[opensuse] no way to get nvidia modules loaded?
I start a new thread to not mix up things. I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE The nvidia drivers are installed, but they are not loaded. I have no idea why. I completely removed everything nvidia, disabling the opensuse.nvidia repository. Then booted in init 3 and installed those drivers again using yast. I hoped this fresh istall would make them appear loaded, but no. It is possible that due to the many things I tried to get this optimus stuff running again (it worked under 13.x using suse-prime) that there is some mess in the things installed or not installed. But I don't know what is needed and what not and where to begin to search.... By the way the is no /etc/X11/xorg.conf file on my system and yast said something about that when installing the nvidia drivers. I have a xorg.conf.bak and a xorg.conf.install, furthermore a xorg.conf.nvidia-post and xorg.conf.nvidia-xconfig-original. I copied xorg.conf.bak (and later xorg.conf.install) to xorg.conf, but with this I could not reach the graphical kde login anymore. So I deleted it again. An here I am. So, if you think there is any hope that I can use the nvidia card on this optimus laptop, please help! I guess the first thing is to have those nvidia drivers loaded. And then go further from there... -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 15/07/2018 à 11:58, Daniel Bauer a écrit :
I guess the first thing is to have those nvidia drivers loaded. And then go further from there...
no real idea. If nvidia is installed, nvidia-setings or nvidia-xconfig could be of some use, but on my own computer this write an xorg.conf file that prevent booting... bumblebee config says also to disable nvidia http://dodin.org/wiki/pmwiki.php?n=Doc.OptimusVideo given so, I tried to make it works with nouveau and sometime it works, sometime not... so I stopped trying :-( jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
15.07.2018 13:14, jdd@dodin.org пишет:
Le 15/07/2018 à 11:58, Daniel Bauer a écrit :
I guess the first thing is to have those nvidia drivers loaded. And then go further from there...
no real idea. If nvidia is installed, nvidia-setings or nvidia-xconfig could be of some use, but on my own computer this write an xorg.conf file that prevent booting...
bumblebee config says also to disable nvidia
Of course it does. Bumblebee physically powers down dGPU when it is not used so you cannot have drivers loaded permanently. They are loaded when you start program using bumblebee.
http://dodin.org/wiki/pmwiki.php?n=Doc.OptimusVideo
given so, I tried to make it works with nouveau and sometime it works, sometime not...
so I stopped trying :-(
jdd
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-15 12:30, Andrei Borzenkov wrote:
15.07.2018 13:14, jdd@dodin.org пишет:
Le 15/07/2018 à 11:58, Daniel Bauer a écrit :
I guess the first thing is to have those nvidia drivers loaded. And then go further from there...
no real idea. If nvidia is installed, nvidia-setings or nvidia-xconfig could be of some use, but on my own computer this write an xorg.conf file that prevent booting...
bumblebee config says also to disable nvidia
Of course it does. Bumblebee physically powers down dGPU when it is not used so you cannot have drivers loaded permanently. They are loaded when you start program using bumblebee.
Ah! Interesting. However, Daniel said on another post that he would be happy to use NVidia permanently. His Intel graphics do not work well, there is flicker and other issues. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
I completely removed everything nvidia, disabling the opensuse.nvidia repository. Then booted in init 3 and installed those drivers again using yast. I hoped this fresh istall would make them appear loaded, but no.
It is possible that due to the many things I tried to get this optimus stuff running again (it worked under 13.x using suse-prime) that there is some mess in the things installed or not installed. But I don't know what is needed and what not and where to begin to search....
By the way the is no /etc/X11/xorg.conf file on my system and yast said something about that when installing the nvidia drivers.
I have a xorg.conf.bak and a xorg.conf.install, furthermore a xorg.conf.nvidia-post and xorg.conf.nvidia-xconfig-original.
I copied xorg.conf.bak (and later xorg.conf.install) to xorg.conf, but with this I could not reach the graphical kde login anymore. So I deleted it again. An here I am.
So, if you think there is any hope that I can use the nvidia card on this optimus laptop, please help!
I guess the first thing is to have those nvidia drivers loaded. And then go further from there...
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt https://www.daniel-bauer.com/test/dmsg2.txt -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows modinfo nvidia nvidia-drm -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found. filename: /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko version: 390.67 supported: external license: MIT srcversion: F3776C9F2655E8F760F5FD9 alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00* alias: pci:v000010DEd*sv*sd*bc03sc02i00* alias: pci:v000010DEd*sv*sd*bc03sc00i00* depends: drm,drm_kms_helper,nvidia-modeset retpoline: Y vermagic: 4.4.138-59-default SMP mod_unload modversions parm: modeset:Enable atomic kernel modesetting (1 = enable, 0 = disable (default)) (bool) -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What rpm -qa '*nvidia*' '*kernel*' says?
filename: /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko> version: 390.67 supported: external license: MIT srcversion: F3776C9F2655E8F760F5FD9 alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00* alias: pci:v000010DEd*sv*sd*bc03sc02i00* alias: pci:v000010DEd*sv*sd*bc03sc00i00* depends: drm,drm_kms_helper,nvidia-modeset retpoline: Y vermagic: 4.4.138-59-default SMP mod_unload modversions parm: modeset:Enable atomic kernel modesetting (1 = enable, 0 = disable (default)) (bool)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 16:13, Andrei Borzenkov wrote:
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет:
I start a new thread to not mix up things.
