I'm browsing howto's at the moment and attempting to get the nvidia drivers
(the non-free ones) the "debian" way.

There doesn't seem to be an avalable version of nvidia-kernel-source for my
particular kernel (I was running 2.6.18-4-k7, but I just upgraded to
2.6.21-2-k7
a few moments ago. I'm using
http://home.comcast.net/~andrex/Debian-nVidia/, and
http://wiki.debian.org/NvidiaGraphicsDrivers as reference documents.

There is also no nvidia-glx in testing. There has been some discussion on
the list recently regarding some updates to X.org video drivers, which have
unfortunately caused my existing nvidia setup to no longer work, so
currently I am using the "nv" driver. From the instructions it would seem my
current card (Geforce FX 5200) should be supported by the "non-legacy" or
regular driver.

According to the testing status page (
http://bjorn.haxx.se/debian/testing....age=nvidia-glx) it would seem
that testing is waiting for the newer driver package, which seems to be ATM
availalbe in sid/unstable.

Question - is it "safe" to retry the nvidia driver at this point? Last I
tried, I ended up with a fairly unusable system and had to renstall most of
X and go back to using the nv driver.

If that is doable, I figure it would be better to do this the "debian way"
and although I posted about this before, I probably would want to go ahead
and add unstable sources to my sources.list and install that way.

Or, I could wait until these are available in testing, but I don't have a
clue low long that would take. (In a previous thread, it was opined that it
would only take a few days or so.)

Hints?

thanks.

On Sat July 28 2007 14:38, David Fox wrote:
> I'm browsing howto's at the moment and attempting to get the nvidia drivers
> (the non-free ones) the "debian" way.
>
> There doesn't seem to be an avalable version of nvidia-kernel-source for my
> particular kernel (I was running 2.6.18-4-k7, but I just upgraded to
> 2.6.21-2-k7
> a few moments ago. I'm using
> http://home.comcast.net/~andrex/Debian-nVidia/, and
> http://wiki.debian.org/NvidiaGraphicsDrivers as reference documents.
>
> There is also no nvidia-glx in testing. There has been some discussion on
> the list recently regarding some updates to X.org video drivers, which have
> unfortunately caused my existing nvidia setup to no longer work, so
> currently I am using the "nv" driver. From the instructions it would seem
> my current card (Geforce FX 5200) should be supported by the "non-legacy"
> or regular driver.
>
> According to the testing status page (
> http://bjorn.haxx.se/debian/testing....age=nvidia-glx) it would seem
> that testing is waiting for the newer driver package, which seems to be ATM
> availalbe in sid/unstable.
>
> Question - is it "safe" to retry the nvidia driver at this point? Last I
> tried, I ended up with a fairly unusable system and had to renstall most of
> X and go back to using the nv driver.
>
> If that is doable, I figure it would be better to do this the "debian way"
> and although I posted about this before, I probably would want to go ahead
> and add unstable sources to my sources.list and install that way.
>
> Or, I could wait until these are available in testing, but I don't have a
> clue low long that would take. (In a previous thread, it was opined that it
> would only take a few days or so.)
>
> Hints?

I don't know if this is "the debian way" but this is how I do it. I am running
testing with kernel-2.6.22-2-amd64 at the moment but have also run the 2.6.21
kernel until recently along with the nvidia drivers without any issues at
all. There are no pre-built kernel modules in testing or unstable, I build
them with module-assistant.

module-assistant prepare or m-a prepare
module-assistant auto-install nvidia or m-a a-i nvidia


Also install nvidia-glx and nvidia-glx-ia32 if appropriate.
Run "dpkg-reconfigure xserver-xorg" and choose the nvidia driver and set the
rest of it up to your liking. If all goes well (and I'm sure it will) restart
x and you should be good to go..


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sat July 28 2007 14:38, David Fox wrote:
> I'm browsing howto's at the moment and attempting to get the nvidia drivers
> (the non-free ones) the "debian" way.
>
> There doesn't seem to be an avalable version of nvidia-kernel-source for my
> particular kernel (I was running 2.6.18-4-k7, but I just upgraded to
> 2.6.21-2-k7

I might suggest that you install the 2.6.22 kernel from unstable. I had
problems building the kernel module for k7 because of a GPL-only export that
isn't an issue with 2.6.22 kernels. You can use the 2.6.21 kernel but you
will need to build your own with paravirt_ops disabled.

> a few moments ago. I'm using
> http://home.comcast.net/~andrex/Debian-nVidia/, and
> http://wiki.debian.org/NvidiaGraphicsDrivers as reference documents.
>
> There is also no nvidia-glx in testing.

They are in testing, I'm using them now..


