Re: Put order on your sound modules on debian

[Thierry Chatelet]

?


-- 
To UNSUBSCRIBE, email to debian-amd64-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201104191505.15366.tchate...@free.fr

Put order on your sound modules on debian

[Marian Corcodel]

On debian without pulseaudio program on everry restart sound card not
assigned correctly.Follow configurations prevent this issue:
Edit and add follow line on /etc/modprobe.d/blacklist.conf as follow ;
echo -e "blacklist first-card\nblacklist second-card\nblacklist third-card "
>> /etc/modprobe.d/blacklist.conf
echo -e "modprobe first-card\nmodprobe second-card\nmodprobe third-card" >>
/etc/rc.local
and put the last one "exit 0".

Re: udev problem Was: kernel modules not loading at boot after upgrade

[Ernesto Domato]

On Fri, Apr 8, 2011 at 16:27, Ernesto Domato  wrote:
> Ok. I downgrade the "udev" package (and "libudev0" dependency) from
> 167-1 to 166-1 and it solved the problem so I guess my behavior could
> come from it.
>
> Looking at the logs it says, when I upgraded to 167-1, "converting old
> udev database" so maybe the problem is on the database conversion.
>

Well, I found the answers here:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=621036

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=620995

Thanks.
Ernesto


-- 
To UNSUBSCRIBE, email to debian-amd64-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/BANLkTi=mqu=7yhnixfx0t_ndr_fuoxg...@mail.gmail.com

udev problem Was: kernel modules not loading at boot after upgrade

[Ernesto Domato]

On Thu, Apr 7, 2011 at 18:34, Ernesto Domato  wrote:
> Hi, I'm using Debian AMD64 Sid/Experimental on a laptop and after
> upgrading it yesterday my touchpad, sound and battery applet stop
> working. I found that the kernel is not autoloading the respective
> modules (psmouse, snd-hda-intel and battery) on boot and even placing
> them on /etc/modules doesn't solve the problem.
>
> The only solution is to load the modules by hand (using modprobe)
> after the system finish loading but I'd like it to behave as before
> :-)
>

Ok. I downgrade the "udev" package (and "libudev0" dependency) from
167-1 to 166-1 and it solved the problem so I guess my behavior could
come from it.

Looking at the logs it says, when I upgraded to 167-1, "converting old
udev database" so maybe the problem is on the database conversion.

Thanks.
Ernesto


-- 
To UNSUBSCRIBE, email to debian-amd64-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/banlktikewhbfh5m+2ysbzbe53_bnalp...@mail.gmail.com

kernel modules not loading at boot after upgrade

[Ernesto Domato]

Hi, I'm using Debian AMD64 Sid/Experimental on a laptop and after
upgrading it yesterday my touchpad, sound and battery applet stop
working. I found that the kernel is not autoloading the respective
modules (psmouse, snd-hda-intel and battery) on boot and even placing
them on /etc/modules doesn't solve the problem.

The only solution is to load the modules by hand (using modprobe)
after the system finish loading but I'd like it to behave as before
:-)

Thanks in advance.
Ernesto


-- 
To UNSUBSCRIBE, email to debian-amd64-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/BANLkTi=we7uq3zmhtyehymzth8t+n67...@mail.gmail.com

New kernel-wedge uploaded, time to upload linux-{kernel,modules}-di

[Otavio Salvador]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hello,

Latest kernel-wedge has been uploaded today to sid and we should start
the updating the installer to 2.6.25 kernel. I have uploaded the
packages for amd64 and i386 and would like to ask porters to handle it
for your pet architecture(s).

Thanks in advance,

- -- 
O T A V I OS A L V A D O R
- -
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
- -
"Microsoft sells you Windows ... Linux gives
 you the whole house."
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iEYEARECAAYFAkhr4t4ACgkQLqiZQEml+FXJCgCeIsUxFXGg5iTcax/NJDLICSET
LdoAoLVvmIc+gnVj08gmV9tD9v5wSwXj
=GTIv
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automatically build and install kernel-modules

[Lennart Sorensen]

On Wed, Jun 25, 2008 at 09:11:39AM +1000, Dean Hamstead wrote:
> ideally the nvidia module should come precompiled

Yeah, we are trying to get that working.

> if the nvidia build fails (normal with new kernels) newbies will be 
> stumped regardless of where it happens

Yep, always an annoying problem with out of kernel drivers.

Really though every new stable release should include all the modules
precompiled for that kernel, which would hopefully help new users a lot.
New users ought not be running testing or unstable of course.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automatically build and install kernel-modules

[Hans-J. Ullrich]

Am Mittwoch, 25. Juni 2008 schrieb Dean Hamstead:
> ideally the nvidia module should come precompiled
>
> if the nvidia build fails (normal with new kernels) newbies will be
> stumped regardless of where it happens
>
> Dean
>
> Lennart Sorensen wrote:
> > On Tue, Jun 24, 2008 at 05:03:51PM +0200, Hans-J. Ullrich wrote:
> >> althoug this is not quite debian-amd-related, I am looking fo a way, to
> >> automatically build and install kernel-modules, which must normally
> >> build by hand and, which got sources in the repository. These are i.e.
> >> madwifi, nvidia-kernel and some other.
> >>
> >> Why not manually ? Well, the problem is, when a new kernel-version was
> >> installed, and the required modules are not build, it might make the
> >> system unusable (i.e. wireless networking will not work any more, or
> >> Xwon't start)
> >>
> >> This is a problem for newbies (and other users with less technical
> >> knowledge).
> >>
> >> So my idea: Change the kernel-installer scripts, so that they build and
> >> install the modules at a new kernel (the script might compare between
> >> "before update" and "installation version")  and then build the required
> >> modules out of a list using module-assistant commandline-tool (or m-a
> >> -i).
> >>
> >> Or miss I something and this is already working at another way ? So let
> >> me know.
> >
> > So you want Ubuntu's dkms system?
> >
> > I tried to do it with a startup script calling m-a if a certain module
> > didn't load, but that didn't always work so well so I went back to doing
> > it manually, although I only have to do it for the nvidia driver now.
> > lirc I use the ubuntu version which uses dkms and is hence automatic,
> > and for ivtv it is now in the kernel and hence no longer a problem.
>
> --
> http://fragfest.com.au

It is not only the nvidia modules, which might be the easiest way to update. 
There are often prebuilt packages in the repository. 


There are a lot of other modules.

There are other cases. For example, if you use cron-apt on laptops (which is 
required as the documentation says), it might be, that a customer/user cannot 
get a wireless access any more, because the new kernel is missing the 
madwifi-module.

This should not happpen. In this case the user even will get no chance, to 
install the missing sources.

You might say, cron-apt on laptops is no good choice. Yes, you might be right, 
because the update process was interrupted , because the user shut down the 
laptop before finish. Shutting down shut be avoided, during an upgrade 
(Windows does it that way, as far as I know).

Anyway, what I want to say is this: These things make the system for normal 
users difficult to handle and can cause problems. I am no coder, but I got 
the hope, those who create packages or are good coders might be able to easy 
change the preinstall or postinstall scripts.

If not, it is o.k., too. Thank you for your patience  and your attention !


Best regards

Hans


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automatically build and install kernel-modules

[Dean Hamstead]

ideally the nvidia module should come precompiled

if the nvidia build fails (normal with new kernels) newbies will be 
stumped regardless of where it happens


Dean

Lennart Sorensen wrote:

On Tue, Jun 24, 2008 at 05:03:51PM +0200, Hans-J. Ullrich wrote:
althoug this is not quite debian-amd-related, I am looking fo a way, to 
automatically build and install kernel-modules, which must normally build by 
hand and, which got sources in the repository. These are i.e. madwifi, 
nvidia-kernel and some other. 

Why not manually ? Well, the problem is, when a new kernel-version was 
installed, and the required modules are not build, it might make the system 
unusable (i.e. wireless networking will not work any more, or Xwon't start)


This is a problem for newbies (and other users with less technical knowledge).

So my idea: Change the kernel-installer scripts, so that they build and 
install the modules at a new kernel (the script might compare between "before 
update" and "installation version")  and then build the required modules out 
of a list using module-assistant commandline-tool (or m-a -i).