I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia running opensuse leap 42.3, KDE
The nvidia drivers are installed, but they are not loaded. I have no idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-( meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
filename: /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko> version: 390.67 supported: external license: MIT srcversion: F3776C9F2655E8F760F5FD9 alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00* alias: pci:v000010DEd*sv*sd*bc03sc02i00* alias: pci:v000010DEd*sv*sd*bc03sc00i00* depends: drm,drm_kms_helper,nvidia-modeset retpoline: Y vermagic: 4.4.138-59-default SMP mod_unload modversions parm: modeset:Enable atomic kernel modesetting (1 = enable, 0 = disable (default)) (bool)
-- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sunday, 15 July 2018 23:59:33 ACST Daniel Bauer wrote:
On 15.07.2018 16:13, Andrei Borzenkov wrote:
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет: > I start a new thread to not mix up things. > > I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia > running opensuse leap 42.3, KDE > > The nvidia drivers are installed, but they are not loaded. I have no > idea why.
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
The latest stable nvidia driver version is 396.24. Have you tried downloading and installing using the nvidia installer instead of the repo version? You'll need to have at least the kernel headers installed to do that. The drawback is that, unless you also setup dkms so that the nvidia drivers are automatically rebuilt when you reboot with an updated kernel, you'll need to manually rebuild and install the modules after a kernel update (and, if you build the drivers into initrd, you'll need to rebuild that too for the new kernel). The latest driver releases are announced here; https://devtalk.nvidia.com/default/topic/533434/linux/current-graphics-drive... You may also find information specific to bumblebee and the optimus chipset on the nvidia forums (although I haven't looked specifically for that). Is there a setting in the bios to choose which gpu to use by default? My guess is that the intel chipset is enabled by default at boot and the nvidia gpu is powered off, thus causing the nvidia modules to fail to load (since they're obviously not compatible with the intel embedded gpu). It may also be worth trying to include the nvidia drivers in initrd using dracut. This thread may be useful: https://forums.opensuse.org/showthread.php/529420-Problems-adding-nvidia-pro... Of course, the standard disclaimer applies: YMMV (your mileage may vary). None of the above is guaranteed to work, but it may be worth a try. -- ============================================================== Rodney Baker VK5ZTV rodney.baker@iinet.net.au CCNA #CSCO12880208 ============================================================== -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 20:02, Rodney Baker wrote:
On Sunday, 15 July 2018 23:59:33 ACST Daniel Bauer wrote:
On 15.07.2018 16:13, Andrei Borzenkov wrote:
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote: > 15.07.2018 12:58, Daniel Bauer пишет: >> I start a new thread to not mix up things. >> >> I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia >> running opensuse leap 42.3, KDE >> >> The nvidia drivers are installed, but they are not loaded. I have no >> idea why. > > Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
The latest stable nvidia driver version is 396.24. Have you tried downloading and installing using the nvidia installer instead of the repo version? You'll need to have at least the kernel headers installed to do that. The drawback is that, unless you also setup dkms so that the nvidia drivers are automatically rebuilt when you reboot with an updated kernel, you'll need to manually rebuild and install the modules after a kernel update (and, if you build the drivers into initrd, you'll need to rebuild that too for the new kernel).
The latest driver releases are announced here;
https://devtalk.nvidia.com/default/topic/533434/linux/current-graphics-drive...
Honestly, I am afraid of that. In old times (really long ago) I had the nvidia drivers directly from nvidia on a desktop machine, and, as you said, with every kernel update, I had to do a lot of install work again, with things that I followed line by line in my thick Linux-book without actually understanding it. I'm just a user... Nowadays, with so many kernel and other updates all the time and with so many changes that make online-instructions obsolete almost the moment they are published, I cannot risk that. Or I am not brace enough.
You may also find information specific to bumblebee and the optimus chipset on the nvidia forums (although I haven't looked specifically for that).
Is there a setting in the bios to choose which gpu to use by default? My guess is that the intel chipset is enabled by default at boot and the nvidia gpu is powered off, thus causing the nvidia modules to fail to load (since they're obviously not compatible with the intel embedded gpu).
Unfortunately there is nothing about that in my BIOS.
It may also be worth trying to include the nvidia drivers in initrd using dracut. This thread may be useful:
https://forums.opensuse.org/showthread.php/529420-Problems-adding-nvidia-pro...
I'll take a look.
Of course, the standard disclaimer applies: YMMV (your mileage may vary). None of the above is guaranteed to work, but it may be worth a try.
:-) -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
15.07.2018 17:29, Daniel Bauer пишет:
On 15.07.2018 16:13, Andrei Borzenkov wrote:
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote:
15.07.2018 12:58, Daniel Bauer пишет: > I start a new thread to not mix up things. > > I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia > running opensuse leap 42.3, KDE > > The nvidia drivers are installed, but they are not loaded. I have no > idea why. >
Show dmesg output immediately after boot.
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
This looks OK. And what find /lib/modules -name 'nvidia*' says? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 20:26, Andrei Borzenkov wrote:
15.07.2018 17:29, Daniel Bauer пишет:
On 15.07.2018 16:13, Andrei Borzenkov wrote:
15.07.2018 15:55, Daniel Bauer пишет:
On 15.07.2018 14:38, Andrei Borzenkov wrote:
15.07.2018 14:47, Daniel Bauer пишет:
On 15.07.2018 12:28, Andrei Borzenkov wrote: > 15.07.2018 12:58, Daniel Bauer пишет: >> I start a new thread to not mix up things. >> >> I have an Asus GL552V laptop, i7 6700, optimus graphics intel/nvidia >> running opensuse leap 42.3, KDE >> >> The nvidia drivers are installed, but they are not loaded. I have no >> idea why. >> > > Show dmesg output immediately after boot. >
after boot, kde-login, open console, su to root: dmesg > dmsg2.txt
Does "modprobe nvidia" work? What shows
modinfo nvidia nvidia-drm
meitli:~ # modprobe nvidia modprobe: ERROR: could not find module by name='nvidia' modprobe: ERROR: could not insert 'nvidia': Function not implemented
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
This looks OK. And what
find /lib/modules -name 'nvidia*'
says?