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

David Fox wrote:
> I'm browsing howto's at the moment and attempting to get the nvidia
> drivers (the non-free ones) the "debian" way.
>
> There doesn't seem to be an avalable version of nvidia-kernel-source
> for my particular kernel (I was running 2.6.18-4-k7, but I just
> upgraded to 2.6.21-2-k7
> a few moments ago. I'm using
> http://home.comcast.net/~andrex/Debian-nVidia/
> <http://home.comcast.net/%7Eandrex/Debian-nVidia/>, and
> http://wiki.debian.org/NvidiaGraphicsDrivers as reference documents.
>
> There is also no nvidia-glx in testing. There has been some discussion
> on the list recently regarding some updates to X.org <http://X.org>
> video drivers, which have unfortunately caused my existing nvidia
> setup to no longer work, so currently I am using the "nv" driver. From
> the instructions it would seem my current card (Geforce FX 5200)
> should be supported by the "non-legacy" or regular driver.
>
> According to the testing status page
> (http://bjorn.haxx.se/debian/testing....age=nvidia-glx) it would
> seem that testing is waiting for the newer driver package, which seems
> to be ATM availalbe in sid/unstable.
>
> Question - is it "safe" to retry the nvidia driver at this point? Last
> I tried, I ended up with a fairly unusable system and had to renstall
> most of X and go back to using the nv driver.
>
> If that is doable, I figure it would be better to do this the "debian
> way" and although I posted about this before, I probably would want to
> go ahead and add unstable sources to my sources.list and install that way.
>
> Or, I could wait until these are available in testing, but I don't
> have a clue low long that would take. (In a previous thread, it was
> opined that it would only take a few days or so.)
>
> Hints?
>
> thanks.
>
>
>
>

Yes, the current kernel will not let you compile the module for the
nvidia drivers. There is a patched driver out that you can use that
will work that gets around this problem:
http://forums.debian.net/viewtopic.php?p=88132#88132

This is what happens when you try to run the official nvidia driver
script: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=432746

Sam

On 2007-07-28 15:10:31 -0700, Alan Ianson wrote:
> I might suggest that you install the 2.6.22 kernel from unstable. I
> had problems building the kernel module for k7 because of a GPL-only
> export that isn't an issue with 2.6.22 kernels.

But for another reason, it doesn't work with 2.6.22:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=434379

--
Vincent Lefèvre <vincent@vinc17.org> - Web: <http://www.vinc17.org/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sat, Jul 28, 2007 at 02:38:44PM -0700, David Fox wrote:
> I'm browsing howto's at the moment and attempting to get the nvidia drivers
> (the non-free ones) the "debian" way.

> According to the testing status page (
> http://bjorn.haxx.se/debian/testing....age=nvidia-glx) it would seem
> that testing is waiting for the newer driver package, which seems to be ATM
> availalbe in sid/unstable.
>

> Or, I could wait until these are available in testing, but I don't have a
> clue low long that would take. (In a previous thread, it was opined that it
> would only take a few days or so.)

IIRC, things go into testing if they haven't changed in Sid for 10 days.

Difficulty with nvidia is the main reason that I'm sticking with Etch.

Good luck,

Doug.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sat July 28 2007 16:35, Vincent Lefevre wrote:
> On 2007-07-28 15:10:31 -0700, Alan Ianson wrote:
> > I might suggest that you install the 2.6.22 kernel from unstable. I
> > had problems building the kernel module for k7 because of a GPL-only
> > export that isn't an issue with 2.6.22 kernels.
>
> But for another reason, it doesn't work with 2.6.22:
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=434379

That's a bummer. I've had that problem before having on older version of gcc
and not a newer one. I managed to get it built by installing a newer gcc, not
to sure how one would get past this one.

Being on amd64 I have safely bypassed all of these issues somehow.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sat July 28 2007 15:45, Douglas Allan Tutty wrote:

> Difficulty with nvidia is the main reason that I'm sticking with Etch.

Are you running an amd64 there? I have had no trouble on the amd64 with the
nvidia drivers. There a couple issues that I know of for the i386 but amd64
has been painless..


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sat, Jul 28, 2007 at 05:23:26PM -0700, Alan Ianson wrote:
> On Sat July 28 2007 15:45, Douglas Allan Tutty wrote:
>
> > Difficulty with nvidia is the main reason that I'm sticking with Etch.
>
> Are you running an amd64 there? I have had no trouble on the amd64 with the
> nvidia drivers. There a couple issues that I know of for the i386 but amd64
> has been painless..
>

Yes I'm running amd64. However, since I'm on slow dialup and the amd64
box is the only one with a CD burner and lots of drive space, it would
be a major headache to have something happen that caused stuff to not
work. My other boxes are a 486 and a PII and etch stuff needs so much
memory that using them to 'rescue' the amd64 is painful. I've tried it
as part of verifying my backup strategy but its difficult.

