When I install Nvidia Drive and Cuda, CentOS 7 can not boot [image]

What is the reason that prevents CentOS7 from initializing itself when installing the Nvidia and Cuda driver.

I've tried it twice. The first time, the driver is installed only when I restart the computer, it hangs on the next screen. The second time, I used yum directly installed Cuda which will also install the driver. Everything will be fine and will be installed successfully, but again the same error has occurred.

enter the description of the image here

One of the tutorials I followed

How to solve this visual problem with "my NVidia Quadro P620 graphics card under Windows 10?"

I have a NVidia Quadro P620 graphics card, provided with my HP Z240 PC, bought new in mid-2019.

It runs Windows 10 that I have installed cleanly from an ISO site Microsoft. "Windows 10.0.18362.476".

I've been running Windows 10 on a number of computers since it was released and I've never had this problem before. These computers mainly use NVidia graphics cards.

I'm using the "dark theme" to save my sight.

There is no problem with the display of images, videos, advanced 3D visuals or everything related to the graphics card, apart from the problem of drawing the GUI that I will describe, especially with "File Explorer". I want it to be very clear. The graphics card can not be logically damaged physically. This must have something to do with the drivers and / or settings and / or Windows 10 itself.

The question is very specific and seems very minor, but it literally drives me crazy. I really think I'm losing my mind because of the frequency with which this happens:

Almost every time I open a File Explorer window, there is one or more frames when the window has a bright white background, just before drawing the background to it. even (which is dark). For the human eye, this sounds like a "flash" net, as if the screen was taking a picture of my face with a camera in the dark.

I went up to close eyes every time I have to open a file explorer (which happens literally thousands of times a day), even if my eyes are closed, I can still "see" the flash because of the extreme contrast between it and the Real background. It burns through my eyelids.

At first I suspected that they (Microsoft) might have ignored quality assurance for the "performance-optimized" settings that I normally use, which ignore all the animations, the eyecandy and other. However, even when I activate all of this, it still displays this split image (or some images) of the white background and sometimes what appear to be partially drawn sub-elements within the window. In other words, we can exclude that as a solution.

When I installed this computer (ie Windows 10), my first instinct was that something was "off". All the computers that I have since the early 2000s, including those in Windows 10, have felt very "lively" and "responsive" in terms of GUI display. In other words, except in case of temporary problems with the drivers of the graphics card. The way this PC draws and behaves looks like these temporary problems of the past, when Windows used some kind of "backup mode" in which things were not optimized and "ready for production", but "was enough to work while you repair the "driver emits".

howeverthis time, I tried this a few times:

  1. I've tried the drivers Microsoft provide via the operating system Windows / GUI, and make sure there are no updates that it can use It literally says that I manage the best / latest drivers.

  2. I have manually downloaded the drivers of HP's website for the exact product: https://support.hp.com/us-en/drivers/selfservice/hp-z240-tower-workstation/8354563/model/8354564?sku=5UD37ES

  3. I have manually downloaded the drivers on Nvidia.com – https://www.nvidia.com/Download/driverResults.aspx/151056/en-us – both "stable" and "blowing edge", making sure that I have chosen the right options.

No matter what I install and the number of reboots the machine, no matter how many times I modify various NVidia "3D settings" in its special control panel, such as "maximize for performance" ( deactivation of the energy saving). stuff), the same problem remains.

I closely monitor the NVidia web page for updated drivers. However, they have not published any updates since I started having these problems (that is, the day I got the machine). I hope of course that they will soon publish an update, but I am skeptical as to whether they are even aware of this problem and even more skeptical about knowing about it. they will solve it in the next update. I do not even understand how that can happen at all.

Speaking of HP and NVidia, I spent a lot of time trying to contact them and chat with their collaborators. I am sad to announce that they have no idea. In fact, I'm pretty sure I did not even communicate to humans because they did not even seem to read or understand what I was typing.

As mentioned, it has affected me very strongly on a psychological level. I will spare you a long "speech" and just say that I can not work like that.

