Kernel Update - lp152.60 Lost NVIDIA Video Resolution - Again

English Install/Boot/Login
1

Current kernel update to lp152.60 from lp152-57 has again cause nvidia proprietary driver resolution back to low def mode from a previously solved condition.

In my last thread: “NVIDIA proprietary driver install lost resolution”, with much help I went through numerous steps to regain 1920x1080 resolution. Was solved without secureboot being enabled in bios and opensuse 15.2. Dual boot w/windows 10 worked. Problem appeared solved at that time.

New kernel install this morning has reverted system back to previous problem of low display resolution. I can regain 1920x1080 resolution by reverting boot back to kernel previous version lp152.57. Apparently previous solution was not a permanent fix.

Suggestions needed as to how to get high resolution video with nvidia proprietary drivers and current kernel version. Also a permanent solution to this problem

System: opensuse 15.2, dual boot w/w10, nvidia proprietary drivers installed using run command, 32gig memory, asus x99 mobo, large disk space available

2

Hi
Because the nvidia driver needs a rebuild, likely in weak-updates with the newer kernel;


/sbin/modinfo nvidia | grep filename

Are you installing the driver from the repositories or the hard way?

3

Here’s comandline results, 1st with boot into current kernel lp152.60,

[DesktopPC:/home/tom # /sbin/modinfo nvidia | grep filename
modinfo: ERROR: Module nvidia not found./CODE]


Then with boot into older lp152.57 that gives high resolution display,

DesktopPC:/home/tom # /sbin/modinfo nvidia | grep filename
filename: /lib/modules/5.3.18-lp152.57-default/kernel/drivers/video/nvidia.ko

4

Nvidia install from last thread was by nvidia *.run using NVIDIA-Linux-x86_64-460.27.04.run

5

If you install it “the hard way” you have to rebuild it on every kernel-update.

Or use dkms.

6

Hi
So, boot/switch to runlevel 3 (multiuser.target) and rebuild or update to the 460.32.03 file…

7

How do I avoid having to rebuild after each kernel update?

DKMS was not installed. I installed dkms using yast. Will that take care of future kernel update situations.

Should I now, boot with updated kernel, boot to level 3, run nvidia *.run file as root, reboot?

Or could I use yast-software G05 install from nvidia repo?

thanks, tom kosvic

8

Hi
You should be able to logout to a VT (tty) via ctrl+alt+F1, login as root and run the driver, then reboot…

I prefer to boot to multi-user.target… and install with the run file.

Sometimes the repo updates can be slow, kernel update, rebuild driver and done… don’t even have time to make a coffee these days :wink: Use the -aq option (accept all and be quiet) then just have to press ok when finished…

9

Now, after grub set to level 3, re-run nvidia *.run as root and reboot, the nvidia install said successful installation (ok ). But, I cannot execute user login using lp152.60 kernel. Upon reboot, It ignores my password and puts the user login screen back up over and over. I think it says user not found. It appears resolution is set properly based upon the size of the curser. All the loading flash screens are not normal. It talks about loading locale service. Says nothing about startx.

If I do a power off restart, grub comes up normally, and if I boot using the older lp152.57 kernel, I get good reboot with right resolution. That’s what I am using now to send this message.

How do I get past login screen not accepting my pw using lp152.60 kernel?

10

Please post

zypper se -si nvidia kernel

There are problems with the new kernel:
https://bugzilla.opensuse.org/show_bug.cgi?id=1180971

But only with the nvidia from Repo:

11

Hi
So a kernel issue… rollback to the .57 and wait for an update to come thorough… FWIW I see the same on my test machine.

12

note: nvidia community repo is not in my software repo list

DesktopPC:/home/tom # zypper se -si nvidia kernel
Loading repository data...
Reading installed packages...

S  | Name                  | Type    | Version              | Arch   | Repository
---+-----------------------+---------+----------------------+--------+-----------------------
i  | jupyter-ipykernel     | package | 5.2.1-lp152.1.2      | noarch | Main Repository
i+ | kernel-default        | package | 5.3.18-lp152.60.1    | x86_64 | Main Update Repository
i+ | kernel-default        | package | 5.3.18-lp152.57.1    | x86_64 | Main Update Repository
i  | kernel-default-devel  | package | 5.3.18-lp152.60.1    | x86_64 | Main Update Repository
i  | kernel-default-devel  | package | 5.3.18-lp152.57.1    | x86_64 | Main Update Repository
i  | kernel-devel          | package | 5.3.18-lp152.60.1    | noarch | Main Update Repository
i  | kernel-devel          | package | 5.3.18-lp152.57.1    | noarch | Main Update Repository
i  | kernel-firmware       | package | 20200107-lp152.2.3.1 | noarch | Main Update Repository
i  | kernel-macros         | package | 5.3.18-lp152.60.1    | noarch | Main Update Repository
i  | kernel-preempt-devel  | package | 5.3.18-lp152.60.1    | x86_64 | Main Update Repository
i  | kernel-preempt-devel  | package | 5.3.18-lp152.57.1    | x86_64 | Main Update Repository
i  | kernel-syms           | package | 5.3.18-lp152.60.1    | x86_64 | Main Update Repository
i  | nfs-kernel-server     | package | 2.1.1-lp152.9.6.1    | x86_64 | Main Update Repository
i  | purge-kernels-service | package | 0-lp152.4.1          | noarch | Main Repository
i  | python3-ipykernel     | package | 5.2.1-lp152.1.2      | noarch | Main Repository
13

Hi
Seems to be working now… updated Tumbleweed host, fired up the qemu Leap test system and all now good…


os152:~> uname -a
Linux grover-os152 5.3.18-lp152.60-default #1 SMP Tue Jan 12 23:10:31 UTC 2021 (9898712) x86_64 x86_64 x86_64 GNU/Linux

os152:~> nvidia-smi
Fri Jan 15 18:00:56 2021       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  GeForce GT 710      Off  | 00000000:00:03.0 N/A |                  N/A |
| 40%   35C    P8    N/A /  N/A |     75MiB /   978MiB |     N/A      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

os152:~> glxinfo | grep "OpenGL renderer"
OpenGL renderer string: GeForce GT 710/PCIe/SSE2
14

I need specific instructions on how to rollback/remove lp152.60 from grub. I tried boot into that kernel again and I still cannot get past login screen. then I can wait for updated kernel

I dont know what “update tumbleweed host and fired up the qemu Leap test system” means from prior message.

I am okay using older lp152.57. I would like that to be default in grub and remove any reference to newer one.

Right now zypper has nothing to do. I will wait for next kernel download.

15

Hi
At boot, in grub select advanced and can boot from here, then once at the desktop, fire up YaST Bootloader and set the default kernel to boot from.

I run test machines in qemu (virtualization, and also use a nvidia gpu and sata controller) real hardware rather than virtualized disk and graphics :wink:

16

malcolm,

I forgot yast bootloader setup. thank for pointing me there. now 15.2 boots using default lp152.57.

tom@DesktopPC:~> uname -a
Linux DesktopPC.lan 5.3.18-lp152.57-default #1 SMP Fri Dec 4 07:27:58 UTC 2020 (7be5551) x86_64 x86_64 x86_64 GNU/Linu

I have too much going on including covid issues but I need to keep this computer going and get things sorted out (records, etc) in case of serious incapacity

thanks, tom kosvic