So, in effect, my Athlon box for me is mission-critical and I'll stick
with stable now that its etch.

Doug.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Douglas Allan Tutty wrote:
> On Sat, Jul 28, 2007 at 05:23:26PM -0700, Alan Ianson wrote:
>> On Sat July 28 2007 15:45, Douglas Allan Tutty wrote:
>>
>>> Difficulty with nvidia is the main reason that I'm sticking with Etch.
>> Are you running an amd64 there? I have had no trouble on the amd64 with the
>> nvidia drivers. There a couple issues that I know of for the i386 but amd64
>> has been painless..
>>
>
> Yes I'm running amd64. However, since I'm on slow dialup and the amd64
> box is the only one with a CD burner and lots of drive space, it would
> be a major headache to have something happen that caused stuff to not
> work.


But what could happen if you tried it on a duplicate partition? So it
doesn't work. You go back to the working partition.



My other boxes are a 486 and a PII and etch stuff needs so much
> memory that using them to 'rescue' the amd64 is painful. I've tried it
> as part of verifying my backup strategy but its difficult.
>
> So, in effect, my Athlon box for me is mission-critical and I'll stick
> with stable now that its etch.
>


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Sun, Jul 29, 2007 at 06:19:47AM -0500, Hugo Vanwoerkom wrote:
>
> >Yes I'm running amd64. However, since I'm on slow dialup and the amd64
> >box is the only one with a CD burner and lots of drive space, it would
> >be a major headache to have something happen that caused stuff to not
> >work.
>
> But what could happen if you tried it on a duplicate partition? So it
> doesn't work. You go back to the working partition.
>

The box uses LVM so I can't give Lenny its own actual disk
partitions. Also, downloading a whole new installation is non-trivial
at 28.8 Kb/s (sometimes 14.4, sometimes 9600). I'm many miles from the
telephone exchange. It took a few days to install an ia32 chroot with
just base, mc, lynx, and iceweasel.

Doug.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Douglas Allan Tutty wrote:
> On Sun, Jul 29, 2007 at 06:19:47AM -0500, Hugo Vanwoerkom wrote:
> >
>>> Yes I'm running amd64. However, since I'm on slow dialup and the amd64
>>> box is the only one with a CD burner and lots of drive space, it would
>>> be a major headache to have something happen that caused stuff to not
>>> work.
>> But what could happen if you tried it on a duplicate partition? So it
>> doesn't work. You go back to the working partition.
>>
>
> The box uses LVM so I can't give Lenny its own actual disk
> partitions. Also, downloading a whole new installation is non-trivial
> at 28.8 Kb/s (sometimes 14.4, sometimes 9600). I'm many miles from the
> telephone exchange. It took a few days to install an ia32 chroot with
> just base, mc, lynx, and iceweasel.
>

True. I forgot: you stated before that you were using LVM. I am running
on a dialup modem myself and the last Sid upgrade took 2.5 days.

Hugo


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Alan Lanson wrote:

module-assistant prepare or m-a prepare
module-assistant auto-install nvidia or m-a a-i nvidia


OK I tried that, it fails with the message that there's no
nvidia-kernel-source available.

On 7/28/07, David Fox <dfox94085@gmail.com> wrote:
>
>
>
> Hints?


OK well, running the patched driver available was a good suggestion, but it
fails miserably on my
system.

here's the attached error log

David Fox wrote:
>
>
> On 7/28/07, *David Fox* <dfox94085@gmail.com
> <mailto:dfox94085@gmail.com>> wrote:
>
>
>
> Hints?
>
>
> OK well, running the patched driver available was a good suggestion,
> but it fails miserably on my
> system.
>
> here's the attached error log
>

Looks like you need to remove the old driver first.

