[opensuse] no X after switch to EFI boot / NVIDIA
Hello, sorry I could not do much recherche.. It's hard without X. Maybe someone has a clue, why Xorg does not launch after switch from BIOS boot to EFI boot.. Thank you for any hint.. maybe it's also obvious for someone... First symptom, I have a different console font (I did never use kms before..). I tried to enable or disable modesetting (nomodeset / drm-nvidia.modesetting=0).. There are two nvidia cards installed.. maybe it is a resource conflict, that did not appear at BIOS boot. Maybe I have some incompatible packages installed... rpm -qa | grep -iE "nv-|nvidia|nouv" nvidia-glG04-390.87-lp150.10.2.x86_64 xf86-video-nv-2.1.21-lp150.1.7.x86_64 x11-video-nvidiaG04-390.87-lp150.10.2.x86_64 nvidia-computeG04-390.87-lp150.10.2.x86_64 php7-iconv-7.2.5-lp150.2.15.1.x86_64 nvidia-gfxG04-kmp-default-390.87_k4.12.14_lp150.11-lp150.10.1.x86_64 libvdpau_nouveau-18.0.2-lp150.17.2.x86_64 Mesa-dri-nouveau-18.0.2-lp150.17.2.x86_64 xf86-video-nouveau-1.0.15-lp150.1.9.x86_64 libdrm_nouveau2-32bit-2.4.91-lp150.1.2.x86_64 unoconv-0.7-lp150.1.8.noarch libdrm_nouveau2-2.4.96-lp150.245.1.x86_64 the output of x.org: https://www.swabian.net/extern/problemnvidia/xsession some important lines: (EE) [drm] Failed to open DRM device for (null): -2 (EE) [drm] Failed to open DRM device for (null): -2 the dmesg output: https://www.swabian.net/extern/problemnvidia/dmesg when I try to start the displaymanager.service I get instantly: [ 99.473246] resource sanity check: requesting [mem 0x000c0000-0x000fffff], which spans more than PCI Bus 0000:00 [mem 0x000c0000-0x000dffff window] [ 99.473347] caller _nv001169rm+0xe3/0x1d0 [nvidia] mapping multiple BARs Thank you for any help Paul Neuwirth -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Hello, sorry I could not do much recherche.. It's hard without X. Maybe someone has a clue, why Xorg does not launch after switch from BIOS boot to EFI boot.. Thank you for any hint.. maybe it's also obvious for someone...
just managed to boot using legacy bios... same problem... It may be related to a recent update..
First symptom, I have a different console font (I did never use kms before..). I tried to enable or disable modesetting (nomodeset / drm-nvidia.modesetting=0).. There are two nvidia cards installed.. maybe it is a resource conflict, that did not appear at BIOS boot. Maybe I have some incompatible packages installed...
rpm -qa | grep -iE "nv-|nvidia|nouv"
nvidia-glG04-390.87-lp150.10.2.x86_64 xf86-video-nv-2.1.21-lp150.1.7.x86_64 x11-video-nvidiaG04-390.87-lp150.10.2.x86_64 nvidia-computeG04-390.87-lp150.10.2.x86_64 php7-iconv-7.2.5-lp150.2.15.1.x86_64 nvidia-gfxG04-kmp-default-390.87_k4.12.14_lp150.11-lp150.10.1.x86_64 libvdpau_nouveau-18.0.2-lp150.17.2.x86_64 Mesa-dri-nouveau-18.0.2-lp150.17.2.x86_64 xf86-video-nouveau-1.0.15-lp150.1.9.x86_64 libdrm_nouveau2-32bit-2.4.91-lp150.1.2.x86_64 unoconv-0.7-lp150.1.8.noarch libdrm_nouveau2-2.4.96-lp150.245.1.x86_64
the output of x.org: https://www.swabian.net/extern/problemnvidia/xsession some important lines: (EE) [drm] Failed to open DRM device for (null): -2 (EE) [drm] Failed to open DRM device for (null): -2
the dmesg output: https://www.swabian.net/extern/problemnvidia/dmesg when I try to start the displaymanager.service I get instantly: [ 99.473246] resource sanity check: requesting [mem 0x000c0000-0x000fffff], which spans more than PCI Bus 0000:00 [mem 0x000c0000-0x000dffff window] [ 99.473347] caller _nv001169rm+0xe3/0x1d0 [nvidia] mapping multiple BARs
Thank you for any help
Paul Neuwirth
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2018-11-02 at 15:22 +0100, Paul Neuwirth wrote:
Hello, sorry I could not do much recherche.. It's hard without X. Maybe someone has a clue, why Xorg does not launch after switch from BIOS boot to EFI boot.. Thank you for any hint.. maybe it's also obvious for someone...
just managed to boot using legacy bios... same problem... It may be related to a recent update..
