Questions tagged [gpu]

Questions pertaining GPU (graphics processing unit), a specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device

Filter by
Sorted by
Tagged with
0 votes
0 answers
14 views

Linux kernel support of old(er) Radeon GPUs

I have a Radeon RX 6700XT and it works flawlessly in Linux. I am looking at getting an older Radeon card and was wondering what is the support for something like a Rx 460. Is it native as my beefy ...
gtludwig's user avatar
  • 873
0 votes
2 answers
41 views

Will Disabling a Discrete Graphics Card Improve GUI Performance on Linux (XWindow/Wayland)?

I'm considering disabling the discrete GPU in my laptop to potentially improve GUI performance on Linux, specifically using XWindow or Wayland. Given the power efficiency of integrated graphics, I'm ...
1729 trace's user avatar
1 vote
0 answers
14 views

Running a program through Gnome GUI shell causes bad GPU behaviour, using the same command in CLI doesn't

If I run Blender by clicking on the icon in the Gnome Launcher I get garbled GPU rubbish. If I execute the command in a terminal window it runs fine. How can I fix it so it runs without having to open ...
stib's user avatar
  • 111
0 votes
0 answers
38 views

At what level is GPU hardware acceleration broken?

I'm using Wayland/Sway/Devuan/Firefox on an iMX8M and I don't manage 1 FPS in the WebGL Aquarium (one fish). CPU load alternates between high and low, which is odd if it is doing software rendering. ...
fadedbee's user avatar
  • 847
0 votes
0 answers
42 views

Multiple NVIDIA RTX GPU for Cuda (arch linux) with EGPU

I've got an arch linux, with two GPU in the laptop (thinkpad P14s Gen 4) + a new RTX 3090 plugged via thunderbolt 4 with the Cool Master EG200 GPU enclosure: ❯ lspci -k | grep -A 2 -E "(VGA|3D)&...
aemonge's user avatar
  • 101
0 votes
1 answer
73 views

Updated Nvidia drivers, Ubuntu OS won't start, can access BIOS and safe-mode

I got a new Nvidia GPU and I read that I had to purge all of the current drivers to avoid collisions. I removed everything, replaced the GPU with the new one, turned the PC on. It worked, I was able ...
David Harar's user avatar
0 votes
1 answer
33 views

Vulcan detecting GPU on TTY but not on gnome

I am on Debian 11 I have a program that works by overriding tty1. So when starting my computer, the program runs right away. This works fine, but I use Gnome as my GUI to develop and test the program. ...
fwan's user avatar
  • 11
0 votes
0 answers
32 views

ElementaryOS Horus 7.1 Performance issues

I am running ElementaryOS Horus 7.1 and until some point, it ran perfectly well (for my class of hardware). But, eventually, today in the morning I discovered my system being laggy, and overall ...
Maks Tamoian's user avatar
0 votes
1 answer
49 views

Hardware controller for Laptop in Linux

I have a Asus gaming laptop and I want to format and use it with Linux (I'm new user) but there is some problem: The main one is Asus doesn't have Armoury Crate and without some kinda of controller (...
Mahyar's user avatar
  • 3
0 votes
0 answers
122 views

How to make picom use the NVIDIA card with PRIME for blurring effects?

I use picom to manage the transparency and blur effects of my terminals, using i3 on Arch Linux. The backend I use is glx, though I don't mind changing it to something else if needed. My laptop has ...
Tristan Nemoz's user avatar
0 votes
0 answers
334 views

What does NVIDIA PRIME actually do?

The way I understand the explanations on the Arch wiki about NVIDIA PRIME is that when rendering an application, the system will ask the NVIDIA card to do it, and will then pass the information to the ...
Tristan Nemoz's user avatar
0 votes
0 answers
83 views

Upgrading NVIDIA L4T (Linux 4 Tegra) GPU modules from 3.10 to Something Newer (Like 4.19 LTS)

I'm an owner of an ARM (armv7h) ChromeBook, in particular, it is the Acer ChromeBook 13 with an NVIDIA Tegra K1 GPU. I've recently have been successfully able to run Ubuntu 22.04 (Jammy Jellyfish) ...
NVIDIALinuxUser's user avatar
0 votes
0 answers
56 views

Can you remove nvidia_modeset without removing the rest of the driver?

