I explicitly mentioned:
but regardless. Did you managed to install and boot from another kernel?
Sorry what do you mean ? is there some part that i have dual OS system ? i got a window and manjaro ?
Ok, will try one more time. Manjaro can have 1 or multiple supported kernels installed, and you can chose to boot from each at a time, from GRUB boot menu, either to test or because one can perform better on a particular hardware than another kernel.
In your case, it happens that you have just one kernel, that is EOL for quite some time. Did you installed and booted from a newer and supported kernel and then remove linux59?
understood, but i didt install or boot from a newer kernels, so mean my kernel is only depends now ? also what does EOL stand for ?
what you mean is i need to install a new kernel and switch to it and remove the linux59 and reinstall again then switch back ?
i checked on the advance menu, i have a fallback initramfs kernel and a normal kernel.
Have you ever read the update announcements ? An example would be:
EOL stands for End Of Life - aka unsupported kernel and as you can see, the linux59 is not even in the repositories anymore.
I told you to do so, install a sported kernel or a LTS one … LTS = Long Time Support.
For the 3rd time, YES. That is what you have to do 
1 Like
Okay after re-read all the post, then finally get what you mean XD.
so i install a new kernel sudo mwhd-kernel -i linux515
* linux515 5.15-rc2
Switch into the new kernel in GRUB boot menu and remove the the linux59 kernel.
after update and last install on the sudo mhwd -f -i pci video-nvidia
Output:
Packages (3) lib32-nvidia-utils-470.63.01.1 linux515-nvidia-470.63.01-0.3 nvidia-utils-470.63.01-1
Total Installed Size: 602.56 MiB
:: Procced with installation [Y/n]
checking keyring...
checking package integrity...
loading package files...
checking for file conflicts...
error: failed to commit transaction (conflicting files)
nvidia-utils: /usr/lib/libEGL_nvidia.so.0 exists in filesystem
nvidia-utils: /usr/llib/libGLESv1_CM_nvidia.so.1 exists in filesystem
nvidia-utils: /usr/llib/libGLESv2_nvidia.so.2 exists in filesystem
nvidia-utils: /usr/lib/libGLX_nvidia.so.0 exists in filesystem
nvidia-utils: /usr/lib/libcuda.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-cfg.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-encode.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-fbc.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ifr.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-m1.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ngx.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-opticalflow.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ptxjitcompiiler.so.1 exists in filesystem
Errors occurred, no packages were upgraded.
Error: pacman failed!
Error: script failed!
the screen still black but linux59 is confirm remove, so what next ? should i test different kernel that could install the video-nvidia ?
What i do not understand is why a RC kernel tho … i mean, linux510 is a curated and LTS kernel. You jumped directly to linux515 … Well …
I’ll propose this, switch branch to testing
sudo pacman-mirrors --api --set-branch 'testing'
then update
pamac update
that way you will get t a better support for nvidia modules for linux515 as they are now 470.74 instead of 470.63, but maybe you will not like that as it has some regressions with G-Sync
Yes, i would go with latest LTS that is linux510
but if that is not working try the branch switch thing i mentioned.
Oh cause linux510 first time i got it install my network seem can’t connect with even with try, but i got it work and now on linux510 kernel
i update everything but it still output the same error as i go for the sudo mhwd -f -i pci video-nvidia
Errors occurred, no packages were upgraded.
Error: pacman failed!
Error: script failed!
but now i ran sudo mhwd -li
Warning: No installed PCI configs!
Warning: No installed USB configs!
so i will keep trying different kernel utill it can be install with the video-nividia or am i missing out anything ?
Remove the driver first. You might have to remove cuda too prior to that.
Then right after that reinstall them without the force -f in the command.
Sorry what drive are we talking about ? sudo mhwd -r video-nvidia ?, also how to remove cuda ? haha thanks being patient with me ya, never face this problem before.
Finally fix the kernel problem, with this 
error: failed to commit transaction (conflicting files)
nvidia-utils: /usr/lib/libEGL_nvidia.so.0 exists in filesystem
nvidia-utils: /usr/llib/libGLESv1_CM_nvidia.so.1 exists in filesystem
nvidia-utils: /usr/llib/libGLESv2_nvidia.so.2 exists in filesystem
nvidia-utils: /usr/lib/libGLX_nvidia.so.0 exists in filesystem
nvidia-utils: /usr/lib/libcuda.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-cfg.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-encode.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-fbc.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ifr.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-m1.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ngx.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-opticalflow.so.1 exists in filesystem
nvidia-utils: /usr/lib/libnvcuvid-ptxjitcompiiler.so.1 exists in filesystem
Errors occurred, no packages were upgraded.
Error: pacman failed!
Error: script failed!
first check each folder isit dependency of some drive
sudo pacman -Qo /etc/path/folder/
( example: /usr/lib/libEGL_nvidia.so.0 )
if they are not under any dependecy drive, unlink them with this
sudo unlink /etc/path/folder/
then update the drive with
sudo mhwd -i pci video-nvidia
reboot the system and they login will be back on
1 Like
Thanks bogdancovaciu very grateful for helping me understand the problem and patiently guide me, please stay health and weathly with all the best for you :D. really appreciate
1 Like
This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.