meitli:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-15 20:33, Daniel Bauer wrote:
On 15.07.2018 20:26, Andrei Borzenkov wrote:
This looks OK. And what
find /lib/modules -name 'nvidia*'
says?
meitli:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
For comparison, this is mine: Telcontar:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.136-56-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia.ko /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/scripts/nvidia /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko Telcontar:~ # The one we are looking for is "/lib/modules/4.4.76-1-default/updates/nvidia.ko" Then on the current one, there is a symlink: Telcontar:~ # uname -a Linux Telcontar 4.4.138-59-default #1 SMP Mon Jun 18 13:48:42 UTC 2018 (f0b8f6b) x86_64 x86_64 x86_64 GNU/Linux Telcontar:~ # l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko lrwxrwxrwx 1 root root 47 Jun 28 16:01 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko -> /lib/modules/4.4.76-1-default/updates/nvidia.ko Telcontar:~ # And you have it. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
15.07.2018 21:50, Carlos E. R. пишет:
On 2018-07-15 20:33, Daniel Bauer wrote:
On 15.07.2018 20:26, Andrei Borzenkov wrote:
This looks OK. And what
find /lib/modules -name 'nvidia*'
says?
meitli:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
For comparison, this is mine:
Telcontar:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.136-56-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia.ko /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/scripts/nvidia /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko Telcontar:~ #
The one we are looking for is "/lib/modules/4.4.76-1-default/updates/nvidia.ko"
No, we do not.
Then on the current one, there is a symlink:
Telcontar:~ # uname -a Linux Telcontar 4.4.138-59-default #1 SMP Mon Jun 18 13:48:42 UTC 2018 (f0b8f6b) x86_64 x86_64 x86_64 GNU/Linux Telcontar:~ # l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko lrwxrwxrwx 1 root root 47 Jun 28 16:01 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko -> /lib/modules/4.4.76-1-default/updates/nvidia.ko Telcontar:~ #
And you have it.
15.07.2018 21:33, Daniel Bauer пишет: ...
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
This looks OK. And what
find /lib/modules -name 'nvidia*'
says?
meitli:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
Well, somehow it fails to add weak updates. Download package nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm from repository on nVidia site and then do under script command to capture full output rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 find /lib/modules -name 'nvidia*' rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm find /lib/modules -name 'nvidia*' (whatever path you have downloaded package in) The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 07/15/2018 02:00 PM, Andrei Borzenkov wrote:
The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere.
Daniel, As a follow-on the Andrei's post (provide that data when you can), your dmesg output does show both the Intel and Nvidia hardware found, but it looks like it is left relying on Intel alone and I can't tell whether it loads anything more than the framebuffer from the dmesg output. I don't have an answer for you, but here are the parts of interest (at least to me) in your dmesg output (blank lines separate non-continuous pars of the dmesg output, '...' represents a short snip of 2 lines or less) [ 0.000000] Reserving Intel graphics stolen memory at 0x59000000-0x78ffffff [ 0.000000] e820: [mem 0x79000000-0xdfffffff] available for PCI devices [ 0.423022] vgaarb: setting as boot device: PCI:0000:00:02.0 [ 0.423023] vgaarb: device added: PCI:0000:00:02.0,decodes=io+mem,owns=io+mem,locks=none [ 0.423026] vgaarb: loaded [ 0.423027] vgaarb: bridge control possible 0000:00:02.0 [ 2.249863] nvidia_modeset: loading out-of-tree module taints kernel. [ 2.249865] nvidia_modeset: module license 'NVIDIA' taints kernel. [ 2.249865] Disabling lock debugging due to kernel taint [ 2.250135] nvidia_modeset: Unknown symbol nv_kthread_q_flush (err 0) [ 2.250168] nvidia_modeset: Unknown symbol nvidia_register_module (err 0) [ 2.250182] nvidia_modeset: Unknown symbol nv_kthread_q_schedule_q_item (err 0) [ 2.250204] nvidia_modeset: Unknown symbol nvidia_get_rm_ops (err 0) [ 2.250225] nvidia_modeset: Unknown symbol nv_kthread_q_item_init (err 0) [ 2.250235] nvidia_modeset: Unknown symbol nv_kthread_q_stop (err 0) [ 2.250243] nvidia_modeset: Unknown symbol nvidia_unregister_module (err 0) [ 2.250265] nvidia_modeset: Unknown symbol nv_kthread_q_init (err 0) [ 2.312641] [drm] Memory usable by graphics device = 4096M [ 2.312644] checking generic (b0000000 7e9000) vs hw (b0000000 10000000) [ 2.312645] fb: switching to inteldrmfb from EFI VGA [ 2.312678] Console: switching to colour dummy device 80x25 [ 2.312916] [drm] Replacing VGA console driver [ 2.313283] AVX2 version of gcm_enc/dec engaged. [ 2.313313] AES CTR mode by8 optimization enabled [ 2.319347] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 2.319349] [drm] Driver supports precise vblank timestamp query. [ 2.320977] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=io+mem:owns=io+mem [ 2.371213] ACPI: Video Device [GFX0] (multi-head: yes rom: no post: no) [ 2.376110] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input4 [ 2.376275] ACPI: Video Device [PEGP] (multi-head: no rom: yes post: no) [ 2.376316] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:12/LNXVIDEO:01/input/input5 [ 2.376455] [drm] Initialized i915 1.6.0 20151010 for 0000:00:02.0 on minor 0 [ 2.422912] [drm] GMBUS [i915 gmbus dpb] timed out, falling back to bit banging on pin 5 [ 2.478859] [drm] GMBUS [i915 gmbus dpd] timed out, falling back to bit banging on pin 6 [ 2.481502] fbcon: inteldrmfb (fb0) is primary device [ 3.578965] Console: switching to colour frame buffer device 240x67 [ 3.586447] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device [ 31.089366] nvidia_modeset: Unknown symbol nv_kthread_q_flush (err 0) [ 31.089488] nvidia_modeset: Unknown symbol nvidia_register_module (err 0) [ 31.089530] nvidia_modeset: Unknown symbol nv_kthread_q_schedule_q_item (err 0) ... [ 31.089637] nvidia_modeset: Unknown symbol nvidia_get_rm_ops (err 0) [ 31.089680] nvidia_modeset: Unknown symbol nv_kthread_q_item_init (err 0) [ 31.089729] nvidia_modeset: Unknown symbol nv_kthread_q_stop (err 0) [ 31.089753] nvidia_modeset: Unknown symbol nvidia_unregister_module (err 0) [ 31.090595] nvidia_modeset: Unknown symbol nv_kthread_q_init (err 0) Maybe another pair of eye here may see something I'm missing. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15.07.2018 21:00, Andrei Borzenkov wrote:
15.07.2018 21:33, Daniel Bauer пишет: ...