I'm trying to prevent nvidia driver from modesetting as an integrated GPU (i915) is doing it. I'm only using this nvidia card on linux for rendering (PRIME), no modesetting. I've tried blacklisting ...
Daniel Krajnik's user avatar
1 vote
0 answers
66 views

Question about libdrm and GPU device drivers

An article summarizes the steps involved in 3D rendering with OpenGL in this way: Your program starts up, using “OpenGL” to draw. A library, “Mesa”, implements the OpenGL API. It uses card-specific ...
apadana's user avatar
  • 123
0 votes
0 answers
133 views

Cannot use hybrid GPU mode in Fedora installation, works fine with the live ISO. Ideas?

My laptop is the Legion 5 Pro 16IAH7H Laptop from Lenovo. It has two graphic cards: iGPU of the Intel i7-12700H processor dGPU Nvidia RTX 3070 For some reason, hybrid mode works fine in Fedora ...
gremo's user avatar
  • 167
0 votes
0 answers
40 views

Mirror/"stream" iGPU display over USB-to-HDMI adapter?

Does anyone know any way for to mirror iGPU display over a USB-to-HDMI adapter? For various reasons my laptop can't use the dedicated Nvidia card and all HDMI and USB-C ports are wired to dGPU. iGPU ...
Daniel Krajnik's user avatar
0 votes
1 answer
310 views

intel_gpu_top output - two valiues for frequency and power consumption

In intel_gpu_top 1.27.1 CLI output, contrary to previous releases detailed in Get output from intel_gpu_top as variable in bash there are two values for each of frequency and power: intel-gpu-top: ...
Alban Browaeys's user avatar
0 votes
1 answer
96 views

How can I check if a GPU is being used for compute or is free?

I am trying to determine which GPUs are currently utilized for compute on a GPU server. Better yet, I'd like to be able to answer whether a GPU is "free" for running a new job. Ideally it'd ...
Shahar's user avatar
  • 1
0 votes
1 answer
345 views

Best way to use Linux with an AMD computer

Goal I have a PC with a good configuration, and I'd like to use it efficiently on Linux for programmation purposes. How can I set it up the best way? The main problem I have in mind is that my CPU/GPU ...
GregoirePelegrin's user avatar
0 votes
0 answers
114 views

Run two xservers on one GPU in headless system

I have a NVIDIA GeForce RTX 2080 Ti in a Linux server without display monitor. Now, I need to run two xservers on this GPU to support two docker containers(each container use one xserver via different ...
Chen Chen's user avatar
0 votes
1 answer
330 views

DP/HDMI monitors show "no signal" - Debian 11

This week I returned to my job after 2 week break and I discovered one important error. Any of my 3 monitors is not responding. All monitors keep displaying only "No signal". I have tried ...
Ondrej Burkot's user avatar
0 votes
0 answers
107 views

Keeping the Screen Turned off While Getting Inputs

I need to run a set of (graphical) benchmarks with the screen disabled. The following command did not work: xset dpms force off Because any keyboard/mouse input would re-enable the screen. The other ...
TheAhmad's user avatar
  • 101
0 votes
1 answer
120 views

GPU usage monitor for FreeBSD?

I am using FreeBSD on a personal desktop computer, and a nvidia GTX 1650 GPU. In my gnome system monitor (similar to task manager of MS Windows), I am able to monitor all the resources, including all ...
rohit4444's user avatar
0 votes
1 answer
243 views

How do I find the GPU device ID on Apple Silicon?

