

[greerd]

Reminder! FAH will be down july 6th and 7th.

https://foldingforum.org/viewtopic.php?f=24&t=30020

Thanks for the heads-up. I'm not sure how one can prepare for this? Can extra WUs be downloaded?

[Citizen229]

yes scroll up this page. I describe how to queue WU when refering to Lazy's 7870.

[LazyDev]

Revisit: IGP for desktop, GPU for F@H

Steps I've taken:
1. Uninstalled everything Nvidia, and Nouvea, leaving in the Intel driver.
2. Plugged in my monitor in to the motherboard, leaving nothing plugged in to the graphics card, and set the IGP as default in the BIOS.
3. Rebooted to make sure the Intel IGP was been used, and the desktop is been rendered correctly.
4. Installed the nvidia-375 drivers.
5. Installed nvidia-settings and nvidia-prime
6. installed nvidia-cuda-toolkit

--From here on forwards, it is trial and error.

7. From the synaptic package manager:
-Installed nvidia-opencl-icd-367
-Installed Cuda Drivers
-Installed clinfo (To see OpenCL and Cuda status)
*These will uninstall previous installations in the above steps
8. Rebooted.

Code: Select all

user@user-MS-7851 ~ $ dpkg -l|grep nvidia
ii  nvidia-375                                  375.66-0ubuntu0.16.04.1                    amd64        NVIDIA binary driver - version 375.66
ii  nvidia-375-dev                              375.66-0ubuntu0.16.04.1                    amd64        NVIDIA binary Xorg driver development files
rc  nvidia-381                                  381.22-0ubuntu0~gpu16.04.2                 amd64        NVIDIA binary driver - version 381.22
rc  nvidia-cuda-toolkit                         7.5.18-0ubuntu1                            amd64        NVIDIA CUDA development toolkit
ii  nvidia-modprobe                             375.26-0ubuntu1                            amd64        Load the NVIDIA kernel driver and create device files
ii  nvidia-opencl-dev:amd64                     7.5.18-0ubuntu1                            amd64        NVIDIA OpenCL development files
ii  nvidia-opencl-icd-367                       375.66-0ubuntu0.16.04.1                    amd64        Transitional package for nvidia-opencl-icd-375
ii  nvidia-opencl-icd-375                       375.66-0ubuntu0.16.04.1                    amd64        NVIDIA OpenCL ICD
ii  nvidia-prime                                0.8.2linuxmint1                            amd64        Tools to enable NVIDIA's Prime
ii  nvidia-settings                             381.22-0ubuntu0~gpu16.04.1                 amd64        Tool for configuring the NVIDIA graphics driver
user@user-MS-7851 ~ $ dpkg -l |grep xserver
ii  x11-xserver-utils                           7.7+7                                      amd64        X server utilities
ii  xserver-common                              2:1.18.4-0ubuntu0.2                        all          common files used by various X servers
ii  xserver-xorg                                1:7.7+13ubuntu3                            amd64        X.Org X server
ii  xserver-xorg-core                           2:1.18.4-0ubuntu0.2                        amd64        Xorg X server - core server
ii  xserver-xorg-input-all                      1:7.7+13ubuntu3                            amd64        X.Org X server -- input driver metapackage
ii  xserver-xorg-input-evdev                    1:2.10.1-1ubuntu2                          amd64        X.Org X server -- evdev input driver
ii  xserver-xorg-input-synaptics                1.8.2-1ubuntu3                             amd64        Synaptics TouchPad driver for X.Org server
ii  xserver-xorg-input-vmmouse                  1:13.1.0-1ubuntu2                          amd64        X.Org X server -- VMMouse input driver to use with VMWare
ii  xserver-xorg-input-wacom                    1:0.32.0-0ubuntu3                          amd64        X.Org X server -- Wacom input driver
ii  xserver-xorg-legacy                         2:1.18.4-0ubuntu0.2                        amd64        setuid root Xorg server wrapper
ii  xserver-xorg-video-amdgpu                   1.1.2-0ubuntu0.16.04.1                     amd64        X.Org X server -- AMDGPU display driver
ii  xserver-xorg-video-ati                      1:7.7.0-1                                  amd64        X.Org X server -- AMD/ATI display driver wrapper
ii  xserver-xorg-video-fbdev                    1:0.4.4-1build5                            amd64        X.Org X server -- fbdev display driver
ii  xserver-xorg-video-intel                    2:2.99.917+git20160325-1ubuntu1.2          amd64        X.Org X server -- Intel i8xx, i9xx display driver
ii  xserver-xorg-video-qxl                      0.1.4-3ubuntu3                             amd64        X.Org X server -- QXL display driver
ii  xserver-xorg-video-radeon                   1:7.7.0-1                                  amd64        X.Org X server -- AMD/ATI Radeon display driver
ii  xserver-xorg-video-vesa                     1:2.3.4-1build2                            amd64        X.Org X server -- VESA display driver
ii  xserver-xorg-video-vmware                   1:13.1.0-2ubuntu3                          amd64        X.Org X server -- VMware display driver
user@user-MS-7851 ~ $ lspci | grep 'VGA'
00:02.0 VGA compatible controller: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller (rev 06)
01:00.0 VGA compatible controller: NVIDIA Corporation Device 1c03 (rev a1)