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 kernel-devel-4.4.138-59.1.noarch kernel-default-devel-4.4.138-59.1.x86_64 nvidia-computeG04-390.67-8.1.x86_64 kernel-firmware-20170530-20.1.noarch kernel-devel-4.4.132-53.1.noarch kernel-default-4.4.132-53.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-default-devel-4.4.132-53.1.x86_64 kernel-default-4.4.138-59.1.x86_64 x11-video-nvidiaG04-390.67-8.1.x86_64 nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
This looks OK. And what
find /lib/modules -name 'nvidia*'
says?
meitli:~ # find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
Well, somehow it fails to add weak updates. Download package nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm from repository on nVidia site and then do under script command to capture full output
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 find /lib/modules -name 'nvidia*' rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm find /lib/modules -name 'nvidia*'
(whatever path you have downloaded package in)
The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere.
ok, I just did that. The output of the 4 commands is here: https://www.daniel-bauer.com/test/nvidia_rpm_output.txt -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-16 10:26, Daniel Bauer wrote:
On 15.07.2018 21:00, Andrei Borzenkov wrote:
The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere.
ok, I just did that. The output of the 4 commands is here: https://www.daniel-bauer.com/test/nvidia_rpm_output.txt
The build seems to go Ok, no errors. But there are errors just after mkinitrd is rebuilt:
+ output='depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_flush depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_register_module depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_schedule_q_item depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_get_rm_ops depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_item_init depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_stop depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_unregister_module depmod: WARNING: //lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_init'
and more. Warning: /lib/modules/4.4.132-53-default is inconsistent Warning: weak-updates symlinks might not be created Finally: find /lib/modules -name 'nvidia*' /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia is the same as you posted yesterday, which Andrei said was missing the weak updates. I have: cer@Telcontar:~> find /lib/modules -name 'nvidia*' /lib/modules/4.4.136-56-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia.ko <=== /lib/modules/4.4.136-56-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko <=== /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-uvm.ko /lib/modules/scripts/nvidia /lib/modules/4.4.76-1-default/updates/nvidia.ko /lib/modules/4.4.76-1-default/updates/nvidia-uvm.ko cer@Telcontar:~> I think he means those I marked with arrows. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On Mon, Jul 16, 2018 at 11:26 AM, Daniel Bauer <linux@daniel-bauer.com> wrote:
Well, somehow it fails to add weak updates. Download package nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm from repository on nVidia site and then do under script command to capture full output
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 find /lib/modules -name 'nvidia*' rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm find /lib/modules -name 'nvidia*'
(whatever path you have downloaded package in)
The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere.
ok, I just did that. The output of the 4 commands is here: https://www.daniel-bauer.com/test/nvidia_rpm_output.txt
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
/usr/src/kernel-modules/nvidia-390.67-default / rm -f -r conftest make[1]: Entering directory '/usr/src/linux-4.4.138-59' make[2]: Entering directory '/usr/src/linux-4.4.138-59-obj/x86_64/default' CLEAN /usr/src/kernel-modules/nvidia-390.67-default CLEAN /usr/src/kernel-modules/nvidia-390.67-default/.tmp_versions CLEAN /usr/src/kernel-modules/nvidia-390.67-default/Module.symvers make[2]: Leaving directory '/usr/src/linux-4.4.138-59-obj/x86_64/default' make[1]: Leaving directory '/usr/src/linux-4.4.138-59' / update-alternatives: warning: alternative /usr/lib/nvidia/alternate-install-present-default (part of link group alternate-install-present) doesn't exist; removing from list of alternatives update-alternatives: warning: /etc/alternatives/alternate-install-present is dangling; it will be updated with best choice
-----------------------
find /lib/modules -name 'nvidia*'
/lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
-----------------------
Well, that is the problem; these files must be symlinks to base modules which they are apparently not. Having output below would be interesting to confirm: ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
+ echo 'Warning: /lib/modules/4.4.138-59-default is inconsistent' Warning: /lib/modules/4.4.138-59-default is inconsistent + echo 'Warning: weak-updates symlinks might not be created' Warning: weak-updates symlinks might not be created
So try now 1. Remove package again 2. Delete files /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko 3. Run (to be on safe side) depmod -a 4.4.132-53-default depmod -a 4.4.138-59-default 4. Reinstall package, verify that you have correct symlinks now. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 16.07.2018 10:49, Andrei Borzenkov wrote:
On Mon, Jul 16, 2018 at 11:26 AM, Daniel Bauer <linux@daniel-bauer.com> wrote:
Well, somehow it fails to add weak updates. Download package nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm from repository on nVidia site and then do under script command to capture full output
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 find /lib/modules -name 'nvidia*' rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm find /lib/modules -name 'nvidia*'
(whatever path you have downloaded package in)
The "rpm -ivv" will produce very verbose output. Upload file typescript created by script command somewhere.