I use a MacBook Pro and Mac Studio with M1 and M2 chips, respectively. I'm trying to use a terminal command that has this parameter --device=/dev/****, which refers to the GPU device ID. How do I find ...
Pyy's user avatar
  • 173
0 votes
0 answers
440 views

Headless factory reset Nvidia Jetson AGX Orin

I want to reset my Nvidia Jetoson AGX Orin from a ssh conected remote desktop. given some GPU settings issues. Due to these, the Jetson does not recognize the GPU, which means does not display any ...
Fernando Cruz Ceravalls's user avatar
0 votes
0 answers
27 views

Radeon card on computer may is defect after current problems. How to check? sddm not start, card detected but no vram

Radeon card on computer may is defect after current problems. How to check? sddm not start, card detected but no vram My raddeon sapphire has 3 fans. These fans were only turning when had gpu loads to ...
Estatistics's user avatar
0 votes
1 answer
231 views

Problem with AMD integrated gpu on new Lenovo laptop

I recently bought a new Lenovo Ideapad Slim 3 laptop and I am having problems getting the amdgpu driver working properly in Arch Linux. The GPU seems to work fine straight away with a Mint live USB (...
Time4Tea's user avatar
  • 2,328
1 vote
0 answers
17 views

GPU fan boosts intermittently at idle

I've recently installed Archbang Linux, although I recently had Manjaro XFCE and had this same problem. My GPU fan (Asus Nvidia GTX 750 ti) keeps boosting up intermittently while the computer is in ...
Magnolias's user avatar
0 votes
1 answer
125 views

Software method to unplug/re-plug a DisplayPort cable to re-initialise the connection?

Is there any software way to have the same effect as unplugging/re-plugging a DisplayPort cable without physically doing so? The reason why I want to know is that my Dell monitor's resolution becomes ...
Damn Vegetables's user avatar
0 votes
0 answers
49 views

Lower the threshold for switching from integrated graphics to discrete graphics

I'm running linux mint cinnamon on a laptop which has both a intel CPU and nvidia GPU. There is an option to use Nvidia on demand for graphics. However, the system is too reluctant to switch to the ...
Pioneer_11's user avatar
3 votes
0 answers
3k views

Horrible lag / stuttering in Debian 12 KDE Plasma Wayland

I just installed Debian 12 with KDE from live installer. Everything seems to work fine until I enter the password on login screen. When I get to the desktop (Wayland Plasma), mouse and keyboard are ...
wololoo's user avatar
  • 63
2 votes
1 answer
514 views

GPU scheduling in Linux

Please, does someone know how GPUs are scheduled in Linux? Is the Linux scheduler responsible for both CPU and GPU scheduler, or is it done in the hardware or by a driver? For example, with DPU (Data ...
S Bitchebe's user avatar
3 votes
1 answer
567 views

Dual GPU Setup: Xorg on Intel Integrated GPU, NVIDIA GPU for Gaming

I recently purchased a new laptop and installed openSUSE Tumbleweed on it. The laptop has an Intel Core i5 processor with integrated graphics and an NVIDIA 3050 Ti. My goal is to configure Xorg to run ...
Tim's user avatar
  • 71
0 votes
0 answers
119 views

CPU spikes to 30% when watching any kind of videos in exodia os (arch distro)

Recently i switched from windows to Linux. I've been trying a lot of distros, but my choice fell on exodia (arch distro based on bspwm). I find it really useful for me as someone who is into ...
Tom Ederson's user avatar
0 votes
1 answer
97 views

"Hello world" to pixels -- everything between OS, driver, and hardware (CPU, GPU, PCIe, USB-C to display)

Can someone explain "Hello World!" in relation to the modern stack of software and hardware on a modern Linux or Unix machine? Doing a $ strace on the binary generated by following ...
codepoet's user avatar
  • 586
3 votes
0 answers
162 views

Is it possible to adjust radeon integrated graphics memory at runtime?

I have a Radeon 680M, an integrated AMD GPU. I do games sometimes, and would prefer to have more memory allocated to GPU when I do. There is a BIOS setting to adjust the GPU memory, but closing all ...
Vin's user avatar
  • 131
2 votes
1 answer
589 views

Troubleshooting Intermittent System Crashes on a Linux PC with Flashing Graphics Card Lights

I have been experiencing random system crashes on your Linux PC, accompanied by a small white light flashing on my graphics card. This issue occurs regularly, but not consistently, and tends to happen ...
George G.'s user avatar
  • 245
1 vote
1 answer
776 views

