[opensuse] Re: Nvidia GFX kernel modules for older 11.1-kernel 2.6.27.7-9, anywhere?
Manfred Hollstein wrote:
On Mon, 29 Jun 2009, 14:04:04 +0200, Joachim Schrod wrote:
Per Jessen wrote:
Joachim Schrod wrote:
I'm looking for the proprietary Nvidia driver (kernel module) for a GeForce FX 5200 graphics card. A rpm is supposed to be at http://download.nvidia.com/opensuse/11.1/, named nvidia-gfxG01-kmp-pae.
But there is only a RPM for the most current 11.1 kernel, version 2.6.27.23_0.1. I need one for kernel 2.6.27.7-9. (I can't use the most current kernel since VMware Server 1 doesn't work properly with it anymore.)
Is there anywhere a repository archive for older versions where I can fetch a respective RPM for my kernel? Alternatively, is there another possibility to install that driver?
If the actual nvidia driver isn't any different, you could just install the latest rpm with --force ?
Of course, I tried that already... :-) That doesn't work, the kernel modules don't fit. (The error message at insmod time is "Invalid module format".)
Actually, I didn't expect them to work, calling insmod was pure mischief. From my VMware problems I know that quite a lot seemed to have happened between 2.6.27.7 and 2.6.27.23.
Dunno for how long it'll work, but the following site helped me:
http://www.insecure.ws/2008/10/20/vmware-specific-specific-55x-and-kernel-26...
That's the patch I'm using that works on 2.6.27.7. Starting with the upgrade to 2.6.27.21 (and now 2.6.27.23), Windows clients use 100% CPU time all the time. Linux clients are OK. Windows clients also have serious problems with clock thrift. Between 2.6.27.7 and 2.6.27.21/2.6.27.23 are quite some changes in SUSE's kernel config that concern the clock, virtualization, and others, IIRC. Maybe if one changes some of them back, it might work again. But my next step will probably be to try Workstation 6.5 on a trial system and see if it has the same problem or not. Actually, the NVidia setup is one of the few things missing to get a new workstation working so that I can use the old ones for such experiments... ;-) Joachim -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Joachim Schrod Email: jschrod@acm.org Roedermark, Germany -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Mon, 29 Jun 2009, 17:51:07 +0200, Joachim Schrod wrote:
On Mon, 29 Jun 2009, 14:04:04 +0200, Joachim Schrod wrote: [...] Dunno for how long it'll work, but the following site helped me:
http://www.insecure.ws/2008/10/20/vmware-specific-specific-55x-and-kernel-26...
That's the patch I'm using that works on 2.6.27.7. Starting with the upgrade to 2.6.27.21 (and now 2.6.27.23), Windows clients use 100% CPU time all the time. Linux clients are OK. Windows clients also have serious problems with clock thrift.
Between 2.6.27.7 and 2.6.27.21/2.6.27.23 are quite some changes in SUSE's kernel config that concern the clock, virtualization, and others, IIRC. Maybe if one changes some of them back, it might work again.
But my next step will probably be to try Workstation 6.5 on a trial system and see if it has the same problem or not. Actually, the NVidia setup is one of the few things missing to get a new workstation working so that I can use the old ones for such experiments... ;-)
FWIW, I've switched to WS 6.5 in the meantime, too, because I felt rather unsafe due to VMware potentially pushing their 2.x server solution more than fixing the 1.x version...
Joachim
Cheers. l8er manfred -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
participants (2)
-
Joachim Schrod
-
Manfred Hollstein