If anyone has a good idea of ​​what might be causing it and that he knows a solution, Please Help me! You are my only hope now.

linux – Fedora 31 uses llvmpipe as the default OpenGL rendering engine instead of Nvidia (nvidia settings do not work either)

I want to use only my Nvidia card on a laptop (Optimus Bumblebee seems horrible) with Fedora 31 (freshly installed). These are my cards:

lspci | egrep 'VGA|3D'
00:02.0 VGA compatible controller: Intel Corporation UHD Graphics 630 (Mobile)
01:00.0 3D controller: NVIDIA Corporation GP107M (GeForce GTX 1050 Ti Mobile) (rev a1)

I have therefore installed the drivers from nvidia.com, deleted New and installed it. This was done in multi-user mode (without graphics). Now I have this module loaded:

lsmod | grep nvidia
nvidia_drm             57344  0
nvidia_modeset       1118208  1 nvidia_drm
nvidia              19116032  1 nvidia_modeset
ipmi_msghandler        73728  2 ipmi_devintf,nvidia
drm_kms_helper        212992  2 nvidia_drm,i915
drm                   512000  5 drm_kms_helper,nvidia_drm,i915

But at the same time, I have this error with nvidia-settings:


ERROR: Unable to load info from any available system

And this OpenGL rendering engine:

inxi -G
Graphics:  Device-1: Intel UHD Graphics 630 driver: i915 v: kernel 
           Device-2: NVIDIA GP107M (GeForce GTX 1050 Ti Mobile) driver: nvidia v: 430.50 
           Display: x11 server: Fedora Project X.org 1.20.5 driver: modesetting,nvidia unloaded: fbdev,vesa 
           resolution: 1920x1080~60Hz 
           OpenGL: renderer: llvmpipe (LLVM 9.0 256 bits) v: 3.3 Mesa 19.2.0 

This llvmpipe starts when I open glxgears:

glxgears -info
GL_RENDERER   = llvmpipe (LLVM 9.0, 256 bits)
GL_VERSION    = 3.1 Mesa 19.2.0
GL_VENDOR     = VMware, Inc.

I guess I can not start the Steam games because of that (I was able to do it in openbox started with the nvidia-xrun script when I had bumblebee). My user is in the "video" group but it does not work.

lot – FFMpeg nVidia Turing best quality

I found this line of code to convert videos to MP4 files by FFMpeg

ffmpeg -i input.avi -c:v h264_nvenc -preset:v slow -profile high 
       -level 4.2 -b:v 1000k -rc-lookahead:v 32 -g 480 
       -f mp4 Turing_output_1000k.mp4

I went through the documentation of unknown parameters to get the following result

-i name of the input file

-c:v h264_nvenc use the nVidia encoder to get a h264 resale

-preset:v slow high quality with 2 passes

-profile high not sure about it or high444p

-level 4.2 this is the "version" of the mp4, 4.2 could cause problems on the old players

-b:v 1000k video bitrate 1000

-rc-lookahead:v 32 number of frames to monitor for rate control

-g 480 no idea, nothing found on nVend doc

-f mp4 sets the output format, not knowing why it should be specified

Question: Could someone fill my knowledge gaps or improve this code to improve quality? Or better use the advantage of an RTX graphics card?

xorg – Ubuntu 18.04 Unrecognized HDMI output with Nvidia Hybrid

I can not use HDMI external display on my IdeaPad S340 with Nvidia GeForce MX230 and Intel integrated graphics processor.

I am currently using nvidia-driver-435 (open source)

With my monitor plugged in, I got the following output:

HDMI is not listed in xrandr:

# xrandr
Screen 0: minimum 8 x 8, current 1920 x 1080, maximum 32767 x 32767
eDP-1-1 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 309mm x 174mm
   1920x1080     60.00*+
DP-1-1 disconnected (normal left inverted right x axis y axis)

Both GPUs are detected

# xrandr --listproviders
Providers: number : 2
Provider 0: id: 0x1de cap: 0x1, Source Output crtcs: 0 outputs: 0 associated providers: 1 name:NVIDIA-0
Provider 1: id: 0x44 cap: 0x6, Sink Output, Source Offload crtcs: 3 outputs: 2 associated providers: 1 name:modesetting

Can any one help me?

The Ubuntu screen shows the broken pixel (Nvidia driver 18.04)

I can not be 100% sure of how to reproduce the following problem. But it seems that if I play video or start a game for a while, the screen freezes and I have to restart it. So I can not say if it is the problem of my driver, the hardware of the graphics card or something else.