I don't know if that is a debian version driver or not. I could not
find any info on the driver version 1.0-9755 that you have installed.
If it is a debian driver, do an "aptitude remove nvidia-glx" and make
sure all it's dependant packages get removed too:
http://packages.debian.org/stable/x11/nvidia-glx


The error you are getting has to do with the old driver installed:/

-> Uninstalling NVIDIA Accelerated Graphics Driver for Linux-x86 (1.0-9755):
WARNING: Unable to restore symbolic link /usr/lib/libGL.so.1 -> libGL.so.1.2
(File exists).
ERROR: Unable to create

'/usr/lib/dosemu/drive_z/tmp/selfgz20192/NVIDIA-Linux-x86-1.0-9755-pkg1/
usr/lib/libGL.so.1.0.9755' for copying (No such file or directory)
WARNING: Unable to restore file

'/usr/lib/dosemu/drive_z/tmp/selfgz20192/NVIDIA-Linux-x86-1.0-9755-pkg
1/usr/lib/libGL.so.1.0.9755'.
ERROR: Unable to create

'/usr/lib/dosemu/drive_z/tmp/selfgz20192/NVIDIA-Linux-x86-1.0-9755-pkg1/
usr/lib/libnvidia-cfg.so.1.0.9755' for copying (No such file or
directory)/




Sam

On 7/30/07, Sam Leon <leon.mailinglist.36@gmail.com> wrote:


> I don't know if that is a debian version driver or not. I could not find
> any info on the driver version 1.0-9755 that you have installed. If it is
> a debian driver, do an "aptitude remove nvidia-glx" and make sure all it's
> dependant packages get removed too:
> http://packages.debian.org/stable/x11/nvidia-glx
>

apt-cache pollicy says there's no installation candidate for
nvidia-glx. Version
1.0.9755 was the nvidia driver that I was using when I was
running Etch, back in May.



The error you are getting has to do with the old driver installed:*
>
> -> Uninstalling NVIDIA Accelerated Graphics Driver for Linux-x86 (1.0-9755
> ):
> WARNING: Unable to restore symbolic link /usr/lib/libGL.so.1 ->
> libGL.so.1.2
> (File exists).
> ERROR: Unable to create
> '/usr/lib/dosemu/drive_z/tmp/selfgz20192/NVIDIA-
> Linux-x86-1.0-9755-pkg1/
> usr/lib/libGL.so.1.0.9755' for copying (No such file or directory)*
>

I don't even have a libGL.so.1 file nlow (and amarok needs it).

I can still run X, barely. (no GL at all).

I just don't understand why this thing won't work. And especially, why is it
trying to place packages in/usr/lib/dosemu?


Sam

On Mon July 30 2007 19:27, David Fox wrote:
> Alan Lanson wrote:
>
> module-assistant prepare or m-a prepare
> module-assistant auto-install nvidia or m-a a-i nvidia
>
>
> OK I tried that, it fails with the message that there's no
> nvidia-kernel-source available.

You will need to have non-free (and probably contrib) along with main in your
main debian archive lines in /etc/apt/sources.list.

Reading into the thread a bit you may have some problems getting the module
built with a 2.6.21 kernel (unless you build your own with config
paravirt_ops diabled). It can be done but it may take an extra push.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On 7/30/07, Alan Ianson <agianson@gmail.com> wrote:
>
>
>
> You will need to have non-free (and probably contrib) along with main in
> your
> main debian archive lines in /etc/apt/sources.list.


I already checked that - I do have non-free and contrib enabled.


Reading into the thread a bit you may have some problems getting the module
> built with a 2.6.21 kernel (unless you build your own with config
> paravirt_ops diabled). It can be done but it may take an extra push.



Supposedly that's what the patehed version of the 100.14.11 driver solves.

On Mon July 30 2007 22:00, David Fox wrote:
> On 7/30/07, Alan Ianson <agianson@gmail.com> wrote:
> > You will need to have non-free (and probably contrib) along with main in
> > your
> > main debian archive lines in /etc/apt/sources.list.
>
> I already checked that - I do have non-free and contrib enabled.

Have you run apt-get update, and is your mirror up to date? I just switched to
2.6.21 from 2.6.22 an hour ago and had no problem with the same version of
the nvidia-kernel-source.

> Reading into the thread a bit you may have some problems getting the module
>
> > built with a 2.6.21 kernel (unless you build your own with config
> > paravirt_ops diabled). It can be done but it may take an extra push.
>
> Supposedly that's what the patehed version of the 100.14.11 driver solves.

Good. I haven't needed that since I haven't had problems with the GPL-only
export on amd64 for reasons unknown. Lucky in this case I guess.

I struggled when etch was testing..


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On 7/30/07, Alan Ianson <agianson@gmail.com> wrote:
>
>
>
> Have you run apt-get update, and is your mirror up to date? I just
> switched to
> 2.6.21 from 2.6.22 an hour ago and had no problem with the same version of
> the nvidia-kernel-source.


Sure have. I'm using ftp.debian.org
, and I just went there and found under debian/pool/n
this stuff: I ran apt-get update (or rather aptitude update) yesterday, and
updated a few things in the
process.

