Problem with offloading with PRIME

Graphics / Linux Linux
1

OK. So i have set everything up as per the arch wiki and the nvidia docs http://us.download.nvidia.com/XFree86/Linux-x86_64/430.14/README/randr14.html and if i execute:

glxinfo | grep vendor

I get

server glx vendor string: SGI
client glx vendor string: Mesa Project and SGI
OpenGL vendor string: Intel Open Source Technology Center

This is the expected response. However if i try prime offloading using

__NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia glxinfo | grep vendor

I get

X Error of failed request:  BadValue (integer parameter out of range for operation)
  Major opcode of failed request:  152 (GLX)
  Minor opcode of failed request:  24 (X_GLXCreateNewContext)
  Value in failed request:  0x0
  Serial number of failed request:  39
  Current serial number in output stream:  40

Yet if I lsmod i can see that the nvidia drivers are indeed loaded.

nvidia_drm             49152  1
drm_kms_helper        212992  2 nvidia_drm,i915
drm                   516096  15 drm_kms_helper,nvidia_drm,i915
nvidia_uvm           1085440  0
nvidia_modeset       1114112  1 nvidia_drm
nvidia              19980288  2 nvidia_uvm,nvidia_modeset
ipmi_msghandler        69632  2 ipmi_devintf,nvidia

FWIW. I initially had my laptop setup to only use the built in nvidia card but it was draining my battery too quickly.

Also the output from nvidia-smi is

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.36       Driver Version: 440.36       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| 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 1050    Off  | 00000000:01:00.0 Off |                  N/A |
| N/A   49C    P0    N/A /  N/A |      0MiB /  4042MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

nvidia-bug-report.log.gz (368 KB)

2

Please run nvidia-bug-report.sh as root and attach the resulting .gz file to your post. Hovering the mouse over an existing post of yours will reveal a paperclip icon.
https://devtalk.nvidia.com/default/topic/1043347/announcements/attaching-files-to-forum-topics-posts/

3

Done

4

Please remove
/usr/share/X11/xorg.conf.d/10-nvidia-drm-outputclass.conf
/etc/X11/xorg.conf.d/10-nvidia-drm-outputclass.conf
If this doesn’t resolve the issue, please create a new nvidia-bug-report.log.

5

I am afriad that didn’t work either
nvidia-bug-report.log.gz (364 KB)

6

The Xserver is always adding the igpu twice instead of the nvidia gpu. Please check if this xorg.conf works:

Section "ServerLayout"
  Identifier "layout"
  Screen 0 "iGPU"
  Option "AllowNVIDIAGPUScreens"
EndSection

Section "Device"
  Identifier "iGPU"
  Driver "modesetting"
  BusID "PCI:0:2:0"
EndSection

Section "Screen"
  Identifier "iGPU"
  Device "iGPU"
EndSection

Section "Device"
  Identifier "nvidia"
  Driver "nvidia"
  BusID "PCI:1:0:0"
EndSection

Xorg had a section order bug at some time.
Also, please post the output of
ls -l /etc/X11

Edit: typo fixed.

7

I seem to be getting a different error now.

Error of failed request:  BadValue (integer parameter out of range for operation)
  Major opcode of failed request:  156 (NV-GLX)
  Minor opcode of failed request:  6 ()
  Value in failed request:  0x0
  Serial number of failed request:  84
  Current serial number in output stream:  84

Output from ls -l /etx/X11

total 20K
drwxr-xr-x 3 root root 4.0K Sep  3  2018 xinit
drwxr-xr-x 2 root root 4.0K Dec 11 13:38 xorg.conf.d
-rw-r--r-- 1 root root  375 Dec 11 16:12 xorg.conf
-rwxr-xr-x 1 root root  685 Dec 10 14:57 xorg.conf.back
-rwxr-xr-x 1 root root  756 Sep  3  2018 xorg.conf.nvidia

nvidia-bug-report.log.gz (382 KB)

8

Ok, looks better, just the sub-module couldn’t be found

[   155.969] (II) LoadModule: "glxserver_nvidia"
[   155.969] (WW) Warning, couldn't open module glxserver_nvidia
[   155.969] (EE) NVIDIA: Failed to load module "glxserver_nvidia" (module does not exist, 0)

Please add to the xorg.conf

Section "Files"
    ModulePath "/usr/lib/xorg/modules, /usr/lib/nvidia/xorg"
EndSection

Furthermore, please remove the files (or rename them)
xorg.conf.back
xorg.conf.nvidia
Some Xserver version have the bug that anything that begins with “xorg.conf” will be taken into account for configuration.

9

This finally solved the problem. Although i had to put the “Files” Section right at the beginning of xorg.conf So this is how the working file looks now

Section "Files"
    ModulePath "/usr/lib/xorg/modules"
    ModulePath "/usr/lib/nvidia/xorg"
EndSection

Section "ServerLayout"
    Identifier "layout"
    Screen 0 "iGPU"
    Option "AllowNVIDIAGPUScreens"
EndSection

Section "Device"
    Identifier "iGPU"
    Driver "modesetting"
    BusID "PCI:0:2:0"
EndSection

Section "Screen"
    Identifier "iGPU"
    Device "iGPU"
EndSection

Section "Device"
    Identifier "nvidia"
    Driver "nvidia"
    BusID "PCI:1:0:0"
EndSection

Finally. Thank you for helping me out
nvidia-bug-report.log.gz (379 KB)