12:44 a.m.
For fun(?) I decided to test those NVIDIA drivers against the latest
rawhide kernel and sundry other apps. Surprise! no worky.
Anyway, that's not this lists problem and I am spamming over to the
nvidia mailing list for guidance, but there was some interesting
behaviour of the system when the module was loaded. Here are some of
the highlights from /var/log/messages:
Pretty Standard here:
Apr 4 22:31:34 localhost init: Switching to runlevel: 5
Apr 4 22:31:37 localhost kernel: nvidia: module license 'NVIDIA' taints
kernel.
Apr 4 22:31:37 localhost kernel: ACPI: PCI interrupt 0000:01:00.0[A] ->
GSI 11 (level, low) -> IRQ 11
Apr 4 22:31:37 localhost kernel: NVRM: loading NVIDIA Linux x86 NVIDIA
Kernel Module 1.0-7174 Tue Mar 22 06:44:39 PST 2005
Then the fun begins. The following three messages appear repeatedly:
Apr 4 22:31:38 localhost kernel: agpgart: Found an AGP 2.0 compliant
device at 0000:00:00.0.
Apr 4 22:31:38 localhost kernel: agpgart: Putting AGP V2 device at
0000:00:00.0 into 4x mode
Apr 4 22:31:38 localhost kernel: agpgart: Putting AGP V2 device at
0000:01:00.0 into 4x mode
Sprinkled with messages about gdm loosing its connection to X(not
surprising if X is having serious issues):
Apr 4 22:31:44 localhost gdm[4251]: gdm_slave_xioerror_handler: Fatal X
error - Restarting :0
Then the final message of Doom:
Apr 4 22:32:29 localhost gdm[4139]: Failed to start X server several
times in a short time period; disabling display :0
Anyhow, I'm running this on a Dell Latitude D800 with an Nvidia GeForce4
Ti 4200 Go AGP 8x. Don't know if anyone on this list really cares at
this point, and I don't think that anyone at Nvidia is going to give a
rip either. Just thought you all might be curious!
Sean
1:22 a.m.
I have the same problem, and cant find a solution to the problem.
I find it strange that every new Fedora release brakes the Nvidia
drivers.
Just a thought.
/Bjorn
man, 04 04 2005 kl. 22:44 -0700, skrev Sean Bruno:
For fun(?) I decided to test those NVIDIA drivers against the latest
rawhide kernel and sundry other apps. Surprise! no worky.
Anyway, that's not this lists problem and I am spamming over to the
nvidia mailing list for guidance, but there was some interesting
behaviour of the system when the module was loaded. Here are some of
the highlights from /var/log/messages:
Pretty Standard here:
Apr 4 22:31:34 localhost init: Switching to runlevel: 5
Apr 4 22:31:37 localhost kernel: nvidia: module license 'NVIDIA' taints
kernel.
Apr 4 22:31:37 localhost kernel: ACPI: PCI interrupt 0000:01:00.0[A] ->
GSI 11 (level, low) -> IRQ 11
Apr 4 22:31:37 localhost kernel: NVRM: loading NVIDIA Linux x86 NVIDIA
Kernel Module 1.0-7174 Tue Mar 22 06:44:39 PST 2005
Then the fun begins. The following three messages appear repeatedly:
Apr 4 22:31:38 localhost kernel: agpgart: Found an AGP 2.0 compliant
device at 0000:00:00.0.
Apr 4 22:31:38 localhost kernel: agpgart: Putting AGP V2 device at
0000:00:00.0 into 4x mode
Apr 4 22:31:38 localhost kernel: agpgart: Putting AGP V2 device at
0000:01:00.0 into 4x mode
Sprinkled with messages about gdm loosing its connection to X(not
surprising if X is having serious issues):
Apr 4 22:31:44 localhost gdm[4251]: gdm_slave_xioerror_handler: Fatal X
error - Restarting :0
Then the final message of Doom:
Apr 4 22:32:29 localhost gdm[4139]: Failed to start X server several
times in a short time period; disabling display :0
Anyhow, I'm running this on a Dell Latitude D800 with an Nvidia GeForce4
Ti 4200 Go AGP 8x. Don't know if anyone on this list really cares at
this point, and I don't think that anyone at Nvidia is going to give a
rip either. Just thought you all might be curious!
Sean
10:45 a.m.
On Apr 5, 2005 2:22 AM, Bjorn Andersen <ba(a)linuxin.dk> wrote:
I have the same problem, and cant find a solution to the problem.
I find it strange that every new Fedora release brakes the Nvidia
drivers.
This is because the NVIDIA drivers are propritary and because they
don't follow Linux driver API conventions (well, the second partially
stems from the first because a big driver API convention is that your
driver be open sourced so that the interface can be painlessly
updated).
