NVidia 6800 Ultra, Solaris 10 x64,1.0-7676 ... guess the driver doesn't work

[rednoise]

I have done all the things that seem necessary, install the driver,
pciscan then update_drv -a -i '"pci10de,040"' nvidia to get nvidia "pci10de,040"
in /etc/driver_aliases, made sure the xorg server was selected, edit xorg.conf,
reboot with reconfigure flag set etc... just won't go... Module doesn't load:

# dmesg|grep -i nvidia
Oct 7 15:15:56 Solaris10-x64 nvidia: [ID 702911 kern.notice] NOTICE: NVRM: loading NVIDIA SunOS x86_64 NVIDIA Kernel Module Fri Jul 29 13:38:55 PDT 2005
Oct 7 15:15:56 Solaris10-x64 pseudo: [ID 129642 kern.info] pseudo-device: nvidia255
Oct 7 15:15:56 Solaris10-x64 genunix: [ID 936769 kern.info] nvidia255 is /pseudo/nvidia@255

So X won't start with nvidia driver:

# grep NV /var/log/Xorg.0.log
(**) | |-->Device "** NVIDIA (generic) [nv]"
(II) Module glx: vendor="NVIDIA Corporation"
(II) Module nvidia: vendor="NVIDIA Corporation"
(II) NVIDIA dlloader X Driver 1.0-7676 Fri Jul 29 13:25:02 PDT 2005
(II) NVIDIA Unified Driver for all NVIDIA GPUs
(--) Chipset NVIDIA GPU found
(++) NVIDIA(0): Depth 24, (--) framebuffer bpp 32
(==) NVIDIA(0): RGB weight 888
(==) NVIDIA(0): Default visual is TrueColor
(==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
(EE) NVIDIA(0): Failed to initialize the NVIDIA kernel module! Please ensure
(EE) NVIDIA(0): that there is a supported NVIDIA GPU in this system, and
(EE) NVIDIA(0): that the NVIDIA device files have been created properly.
(EE) NVIDIA(0): Please consult the NVIDIA README for details.
(EE) NVIDIA(0): *** Aborting ***

and it seems a few devices are missing:

# ls -l /dev/nv* /dev/fbs/nv*
/dev/fbs/nv*: No such file or directory
lrwxrwxrwx 1 root other 11 Oct 7 15:04 /dev/nvidia0 -> fbs/nvidia0
lrwxrwxrwx 1 root other 38 Oct 7 15:04 /dev/nvidiactl -> ../devices/pseudo/nvidia@255:nvidiactl

wonder if I should tinker with mknod... memories. been a while for me using solaris

bash-3.00$ ls -l /dev/fbs
total 2
lrwxrwxrwx 1 root root 51 Oct 4 14:06 text-0 -> ../../devices/pci@0,0/pci10de,e2@b/display@0:text-0
bash-3.00$ ls -l /devices/pci\@0\,0/pci10de\,e2\@b/dis*
crw------- 1 rednoise other 99, 0 Oct 4 14:15 /devices/pci@0,0/pci10de,e2@b/display@0:text-0

/devices/pci@0,0/pci10de,e2@b/display@0:
total 0
bash-3.00$

So which comes first, the module or the device ?
Is the device not being made because the module fails, or does the module fail because the device wasn't made ?

[casevh]

Try using

update_drv -a -i '"pci10de,40"' nvidia

instead. Note the missing leading zero - "40" instead of "040".

[rednoise]

hmm... worth a try I guess

the actual pci scan address was 0x0040 - why I used 040

I'll give it a try and update Thanks for suggestion

[catalan]

You need to avoid any leading zeros in the command:

update_drv -a -i '"pci10de,040"' nvidia

should read

update_drv -a -i '"pci10de,40"' nvidia

I just installed the new driver today on a Dell D800 an it is working.

Alan

[rednoise]

Hey thanks a ton guys, it worked !

I guess I was just under the impression that the 0x0 was what was being removed
in the update command