ok, I just did that. The output of the 4 commands is here: https://www.daniel-bauer.com/test/nvidia_rpm_output.txt
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
/usr/src/kernel-modules/nvidia-390.67-default / rm -f -r conftest make[1]: Entering directory '/usr/src/linux-4.4.138-59' make[2]: Entering directory '/usr/src/linux-4.4.138-59-obj/x86_64/default' CLEAN /usr/src/kernel-modules/nvidia-390.67-default CLEAN /usr/src/kernel-modules/nvidia-390.67-default/.tmp_versions CLEAN /usr/src/kernel-modules/nvidia-390.67-default/Module.symvers make[2]: Leaving directory '/usr/src/linux-4.4.138-59-obj/x86_64/default' make[1]: Leaving directory '/usr/src/linux-4.4.138-59' / update-alternatives: warning: alternative /usr/lib/nvidia/alternate-install-present-default (part of link group alternate-install-present) doesn't exist; removing from list of alternatives update-alternatives: warning: /etc/alternatives/alternate-install-present is dangling; it will be updated with best choice
-----------------------
find /lib/modules -name 'nvidia*'
/lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.138-59-default/kernel/drivers/net/ethernet/nvidia /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/kernel/drivers/net/ethernet/nvidia
-----------------------
Well, that is the problem; these files must be symlinks to base modules which they are apparently not. Having output below would be interesting to confirm:
ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
meitli:~ # ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko lrwxrwxrwx 1 root root 51 Jul 14 13:49 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko
+ echo 'Warning: /lib/modules/4.4.138-59-default is inconsistent' Warning: /lib/modules/4.4.138-59-default is inconsistent + echo 'Warning: weak-updates symlinks might not be created' Warning: weak-updates symlinks might not be created
So try now
Do I understand correct?:
1. Remove package again
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 ?
2. Delete files
/lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
delete them directly with rm ... ?
3. Run (to be on safe side)
depmod -a 4.4.132-53-default depmod -a 4.4.138-59-default
4. Reinstall package, verify that you have correct symlinks now.
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 ? ? -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-16 11:00, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote:
Well, that is the problem; these files must be symlinks to base modules which they are apparently not. Having output below would be interesting to confirm:
ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
meitli:~ # ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko lrwxrwxrwx 1 root root 51 Jul 14 13:49 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko
It was to be one long line, four paths: ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On 16.07.2018 11:15, Carlos E. R. wrote:
On 2018-07-16 11:00, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote:
Well, that is the problem; these files must be symlinks to base modules which they are apparently not. Having output below would be interesting to confirm:
ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
meitli:~ # ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko lrwxrwxrwx 1 root root 51 Jul 14 13:49 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko
It was to be one long line, four paths:
ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
Thank you Carlos. So this now gives: meitli:~ # ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko lrwxrwxrwx 1 root root 51 Jul 16 11:16 /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko lrwxrwxrwx 1 root root 55 Jul 16 11:16 /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko lrwxrwxrwx 1 root root 51 Jul 16 11:16 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko lrwxrwxrwx 1 root root 55 Jul 16 11:16 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-modeset.ko -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
* Daniel Bauer <linux@daniel-bauer.com> [07-16-18 05:01]:
On 16.07.2018 10:49, Andrei Borzenkov wrote:
[...]
4. Reinstall package, verify that you have correct symlinks now.
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 ?
zypper -v in nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 rpm -e deletes package, not install and --nodeps does not check that other packages may be affected -- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Registered Linux User #207535 @ http://linuxcounter.net Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-16 13:34, Patrick Shanahan wrote:
* Daniel Bauer <> [07-16-18 05:01]:
On 16.07.2018 10:49, Andrei Borzenkov wrote:
[...]
4. Reinstall package, verify that you have correct symlinks now.
rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 ?
zypper -v in nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64
rpm -e deletes package, not install and --nodeps does not check that other packages may be affected
Yes, exactly what was intended. First remove with no checks, then some manual file removes and checks, then another step that was not in the quote, install again with rpm command from local file, with certain options to get a very verbose output ("rpm -ivv"). No, zypper is not adequate for this particular set of operations, sorry. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On 16.07.2018 10:49, Andrei Borzenkov wrote
Well, that is the problem; these files must be symlinks to base modules which they are apparently not. Having output below would be interesting to confirm:
ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
+ echo 'Warning: /lib/modules/4.4.138-59-default is inconsistent' Warning: /lib/modules/4.4.138-59-default is inconsistent + echo 'Warning: weak-updates symlinks might not be created' Warning: weak-updates symlinks might not be created
So try now
1. Remove package again 2. Delete files
/lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko
3. Run (to be on safe side)
depmod -a 4.4.132-53-default depmod -a 4.4.138-59-default
4. Reinstall package, verify that you have correct symlinks now.
So I did: rpm -e --nodeps nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 rm /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko rm /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko rm /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-drm.ko rm /lib/modules/4.4.132-53-default/weak-updates/updates/nvidia-modeset.ko depmod -a 4.4.132-53-default depmod -a 4.4.138-59-default rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm the output of the last command is here: https://www.daniel-bauer.com/test/nvidia_rpm_output_2.txt how do I verify the symlinks? ls -l /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko still gives: lrwxrwxrwx 1 root root 51 Jul 16 11:16 /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-drm.ko -> /lib/modules/4.4.76-1-default/updates/nvidia-drm.ko -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-16 11:29, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote
rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm
the output of the last command is here: https://www.daniel-bauer.com/test/nvidia_rpm_output_2.txt
how do I verify the symlinks?