Or miss I something and this is already working at another way ? So let me 
know.


So you want Ubuntu's dkms system?

I tried to do it with a startup script calling m-a if a certain module
didn't load, but that didn't always work so well so I went back to doing
it manually, although I only have to do it for the nvidia driver now.
lirc I use the ubuntu version which uses dkms and is hence automatic,
and for ivtv it is now in the kernel and hence no longer a problem.



--
http://fragfest.com.au


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automatically build and install kernel-modules

[Lennart Sorensen]

On Tue, Jun 24, 2008 at 05:03:51PM +0200, Hans-J. Ullrich wrote:
> althoug this is not quite debian-amd-related, I am looking fo a way, to 
> automatically build and install kernel-modules, which must normally build by 
> hand and, which got sources in the repository. These are i.e. madwifi, 
> nvidia-kernel and some other. 
> 
> Why not manually ? Well, the problem is, when a new kernel-version was 
> installed, and the required modules are not build, it might make the system 
> unusable (i.e. wireless networking will not work any more, or Xwon't start)
> 
> This is a problem for newbies (and other users with less technical knowledge).
> 
> So my idea: Change the kernel-installer scripts, so that they build and 
> install the modules at a new kernel (the script might compare between "before 
> update" and "installation version")  and then build the required modules out 
> of a list using module-assistant commandline-tool (or m-a -i).
> 
> Or miss I something and this is already working at another way ? So let me 
> know.

So you want Ubuntu's dkms system?

I tried to do it with a startup script calling m-a if a certain module
didn't load, but that didn't always work so well so I went back to doing
it manually, although I only have to do it for the nvidia driver now.
lirc I use the ubuntu version which uses dkms and is hence automatic,
and for ivtv it is now in the kernel and hence no longer a problem.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Automatically build and install kernel-modules

[Hans-J. Ullrich]

Hi all, 

althoug this is not quite debian-amd-related, I am looking fo a way, to 
automatically build and install kernel-modules, which must normally build by 
hand and, which got sources in the repository. These are i.e. madwifi, 
nvidia-kernel and some other. 

Why not manually ? Well, the problem is, when a new kernel-version was 
installed, and the required modules are not build, it might make the system 
unusable (i.e. wireless networking will not work any more, or Xwon't start)

This is a problem for newbies (and other users with less technical knowledge).

So my idea: Change the kernel-installer scripts, so that they build and 
install the modules at a new kernel (the script might compare between "before 
update" and "installation version")  and then build the required modules out 
of a list using module-assistant commandline-tool (or m-a -i).

Or miss I something and this is already working at another way ? So let me 
know.


Regards 

Hans-J. Ullrich


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Building kernel modules for any kernel ?

[Lennart Sorensen]

On Fri, May 16, 2008 at 12:27:31PM +0200, Hans-J. Ullrich wrote:
> is it possible, to build kernel modules for kernels, which are not 
> running/installed ? I have installed the kernel headers for 2.6.25 and 
> running 2.6.24. 
> 
> Before I change to 2.6.25, I want to make sure, all needed kernel modules 
> (madwifi, nvidia, kvm, kqemu, openvbox-ose etc. ) can be built. In the past I 
> got trouble with this (there is i.e. a report about "nvidia-kernel" relating 
> to this). 

m-a a-i -t modulename -l 2.6.25-2-amd64

The current 169.12 nvidia driver will not compile.  A patch is needed
which we are adding at the moment.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Building kernel modules for any kernel ?

[Goswin von Brederlow]

"Hans-J. Ullrich" <[EMAIL PROTECTED]> writes:

> Hi all,
>
> is it possible, to build kernel modules for kernels, which are not 
> running/installed ? I have installed the kernel headers for 2.6.25 and 
> running 2.6.24. 

yes. Look into module-assistant.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Building kernel modules for any kernel ?

[Hans-J. Ullrich]

Hi all,

is it possible, to build kernel modules for kernels, which are not 
running/installed ? I have installed the kernel headers for 2.6.25 and 
running 2.6.24. 

Before I change to 2.6.25, I want to make sure, all needed kernel modules 
(madwifi, nvidia, kvm, kqemu, openvbox-ose etc. ) can be built. In the past I 
got trouble with this (there is i.e. a report about "nvidia-kernel" relating 
to this). 

regards

Hans


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Lennart Sorensen]

On Wed, May 14, 2008 at 03:13:24PM -0500, Jaime Ochoa Malagón wrote:
> Someone have a prepacked debian way version with that patch and this
> work with 2.6.25-2-amd64 too...
> 
> it's stable enough 173.08?
> 
> anyone have this debian way packed?

I do, running on my mtytv box for testing.  I was trying to see if it
helped the seg fault problem on my 8600GT, although I have found out
that problem is caused by using nvidia's libwfb rather than xorg's
libwfb so that will be fixed in the 169.12 package soon (I hope).  No
issues with the 173.08 driver in the first 30 hours though.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Lennart Sorensen]

On Wed, May 14, 2008 at 06:58:01PM +, Jack Malmostoso wrote:
> That's funny. It still works just fine for me:

Yeah it does.  I screwed up when I tried it.  My mistake.

> [EMAIL PROTECTED]:~$ COLUMNS=200 dpkg -l nvidia-kernel-2.6.25*
> [cut]
> 
> ii  nvidia-kernel-2.6.25-1-amd64 169.12-1
> +2.6.25-1NVIDIA binary kernel module for 
> Linux 2.6.25-1-amd64
> 
> ii  nvidia-kernel-2.6.25-2-amd64 169.12-1
> +2.6.25-2NVIDIA binary kernel module for 
> Linux 2.6.25-2-amd64
> 
> I mean, it's not a big deal since you're working on the new driver 
> version, but it would surely be interesting understanding why it works 
> for me and not for others.

Well nvidia still lists 169.12 as the current stable version.  I am not
quite sure what they think 173.08 is, other than a driver that also
happens to support 9xxx series cards and 2.6.25 kernel.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Jaime Ochoa Malagón]

Someone have a prepacked debian way version with that patch and this
work with 2.6.25-2-amd64 too...

it's stable enough 173.08?

anyone have this debian way packed?

On 5/14/08, Lennart Sorensen <[EMAIL PROTECTED]> wrote:
> On Wed, May 14, 2008 at 04:33:07PM +, Jack Malmostoso wrote:
> > On Wed, 14 May 2008 15:50:15 +0200, Lennart Sorensen wrote:
> >
> > > I am working on it.  173.08 builds fine with 2.6.25 so if nothing else I
> > > can compare 169.12 with 173.08 and see what nvidia changed to get around
> > > the change.
> >
> > Lennart,
> >
> > bug #476504, and in particular this comment:
> >
> > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=476504#32
> >
> > worked just fine for me. Maybe you should just include the patch from
> > nvidia?
>
> That patch does NOT compile for me with 2.6.25-2-amd64 which just
> entered Debian unstable.  I made a patch based on 173.08's changes which
> does compile.  It's a bit longer than the patch mentioned in that bug
> report but apparently the extra changes are relevant to some things.  So
> it might have been fine for 2.6.25-1 but isn't good enough for 2.6.25-2.
>
> --
> LeN Sorensen
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
>


-- 
Perhaps the depth of
love can be calibrated by the number of different selves that are
actively involved in a given relationship.

Carl Sagan (Contact)

Jaime Ochoa Malagón
Arquitecto de 
SolucionesCel: +52 (55) 
1021 0774


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Jack Malmostoso]

On Wed, 14 May 2008 20:20:15 +0200, Lennart Sorensen wrote:

> That patch does NOT compile for me with 2.6.25-2-amd64 which just
> entered Debian unstable.  I made a patch based on 173.08's changes which
> does compile.  It's a bit longer than the patch mentioned in that bug
> report but apparently the extra changes are relevant to some things.  So
> it might have been fine for 2.6.25-1 but isn't good enough for 2.6.25-2.