Here is my configuration:

enter the description of the image here

enter the description of the image here

~  sudo lshw -c display
(sudo) password for yuri: 
       description: VGA compatible controller
       product: GP104 (GeForce GTX 1080)
       vendor: NVIDIA Corporation
       physical id: 0
       bus info: pci@0000:01:00.0
       version: a1
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
       configuration: driver=nvidia latency=0
       resources: irq:139 memory:de000000-deffffff memory:c0000000-cfffffff memory:d0000000-d1ffffff ioport:e000(size=128) memory:c0000-dffff
       description: VGA compatible controller
       product: GP104 (GeForce GTX 1080)
       vendor: NVIDIA Corporation
       physical id: 0
       bus info: pci@0000:02:00.0
       version: a1
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
       configuration: driver=nvidia latency=0
       resources: irq:142 memory:dc000000-dcffffff memory:a0000000-afffffff memory:b0000000-b1ffffff ioport:d000(size=128) memory:dd000000-dd07ffff

~  sudo ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:1c.0/0000:04:00.0 ==
modalias : pci:v000014E4d000043A0sv00001043sd000085DFbc02sc80i00
vendor   : Broadcom Limited
model    : BCM4360 802.11ac Wireless Network Adapter
driver   : bcmwl-kernel-source - distro non-free

== /sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0 ==
modalias : pci:v000010DEd00001B80sv00001462sd00003367bc03sc00i00
vendor   : NVIDIA Corporation
model    : GP104 (GeForce GTX 1080)
driver   : nvidia-driver-390 - distro non-free
driver   : nvidia-driver-430 - distro non-free recommended
driver   : xserver-xorg-video-nouveau - distro free builtin

apt – How can I remove nvidia drivers with patches in my kernel?

I've removed all nvidia drivers with apt: apt list –installed | Grep Nvidia do not return anything. But if I look at dmesg, I see for example:

[    4.148454] NVRM: API mismatch: the client has the version 410.104, but
               NVRM: this kernel module has the version 430.50.  Please
               NVRM: make sure that this kernel module and all NVIDIA driver
               NVRM: components have the same version.

and also if I run lsmod | Grep Nvidia:

nvidia_drm             49152  1
nvidia_modeset       1114112  1 nvidia_drm
nvidia              19038208  1 nvidia_modeset
drm_kms_helper        180224  1 nvidia_drm
drm                   483328  4 drm_kms_helper,nvidia_drm
ipmi_msghandler       102400  2 ipmi_devintf,nvidia

How can I remove all these items from my system?

Unable to change the screen resolution after trying to overclock the nvidia 1050ti graphics

I was trying to overclock my nvidia 1050ti graphics card but I was not able to.
After restarting my computer, the resolution of my screen is completely wrong, all the icons and the slider are really big and I can not change the new driver into nvidia driver.

I am also unable to change the screen resolution with the settings

-Ubuntu 18.04

-Nvidia 1050ti

-Amd ryzen 1600

nvidia – How to update / update the driver after changing graphics card?

While my old graphics card (GT 730) was working normally, I removed it and installed a new one (GT 630). On Windows and Ubuntu, the new driver has started with the default generic driver supporting basic screen resolutions (800×600 and 1024×768).

The driver selection grayens the other options

I have 2 questions:

  1. The Nvidia v390 driver supports Fermi and Kepler. Why does not it work with both graphics cards after a change?
  2. The generic driver only runs for a short time on Windows (1 to 2 minutes) until Windows automatically recognizes and reconfigures the driver so that it works with the new graphics card without it. you have to reinstall it. Ubuntu remains faithful forever. How to manually operate the new card?

Stuck on startup after installing the new nvidia graphics card

After upgrading rtx 2070 super, I was able to start and attempt to install the drivers. During installation, the system is frozen and does not move. When I restarted, I get the following lines:

"/ dev / sdc1: clean, 700021/15269888 files, 23005431/61049344 blocks"
"Started bpfilter"

I let it sit for more than 15 minutes without anything happening. I went into recovery mode and purged Nvidia driver packages, I tried to put "nomodeset" in grunt. I've also loaded from a USB drive and run the startup repair. The pastebin URL for the result is as follows: