Can only Install when using onboard video

This problem is on again, off again. I can install with my pci FX 5200 on openSUSE 10.2, 11.0 but not on 10.3 or 11.1. I have to revert to the onboard video. Forget the basics here. I have tried just about everything and need help trying to capture and decipher the boot process.

Using the PCI card (ANY make or model on my Tyan 2098 mb) fails for 11.1. Dies trying to start device detection. Here is the only way I can get it to work…

  1. install using onboard video with bios set to boot with onboard as initial.
  2. install all drivers for nVidia card, change boot init vga to pci

At this point if it boots to gnome it take a LONG time and essentially is very unstable. Hitting F2 during boot shows a ton of stuff in the udev process that is throwing up. Activating the logs for udev don’t show me anything I understand to be a problem. I suppose recompiling the kernel is an option but while not a novice by a stretch, doing this is beyond me. Besides, I still cannot tell what is causing the issue. Whatever it is it’s not happening in 11.0.

If it helps at all, running sax2 in command (after posing runstate 3 at boot) to configure and use the PCI card always brings up the profile and settings for the onboard. This happens for either command (sax2 -c1, sax2 -c0). To make things even more fun, on either the onboard or the pci, I cannot get to the command screen using any of the ctrl-alt-fkeys. I get a blank screen and cannot get back to the desktop using ctrl-alt-F7.

I know this probably resides with something in the kernel, but I cannot get information to confirm or recompile if I were even brave enough…

I know the board is a little older but here is the kicker, My brother has a board that is a little newer (last 3 years) and a FX 5500 and has the same problems with 10.3… there must be something with the BIOS I suppose but there are no firmware updates and I really don’t see why I should chuck the board (and everything else) for something that was working in 10.2, and 11.0. Besides, I have 4 identical systems using these boards and video cards. Too much of an investment.

I guess I am pleading for a direction at this point. Last time (10.3) I got entry level troubleshooting questions and answers and ultimately was left hanging without a solution. I cannot be the only one having this onboard vs. pci vga boot/install problem and someone should know what was changed in the kernel or boot rules that could effect it. Especially since the team has to turn on and off common features for the shipping distro…

Anyone know what I can do to get a bead on this problem? I am about to install from scratch with 11.0 and then use the upgrade option but I am sure that will fail. I have tried every turn in the install process between 10.3 and 11.1. (failsafe, text mode, turn off automatic configuration, check for updates, install current drivers, etc…)

So yes it’s down to the onboard VGA vs. PCI but it’s leaning towards the linux vs. mb bios issues…

HELP!!! PLEASE!!!

My advice
Install using on-board
Then download this: Linux Display Driver - x86

This assumes you are 32bit.

Place the file in: /home/*username

Follow these directions:

Alright so here is how to install the nvidia driver manually, in case the one in the repo doesn’t work or u just want to use the latest.
Go to Yast>Software>Software Management
Search for and install if you don’t have these:
make
gcc
kernel-source
Now download the latest Nvidia driver:
Place the file in your /home/username
Now restart and at the boot screen, pause the boot by moving the down button, then move back up and clear any text in the boot arguments by holding backspace. Then just type the number: 3
At the login
Type “root” then enter and then your root password and press enter.
now type
cd /home/username
*Now remember you can use the {TAB} key to auto complete
so type:
sh NVIDIA{TAB}
and the whole file name should auto complete
eg: sh NVIDIA-Linux-x86-173.14.09-pkg1.run
Follow the installer and let it compile the kernel module for you.
Say Yes to everything
Use TAB to move around
reboot

OK, That didn’t work. I got an error message and here is partof the error log it produced. Doesn’t make any sense to me, hopefully it will to you…