That's funny. It still works just fine for me:

[EMAIL PROTECTED]:~$ COLUMNS=200 dpkg -l nvidia-kernel-2.6.25*
[cut]

ii  nvidia-kernel-2.6.25-1-amd64 169.12-1
+2.6.25-1NVIDIA binary kernel module for 
Linux 2.6.25-1-amd64

ii  nvidia-kernel-2.6.25-2-amd64 169.12-1
+2.6.25-2NVIDIA binary kernel module for 
Linux 2.6.25-2-amd64

I mean, it's not a big deal since you're working on the new driver 
version, but it would surely be interesting understanding why it works 
for me and not for others.

-- 
Best Regards, Jack
Linux User #264449
Powered by Debian GNU/Linux on AMD64


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Lennart Sorensen]

On Wed, May 14, 2008 at 04:33:07PM +, Jack Malmostoso wrote:
> On Wed, 14 May 2008 15:50:15 +0200, Lennart Sorensen wrote:
> 
> > I am working on it.  173.08 builds fine with 2.6.25 so if nothing else I
> > can compare 169.12 with 173.08 and see what nvidia changed to get around
> > the change.
> 
> Lennart,
> 
> bug #476504, and in particular this comment:
> 
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=476504#32
> 
> worked just fine for me. Maybe you should just include the patch from 
> nvidia?

That patch does NOT compile for me with 2.6.25-2-amd64 which just
entered Debian unstable.  I made a patch based on 173.08's changes which
does compile.  It's a bit longer than the patch mentioned in that bug
report but apparently the extra changes are relevant to some things.  So
it might have been fine for 2.6.25-1 but isn't good enough for 2.6.25-2.

-- 
LeN Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Jack Malmostoso]

On Wed, 14 May 2008 15:50:15 +0200, Lennart Sorensen wrote:

> I am working on it.  173.08 builds fine with 2.6.25 so if nothing else I
> can compare 169.12 with 173.08 and see what nvidia changed to get around
> the change.

Lennart,

bug #476504, and in particular this comment:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=476504#32

worked just fine for me. Maybe you should just include the patch from 
nvidia?

-- 
Best Regards, Jack
Linux User #264449
Powered by Debian GNU/Linux on AMD64


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: nvidia-kernel modules and >2.6.25?

[Lennart Sorensen]

On Wed, May 14, 2008 at 01:59:33PM +0200, Giacomo Mulas wrote:
> It looks like the exported kernel function global_flush_tlb disappeared
> going from 2.6.24 to 2.6.25, and the change_page_attr function has been
> throughout replaced with new functions. The nvidia module depends on them,
> so it doesn't build anymore on amd64. Is an experimental new
> nvidia-kernel-source version fixing this available somewhare? Or, as an
> alternative, does any kernel guru know how to replace the calls to these
> functions in the nvidia kernel module to use the new kernel API?

I am working on it.  173.08 builds fine with 2.6.25 so if nothing else I
can compare 169.12 with 173.08 and see what nvidia changed to get around
the change.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

nvidia-kernel modules and >2.6.25?

[Giacomo Mulas]

It looks like the exported kernel function global_flush_tlb disappeared
going from 2.6.24 to 2.6.25, and the change_page_attr function has been
throughout replaced with new functions. The nvidia module depends on them,
so it doesn't build anymore on amd64. Is an experimental new
nvidia-kernel-source version fixing this available somewhare? Or, as an
alternative, does any kernel guru know how to replace the calls to these
functions in the nvidia kernel module to use the new kernel API?

Thanks in advance
Giacomo

--
_

Giacomo Mulas <[EMAIL PROTECTED]>
_

OSSERVATORIO ASTRONOMICO DI CAGLIARI
Str. 54, Loc. Poggio dei Pini * 09012 Capoterra (CA)

Tel. (OAC): +39 070 71180 248 Fax : +39 070 71180 222
Tel. (UNICA): +39 070 675 4916
_

"When the storms are raging around you, stay right where you are"
 (Freddy Mercury)
_


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: building kernel-modules at new kernels

[Goswin von Brederlow]

Jack Malmostoso <[EMAIL PROTECTED]> writes:

> On Sat, 10 May 2008 20:30:14 +0200, Hans-J. Ullrich wrote:
>
>> Is there a debian way, to automatically build additional kernel-modules
>> (i.e. madwifi, nvidia, virtualbox-ose etc.), when installing a newer
>> kernelversion and headers ?
>
> I wrote a (rather horrible) script once, that I invoke with cron. You can 
> find it here:
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=459373

A better place would be /etc/apt/apt.conf.d/ as a post update or dpkg
hook.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: building kernel-modules at new kernels

[Goswin von Brederlow]

Harald Dunkel <[EMAIL PROTECTED]> writes:

> Jack Malmostoso wrote:
>> On Sat, 10 May 2008 20:30:14 +0200, Hans-J. Ullrich wrote:
>>
>>> Is there a debian way, to automatically build additional kernel-modules
>>> (i.e. madwifi, nvidia, virtualbox-ose etc.), when installing a newer
>>> kernelversion and headers ?
>>
>
> Maybe you should consider to look at kernel-package. It allows you
> to build *.deb files for both a new kernel and a list of modules
> within one run.
>
>
> Hope this helps. Regards
>
> Harri

I'm assuming he means building extra modules for the prebuild debian
kernels.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: building kernel-modules at new kernels

[Harald Dunkel]

Jack Malmostoso wrote:

On Sat, 10 May 2008 20:30:14 +0200, Hans-J. Ullrich wrote:


Is there a debian way, to automatically build additional kernel-modules
(i.e. madwifi, nvidia, virtualbox-ose etc.), when installing a newer
kernelversion and headers ?




Maybe you should consider to look at kernel-package. It allows you
to build *.deb files for both a new kernel and a list of modules
within one run.


Hope this helps. Regards

Harri


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: building kernel-modules at new kernels

[Jack Malmostoso]

On Sat, 10 May 2008 20:30:14 +0200, Hans-J. Ullrich wrote:

> Is there a debian way, to automatically build additional kernel-modules
> (i.e. madwifi, nvidia, virtualbox-ose etc.), when installing a newer
> kernelversion and headers ?

I wrote a (rather horrible) script once, that I invoke with cron. You can 
find it here:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=459373

-- 
Best Regards, Jack
Linux User #264449
Powered by Debian GNU/Linux on AMD64


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: building kernel-modules at new kernels

[Lennart Sorensen]

On Sat, May 10, 2008 at 08:22:17PM +0200, Hans-J. Ullrich wrote:
> Is there a debian way, to automatically build additional kernel-modules (i.e. 
> madwifi, nvidia, virtualbox-ose etc.), when installing a newer kernelversion 
> and headers ?
> 
> (I mean, not by using module-assistant, but just automatically build during 
> the upgrade process).

None I know of.  Of course there are new packages
(linux-modules-[extra,contrib,nonfree]-2.6 being done that will auto
build all the modules it knows about whenever a new Debian kernel is
made, so unless make your own kernel you probably won't have to do
anything in the future.

-- 
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

building kernel-modules at new kernels

[Hans-J. Ullrich]

Hello !

Is there a debian way, to automatically build additional kernel-modules (i.e. 
madwifi, nvidia, virtualbox-ose etc.), when installing a newer kernelversion 
and headers ?

(I mean, not by using module-assistant, but just automatically build during 
the upgrade process).

Regards

Hans




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: how-to compile 64 bit modules on 32 bit system with m-a ?

[Jamil Djadala]

В пн, 2007-07-02 в 14:26 +0200, Goswin von Brederlow написа:
> Jamil Djadala <[EMAIL PROTECTED]> writes:
> 
> > Hi all,
> > I have 32 bit debian(testing)  with 64 bit stock kernel
> > (linux-image-2.6.18-4-amd64)
> >
> > (of course i have 64 bit processor)
> >
> > is there standard way to build 64 bit modules (nvidia, rt2500)  with
> > module-assistant ?
> >
> > i guess 64bit chroot may be solution, but any other ?
> 
> Doesn't it just work? Works for the kernel as it automatically adds
> -m64 as needed.

it doesn't work.
after some testing i discover these results:

under 32 bit kernel, command

m-a  -l 2.6.21-2-k7 a-i kqemu

end successfully, but

m-a  -l 2.6.21-2-amd64 a-i kqemu

end with error:
make[4]: *** No rule to make target
`/usr/src/modules/kqemu/kqemu-mod-x86_64.o', needed by
`/usr/src/modules/kqemu/kqemu-mod.o'.  Stop.


under 64 bit kernel, both 
m-a  -l 2.6.21-2-k7 a-i kqemu
m-a  -l 2.6.21-2-amd64 a-i kqemu
fail to compile.


