Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [gpu]

The tag has no usage guidance.

7
votes
0answers
1k views

11 GB of GPU RAM used, and no process listed by nvidia-smi

In my GPU #0, 11341MiB of GPU RAM is used, and no process is listed by nvidia-smi. How is that possible, and how can I get my memory back? Thu Aug 18 14:27:58 2016 +-----------------------------------...
7
votes
0answers
474 views

User-based GPU priority

I have a computer running Linux (Kubuntu 14.04.3 LTS x64), where I use the CUDA toolkit to perform some computations on some Nvidia GPUs. I share the computer with a few other users. How can I define ...
4
votes
0answers
404 views

Intel, DVMT & Linux: a question

In my Intel notebook BIOS I can change the amount of memory allocated to the integrated GPU (from 32MB to 512MB, via the selector "Pre-Allocated Memory"). My question is: Does Linux change the ...
3
votes
0answers
1k views

How to swap integrated Intel graphics card to discrete AMD

I am unable to change graphics card on Kali Linux 4.6.0, Debian 4.6.4-1kali1 (2016-07-21) x86_64 GNU/Linux with Gnome. Machine running is a HP Pavillion DV6. I have tried: Reinstalling AMD drivers: ...
3
votes
0answers
2k views

Installing Mali Drivers? (ArchLinuxARM)

I've recently gotten hold of an ARM-based Chromebook C100P. After successfully installing ArchLinuxARM, I've struggled to enable the Mali-T764 GPU. The recommended drivers for the device, (xf86-video-...
3
votes
0answers
1k views

Why I need a framebuffer size larger then monitor size when transform destination area is smaller

I tried to remove 100px left & right the top of a 1366x768 screen (as below) using xrandr transformation. But I got an error about not enough framebuffer size. $ xrandr --output LVDS1 --fb ...
3
votes
0answers
1k views

Virtualbox and GPU Pass-trough: Device with host PCI address already attached to this VM

I have server with several GPUs. My idea is to attach them to different VMs in order to use that VMs in CUDA (or OpenCL) calculations. My setup: CPU: Intel Xeon E5-2670 GPUs: 1 x Radeon HD 7970 $ ...
2
votes
0answers
48 views

Is it possible to modify Ubuntu 18.10 System Monitor to include GPU statistics?

I think I can throw together a fairly simple script to poll the usage data every few seconds. I am fluent enough with node/react that I could get that data and display it in a live graph that polls ...
2
votes
0answers
101 views

SLURM: restrict GPU access only to SLURM

I have a single machine (Ubuntu 16.04 Server) with 4 TitanX GPUs. This will be a lab machine on which students will learn about CUDA and stuff. I installed SLURM because I want a tool to schedule and ...
2
votes
0answers
175 views

MacBook Pro 9.1 (mid 2012) enable GPU switching in Debian 9.3

I work in deep learning projects and I need to run my MBP on Linux for running demos. I have installed Debian Linux 9.3 and everything is fine, except for the GPU which is always on. I cannot find a ...
2
votes
0answers
122 views

How can I tell if my GPU is a bottleneck?

I already know of utilities like htop that can be used to (roughly) tell if a process is primarily bottlenecking at disk read/write speeds, network IO, not being parallel, or being parallel but ...
2
votes
0answers
700 views

PRIME on Wayland

Is there a way to get PRIME working on a Wayland powered display? I am using Arch Linux with Integrated Skylake Graphics as well as a mobile AMD card. The instructions on the Arch wiki rely on xrandr,...
2
votes
0answers
477 views

How to enable browser vsync on Linux arm?

I'm playing an Odroid-C2 and I'm having a hard time trying find a linux + browser combination that will work with vsync enabled. So far I've tested using this distribution of Ubuntu 16.04 LTS which ...
2
votes
0answers
514 views

Why nvidia-smi's GPU so rarely reaches 100%?

I have logged the GPU utilization of my 3 GPUs (GeForce GTX Titan X) for 5 days with nvidia-smi, every 10 seconds. The GPUs were all being utilized: I count 1307 occurrences of 99 % while only 8 ...
2
votes
0answers
451 views

Per-user GPU usage on a Linux machine (CUDA)

I use the CUDA toolkit to perform some computations on my Nvidia GPUs. How can I see the per-user GPU usage on a Linux machine (CUDA)? nvidia-smi doesn't seem to contain information: I am aware I ...
2
votes
0answers
2k views

How to boot Arch Linux after changing from AMD Radeon HD to Nvidia GTX video card?

