Nivida module fails to load for Quadro FX 3700

[napopa]

Hi,
I'm trying to install the Nvidia driver form my Nvidia Quadro FX 3700 (10de:061a) on an ArchLinux system. The card is listed as supported both on Nvidia website and in the README.txt that comes with the driver.

I disabled the "nouveau" module system-wide and I compiled the Nvidia module flawlessly but when I try to load the it I got this

#modprobe nvidia

FATAL: Error inserting nvidia (/lib/modules/2.6.38-ARCH/kernel/drivers/video/nvidia.ko): No such device.

#dmesg

[ 3175.210636] nvidia 0000:03:00.0: PCI INT A -> Link[APC5] -> GSI 16 (level, low) -> IRQ 16
[ 3175.210647] nvidia 0000:03:00.0: setting latency timer to 64
[ 3175.210651] vgaarb: device changed decodes: PCI:0000:03:00.0,olddecodes=none,decodes=none:owns=io+mem
[ 3175.210676] NVRM: The NVIDIA GPU 0000:03:00.0 (PCI ID: 10de:061a) installed
[ 3175.210677] NVRM: in this system is not supported by the 275.09 NVIDIA Linux
[ 3175.210678] NVRM: graphics driver release.  Please see 'Appendix A -
[ 3175.210679] NVRM: Supported NVIDIA GPU Products' in this release's README,
[ 3175.210680] NVRM: available on the Linux graphics driver download page at
[ 3175.210681] NVRM: www.nvidia.com.
[ 3175.210688] nvidia 0000:03:00.0: PCI INT A disabled
[ 3175.210695] nvidia: probe of 0000:03:00.0 failed with error -1
[ 3175.210712] NVRM: The NVIDIA probe routine failed for 1 device(s).
[ 3175.210714] NVRM: None of the NVIDIA graphics adapters were initialized!

As I said, the card is listed as supported by the driver...
I tried several recent versions of the driver itself (even the beta) but I can't pick up too old ones since I need the 3d Stereo mode support.

Is there a way to force the driver to load anyways? or to disable the revision control of the card?
thanks!

[AaronP]

This sounds a lot like a problem reported recently on Ubuntu: https://bugs.launchpad.net/ubuntu/+s...rs/+bug/661248

This is not an NVIDIA driver bug; rather, the kernel or system BIOS is breaking access to the GPU somehow before the NVIDIA driver is even loaded. This manifests itself as a "this GPU is unsupported" error but the real problem is that the GPU is simply accessible to the driver.

[napopa]

Hi,
thanks for your answer.

Flipping through the comments on the bug report on launchpad i found a link to this topic

http://www.nvnews.net/vbulletin/showthread.php?t=134271

it looks like the same problem I'm having here but i have different values for the memory addresses

(...)
[virtual] Memory at f8000000 (32-bit, non-prefetchable) [size=16M]
Memory at d0000000 (64-bit, prefetchable) [size=256M]
Memory at f6000000 (64-bit, non-prefetchable) [size=32M]
[virtual] Expansion ROM at f9fe0000 [disabled] [size=128K]
(...)

applying the same solution proposed in that topic allows me to load the module, but if I try to start the X server I end up with now graphics and I cant go back to the console.

Should I set another memory address via the setpci command given my situation ?

EDIT:
I think i figured out that the command I should give is

setpci -s 03:00.0 BASE_ADDRESS_0=0xf8000000

with this command the [virtual] in the previous output goes away for the first line, but not for the Expansion ROM.
the results modprobing the nvidia modules are the same as before. the module gets loaded now but I lose all the graphics and i have to reboot the machine via ssh.