You are not logged in.
It seems that it is a kernel bug after all : http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=613078.
There is a patch, but like said Sven in the bug report,
"The mentioned patch was rather a hack, 2.6.38-rc6 has a proper patch that supposedly fixes the problem"
and that implies a rebuilt of the kernel or, in the other hand, playing with an experimental kernel.
The link to the patch : https://bugs.freedesktop.org/attachment.cgi?id=43011
EDIT :
I forgot to say that, some people applied the patch and it seems to work, samething with the kernel 2.6.38-rc6.
I didn't tested personnally any of these because I got a heavy load of works this weekend.
Last edited by CPU Gastronomy (2011-03-05 19:01:26)
Offline
I confirm that it is working with the kernel 2.6.38-rc8 without any problem to now.
Offline
Hello folks,
I'm having the same problem since I updated today.
Is there any applicable solution for a novice like me? => Means: I don't want to compile my own kernel, I feel a little nervous to result in a system that doesn't boot anything at all (e.g. encrypted harddiscs, etc).
Any urgent help, please!
Offline
As he said, you could downgrade the packages :
I had to downgrade the drm and nouveau related packages to recover the graphical environment :
libdrm (2.4.23-1 -> 2.4.22-3)
libgl (7.10-1 -> 7.9.0.git20101207-2)
libva (1.0.8-1 -> 1.0.6-1)
mesa (7.10-1 -> 7.9.0.git20101207-2)
xf86-video-nouveau (0.0.16_git20101217-1 -> 0.0.16_git20100819-2)running the commands under root :
cd /var/cache/pacman/pkg pacman -U libdrm-2.4.22-3-x86_64.pkg.tar.xz libgl-7.9.0.git20101207-2-x86_64.pkg.tar.xz libva-1.0.6-1-x86_64.pkg.tar.xz mesa-7.9.0.git20101207-2-x86_64.pkg.tar.xz xf86-video-nouveau-0.0.16_git20100819-2-x86_64.pkg.tar.xz
As you have the nouveau-dri package you will have to downgrade it too.
You should do that after having killed the graphical processes.
Sure, maybe you'll have to change the package name to the previous ones you had if needed.
Compiling a new kernel isn't that dangerous as long you don't save it over your current one (and don't play blindly in the config).
For your encrypted hardisks, don't worry, it doesn't modify your main configurations (ex : most of all in /etc).
I compiled my first kernel when I got this problem with my display and I had encrypted disks too.
Just check a little this page :
https://wiki.archlinux.org/index.php/Ke … rom_Source
It explains very well how to compile it safely without erasing your current one (meaning that if the new kernel isn't correct, you still can go with your current one).
But if it is an emergency, try to downgrade.
Last edited by CPU Gastronomy (2011-03-27 20:40:18)
Offline
Hello CPU Gastronomy, thanks for your quick answer.
Compiling a new kernel isn't that dangerous as long you don't save it over your current one (and don't play blindly in the config).
For your encrypted hardisks, don't worry, it doesn't modify your main configurations (ex : most of all in /etc).I compiled my first kernel when I got this problem with my display and I had encrypted disks too.
Just check a little this page :
https://wiki.archlinux.org/index.php/Ke … rom_Source
It explains very well how to compile it safely without erasing your current one (meaning that if the new kernel isn't correct, you still can go with your current one).
Well I tried to downgrade first, but this didn't take any effects. I also tried out some of the git-packages from AUR, e.g. nouveau-dri-git, but this even did not solve the problem.
But your words sound very helpful and stressless to build a kernel on my own, so I'll try this. (yet, I only had build Enlightenment, my favorite window manager on my own).
One more question to you: There is a kernel package in AUR, called aur/kernel26-nouveau-git, may I use this one or should I better do it the way described in https://wiki.archlinux.org/index.php/Ke … rom_Source?
But if it is an emergency, try to downgrade.
For now, I switched to the legacy closed source drivers. Bad console resolution, slow tty switching, but it works for now.
I really miss nouveau, but have some more calmness now to try the kernel build.
Thanks a lot, and hopefully this problem will also be fixed soon in the standard packages of Arch.
So long,
beachcoder
Offline
ok
kernel 38 solve
Offline
ok
kernel 38 solve
I still have
(EE) [drm] No DRICreatePCIBusID symbol
with current official state (kernel 2.6.38.2 and so on) with this card:
$ lspci | grep VGA
01:00.0 VGA compatible controller: nVidia Corporation G98 [GeForce 8400 GS] (rev a1)
"I exist" is the best myth I know..
Offline
I use driver: nouveau
with: GeForce 6150SE nForce 430
Offline
I use driver: nouveau
Yes, I'm saying about nouveau.
"I exist" is the best myth I know..
Offline
Offline
Can you be more elaborate? What does this question mark mean?
"I exist" is the best myth I know..
Offline
...only ... if you have read that link
Offline
...only ... if you have read that link
I have. What must I understand?
There is an information it is a kernel bug (I mean "no symbol"): https://bugs.archlinux.org/task/22700 OTOH some archers have downgraded noveau driver and are happy. So, I don't understand where is the reason of the symbol error. I have understrood your message as if with last driver and kernel all does work.
"I exist" is the best myth I know..
Offline
Deleted.
Last edited by diegoviola (2023-04-17 22:42:11)
Offline
@student975
I think you should create another thread if the bug persists, because it is not the same problem as the one that started the thread (that is solved).
Offline
@diegoviola
Thanks for informing! At last I have successfully started X with nouveau driver (being under one just now). I have 'G98 [GeForce 8400 GS] (rev a1)' fanless card.
@CPU Gastronomy
As you see, I'm here That error with symbol appears with 'sudo Xorg -configure' only, not preventing to start X. Cirrent X11 log hasn't the error.
"I exist" is the best myth I know..
Offline
Deleted.
Last edited by diegoviola (2023-04-17 22:41:49)
Offline
You should report upstream if the new nouveau/kernel fixed your issue.
I'm sure, I shouldn't At first, it isn't "my" issue - just have got a screen similar to one presented in the first topic post. Next, why I must bother developers not being a reporter (or even subsciber) of a bug when there isn't any problems? There are thousands of packages, and millions of users knowing "aha, with this release of that package that issue has gone, nice". Must all these millions notify developers at such situation?
Agree?
Last edited by student975 (2011-04-25 20:23:17)
"I exist" is the best myth I know..
Offline
Deleted.
Last edited by diegoviola (2023-04-17 22:41:14)
Offline
@diegoviola
Have you read my message? Again: why must I (or anybody) bother developers when there is not a problem? As for filing new issues - please, believe me, during ten years with single-boot linux workstation I have reported plenty of them "student" means desire to learn rather age
"I exist" is the best myth I know..
Offline
Deleted.
Last edited by diegoviola (2023-04-17 22:40:58)
Offline