After swapping video cards my PC boots into Windows 7 just fine, but Arch Linux gets stuck before entering graphical mode. /var/log/Xorg.0.log contains some message listing all the supported AMD (aka. ...
2
votes
0answers
1k views

How to setup 3 screens with 2 ATI GPU Debian Squeeze

I want to use 3 screens on my computers with 2 graphics card ATI Radeon HD 5450. I've installed fglrx ATI drivers. Here is the result of lspci : root@dev1:~# lspci | grep VGA 01:00.0 VGA compatible ...
1
vote
0answers
35 views

Nvidia drivers on debian

I'm on an ASUS laptop, with the following GPUs $ inxi -Gx Graphics: Card-1: Intel Device 5917 bus-ID: 00:02.0 Card-2: NVIDIA Device 1d10 bus-ID: 01:00.0 Display Server: X.Org 1....
1
vote
0answers
14 views

X doesn't detect secondary nvidia GPU

My current setup is a 2 GPU setup, with an RX 480 as primary and a GTX 960 as secondary. I used to run nouveau just fine, X detected it and I could connect monitors or offload workloads onto it with ...
1
vote
0answers
618 views

Linux mint Tara 19 problems with Nvidia Drivers [MX150]

I am trying to get my system to use the mx150 graphics card Problem: Nvidia x Server settings show blank The driver manager shows that i'm using nvidia-driver-390 and also says "No proprietary ...
1
vote
0answers
35 views

The standard mechanism to switch GPUs isn't undocumented, is it?

My laptop is a normal, uninteresting machine with two standard, unmultiplexed GPUs and an ordinary Debian stretch installation. The secondary GPU (a Radeon) is usually powered down but I can activate ...
1
vote
0answers
485 views

Arch Linux - Unknown controller version on booting

I recently installed Arch Linux. But Every time I boot, I see this message for few seconds: mmc0: Unknown controller version (3). You may experience problems. After more few seconds I see: radeon ...
1
vote
0answers
45 views

GPU Temperature Applet interfering with GPU function

I have a dual booting PC (Mint 18.3 with Windows 10). My graphics card is a NVIDIA GeForce GT710. While using Linux, I noticed that the GPU Temperature reached 57-60°C with simple Internet browsing (...
1
vote
0answers
120 views

How to reduce CPU consumption of QWebEngine?

I wrote an app in Qt using the QtWebEngine component to display a simple Web page with SVG elements on a 7" LVDS screen. The app is running on a UDOO Neo board and over a custom Linux installation ...
1
vote
0answers
684 views

Nvidia Overclock and Fan control

I am trying to overclock (or at least control the fan speed on) a couple 1080tis I am using to mine on a ubuntu 18.04 system. The problem is to enable overclocking coolbits need to be set in Xorg....
1
vote
0answers
180 views

HDMI port stop recognizes external monitor after a system update with one Intel GPU

Yesterday I updated my system (DELL laptop with Arch linux), and unfortunately today I realized that HDMI port cannot recognize the external monitor. Until now I use the following command inside ....
1
vote
0answers
625 views

Is there a GPU system monitor for Debian?

I use a system monitor to watch CPU, memory and network usage on my Debian 8: I need a GPU (graphic) monitor. I tried differentes monitors with and without GIU but they don't show GPU. Why the ...
1
vote
0answers
729 views

Dual GPU Dual Display setup: monitor is black but detected

I've recently installed linux alognside Windows 10 and I'm trying to use a second monitor the way I use it on Windows. The first monitor is connected to a GTX 770 and the second one to the iGPU. I've ...
1
vote
0answers
797 views

Kill running process

I own some old processes on a shared compute server. They consume a lot of CPU and according to htop they're in the running-state: PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ ...
1
vote
0answers
188 views

Permision denied when change GPU clock rate in Nvidia Jetson TX1

I have a Nvidia TX1 development kit which I've installed ubuntu 16.04 using JetPack-L4T-3.1-linux-x64.run (in full installation mode) on it. The installation procedure are based on the following link: ...
1
vote
0answers
671 views

Extremely high CPU usage and laggy behaviour with 3D-acceleration in Linux Mint (Virtual Box) but not with software rendering

I'm currently trying to get Linux Mint 18.2 to work in Oracle Virtual Box. What have I done so far: Installed VBoxAdditions 5.0.40 or 5.1.26 (first one through apt-get, second one through the ...
1
vote
0answers
440 views

Linux: Boot Set boot GPU

My gaming rig runs Arch Linux and utilizes PCI passthrough to expose my GTX 980 to KVM. The other day I switched motherboards to aid in a CPU upgrade. I'm running into a serious problem however that ...
1
vote
0answers
772 views

How to downgrade CuDNN in Fedora?

I am installing tensorflow. After installation, I got compatibility issue. tensorflow library I have installed is not compatible with the CuDNN I think. How to solve this issue. I am using Fedora 25....
1
vote
0answers
225 views

Which nvidia driver deb-package should I install for gtx1050ti?