Ah. I was going to say that it might be a firmware (EFI) bug, thus try to update it. But if going back to bios did not solve things, then no.
First symptom, I have a different console font (I did never use kms before..). I tried to enable or disable modesetting (nomodeset / drm-nvidia.modesetting=0).. There are two nvidia cards installed.. maybe it is a resource conflict, that did not appear at BIOS boot. Maybe I have some incompatible packages installed...
rpm -qa | grep -iE "nv-|nvidia|nouv"
If you did some updates before the problem, try to undo them. I try not to never update the nvidia drivers since the system install day. - -- Cheers, Carlos E. R. (from openSUSE 42.3 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlvdpSQACgkQtTMYHG2NR9UKhACgi+pB7DbMv3dVcEqlqoXTU/CY 1k4An36/CdPzY4vUatRDyC6pXyg0l3gH =RDDV -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
If you did some updates before the problem, try to undo them.
I try not to never update the nvidia drivers since the system install day.
finally resolved it, by completely installing any video driver.. then installed the recent nvidia driver again.. there was a long postinstall script which was not executed when just updating or switching version.. thank you -----BEGIN PGP SIGNATURE----- Comment: Topal (http://freshmeat.net/projects/topal) iQIzBAEBCAAdFiEEFXicsLbzJdYXuR4eiJpr5jxgM+4FAlvdtVgACgkQiJpr5jxg M+5k1RAAumAuwuwGJIFkTzh+BVYIm/H9AMDE7NCHYiL9CoYp1yrMadkri+kiTufo ngooeiuSP6E7ifUVC8zZMv4sTFGB+a9xdxoPAyKcoEzMtXHEv+Yz2q4hF3Zgl2Vx 6xDQCbAyw9ZBJwtN/mBYXp/0e7dkX4ed0beRFTz9zVx95EcfhoeUErQ//DJT9gll g2SKY7zn4hlwCs6r7AQNWf8K0HYxuHO3pNeq7k9zIGTmSnjxbHiOj488/NGjfeuQ q1RMcGMmvDl26HBx5A7JJCag9kwsQSuNW3wEZoUicKhEtwsbqOK3hgkBiGavjWAC eSrof6cZfJLVr1EhRPlTKBqmeyN4wikFUyj5kpHcOniRe0XM2itCj0dZO3DIZpQ7 KbnoTiI0nV6p0Ns5lv4K/dNzJRQushJT3maWE01AHyLgX7TP6yG7UECxlf2Mqa2c PKvzm7J0HnYYSAQYC6bHw0hGqbSEEX0PIFS/Dbmk0pjPNe2Z4+HveC4wXH2Z5Dul pLG3EMS3TwIN4zPQnkWZlt/aumojXWLRh+EjgMadntkEaEMbmTKE8zYeYCfPgc3m MPlSYU/vVoNPMc9SayPJp2Ut8WcmuJXNBQISb3ZO9unLRL9srlD0OWbwiGpHpMV1 dxHwdZLBIrcRL2UvBH5I3ts+0R2nx7KA0kd6MB7XWcwxJtTEkIw= =JYeV -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
W dniu 03.11.2018 o 15:47, Paul Neuwirth pisze:
If you did some updates before the problem, try to undo them.
I try not to never update the nvidia drivers since the system install day.
finally resolved it, by completely installing any video driver.. then installed the recent nvidia driver again.. there was a long postinstall script which was not executed when just updating or switching version..
thank you
Thanks for the hint. I had seemingly unrelated problem. After last update my Xorg was working, but llvmpipe was used instead of hardware acceleration. Reinstalling Mesa, xorg-server and nvidia packages fixed the issue. Snapper diff shows two new files created: /usr/lib/tmpfiles.d/nvidia-logind-acl-trick.conf /usr/src/kernel-modules/nvidia-390.87-default/conftest26468.c They don't belong to any rpm package, so they must be created by some rpm post scripts. -- Adam Mizerski
participants (3)
-
Adam Mizerski -
Carlos E. R. -
Paul Neuwirth