Can I use opensource drivers to code in Cuda?

Specific Domains GPU
1

Hi all

can I use xserver-xorg-video-nouveau v 1:1.0.16.1 display drivers to code in Cuda and NOT the official nvidia 470?

I have had a bit of bad news. I am using linux mint 20.2 ( based on Ubuntu) and am using open source drivers instead of official nvidia. The opensource drivers loaded because I get 3840x2160~30Hz via the display port hooked by a DP cable to the hdmi of a tv.

The official nvidia drivers only get me 1920*1080. My card is the Nvidia k5000 card.

the official nvidia 470 drivers

raphics:  Device-1: NVIDIA GK104GL [Quadro K5000] vendor: Hewlett-Packard driver: nvidia v: 470.74 bus ID: 04:00.0 
           chip ID: 10de:11ba 
           Display: server: X.Org 1.20.11 driver: nvidia unloaded: fbdev,modesetting,nouveau,vesa resolution: 1920x1080~60Hz 
           OpenGL: renderer: Quadro K5000/PCIe/SSE2 v: 4.6.0 NVIDIA 470.74 direct render: Yes 

**Whereas the OPENSOURCE goes up to 3840x2160~30Hz**

Graphics:  Device-1: NVIDIA GK104GL [Quadro K5000] vendor: Hewlett-Packard driver: nouveau v: kernel 
           bus ID: 04:00.0 chip ID: 10de:11ba 
           Display: server: X.Org 1.20.11 driver: modesetting unloaded: fbdev,vesa 
           resolution: 3840x2160~30Hz 
           OpenGL: renderer: NVE4 v: 4.3 Mesa 21.0.3 direct render: Ye

ALSO the nvidia 470 drivers might NOT support cuda compute capability 3.0. I think CUDA 10.2 is the last CUDA version with support for CC 3.0

I am just starting out with Cuda so please point out anything I am missing.
thank you
theakson

2

problem solved. Here’s what I did

use the nvidia 470 drivers.
Shut down. Take out DP to Hdmi cable.
replace with Plugable displayport to hdmi 2.0 active adapter (model DP-HDMI), HDMI cable to HDMI1 (arc) on Hisense.
Reboot.

tave@deepthought:~$ inxi -Gx
Graphics:
  Device-1: NVIDIA GK104GL [Quadro K5000] vendor: Hewlett-Packard
  driver: nvidia v: 470.74 bus ID: 04:00.0
  Display: x11 server: X.Org 1.20.11 driver: nvidia
  unloaded: fbdev,modesetting,nouveau,vesa resolution: 3840x2160~30Hz
  OpenGL: renderer: Quadro K5000/PCIe/SSE2 v: 4.6.0 NVIDIA 470.74
  direct render: Yes


dave@deepthought:~$ xrandr
Screen 0: minimum 8 x 8, current 3840 x 2160, maximum 16384 x 16384
DVI-I-0 disconnected (normal left inverted right x axis y axis)
DVI-I-1 disconnected (normal left inverted right x axis y axis)
DP-0 disconnected (normal left inverted right x axis y axis)
DP-1 disconnected (normal left inverted right x axis y axis)
DVI-D-0 disconnected (normal left inverted right x axis y axis)
DP-2 connected primary 3840x2160+0+0 (normal left inverted right x axis y axis) 1872mm x 1053mm
   3840x2160     30.00*+  29.97    25.00    23.98
   4096x2160     29.97    25.00    24.00    23.98
   1920x1080     60.00    59.94    50.00    29.97    25.00 23.98
   1280x1024     60.02
   1280x960      60.00
   1280x720      60.00    59.94    50.00
   1024x768      60.00
   800x600       60.32
   720x576       50.00
   720x480       59.94
   640x480       59.94    59.93
DP-3 disconnected (normal left inverted right x axis y axis)
dave@deepthought:~$
3

In case anyone is confused: No, you cannot use CUDA with nouveau FOSS drivers. CUDA only works with the proprietary NVIDIA driver.

4

fair point. I didn’t make that clear as I got the answer from the Cuda.jl docs

The Julia CUDA stack only requires a working NVIDIA driver;

I put this up just in case someone else meets the same issue with another old GPU card. They might consider using them given the costs of new ones and hit the same issue. It’s not obvious that A solution is to put a DP active adapter into the cabling some time can be wasted messing with the linux config files. Sorry for the confusion.