-- 
Jamil Djadala <[EMAIL PROTECTED]>

Re: how-to compile 64 bit modules on 32 bit system with m-a ?

[Goswin von Brederlow]

Jamil Djadala <[EMAIL PROTECTED]> writes:

> Hi all,
> I have 32 bit debian(testing)  with 64 bit stock kernel
> (linux-image-2.6.18-4-amd64)
>
> (of course i have 64 bit processor)
>
> is there standard way to build 64 bit modules (nvidia, rt2500)  with
> module-assistant ?
>
> i guess 64bit chroot may be solution, but any other ?

Doesn't it just work? Works for the kernel as it automatically adds
-m64 as needed.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

how-to compile 64 bit modules on 32 bit system with m-a ?

[Jamil Djadala]

Hi all,
I have 32 bit debian(testing)  with 64 bit stock kernel
(linux-image-2.6.18-4-amd64)

(of course i have 64 bit processor)

is there standard way to build 64 bit modules (nvidia, rt2500)  with
module-assistant ?

i guess 64bit chroot may be solution, but any other ?

-- 
Jamil Djadala <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: memory modules

[Francesco Pietra]

--- Micha <[EMAIL PROTECTED]> wrote:

> If there's enough money (and reason) i'd go for
> prepackaged quality
> sets of exactly the same brand, even from the same
> fabrication plant,
> and production series. Else, try to have a deal
> allwoing you to return 
> mem bars which doesn't work as expected.

Thank you. That's exactly what I did, though
expensive.
> 
> Also, you may compare lshw to 'dmidecode'
> though i'd expect they just do the same system
> (BIOS) calls.
> 
> 
> -- 
> To UNSUBSCRIBE, email to
> [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]
> 
> 



 

Don't get soaked.  Take a quick peak at the forecast
with the Yahoo! Search weather shortcut.
http://tools.search.yahoo.com/shortcuts/#loc_weather


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: memory modules

[Micha]

If there's enough money (and reason) i'd go for prepackaged quality
sets of exactly the same brand, even from the same fabrication plant,
and production series. Else, try to have a deal allwoing you to return 
mem bars which doesn't work as expected.

Also, you may compare lshw to 'dmidecode'
though i'd expect they just do the same system
(BIOS) calls.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: memory modules

[Lennart Sorensen]

On Thu, Feb 01, 2007 at 03:05:08AM -0800, Francesco Pietra wrote:
> With the old i386 machine very little info on (cheap)
> ram. 
> 
> With tyan2895 K8WE/dual-opteron amd64 machine far more
> info on the 8 mem modules (Kingston DDR ECC): width
> 64bit, though non info on frequency. What is given is
> the controller clock (66MHz) at surprisingly 32bit. 
> 
> Is that worthwhile, under such conditions, to pay for
> expensive mem modules (in terms of speed I mean, the
> security of ECC is another affair)

Well I have seen many messages on this list from people with various
problems, all claiming to have run memtest for days on end and being
absolutely sure their ram was perfect.  When they then tried removing
half their ram, had the problem go away, swapping in the opposite half
of ram and having the problem come back, they finally understood that
memtest doesn't catch all types of errors, and that the opteron demands
quality memory.  It doesn't have to be expensive, but it absolutely must
be made to the spec it claims to be.  If you claim a latency of 3
cycles, and you are really 3.02 cycles latency, figuring the memory
controller isn't going to run the ram at exactly the spec you put in
your SPD chip, well then the opteron will likely prove you wrong.  It
really does try to get the most from the ram.  Well that's my theory on
it at least.

So far I have not personally run opterons, just athlon 64s, but kingston
value DDR 400 ram worked perfectly.  Given name brand ram doesn't have
to cost particularly more than generic ram, why use generic when you can
pay a few percent more for lifetime warenty ram that works.  Even stuff
like crucial and corsair isn't really that much more expensive (as long
as you don't buy the crazy overclocking ram running at 1000MHz).

Of course you could always follow the advice of the system/motherboard
maker and buy what they have tested and qualified to work.

In terms of using 333MHz or 400MHz or whichever ram, well I guess that
depends a lot on what type of work you do.  If it is very memory
intensive, then the extra bandwidth of 400Mhz is probably worth it.  If
you are mainly running cpu intensive, with not too much actual memory
access, then you may want to save a bit (if 333MHz is cheaper at all
anymore) on the ram, and get a faster cpu instead.  Of course buying the
fastest posible cpu and putting in the cheapest slowest ram just doesn't
make sense in general, but I guess it might for some people.

--
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: memory modules

[Francesco Pietra]

With the old i386 machine very little info on (cheap)
ram. 

With tyan2895 K8WE/dual-opteron amd64 machine far more
info on the 8 mem modules (Kingston DDR ECC): width
64bit, though non info on frequency. What is given is
the controller clock (66MHz) at surprisingly 32bit. 

Is that worthwhile, under such conditions, to pay for
expensive mem modules (in terms of speed I mean, the
security of ECC is another affair)

Thanks a lot

francesco



--- Matteo Vescovi <[EMAIL PROTECTED]> wrote:

> 2007/2/1, Francesco Pietra <[EMAIL PROTECTED]>:
> > On amd64, especially for the dual-opteron
> arrangement,
> > is any bash command to learn more (frequency,
> brand,
> > version, etc) about memory modules than what is
> > reported by dmesg? If modules are identical in
> > couples, though not couples identical?
> >
> > Thanks for answering
> >
> > francesco pietra
> 
> lshw.
> It's an optional tool that has to be installed.
> 
> Hope it helps.
> 
> Matteo
> 
> --
> Imagine a red swirl here.
> 
> 
> -- 
> To UNSUBSCRIBE, email to
> [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]
> 
> 



 

Want to start your own business?
Learn how on Yahoo! Small Business.
http://smallbusiness.yahoo.com/r-index


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: memory modules

[Matteo Vescovi]

2007/2/1, Francesco Pietra <[EMAIL PROTECTED]>:

On amd64, especially for the dual-opteron arrangement,
is any bash command to learn more (frequency, brand,
version, etc) about memory modules than what is
reported by dmesg? If modules are identical in
couples, though not couples identical?

Thanks for answering

francesco pietra


lshw.
It's an optional tool that has to be installed.

Hope it helps.

Matteo

--
Imagine a red swirl here.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

memory modules

[Francesco Pietra]

On amd64, especially for the dual-opteron arrangement,
is any bash command to learn more (frequency, brand,
version, etc) about memory modules than what is
reported by dmesg? If modules are identical in
couples, though not couples identical?