I would reboot :-) -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On Mon, Jul 16, 2018 at 12:45 PM, Carlos E. R. <robin.listas@telefonica.net> wrote:
On 2018-07-16 11:29, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote
rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm
the output of the last command is here: https://www.daniel-bauer.com/test/nvidia_rpm_output_2.txt
how do I verify the symlinks?
I would reboot :-)
modinfo nvidia modprobe nvidia should be enough. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 16.07.2018 11:53, Andrei Borzenkov wrote:
On Mon, Jul 16, 2018 at 12:45 PM, Carlos E. R. <robin.listas@telefonica.net> wrote:
On 2018-07-16 11:29, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote
rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm
the output of the last command is here: https://www.daniel-bauer.com/test/nvidia_rpm_output_2.txt
how do I verify the symlinks?
I would reboot :-)
I did
modinfo nvidia
modinfo nvidia filename: /lib/modules/4.4.138-59-default/weak-updates/updates/nvidia.ko alias: char-major-195-* version: 390.67 supported: external license: NVIDIA srcversion: 209B1D0CB123DE466F700AD alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00* alias: pci:v000010DEd*sv*sd*bc03sc02i00* alias: pci:v000010DEd*sv*sd*bc03sc00i00* depends: ipmi_msghandler retpoline: Y vermagic: 4.4.138-59-default SMP mod_unload modversions parm: NVreg_Mobile:int parm: NVreg_ResmanDebugLevel:int parm: NVreg_RmLogonRC:int parm: NVreg_ModifyDeviceFiles:int parm: NVreg_DeviceFileUID:int parm: NVreg_DeviceFileGID:int parm: NVreg_DeviceFileMode:int parm: NVreg_UpdateMemoryTypes:int parm: NVreg_InitializeSystemMemoryAllocations:int parm: NVreg_UsePageAttributeTable:int parm: NVreg_MapRegistersEarly:int parm: NVreg_RegisterForACPIEvents:int parm: NVreg_CheckPCIConfigSpace:int parm: NVreg_EnablePCIeGen3:int parm: NVreg_EnableMSI:int parm: NVreg_TCEBypassMode:int parm: NVreg_UseThreadedInterrupts:int parm: NVreg_EnableStreamMemOPs:int parm: NVreg_EnableBacklightHandler:int parm: NVreg_EnableUserNUMAManagement:int parm: NVreg_EnableIBMNPURelaxedOrderingMode:int parm: NVreg_MemoryPoolSize:int parm: NVreg_IgnoreMMIOCheck:int parm: NVreg_RegistryDwords:charp parm: NVreg_RegistryDwordsPerDevice:charp parm: NVreg_RmMsg:charp parm: NVreg_AssignGpus:charp So, I think the modules are here and loaded... Now how to continue with prime-select? If I run prime-select nvidia (after reboot) the script runs but then something (?) crashes - a message appears on screen but due to the flickering I could not read it. I cannot logout anymore, nor reboot with the kde-menu-button. I rebooted with shutdown from console... -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-16 12:09, Daniel Bauer wrote:
On 16.07.2018 11:53, Andrei Borzenkov wrote:
On Mon, Jul 16, 2018 at 12:45 PM, Carlos E. R. <robin.listas@telefonica.net> wrote:
On 2018-07-16 11:29, Daniel Bauer wrote:
On 16.07.2018 10:49, Andrei Borzenkov wrote
rpm -ivvh ./nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm
the output of the last command is here: https://www.daniel-bauer.com/test/nvidia_rpm_output_2.txt
how do I verify the symlinks?
I would reboot :-)
I did
...
So, I think the modules are here and loaded...
Do you see improvements at this point? I do not know, I don't have optimus hardware.
Now how to continue with prime-select? If I run prime-select nvidia (after reboot) the script runs but then something (?) crashes - a message appears on screen but due to the flickering I could not read it. I cannot logout anymore, nor reboot with the kde-menu-button.
I rebooted with shutdown from console...
Oh :-( -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
16.07.2018 13:09, Daniel Bauer пишет:
So, I think the modules are here and loaded...
Now how to continue with prime-select?