I already installed the Nvidia drivers for gtx1050ti once manually, but it made some problems occassionaly, like suddenly the driver being erased out of thin air. So, I would like to install it this ...
1
vote
0answers
822 views

How do I disable GPU acceleration in my arch linux?

I have dual boot arch linux with Windows 8. I have AMD Radeon GPU. Recently, my GPU has broken. And now, I cannot boot into my arch linux. I am able to boot into live CD, and it says: "disabling GPU ...
1
vote
0answers
1k views

PCI passthrough: vfio-pci ignores ids of devices

I have 3 GPUs in my dual XEON server. I followed instructions on Arch wiki and set up vfio-pci with ids=10de:100c,10de:0e1a: $ modprobe -c | grep vfio options vfio_iommu_type1 ...
1
vote
0answers
118 views

Enable cuda support in libbsc

I am interested in using libbsc in CUDA support. libbsc is basically used for compression in GPU (as well as CPU). As far as I tested, it is running in CPU (openmp) not in NVIDIA GPU (CUDA driver ...
1
vote
0answers
64 views

Multi GPU Compiz?

Is it possible to enable OpenGL compositing in multi GPU setup? I have Quadro 4000 and GTX580. Atm I'm using xinerama and 4 X servers but it doesn't support Compiz. On the other hand I can't find a ...
1
vote
0answers
564 views

Attempting to use OpenCL with Nvidia on Arch Linux

I have an Nvidia Quadro 400 GPU. In order to use OpenCL, I installed the following packages: nvidia-340xx nvidia-340xx-libgl nvidia-340xx-utils opencl-nvidia-340xx ocl-icd opencl-headers pyopencl-...
1
vote
0answers
605 views

Why do interrupted processes show up in nvidia-smi?

When ​I kill processes in terminal using C-c and run nvidia-smi, the process are still shown as "compute processes". These processes are no longer in the top resu​lts. I have to kill -9 to clean them (...
1
vote
0answers
157 views

Possibilty of DisplayPort over PCIe

I'm trying to figure out if it would be possible to add a DisplayPort to my Linux PC using a PCIe slot. I have found cards like http://www.ebay.com/itm/PCI-E-1x-Express-to-USB-3-1-Type-C-Dual-Port-...
1
vote
0answers
65 views

High GPU usage before starting X-Server

When starting my Debian Jessie, my GPU (ATI 5770) uses a lot of power (it is getting hot), although I am only displaying the terminal. Then, if I start an x-server, the GPU cools down again, also the ...
1
vote
0answers
138 views

how to stop GPU NVIDIA calculation

I'm currently into learning using NVIDIA for parallel calculation. Sometimes I run too many tasks in parallel and my computer stops responding (I use both CPU and GPU for calculations). So, I cannot ...
1
vote
0answers
813 views

How to switch between GPUs on Mint 17.2 ASUS laptop?

I have NVIDIA GeForce GTX 960M and Intel HD Graphics 4600. In CS:GO I have frame-drops constantly, it jumps all the time between 5 and 40 fps - this makes game unplayable. And on 960m it should work ...
1
vote
0answers
161 views

multiseat dual gpu, can I have 2 different drivers?

I have 2 radeon 7950, set up my multiseat, everything is working fine. I use the free radeon drivers, and I'm pretty happy except for one thing, I can't test the gpu mining of bitcoins.... I'd like to ...
1
vote
0answers
3k views

Debian 8 Jessie and NVIDIA drivers

I am in the process of installing Debian 8.0 on a laptop with NVIDIA graphics cards. After a successful installation and reboot I use this guide to install the proprietary NVIDIA drivers; essentially ...
1
vote
0answers
21 views

Booting with different X settings

My laptop has the Optimus technology. In order to save battery life I run graphics out of my CPU and only turn the GPU on on occasions using bumblebee. This works relatively well as long as I don't ...
1
vote
0answers
173 views

2 monitors on GPU, 1 monitor on integrated graphics

I am on linux mint 17.1 cinnamon. Here is my xorg.conf: http://pastebin.com/6xCncZJ5 I can move my mouse around on the 3 monitors but can not drag the windows over the borders. Cinnamon is running in ...
1
vote
0answers
404 views

Zero graphic card fan speed when no calculations are performed

I have fresh install of ubuntu 12.04.4 LTS and CUDA 5.5 with nvidia-331.20 driver. There are 4 GeForce GTX 780 on board. X runs on integrated card, while Nvidia cards are superposed to do some ...
1
vote
0answers
1k views

Dual Monitor with integrated and discrete GPU on Debian

I've got an Asrock Extreme 3 motherboard and a Nvidia 660ti GPU. I was trying to configure dual monitor on debian 7 connecting one monitor to the mb and the other to the GPU. This configuration works ...