Because of this, the repeated brokeness of NVIDIA's driver is entirely
off topic for the fedora-devel list... Please bring the subject back
up here to ask for the inclusion of the driver in the distro once
nvidia has released its source under a free license.
1:26 a.m.
On Mon, 2005-04-04 at 22:44 -0700, Sean Bruno wrote:
For fun(?) I decided to test those NVIDIA drivers against the latest
rawhide kernel and sundry other apps. Surprise! no worky.
Works for me, using stock 7174 on a GeforceFX 5600XT. There aren't any
patches for the current release. I suspect the problem is on your end.
[snip]
Then the final message of Doom:
Apr 4 22:32:29 localhost gdm[4139]: Failed to start X server several
times in a short time period; disabling display :0
The last time I got this, I updated my kernel and then booted into that
kernel without going to runlevel 3 first and installing the kernel
module using the -K option (-A has it and the other advanced options).
Did you upgrade your kernel and not rebuild the drivers?
--
Aaron Kurtz <a.kurtz(a)hardsun.net> GPG Key ID: ED588CF2
1:39 a.m.
On Mon, 2005-04-04 at 23:26 -0700, Aaron Kurtz wrote:
The last time I got this, I updated my kernel and then booted into
that
kernel without going to runlevel 3 first and installing the kernel
module using the -K option (-A has it and the other advanced options).
Did you upgrade your kernel and not rebuild the drivers?
This is the first kernel that I have even bothered testing with the
Nvidia drivers(2.6.11-1.1225_FC4). What would u suggest I try to get
the module to work(if u r running this kernel)?
Sean
3:49 a.m.
On Tue, 2005-04-05 at 08:39, Sean Bruno wrote:
On Mon, 2005-04-04 at 23:26 -0700, Aaron Kurtz wrote:
> The last time I got this, I updated my kernel and then booted into that
> kernel without going to runlevel 3 first and installing the kernel
> module using the -K option (-A has it and the other advanced options).
> Did you upgrade your kernel and not rebuild the drivers?
This is the first kernel that I have even bothered testing with the
Nvidia drivers(2.6.11-1.1225_FC4). What would u suggest I try to get
the module to work(if u r running this kernel)?
I had similar trouble getting a FX5700 card up and running on FC3. What
did the trick was to put:
Option "NvAGP" "0"
in the "Device" section (this disables AGP which seems to be the
problem). I'm not sure what would be a better/more permanent solution to
the problem though - not sure how much (or even if) this affects 3D
performance. From the NVidia README:
Option "NvAGP" "integer"
Configure AGP support. Integer argument can be one of:
0 : disable agp
1 : use NVIDIA's internal AGP support, if possible
2 : use AGPGART, if possible
3 : use any agp support (try AGPGART, then NVIDIA's AGP)
Please note that NVIDIA's internal AGP support cannot
work if AGPGART is either statically compiled into your
kernel or is built as a module, but loaded into your
kernel (some distributions load AGPGART into the kernel
at boot up). Default: 3 (the default was 1 until after
1.0-1251).
Hope this helps,
--
Tarjei
5:14 a.m.
Works for me: FC4t1 Geforce 6600GT PCI-E and 7174 - x86_64.
Have you followed all xorg.conf changes in the readme?
9:35 a.m.
On Tue, 2005-04-05 at 12:14 +0200, Cimmo wrote:
Works for me: FC4t1 Geforce 6600GT PCI-E and 7174 - x86_64.
Have you followed all xorg.conf changes in the readme?
Thanks for the suggestion. I did in fact modify my xorg.conf, however I
may not have done it correctly.
I made two changes. I commented out the "dri" section and changed my
driver from "nv" to "nvidia"
In the past this has been all that was required after a build. The
module is loaded(seen through lsmod) and the X server doesn't seem to be
complaning about the module itself.
I think the issue is more along the lines of agpart attempting to do
something that it cannot. This may be a special case and only happening
on my H/W (Dell Latitude D800 -- laptop).
Sean
1:44 a.m.
New subject: The "miracle" of the nvidia drivers in action: Was "beauty?"
The "miracle" of the nvidia drivers appears to have occured on this last
batch of updates to rawhide. There were many updates(one kernel, some
xorg), so I can't really tell why it started working, but it is for me.
Anonymous thanks to whomever intentionally or unintentionally made this
work for me(sounds of Angels playing harps).
Sean
6998
days inactive
6999
days old
8 comments
6 participants
participants (6)
-
Aaron Kurtz -
Bjorn Andersen -
Cimmo -
Gregory Maxwell -
Sean Bruno -
Tarjei Knapstad