Thanks for answering

francesco pietra



 

Need Mail bonding?
Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
http://answers.yahoo.com/dir/?link=list&sid=396546091


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[dam]

I am writing to you in emacs on a debian unstable install to an
e6600/asusP5W-Deluxe... I tried several boot images, and I think the
mini.iso of the unstable netboot was working, but I actually installed
using the chroot method in the amd64 howto off of debian.org.  ubuntu
latest properly detects sata and sets up ok, so I installed that, then
did the cdebootstrap into another mounted partition, set up the fstab
and the network interfaces, and it comes right up, sata, ethernet...
I then built a kernel and booted with it and it has been quite stable
since.  The ubuntu 6.0.5 LTS did seem to have problems, would
lock up completely during some of the gl games... have not had the
same problem with the debian install. 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Re: Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[Eric Valette]

>
> We have a P5W Delux at work which had various problems too.  It is
> running windows XP, and after about a week of time it stops booting.  It
> shows the splash screen of windows, and at the point it should change to
> a higher resolution you just get a black screen.
>
> Sometimes in the event log it shows disks errors.  This is with an SATA
> disk.
>
> We've replaced the disk, still the same problem.  We've replaced the
> motherboard, power supply, and still have the same problem.
>
> I've also tried it using Debian (i386 netinst etch beta3), I've ran a
> badblock read-only test, didn't find any problems.  Then I ran a 
> read-write test, after a short time the disks basicly stops responding.
> The led of the disk on the PC is constatly on, instead of most of the
> time on, blinking.  At that point, the tests only shows failed sectors.
>
> We've replaced the disk with an PATA disk, and didn't see a problem so
> far.
>   

So ther might be a problem with the sata disk controller. I guess I
would try to upgrade the BIOS as first thing. But my problem was
actually entirely related to Linux : chnaging the kernel to a newer
version does indeed fix the disk proble (although I admit, I haven't yet
formatted the disk because it was too late last nigh I tried).
> We also have an identical PC that doesn't show any problems.
>
> I really can't do any tests with this, since someone is using it.
>   
OK. Thansk for the hint about the possible SATA disk problems.

-- eric


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[Kurt Roeckx]

On Wed, Nov 01, 2006 at 05:08:21PM +0100, Eric Valette wrote:
> Kurt Roeckx wrote:
> > On Mon, Oct 30, 2006 at 09:46:03AM +0100, Eric Valette wrote:
> >>2) Downloaded 
> >> ftp://ftp.fr.debian.org/debian/dists/unstable/main/installer-amd64/20061022/images/gtk-miniiso
> >> burned it and booted it. Starts well but hangs after a while because of 
> >> spurious interrupts related to disk activities (SATA) endless reset of 
> >> controller. Have also spurious irq listed in dmesg. Did not know if a 
> >> kernel was correctly running on this mtherboard + CPU combination. So 
> >> tried to find another instable AMD64 distrib.
> > 
> > Could you please file an installation-report?  There are some tools in
> > the installer that can show us more information about the hardare you're
> > using.  You can probably also run them on an installed system.
> 
> If you hunt asus you can download the user manual with all chipsets
> listed. But I will report once I managed to install. Nothing really
> special except PCI-E and sky2 ethernet driver.

We have a P5W Delux at work which had various problems too.  It is
running windows XP, and after about a week of time it stops booting.  It
shows the splash screen of windows, and at the point it should change to
a higher resolution you just get a black screen.

Sometimes in the event log it shows disks errors.  This is with an SATA
disk.

We've replaced the disk, still the same problem.  We've replaced the
motherboard, power supply, and still have the same problem.

I've also tried it using Debian (i386 netinst etch beta3), I've ran a
badblock read-only test, didn't find any problems.  Then I ran a 
read-write test, after a short time the disks basicly stops responding.
The led of the disk on the PC is constatly on, instead of most of the
time on, blinking.  At that point, the tests only shows failed sectors.

We've replaced the disk with an PATA disk, and didn't see a problem so
far.

We also have an identical PC that doesn't show any problems.

I really can't do any tests with this, since someone is using it.


Kurt


> 
> 
> > Can you please show the messages the kernel reports?
> 
> No. No time to do that as it would require booting with a serial line
> connected + compiling a newer kernel solve the problem. Will do once the
> system is operationnal.
> 
> > 
> > Is what you're seeing simular to:
> > http://bugs.debian.org/309964
> > 
> > Or:
> > http://bugs.debian.org/393977
> 
> For the irq message yes. For the SATA controller reset no.
> 
> 
> -- eric


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[Kurt Roeckx]

On Mon, Oct 30, 2006 at 09:46:03AM +0100, Eric Valette wrote:
>   2) Downloaded 
> ftp://ftp.fr.debian.org/debian/dists/unstable/main/installer-amd64/20061022/images/gtk-miniiso
> burned it and booted it. Starts well but hangs after a while because of 
> spurious interrupts related to disk activities (SATA) endless reset of 
> controller. Have also spurious irq listed in dmesg. Did not know if a 
> kernel was correctly running on this mtherboard + CPU combination. So 
> tried to find another instable AMD64 distrib.

Could you please file an installation-report?  There are some tools in
the installer that can show us more information about the hardare you're
using.  You can probably also run them on an installed system.

Can you please show the messages the kernel reports?

Is what you're seeing simular to:
http://bugs.debian.org/309964

Or:
http://bugs.debian.org/393977

Have you tried booting with options like noacpi and acpi=off?


Kurt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[Eric Valette]

Eric Valette wrote:
> Please CC me when anwsering: I'm not suscribed.
>
> 1) Did anybody else tried to replace a kernel with all drivers build in?
> Why did the "modprobe -q evdev ||true" fail? Has it anything to do with
> SMP support as the shell will probably spawn the two child on different
> proc but should receive exit status as on UP? Sound like a bug in d-i
> isn't it?

Found the culprit : S61mouse-support-x86 is incorrectly written  : it
does not use -q option for modeprobe not the || true.

Will fill a bug because it almost prevent to easilly upgrade to a
newer/better configured kernel.


-- eric


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Installing Debian amd64 on an asus P5W DH Deluxe + E6600 : kernel on installer broken + bugs in d-i when changing kernel + modules

[Eric Valette]

Please CC me when anwsering: I'm not suscribed.

OK I knew this would be painful although I've installed debian on 
hundred of machines over the past ten years. No deception on this point.


At least I will report the actual status and seek for advices for 
finishing. What I did so far:


	1) Took a knoppix 5.01 DVD booted it. Works like a charm except the X 
nv driver. Defaults to vesa after some trials. Unfortunately kernel is 
2.6.15 and I need a SKY2 ethernet driver for my dual marvell Gigabit 
ethernet onbaord + I want to run in 64 bit mode. At least I noted the 
drivers used by doing an lsmod.
	2) Downloaded 
ftp://ftp.fr.debian.org/debian/dists/unstable/main/installer-amd64/20061022/images/gtk-miniiso
burned it and booted it. Starts well but hangs after a while because of 
spurious interrupts related to disk activities (SATA) endless reset of 
controller. Have also spurious irq listed in dmesg. Did not know if a 
kernel was correctly running on this mtherboard + CPU combination. So 
tried to find another instable AMD64 distrib.
	3) Downloaded the brand new opensuze 10.2 beta1 ISO because it includes 
almost all the up-to-date stuff. Booted installer. Works like a charm 
kernel wise.
	3) Downloaded 2.6.19-rc3-git5, configured it with all needed drivers IN 
+ SMP. Made a modules_install to get the module directory. Then replaced 
the kernel, replaced the /lib/modules by the correct modules directory 
in initrd.img, rebuild the CD iso. Booted it. Boots fine, no freeze, 
detected almost all hardware part necessary to installation (network, 
disks, DVD,s, ...). Unfortunately d-i fails with a series of error when 
trying to load evdev modules (that I have'nt because its compiled in). 
Found the scripts that tries to load the modules and saw that it is 
called like "modprobe -q evdev ||true" so do not know why it loops with 
an error.


So now I have some questions because I do not want to explore all the 
dead ends by myself if other s have already gone struggling this path.


1) Did anybody else tried to replace a kernel with all drivers build in? 
Why did the "modprobe -q evdev ||true" fail? Has it anything to do with 
SMP support as the shell will probably spawn the two child on different 
proc but should receive exit status as on UP? Sound like a bug in d-i 
isn't it?


2) If I recompile my kernel with all code as modules actually loaded 
with modprobe by d-i, will it have a change to work? I personnaly would 
prefer to comment all the modprobe in the scripts...


3) Shall I build the kernel without SMP? Is the installer known to be 
broken when using SMP kernel?


PS : the documentation to make the kind of trick like in 3) above is 
rather sparse. Finding the right cpio command to rebuild the initrd 
image is rather difficult! and the mkisofs command also. Annd  no I do 
not want to use the debvian scripts like bootcd or initrd-tools because 
its not convenient in my case.





-- eric


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel 2.6.18: agpgart-modules ???