You should either start new thread or at least change subject here. You original problem in subject is solved. I do not have spare system to test Leap 42 and it does not have live option. I tested live Leap 15 on optimus system and it worked but it does not mean it will work in Leap 15 (completely different Xorg stack). Still it may give enough starting point (of course Leap 42 still needs ldconfig mangling that glvnd made obsolete on Leap 15). -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 07/16/2018 03:26 AM, Daniel Bauer wrote:
Well, somehow it fails to add weak updates. Download package nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64.rpm from repository on nVidia site and then do under script command to capture full output
I'm not sure what is going on, but from my earlier post looking at what was interesting in the dmesg output, I picked up on: [ 2.250135] nvidia_modeset: Unknown symbol nv_kthread_q_flush (err 0) [ 2.250168] nvidia_modeset: Unknown symbol nvidia_register_module (err 0) [ 2.250182] nvidia_modeset: Unknown symbol nv_kthread_q_schedule_q_item (err 0) [ 2.250204] nvidia_modeset: Unknown symbol nvidia_get_rm_ops (err 0) [ 2.250225] nvidia_modeset: Unknown symbol nv_kthread_q_item_init (err 0) [ 2.250235] nvidia_modeset: Unknown symbol nv_kthread_q_stop (err 0) [ 2.250243] nvidia_modeset: Unknown symbol nvidia_unregister_module (err 0) [ 2.250265] nvidia_modeset: Unknown symbol nv_kthread_q_init (err 0) Now from the current output Daniel posted, it indicates those will be provided by the driver: D: Requires: ksym(default:nv_kthread_q_flush) = a5a6ef41 YES (added provide) D: Requires: ksym(default:nv_kthread_q_init) = 1ad58eb YES (added provide) D: Requires: ksym(default:nv_kthread_q_item_init) = 519bbbeb YES (added provide) D: Requires: ksym(default:nv_kthread_q_run_self_test) = dc74e56f YES (added provide) D: Requires: ksym(default:nv_kthread_q_schedule_q_item) = 6f8d1d81 YES (added provide) D: Requires: ksym(default:nv_kthread_q_stop) = b3ad9cd7 YES (added provide) D: Requires: ksym(default:nvidia_get_rm_ops) = db5880e5 YES (added provide) D: Requires: ksym(default:nvidia_register_module) = e7da488 YES (added provide) D: Requires: ksym(default:nvidia_unregister_module) = 6f96aacb YES (added provide) But for some reason appear not to be found: ++ /sbin/depmod -b / -ae -E /var/tmp/weak-modules2.Wyd2zP/symvers-4.4.138-59-default 4.4.138-59-default + output='depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_flush depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_register_module depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_schedule_q_item depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_get_rm_ops depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_item_init depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_stop depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nvidia_unregister_module depmod: WARNING: //lib/modules/4.4.138-59-default/weak-updates/updates/nvidia-modeset.ko needs unknown symbol nv_kthread_q_init' ... This is just a guess looking at what parts of the dmesg and current output don't make sense. Hopefully this will all be corrected by the deletion suggested by Andrei. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2018-07-15 16:29, Daniel Bauer wrote:
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 Ok kernel-devel-4.4.138-59.1.noarch Ok kernel-default-devel-4.4.138-59.1.x86_64 Ok nvidia-computeG04-390.67-8.1.x86_64 Ok kernel-firmware-20170530-20.1.noarch Ok kernel-devel-4.4.132-53.1.noarch previous kernel-default-4.4.132-53.1.x86_64 previous kernel-macros-4.4.138-59.1.noarch Ok kernel-default-devel-4.4.132-53.1.x86_64 previous kernel-default-4.4.138-59.1.x86_64 ok x11-video-nvidiaG04-390.67-8.1.x86_64 ok nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 Ok
For comparison, this is mine - desktop with just nvidia hardware; and not the same card, so I use G03: Telcontar:~ # rpm -qa '*nvidia*' '*kernel*' kernel-source-4.4.136-56.1.noarch not needed nvidia-glG03-340.106-11.1.x86_64 kernel-default-4.4.138-59.1.x86_64 nfs-kernel-server-1.3.0-29.3.1.x86_64 kernel-default-devel-4.4.136-56.1.x86_64 texlive-l3kernel-doc-2016.122.svn_6512svn41246-25.5.noarch texlive-l3kernel-2016.122.svn_6512svn41246-25.5.noarch kernel-source-4.4.138-59.1.noarch not needed kernel-syms-4.4.136-56.1.x86_64 kernel-devel-4.4.138-59.1.noarch nvidia-uvm-gfxG03-kmp-default-340.106_k4.4.76_1-11.1.x86_64 ?? nvidia-gfxG03-kmp-default-340.106_k4.4.76_1-11.1.x86_64 kernel-syms-4.4.138-59.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-firmware-20170530-20.1.noarch kernel-default-4.4.136-56.1.x86_64 kernel-devel-4.4.136-56.1.noarch nvidia-computeG03-340.106-11.1.x86_64 kernel-docs-4.4.138-59.1.noarch x11-video-nvidiaG03-340.106-11.1.x86_64 kernel-default-devel-4.4.138-59.1.x86_64 Telcontar:~ # The only difference that might be important is that I have nvidia-uvm-gfxG03-kmp-default and you don't have nvidia-uvm-gfxG04-kmp-default nvidia-uvm-gfxG03-kmp-default - NVIDIA Unified Memory kernel module The package provides the NVIDIA Unified Memory kernel module. I see in the repository that it only exists for G03. :-? Then it may be nothing. I would reinstall the rpm "nvidia-glG03-340.106-11.1.x86_64" from the command line, (zypper) to see the exact output it prints. Maybe there is some error message that you do not see in YaST. Paste it here. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)
On 15.07.2018 20:42, Carlos E. R. wrote:
On 2018-07-15 16:29, Daniel Bauer wrote:
meitli:~ # modinfo nvidia nvidia-drm modinfo: ERROR: Module nvidia not found.
Something is seriously wrong. What
rpm -qa '*nvidia*' '*kernel*'
says?