-f /usr/src/linux-2.6.27.7-9/Makefile
modules
test -e include/linux/autoconf.h -a -e include/config/auto.conf || (
echo;
echo " ERROR: Kernel configuration is invalid.";
echo " include/linux/autoconf.h or include/config/auto.conf are mis
sing.";
echo " Run ‘make oldconfig && make prepare’ on kernel src to fix it
.";
echo;
/bin/false)
mkdir -p /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/.tmp_ver
sions ; rm -f /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/.tm
p_versions/*
make -f /usr/src/linux-2.6.27.7-9/scripts/Makefile.build obj=/tmp/selfgz3858
/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv
cc -Wp,-MD,/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/.nv.
o.d -nostdinc -isystem /usr/lib/gcc/i586-suse-linux/4.3/include -D__KERNEL_
_ -Iinclude -Iinclude2 -I/usr/src/linux-2.6.27.7-9/include -I/usr/src/linux-
2.6.27.7-9/arch/x86/include -include include/linux/autoconf.h -I/tmp/selfgz
3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv -Wall -Wundef -Wstrict-proto
types -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-funct
ion-declaration -Os -m32 -msoft-float -mregparm=3 -freg-struct-return -mpref
erred-stack-boundary=2 -march=i586 -mtune=generic -ffreestanding -DCONFIG_AS
CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare -mno-sse -mno-
mmx -mno-sse2 -mno-3dnow -I/usr/src/linux-2.6.27.7-9/include/asm-x86/mach-ge
neric -Iinclude/asm-x86/mach-generic -I/usr/src/linux-2.6.27.7-9/include/asm
-x86/mach-default -Iinclude/asm-x86/mach-default -fno-stack-protector -fomit
-frame-pointer -fasynchronous-unwind-tables -Wdeclaration-after-statement -W
no-pointer-
sign -I/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv -Wall -Wi
mplicit -Wreturn-type -Wswitch -Wformat -Wchar-subscripts -Wparentheses -Wpo
inter-arith -Wno-multichar -Werror -MD -Wsign-compare -Wno-cast-qual -Wno-er
ror -D__KERNEL
_ -DMODULE -DNVRM -DNV_VERSION_STRING=“173.14.12” -UDEBUG -
U_DEBUG -DNDEBUG -DMODULE -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR
(nv)" -D"KBUILD_MODNAME=KBUILD_STR(nvidia)" -D"DEBUG_HASH=11" -D"DEBUG_HASH
2=20" -c -o /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/.tmp_
nv.o /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv.c
In file included from /usr/src/linux-2.6.27.7-9/include/linux/bitops.h:17,
from /usr/src/linux-2.6.27.7-9/include/linux/kernel.h:15,
from /usr/src/linux-2.6.27.7-9/include/linux/sched.h:52,
from /usr/src/linux-2.6.27.7-9/include/linux/utsname.h:35,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:19,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
include2/asm/bitops.h: In function ‘set_bit’:
include2/asm/bitops.h:60: warning: pointer of type ‘void *’ used in arit
hmetic
include2/asm/bitops.h: In function ‘clear_bit’:
include2/asm/bitops.h:97: warning: pointer of type ‘void *’ used in arit
hmetic
In file included from /usr/src/linux-2.6.27.7-9/include/linux/list.h:6,
from /usr/src/linux-2.6.27.7-9/include/linux/preempt.h:11,
from /usr/src/linux-2.6.27.7-9/include/linux/spinlock.h:50,
from /usr/src/linux-2.6.27.7-9/include/linux/seqlock.h:29,
from /usr/src/linux-2.6.27.7-9/include/linux/time.h:8,
from /usr/src/linux-2.6.27.7-9/include/linux/timex.h:57,
from /usr/src/linux-2.6.27.7-9/include/linux/sched.h:54,
from /usr/src/linux-2.6.27.7-9/include/linux/utsname.h:35,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:19,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/usr/src/linux-2.6.27.7-9/include/linux/prefetch.h: In function ‘prefetch_
range’:
/usr/src/linux-2.6.27.7-9/include/linux/prefetch.h:57: warning: pointer of t
ype ‘void *’ used in arithmetic
In file included from include2/asm/mmzone.h:2,
from /usr/src/linux-2.6.27.7-9/include/linux/mmzone.h:701,
from /usr/src/linux-2.6.27.7-9/include/linux/gfp.h:4,
from /usr/src/linux-2.6.27.7-9/include/linux/slab.h:12,
from /usr/src/linux-2.6.27.7-9/include/linux/percpu.h:5,
from /usr/src/linux-2.6.27.7-9/include/linux/rcupdate.h:39,
from /usr/src/linux-2.6.27.7-9/include/linux/sem.h:81,
from /usr/src/linux-2.6.27.7-9/include/linux/sched.h:69,
from /usr/src/linux-2.6.27.7-9/include/linux/utsname.h:35,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:19,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
include2/asm/mmzone_32.h: In function ‘pfn_valid’:
include2/asm/mmzone_32.h:89: warning: comparison between signed and unsigned
In file included from /usr/src/linux-2.6.27.7-9/include/linux/utsname.h:35,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:19,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/usr/src/linux-2.6.27.7-9/include/linux/sched.h: In function ‘object_is_on
stack’:
/usr/src/linux-2.6.27.7-9/include/linux/sched.h:1989: warning: pointer of ty
pe ‘void *’ used in arithmetic
In file included from include2/asm/dma-mapping.h:9,
from /usr/src/linux-2.6.27.7-9/include/linux/dma-mapping.h:
52,
from /usr/src/linux-2.6.27.7-9/include/asm-generic/pci-dma-
compat.h:7,
from include2/asm/pci.h:96,
from /usr/src/linux-2.6.27.7-9/include/linux/pci.h:1009,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:86,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/usr/src/linux-2.6.27.7-9/include/linux/scatterlist.h: In function ‘sg_vir
t’:
/usr/src/linux-2.6.27.7-9/include/linux/scatterlist.h:199: warning: pointer
of type ‘void *’ used in arithmetic
In file included from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv-linux.h:107:27
: error: asm/semaphore.h: No such file or directory
In file included from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv-linux.h:109,
from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h: In function ‘zero_user

segments’:
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:138: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:138: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:138: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:138: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:141: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:141: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:141: warning: pointer of t
ype ‘void *’ used in arithmetic
/usr/src/linux-2.6.27.7-9/include/linux/highmem.h:141: warning: pointer of t
ype ‘void *’ used in arithmetic
In file included from /tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv/nv.c:14:
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv-linux.h: In fu
nction ‘nv_execute_on_all_cpus’:
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv-linux.h:674: e
rror: too many arguments to function ‘on_each_cpu’
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv.c: In function
‘nv_kern_cpu_callback’:
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv.c:1299: error:
too many arguments to function ‘smp_call_function’
/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv.c:1306: error:
too many arguments to function ‘smp_call_function’
make[4]: *** [/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/src/nv/nv.
o] Error 1
make[3]: *** [module/tmp/selfgz3858/NVIDIA-Linux-x86-173.14.12-pkg1/usr/sr
c/nv] Error 2
make[2]: *** [sub-make] Error 2
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[1]: *** [module] Error 1
make: *** [module] Error 2
-> Error.
ERROR: Unable to build the NVIDIA kernel module.
ERROR: Installation has failed. Please see the file
‘/var/log/nvidia-installer.log’ for details. You may find suggestions
on fixing installation problems in the README available on the Linux
driver download page at Welcome to NVIDIA - World Leader in Visual Computing Technologies.

Make sure you have the matching kernel-source for the kernel.

But it looks not so good from what I read in a google search. Try some different drivers from here.
NVIDIA Driver Downloads - Advanced Search

There are plenty to go at.

If you have to switch off on-board in BIOS, do so before install.