This is were I stopped for now. Currently,

Prime is set for the Nvidia GPU,
clinfo reports: Platform Version OpenCL 1.2 CUDA 8.0.0
F@H is currently running, with only my monitor plugged into the motherboard. Desktop is a little more laggy and my PPD is lower.

References:
viewtopic.php?f=59&t=236026&p=1256065#p1256065

Edit: I think I nailed it.
After watching this video https://www.youtube.com/watch?v=5nGbWE-pvIE
I followed this video and rebooted.

Code: Select all

user@user-MS-7851 ~ $ optirun glxgears
16581 frames in 5.0 seconds = 3316.030 FPS
15002 frames in 5.0 seconds = 3000.257 FPS
15690 frames in 5.0 seconds = 3137.870 FPS
[VGL] ERROR: in readback--
[VGL]    256: Window has been deleted by window manager
user@user-MS-7851 ~ $ glxgears
Running synchronized to the vertical refresh.  The framerate should be
approximately the same as the monitor refresh rate.
483 frames in 5.0 seconds = 96.346 FPS
300 frames in 5.0 seconds = 59.998 FPS
XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
      after 2441 requests (2441 known processed) with 0 events remaining.
user@user-MS-7851 ~ $ 

Only issue now, I have to run F@H with optirun. For example.
sudo optirun /etc/init.d/FAHClient start / stop

I don't know where to set this on the system boot up.

What's more, my desktop is no longer laggy.



Will update my PPD soon.

[LazyDev]

PPD Update. It's very eratic, but already averaging higher than previously.


(Just updated) 376,008 PPD.

Also to note, my power consumption has dropped, averaging at 92 Watts.

[LazyDev]

Final PPD update.

429K is a record high. I didn't even achieve this with my 980.



Some benefits I now get:
I can watch netfilx without having to pause folding, otherwise Firefox lags too much.
I can play basic games, such as Paalpeli (Jigsaw)

Edit:
Steam games won't work straight away. To fix,

Code: Select all

LD_PRELOAD="libpthread.so.0 libGL.so.1" __GL_THREADED_OPTIMIZATIONS=1 optirun %command%

https://unix.stackexchange.com/question ... e0bcbd85f5

Currently averaging 430K PPD with my 1060.

[900i]

Started building this open air folder when i got my 2nd GTX960, unfortunately my 1st GTX960 decided to self destruct so I am back down to my original config of just 2 cards. Still I now have room for expansion.

[LazyDev]

Could try getting a 1060, 120W TDP with it been 10% slower than a 980. I don't think folding uses lot of vram.. so I think the 3GB versions are ok. You can get them dirt cheap at around £150 used

Edit: http://gpu.userbenchmark.com/Compare/Nv ... 3646vs3165