[Stephen Olander Waters]

On Wed, 2006-10-25 at 16:14 +0200, Hans-J. Ullrich wrote:
> I am missing some agpgart-modules in the kernel-sources.
> In earlier days, I could build them as modules, now I only find intel-agp, 
> sis-agp and ali-agp. 

There is only one AGP module for the AMD64 architecture cleverly named
"amd64-agp" -- however it is built-in on Debian kernels. I assume this
is because it's built into all AMD64 CPUs or is required for AMD64
northbridges. 

-s



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Kernel 2.6.18: agpgart-modules ???

[Hans-J. Ullrich]

Hi folks,

I am missing some agpgart-modules in the kernel-sources.
In earlier days, I could build them as modules, now I only find intel-agp, 
sis-agp and ali-agp. It is wiered, as on another computer, I can see the 
others and activate them. I supposed, it has something to do with 
the ".config" in the "/usr/src/linux"-directory, so I made a fresh 
installation of the related packages (sources and header-files) 

How can I see them again ? Is this related on the hardware the running kernel 
does see ? Are there some temoporeral files I should delete ? I am running 
the headers and image "*-2.6.18-1-amd64". 

Is there a difference between 32-bit and 64-bit ? (I have no second 64-bit 
system available) 

Thanks for help ! 

Regards

Hans


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: lirc modules needing complete kernel source

[Hamish Moffatt]

On Tue, Oct 03, 2006 at 09:59:14AM -0400, Lennart Sorensen wrote:
> They can.  Unfortunately last time I tried them they were incorrectly
> packaged and hardcoded to build against 2.6.16 rather than the kernel
> headers you point them at.  I really should file a bug report.  The

Compiled OK here with 2.6.17-{1,2}, though not .18-1.
(Already reported as #390907.)

Actually you do need a patch from the BTS to get the gpio and/or bttv
modules to compile with .17.

Hamish
-- 
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: lirc modules needing complete kernel source

[Lennart Sorensen]

On Tue, Oct 03, 2006 at 11:30:46PM +1000, garrone wrote:
> I'm thinking of giving lirc a try.
> I find I need the lirc-modules-source package
> because lirc_dev is absent.
> 
> The package in dselect says that kernel sources
> must be installed to compile these modules.
> 
> Why cannot they simply be compiled with headers,
> like the nvidia module for example?

They can.  Unfortunately last time I tried them they were incorrectly
packaged and hardcoded to build against 2.6.16 rather than the kernel
headers you point them at.  I really should file a bug report.  The
package also modifies itself on debian/rules clean, so overall it is
really in an awful state.

--
Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: lirc modules needing complete kernel source

[Hamish Moffatt]

On Tue, Oct 03, 2006 at 11:30:46PM +1000, garrone wrote:
> I'm thinking of giving lirc a try.
> I find I need the lirc-modules-source package
> because lirc_dev is absent.
> 
> The package in dselect says that kernel sources
> must be installed to compile these modules.
> 
> Why cannot they simply be compiled with headers,
> like the nvidia module for example?

They can, with module-assistant.

Note that lirc-modules-source doesn't seem to compile with 2.6.18
currently. Works with 2.6.17-2.

Hamish
-- 
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

lirc modules needing complete kernel source

[garrone]

I'm thinking of giving lirc a try.
I find I need the lirc-modules-source package
because lirc_dev is absent.

The package in dselect says that kernel sources
must be installed to compile these modules.

Why cannot they simply be compiled with headers,
like the nvidia module for example?

Peter Garrone


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Upgrade to linux-image-2.6.14-2 breaks sound modules

[Lee Begg]

On Sat, 24 Dec 2005 01:40, Frederik Schueler wrote:
> Hello,
>
> On Fri, Dec 23, 2005 at 11:17:01PM +1300, Lee Begg wrote:
> > I've just upgraded from 2.6.12 to 2.6.14-2 (-amd64-k8-smp).
> >
> > Is there something I can do?
>
> Yes: rerun alsaconf, this will fix it.

Ahh, that easy.
Thanks, works now.

> Best regards
> Frederik Schueler

Later
Lee Begg


pgpT7luAIsNMK.pgp
Description: PGP signature

Re: Upgrade to linux-image-2.6.14-2 breaks sound modules

[Frederik Schueler]

Hello,


On Fri, Dec 23, 2005 at 11:17:01PM +1300, Lee Begg wrote:
> I've just upgraded from 2.6.12 to 2.6.14-2 (-amd64-k8-smp).
> 
> Is there something I can do?

Yes: rerun alsaconf, this will fix it.

Best regards
Frederik Schueler

-- 
ENOSIG


signature.asc
Description: Digital signature

Upgrade to linux-image-2.6.14-2 breaks sound modules

[Lee Begg]

I've just upgraded from 2.6.12 to 2.6.14-2 (-amd64-k8-smp).

The snd module and the modules for my sound card refuse to load.  Below is the 
dmesg log.

Is there an update coming? Is there something I can do?

Later
Lee Begg



(On "modprobe snd")
snd: Unknown parameter `device_mode'

(On "modprobe snd_emu10k1")
snd: Unknown parameter `device_mode'
snd_hwdep: Unknown symbol snd_info_register
snd_hwdep: Unknown symbol snd_info_create_module_entry
snd_hwdep: Unknown symbol snd_ctl_unregister_ioctl_compat
snd_hwdep: Unknown symbol snd_info_free_entry
snd_hwdep: Unknown symbol snd_unregister_oss_device
snd_hwdep: Unknown symbol snd_register_oss_device
snd_hwdep: Unknown symbol snd_ctl_register_ioctl
snd_hwdep: Unknown symbol snd_card_file_add
snd_hwdep: Unknown symbol snd_iprintf
snd_hwdep: Unknown symbol snd_unregister_device
snd_hwdep: Unknown symbol snd_device_new
snd_hwdep: Unknown symbol snd_ctl_unregister_ioctl
snd_hwdep: Unknown symbol snd_ctl_register_ioctl_compat
snd_hwdep: Unknown symbol snd_card_file_remove
snd_hwdep: Unknown symbol snd_info_unregister
snd_hwdep: Unknown symbol snd_register_device
snd_timer: Unknown symbol snd_info_register
snd_timer: Unknown symbol snd_info_create_module_entry
snd_timer: Unknown symbol snd_info_free_entry
snd_timer: Unknown symbol snd_iprintf
snd_timer: Unknown symbol snd_ecards_limit
snd_timer: Unknown symbol snd_oss_info_register
snd_timer: Unknown symbol snd_unregister_device
snd_timer: Unknown symbol snd_device_new
snd_timer: Unknown symbol snd_info_unregister
snd_timer: Unknown symbol snd_register_device
snd: Unknown parameter `device_mode'
snd_timer: Unknown symbol snd_info_register
snd_timer: Unknown symbol snd_info_create_module_entry
snd_timer: Unknown symbol snd_info_free_entry
snd_timer: Unknown symbol snd_iprintf
snd_timer: Unknown symbol snd_ecards_limit
snd_timer: Unknown symbol snd_oss_info_register
snd_timer: Unknown symbol snd_unregister_device
snd_timer: Unknown symbol snd_device_new
snd_timer: Unknown symbol snd_info_unregister
snd_timer: Unknown symbol snd_register_device
snd_pcm: Unknown symbol snd_info_register
snd_pcm: Unknown symbol snd_info_create_module_entry
snd_pcm: Unknown symbol snd_ctl_unregister_ioctl_compat
snd_pcm: Unknown symbol snd_timer_notify
snd_pcm: Unknown symbol snd_timer_interrupt
snd_pcm: Unknown symbol snd_info_free_entry
snd_pcm: Unknown symbol snd_info_get_str
snd_pcm: Unknown symbol snd_ctl_register_ioctl
snd_pcm: Unknown symbol snd_card_file_add
snd_pcm: Unknown symbol snd_iprintf
snd_pcm: Unknown symbol snd_major
snd_pcm: Unknown symbol snd_unregister_device
snd_pcm: Unknown symbol snd_timer_new
snd_pcm: Unknown symbol snd_device_new
snd_pcm: Unknown symbol snd_ctl_unregister_ioctl
snd_pcm: Unknown symbol snd_ctl_register_ioctl_compat
snd_pcm: Unknown symbol snd_info_create_card_entry
snd_pcm: Unknown symbol snd_power_wait
snd_pcm: Unknown symbol snd_device_free
snd_pcm: Unknown symbol snd_card_file_remove
snd_pcm: Unknown symbol snd_info_unregister
snd_pcm: Unknown symbol snd_device_register
snd_pcm: Unknown symbol snd_register_device
snd_pcm: Unknown symbol snd_info_get_line
snd_ac97_codec: Unknown symbol snd_info_register
snd_ac97_codec: Unknown symbol snd_ctl_add
snd_ac97_codec: Unknown symbol snd_info_free_entry
snd_ac97_codec: Unknown symbol snd_interval_refine
snd_ac97_codec: Unknown symbol snd_ctl_find_id
snd_ac97_codec: Unknown symbol snd_ctl_new1
snd_ac97_codec: Unknown symbol snd_ctl_remove_id
snd_ac97_codec: Unknown symbol snd_component_add
snd_ac97_codec: Unknown symbol snd_pcm_hw_rule_add
snd_ac97_codec: Unknown symbol snd_iprintf
snd_ac97_codec: Unknown symbol snd_device_new
snd_ac97_codec: Unknown symbol snd_info_create_card_entry
snd_ac97_codec: Unknown symbol snd_info_unregister
snd_seq_device: Unknown symbol snd_info_register
snd_seq_device: Unknown symbol snd_info_create_module_entry
snd_seq_device: Unknown symbol snd_info_free_entry
snd_seq_device: Unknown symbol snd_seq_root
snd_seq_device: Unknown symbol snd_iprintf
snd_seq_device: Unknown symbol snd_device_new
snd_seq_device: Unknown symbol snd_info_unregister
snd_rawmidi: Unknown symbol snd_info_register
snd_rawmidi: Unknown symbol snd_ctl_unregister_ioctl_compat
snd_rawmidi: Unknown symbol snd_seq_device_new
snd_rawmidi: Unknown symbol snd_info_free_entry
snd_rawmidi: Unknown symbol snd_unregister_oss_device
snd_rawmidi: Unknown symbol snd_register_oss_device
snd_rawmidi: Unknown symbol snd_ctl_register_ioctl
snd_rawmidi: Unknown symbol snd_card_file_add
snd_rawmidi: Unknown symbol snd_iprintf
snd_rawmidi: Unknown symbol snd_oss_info_register
snd_rawmidi: Unknown symbol snd_unregister_device
snd_rawmidi: Unknown symbol snd_device_new
snd_rawmidi: Unknown symbol snd_ctl_unregister_ioctl
snd_rawmidi: Unknown symbol snd_ctl_register_ioctl_compat
snd_rawmidi: Unknown symbol snd_info_create_card_entry
snd_rawmidi: Unknown symbol snd_device_free
snd_rawmidi: Unknown symbol snd_card_file_remove
snd_raw