Sharing GPU between host and guest

Is there a way to make both host and guest use a singular GPU without any paid software (vGPU)? Bonus points if it: Dynamically allocates GPU resources Works for AMD and Intel GPUS I have an nVidia ...
Anm's user avatar
  • 113
1 vote
1 answer
181 views

GreenWithEnvy's overclocking menu has no overclocking sliders

So, I did what a tutorial said and I set coolbits to 8. As expected, GreenWithEnvy's overclocking menu became available. However, when I tried to make a custom overclock setting, the overclock ...
Epic's user avatar
  • 11
2 votes
0 answers
79 views

System unresponsive even when running command with nice

Debian 10. I am trying to render with blender in the background, like so: nice -n 19 blender -b --python my_script.py This renders to a file using the GPU. When rendering, I can see GPU usage spike ...
cat pants's user avatar
  • 405
1 vote
1 answer
749 views

Integrated AMD Radeon 660M does not seem to be used on Debian Bookworm

I have a new Lenovo Yoga Slim 7 with an AMD Ryzen 5 6600HS processor. This processor has a Radeon 660M integrated graphics controller, and I don't have a dedicated GPU. I have several issues which I ...
user505344's user avatar
0 votes
0 answers
178 views

Different versions of application use different OpenGL renderer

I have two versions of the same application (same source code, just built with a different version of the wxWidgets library) that use different OpenGL renderers for some reason. I run them using ...
Lukas Lang's user avatar
1 vote
0 answers
459 views

Vulkan, DRI, X11, and KMS: how do they relate?

I think I have a decent understanding of the chain of software that gets graphics code from a program to your hardware, but I would like someone to correct me on any mistakes. When I write a program (...
Lennon McLean's user avatar
1 vote
1 answer
519 views

How to run udev rule before a certain module loads?

I am trying to get the nvidia GPU driver working on a Macbook Pro (running Arch Linux). According to the instructions here, some PCI registers need to be set before the nvidia module gets loaded, ...
Time4Tea's user avatar
  • 2,328
2 votes
1 answer
955 views

Restricting user access to nvidia GPU?

On a server with Tesla Nvidia Card we decide to Restrict user access to GPU. In our server 2 GPU. # ls -las /dev/nvidia* 0 crw-rw-rw-. 1 root root 195, 0 Dec 2 22:02 /dev/nvidia0 0 crw-rw-rw-. 1 ...
Nikolay Baranenko's user avatar
1 vote
0 answers
35 views

North Bridge stuck in P1 state

I have AMD E-350 processor stuck in NBP1 (north bridge P1) state in Linux (Debian 11) and doesn't transit into NBP0 state under CPU/GPU load. The problem doesn't take place in Windows. Let's follow ...
Trudy's user avatar
  • 844
3 votes
1 answer
324 views

How is Linux able to display when driver is missing?

I tried installing Linux on a computer with an Nvidia GPU without bundling the necessary driver just for fun. Although I expected an error or a blank screen, to my surprise, I could install Linux just ...
ianc47859's user avatar
1 vote
1 answer
2k views

Problem using Intel integrated graphics GPU (Xorg)

I am trying to get the Intel integrated GPU working with my Parabola (Arch variant) desktop PC. According to lspci, the GPU is: 00:02.0 Display controller: Intel Corporation Xeon E3-1200 v2/3rd Gen ...
Time4Tea's user avatar
  • 2,328
2 votes
1 answer
457 views

Stuck at booting after upgrading

My GPU is NVIDIA - GeForce RTX 3090 Ti, and the OS is Ubuntu 18.04. As my code didn’t work, I checked the versions of python, pytorch, cuda, and cudnn. Python: 3.6 torch. version : 1.4.0 torch....
user12077723's user avatar
1 vote
0 answers
819 views

How can I kill a process in Linux when kill -9 fails?

I have a process (actually a pair of them) numbered 1234 which uses my NVIDIA GPUs. I want to kill it. I tried kill 1234; I tried kill -9 1234 I tried kill -KILL 1234 - no effect. Tried Ctrl+C and ...
einpoklum's user avatar
  • 9,462

1
2 3 4 5
8