Looking at a debian ftp site that I use (#1 listing in my sources.list) I
see:


debian/pool/non-free/n > ls -l nvi*
drwxr-sr-x 2 1176 1176 4096 Jul 7 08:35 nvidia-graphics-drivers
drwxr-sr-x 2 1176 1176 4096 Dec 7 2006 nvidia-graphics-drivers-legacy
drwxr-xr-x 2 1176 1176 4096 Jun 21 20:34 nvidia-graphics-drivers-legacy-71xx
drwxr-xr-x 2 1176 1176 4096 Jun 21 20:34 nvidia-graphics-drivers-legacy-96xx
drwxr-xr-x 2 1176 1176 4096 Jun 22 08:42
nvidia-graphics-legacy-71xx-modules-amd64
drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
nvidia-graphics-legacy-71xx-modules-i386
drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
nvidia-graphics-legacy-96xx-modules-amd64
drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
nvidia-graphics-legacy-96xx-modules-i386
drwxr-xr-x 2 1176 1176 4096 Jun 27 08:34
nvidia-graphics-legacy-modules-amd64
drwxr-sr-x 2 1176 1176 4096 Jun 27 08:34 nvidia-graphics-legacy-modules-i386
drwxr-xr-x 2 1176 1176 4096 Jun 27 08:34 nvidia-graphics-modules-amd64
drwxr-sr-x 2 1176 1176 4096 Jun 27 08:34 nvidia-graphics-modules-i386
drwxr-sr-x 2 1176 1176 4096 Nov 21 2006 nvidia-modules-i386

Note there is no "nvidia-kernel-source".

Poking around with various 'apt-cache policy' commands on some of those
files, all I get are "unable to locate package
such and such" errors. I've tried nvidia-graphics-drivers,
nvidia-graphics-legacy-96xx.

If it is in 'pool" that doesn't necessarily mean that these files are in
"testing", does it? OTOH, look at the file dates.

The "why is package X not in testing" page does have some info - says there
is no new version in testing, and is trying to add, not update.

On Tue July 31 2007 18:17, David Fox wrote:

> Looking at a debian ftp site that I use (#1 listing in my sources.list) I
> see:
>
>
> debian/pool/non-free/n > ls -l nvi*
> drwxr-sr-x 2 1176 1176 4096 Jul 7 08:35 nvidia-graphics-drivers
> drwxr-sr-x 2 1176 1176 4096 Dec 7 2006 nvidia-graphics-drivers-legacy
> drwxr-xr-x 2 1176 1176 4096 Jun 21 20:34
> nvidia-graphics-drivers-legacy-71xx drwxr-xr-x 2 1176 1176 4096 Jun 21
> 20:34 nvidia-graphics-drivers-legacy-96xx drwxr-xr-x 2 1176 1176 4096 Jun
> 22 08:42
> nvidia-graphics-legacy-71xx-modules-amd64
> drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
> nvidia-graphics-legacy-71xx-modules-i386
> drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
> nvidia-graphics-legacy-96xx-modules-amd64
> drwxr-xr-x 2 1176 1176 4096 Jun 25 20:30
> nvidia-graphics-legacy-96xx-modules-i386
> drwxr-xr-x 2 1176 1176 4096 Jun 27 08:34
> nvidia-graphics-legacy-modules-amd64
> drwxr-sr-x 2 1176 1176 4096 Jun 27 08:34
> nvidia-graphics-legacy-modules-i386 drwxr-xr-x 2 1176 1176 4096 Jun 27
> 08:34 nvidia-graphics-modules-amd64 drwxr-sr-x 2 1176 1176 4096 Jun 27
> 08:34 nvidia-graphics-modules-i386 drwxr-sr-x 2 1176 1176 4096 Nov 21 2006
> nvidia-modules-i386
>
> Note there is no "nvidia-kernel-source".

Your right. I just had a look at packages.debian.org and had a look for it. It
isn't in testing at the moment. I'm not sure why.

> Poking around with various 'apt-cache policy' commands on some of those
> files, all I get are "unable to locate package
> such and such" errors. I've tried nvidia-graphics-drivers,
> nvidia-graphics-legacy-96xx.
>
> If it is in 'pool" that doesn't necessarily mean that these files are in
> "testing", does it? OTOH, look at the file dates.

That's right, they may be in stable or unstable.

> The "why is package X not in testing" page does have some info - says there
> is no new version in testing, and is trying to add, not update.

I also have unstable lines in my sources.list so that must be where I got them
from. If you don't mind doing the same that may work better for you. I always
add the following to my /etc/apt/apt.conf so I don't get stuff from unstable
unexpectedly (mind you I was unaware I got the nvidia stuff from unstable!)

-- /etc/apt/apt.conf --
APT:efault-Release "testing";

-- end --


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org