yes :-(
meitli:~ # rpm -qa '*nvidia*' '*kernel*' nvidia-glG04-390.67-8.1.x86_64 Ok kernel-devel-4.4.138-59.1.noarch Ok kernel-default-devel-4.4.138-59.1.x86_64 Ok nvidia-computeG04-390.67-8.1.x86_64 Ok kernel-firmware-20170530-20.1.noarch Ok kernel-devel-4.4.132-53.1.noarch previous kernel-default-4.4.132-53.1.x86_64 previous kernel-macros-4.4.138-59.1.noarch Ok kernel-default-devel-4.4.132-53.1.x86_64 previous kernel-default-4.4.138-59.1.x86_64 ok x11-video-nvidiaG04-390.67-8.1.x86_64 ok nvidia-gfxG04-kmp-default-390.67_k4.4.76_1-8.1.x86_64 Ok
For comparison, this is mine - desktop with just nvidia hardware; and not the same card, so I use G03:
Telcontar:~ # rpm -qa '*nvidia*' '*kernel*' kernel-source-4.4.136-56.1.noarch not needed nvidia-glG03-340.106-11.1.x86_64 kernel-default-4.4.138-59.1.x86_64 nfs-kernel-server-1.3.0-29.3.1.x86_64 kernel-default-devel-4.4.136-56.1.x86_64 texlive-l3kernel-doc-2016.122.svn_6512svn41246-25.5.noarch texlive-l3kernel-2016.122.svn_6512svn41246-25.5.noarch kernel-source-4.4.138-59.1.noarch not needed kernel-syms-4.4.136-56.1.x86_64 kernel-devel-4.4.138-59.1.noarch nvidia-uvm-gfxG03-kmp-default-340.106_k4.4.76_1-11.1.x86_64 ?? nvidia-gfxG03-kmp-default-340.106_k4.4.76_1-11.1.x86_64 kernel-syms-4.4.138-59.1.x86_64 kernel-macros-4.4.138-59.1.noarch kernel-firmware-20170530-20.1.noarch kernel-default-4.4.136-56.1.x86_64 kernel-devel-4.4.136-56.1.noarch nvidia-computeG03-340.106-11.1.x86_64 kernel-docs-4.4.138-59.1.noarch x11-video-nvidiaG03-340.106-11.1.x86_64 kernel-default-devel-4.4.138-59.1.x86_64 Telcontar:~ #
The only difference that might be important is that I have nvidia-uvm-gfxG03-kmp-default and you don't have nvidia-uvm-gfxG04-kmp-default
nvidia-uvm-gfxG03-kmp-default - NVIDIA Unified Memory kernel module
The package provides the NVIDIA Unified Memory kernel module.
I see in the repository that it only exists for G03. :-?
Then it may be nothing.
I would reinstall the rpm "nvidia-glG03-340.106-11.1.x86_64" from the command line, (zypper) to see the exact output it prints. Maybe there is some error message that you do not see in YaST. Paste it here.
When I remove the G04 drivers, automatically the G03-drivers are installed - BUT: suse-prime is removed. When I install suse-prime automatically the G04 drivers are installed... I don't know why suse-prime insists on G04, but I guess there is a reason ? -- Daniel Bauer photographer Basel Málaga https://www.patreon.com/danielbauer http://www.daniel-bauer.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
El 2018-07-15 a las 20:51 +0200, Daniel Bauer escribió:
I would reinstall the rpm "nvidia-glG03-340.106-11.1.x86_64" from the command line, (zypper) to see the exact output it prints. Maybe there is some error message that you do not see in YaST. Paste it here.
When I remove the G04 drivers, automatically the G03-drivers are installed - BUT: suse-prime is removed. When I install suse-prime automatically the G04 drivers are installed...
I don't know why suse-prime insists on G04, but I guess there is a reason ?
I don't understant that logic. Do what Andrei has written at 21 hours. You will have to remove packages using rpm directly, not zypper, because as you see zypper tries to resolve deps some other way. You have to manually download the rpms to a directory previously. I think you may run the commands to get the output to paste this way: rm outputlogfile program 2>&1 | tee -a outputlogfile echo >> outputlogfile echo "-----------------------" >> outputlogfile echo >> outputlogfile other_program 2>&1 | tee -a outputlogfile Finally: susepaste -n "Daniel" -t "rpm output" -e 40320 outputlogfile and paste here the link you obtain. Another method is in the terminal, select edit/select all, copy, then in an editor. But if the output is really verbose, the buffer may overflow and not capture everything. -- Cheers Carlos E. R. (from openSUSE 15.0 (Legolas))
* Carlos E. R. <robin.listas@telefonica.net> [07-15-18 17:49]:
El 2018-07-15 a las 20:51 +0200, Daniel Bauer escribió:
I would reinstall the rpm "nvidia-glG03-340.106-11.1.x86_64" from the command line, (zypper) to see the exact output it prints. Maybe there is some error message that you do not see in YaST. Paste it here.
When I remove the G04 drivers, automatically the G03-drivers are installed - BUT: suse-prime is removed. When I install suse-prime automatically the G04 drivers are installed...
I don't know why suse-prime insists on G04, but I guess there is a reason ?
I don't understant that logic.
Do what Andrei has written at 21 hours. You will have to remove packages using rpm directly, not zypper, because as you see zypper tries to resolve deps some other way.
no, not if you use "force" zypper -v rm -f <package> -- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Registered Linux User #207535 @ http://linuxcounter.net Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
* Carlos E. R. <robin.listas@telefonica.net> [07-15-18 17:49]:
El 2018-07-15 a las 20:51 +0200, Daniel Bauer escribió:
I would reinstall the rpm "nvidia-glG03-340.106-11.1.x86_64" from the command line, (zypper) to see the exact output it prints. Maybe there is some error message that you do not see in YaST. Paste it here.
When I remove the G04 drivers, automatically the G03-drivers are installed - BUT: suse-prime is removed. When I install suse-prime automatically the G04 drivers are installed...
I don't know why suse-prime insists on G04, but I guess there is a reason ?
I don't understant that logic.
Do what Andrei has written at 21 hours. You will have to remove packages using rpm directly, not zypper, because as you see zypper tries to resolve deps some other way.
no, not if you use "force" zypper -v rm -f <package> ps: that is incorrect, s/b zypper -v rm --force <package> -- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Registered Linux User #207535 @ http://linuxcounter.net Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (7)
-
Andrei Borzenkov -
Carlos E. R. -
Daniel Bauer -
David C. Rankin -
jdd＠dodin.org -
Patrick Shanahan -
Rodney Baker