Re: Compiling third-party kernel modules for amd64 kernel onDebian/386

[Shaun Jackman]

On 2005-07-22, Xavier Bestel wrote...
> Ok, I finally found a solution: I rebuilt the script/mod/modpost
> executable and copied it over the one in the kernel-build directory,
> and the installation went perfectly.
>
> Thanks,
> Xav
http://lists.debian.org/debian-amd64/2005/07/msg00951.html

I also tried rebuilding script/mod/modpost, but it didn't solve my
problem. How did you rebuild it, and did you have to do modify
anything? I tried...
$ gcc modpost.c file2alias.c sumversion.c -o modpost

I did some debugging in modpost, and found in my case that in
parse_elf, hdr->e_shnum was always zero, which is clearly not correct
or good. readelf says, correctly, that...
  Number of section headers: 20

Cheers,
Shaun

ELF Header:
  Magic:   7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00
  Class: ELF64
  Data:  2's complement, little endian
  Version:   1 (current)
  OS/ABI:UNIX - System V
  ABI Version:   0
  Type:  REL (Relocatable file)
  Machine:   Advanced Micro Devices X86-64
  Version:   0x1
  Entry point address:   0x0
  Start of program headers:  0 (bytes into file)
  Start of section headers:  23432 (bytes into file)
  Flags: 0x0
  Size of this header:   64 (bytes)
  Size of program headers:   0 (bytes)
  Number of program headers: 0
  Size of section headers:   64 (bytes)
  Number of section headers: 20
  Section header string table index: 17

Re: remove unwanted modules

[Lennart Sorensen]

On Wed, Sep 21, 2005 at 09:06:44PM -0500, Marc DM wrote:
> Actually, I wanted to know just for knowing purposes.
> 
> The other reason I wanted to know is because I'm using Debian with a 
> single Opteron246 to create a router to handle traffic between 4 vlans 
> and the internet. So I wanted to make sure that I didn't have any 
> modules in there that might be a potential security threat nor any that 
> would degrade performance solely due to its presence.

Unused modules on disk won't hurt performance any more than 5 jpegs
on the drive does.  It just takes diskspace that is all.

Modules usually only get loaded if there is a piece of hardware present
that requires it.  Or some program requires a certain feature.

As for security, well unless someone is already root they can't load
modules, so I can't really see much chance of a security problem.

Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Jim Crilly]

On 09/21/05 09:06:44PM -0500, Marc DM wrote:
> 
> 
> Jim Crilly wrote:
> 
> >But why do you want to do this? A full modules directory in /lib/`uname -r`
> >only takes up ~40M. And who knows when you'll plug in some new USB device
> >or something and wish you had that module handy.
> > 
> >
> Actually, I wanted to know just for knowing purposes.
> 
> The other reason I wanted to know is because I'm using Debian with a 
> single Opteron246 to create a router to handle traffic between 4 vlans 
> and the internet. So I wanted to make sure that I didn't have any 
> modules in there that might be a potential security threat nor any that 
> would degrade performance solely due to its presence.
> 
> Know of any?

Just a guess, but if a module was known to be a security problem it would
most likely have been removed or fixed =) And since you need to be root (or
at least have CAP_SYS_MODULE) to load/unload modules, the box will already
be compromised by the time they can load any potentially malicious modules.

And as for performance, I really doubt any modules would slow anything down
to the point where you would notice. Most of the modules that might affect
performance require you to do something to activate them, like even if you
load every iptables module available it won't matter unless you have rules
to make them do something.  Especially with a box as fast as an Opteron. 
You might end up with a little less free memory if you load a few modules 
that you don't plan on using, but most modules are only few K each anyway.

> 
> Marc DM
> 

Jim.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Marc DM]

Jim Crilly wrote:


But why do you want to do this? A full modules directory in /lib/`uname -r`
only takes up ~40M. And who knows when you'll plug in some new USB device
or something and wish you had that module handy.
 


Actually, I wanted to know just for knowing purposes.

The other reason I wanted to know is because I'm using Debian with a 
single Opteron246 to create a router to handle traffic between 4 vlans 
and the internet. So I wanted to make sure that I didn't have any 
modules in there that might be a potential security threat nor any that 
would degrade performance solely due to its presence.


Know of any?

Marc DM



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Jim Crilly]

On 09/21/05 12:24:06AM -0500, Marc DM wrote:
> Stupid questions :
> 
> How can I find out which modules my system actually needs and disable 
> the ones I don't need.
> 
> How can I know if a module I'm disabling at startup isn't needed for 
> another module that I plan to load?
> 
> Thanks. And I won't ask anymore stupid questions for the rest of the week.

But why do you want to do this? A full modules directory in /lib/`uname -r`
only takes up ~40M. And who knows when you'll plug in some new USB device
or something and wish you had that module handy.

The only way you're going to get an accurate list is if you know what all
of the modules are used for. And stripping out the modules with a 0
reference count won't be enough, for instance ide-cd currently has a ref
count of 0 on this machine because there's no disc mounted. Hmmm and
somehow tulip has a ref count of 0 as well even though I know I'm using
that NIC...

Anyway, you'll pretty much have to go through the lsmod output and run
modinfo on each module and decide whether it's important enough to keep or
not.

> 
> Regards,
> 
> Marc DM

Jim.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Marc D. Murray]

You can check the use count from lsmod:

Module  Size  Used by
nls_utf82432  0
it87   29472  0
i2c_sensor  3712  1 it87
i2c_isa 2688  0
i2c_dev12288  0
powernow_k811088  0
freq_table  5192  1 powernow_k8
 ^ This column

If the count is 0, it means the module is not being used, either because
the device it supplies isn't currently being used, or because you don't
need it.
   


This is useful info. Thanks.



Or with regards to the modules loaded in the initrd, 
change /etc/mkinitrd/mkinitrd.conf to have the line

   MODULES=dep
instead of =[all,most] and then regenerate the initrd.
 

I actually had to do this on a Tyan Thunder K8W (S2885) in order to get 
it to mount /home on the sata controller at boot (LVM over Raid1). It 
would always complain of not being able to find any md devices when 
loading lvm.


I think this is because the sil3114 module was being loaded by hotplug 
later on in the boot process, long after disks are mounted.



to not add eth1394 I added this line into /etc/modprobe.conf :

install eth1394 /bin/true


I try to stay away from modprobe.conf because of the commemts at the top 
of the file, so I'm working with modutils.d and mkinitrd.conf


Marc DM


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Marc DM]

You can check the use count from lsmod:

Module  Size  Used by
nls_utf82432  0
it87   29472  0
i2c_sensor  3712  1 it87
i2c_isa 2688  0
i2c_dev12288  0
powernow_k811088  0
freq_table  5192  1 powernow_k8
 ^ This column

If the count is 0, it means the module is not being used, either because
the device it supplies isn't currently being used, or because you don't
need it.
  



This is useful info. Thanks.



Or with regards to the modules loaded in the initrd, change 
/etc/mkinitrd/mkinitrd.conf to have the line

   MODULES=dep
instead of =[all,most] and then regenerate the initrd.
 

I actually had to do this on a Tyan Thunder K8W (S2885) in order to get 
it to mount /home on the sata controller at boot (LVM over Raid1). It 
would always complain of not being able to find any md devices when 
loading lvm.


I think this is because the sil3114 module was being loaded by hotplug 
later on in the boot process, long after disks are mounted.



to not add eth1394 I added this line into /etc/modprobe.conf :


install eth1394 /bin/true



I try to stay away from modprobe.conf because of the commemts at the top 
of the file, so I'm working with modutils.d and mkinitrd.conf


Marc DM




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Lennart Sorensen]

On Wed, Sep 21, 2005 at 11:16:48AM -0700, Joel Johnson wrote:
> But that still doesn't get to the issue of how mkinitrd decides which IDE 
> modules should be included when set to only include dependent modules.
> 
> Any insights?

That would be controlled by mkinitrd's config.  You can actually
manually list excactly which modules to load if you want.  You can also
write a new better mkinitrd if you want (yaird is one other option for
example) and it sounds like mkinitrd may need replacing anyhow for
initramfs and no devfs in 2.6.13 for some setups in the future.

Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Joel Johnson]

On Wednesday 21 September 2005 11:03 am, Lennart Sorensen wrote:
> On Wed, Sep 21, 2005 at 10:22:00AM -0500, Mike Dobbs wrote:
> > to not add eth1394 I added this line into /etc/modprobe.conf :
> > install eth1394 /bin/true
>
> If you create /etc/modprobe.conf you disable the use of /etc/modprobe.d
> which is what debian uses.  This is not a good idea.
>
> Make your changes in a new file under /etc/modprobe.d instead.
>
> Also the common way that I know of ti disable someting is to either
> blacklist it from discover/hotplug (so you can still load it manually if
> you want to), or do 'alias eth1394 off' in a modprobe.d/ file.
>
> Len Sorensen

But that still doesn't get to the issue of how mkinitrd decides which IDE 
modules should be included when set to only include dependent modules.

Any insights?

Joel Johnson


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Lennart Sorensen]

On Wed, Sep 21, 2005 at 10:22:00AM -0500, Mike Dobbs wrote:
> to not add eth1394 I added this line into /etc/modprobe.conf :
> install eth1394 /bin/true

If you create /etc/modprobe.conf you disable the use of /etc/modprobe.d
which is what debian uses.  This is not a good idea.

Make your changes in a new file under /etc/modprobe.d instead.

Also the common way that I know of ti disable someting is to either
blacklist it from discover/hotplug (so you can still load it manually if
you want to), or do 'alias eth1394 off' in a modprobe.d/ file.

Len Sorensen


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Mike Dobbs]

to not add eth1394 I added this line into /etc/modprobe.conf :
install eth1394 /bin/true


On Wed, 2005-09-21 at 07:55 -0700, Joel Johnson wrote:
> On Wednesday 21 September 2005 12:53 am, Hamish Moffatt wrote:
> > On Wed, Sep 21, 2005 at 12:24:06AM -0500, Marc DM wrote:
> > > Stupid questions :
> > >
> > > How can I find out which modules my system actually needs and disable
> > > the ones I don't need.
> > >
> > > How can I know if a module I'm disabling at startup isn't needed for
> > > another module that I plan to load?
> > >
> > > Thanks. And I won't ask anymore stupid questions for the rest of the
> > > week.
> >
> > You can check the use count from lsmod:
> >
> > Module  Size  Used by
> > nls_utf82432  0
> > it87   29472  0
> > i2c_sensor  3712  1 it87
> > i2c_isa 2688  0
> > i2c_dev12288  0
> > powernow_k811088  0
> > freq_table  5192  1 powernow_k8
> >   ^ This column
> >
> > If the count is 0, it means the module is not being used, either because
> > the device it supplies isn't currently being used, or because you don't
> > need it.
> 
> Or with regards to the modules loaded in the initrd, 
> change /etc/mkinitrd/mkinitrd.conf to have the line
> MODULES=dep
> instead of =[all,most] and then regenerate the initrd.
> 
> That's what I've done anyway and it works great. One issue with newer kernels 
> is that is seems to load all IDE drivers regardless since they aren't yet 
> tagged as unloadable (shows [permanent] in lsmod). How can one indicate to 
> initrd (or yaird) selectively which ide modules to load?
> 
> Joel Johnson
> 
> 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Joel Johnson]

On Wednesday 21 September 2005 12:53 am, Hamish Moffatt wrote:
> On Wed, Sep 21, 2005 at 12:24:06AM -0500, Marc DM wrote:
> > Stupid questions :
> >
> > How can I find out which modules my system actually needs and disable
> > the ones I don't need.
> >
> > How can I know if a module I'm disabling at startup isn't needed for
> > another module that I plan to load?
> >
> > Thanks. And I won't ask anymore stupid questions for the rest of the
> > week.
>
> You can check the use count from lsmod:
>
> Module  Size  Used by
> nls_utf82432  0
> it87   29472  0
> i2c_sensor  3712  1 it87
> i2c_isa 2688  0
> i2c_dev12288  0
> powernow_k811088  0
> freq_table  5192  1 powernow_k8
>   ^ This column
>
> If the count is 0, it means the module is not being used, either because
> the device it supplies isn't currently being used, or because you don't
> need it.

Or with regards to the modules loaded in the initrd, 
change /etc/mkinitrd/mkinitrd.conf to have the line
MODULES=dep
instead of =[all,most] and then regenerate the initrd.

That's what I've done anyway and it works great. One issue with newer kernels 
is that is seems to load all IDE drivers regardless since they aren't yet 
tagged as unloadable (shows [permanent] in lsmod). How can one indicate to 
initrd (or yaird) selectively which ide modules to load?

Joel Johnson


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: remove unwanted modules

[Hamish Moffatt]

On Wed, Sep 21, 2005 at 12:24:06AM -0500, Marc DM wrote:
> Stupid questions :
> 
> How can I find out which modules my system actually needs and disable 
> the ones I don't need.
> 
> How can I know if a module I'm disabling at startup isn't needed for 
> another module that I plan to load?
> 
> Thanks. And I won't ask anymore stupid questions for the rest of the week.

You can check the use count from lsmod:

Module  Size  Used by
nls_utf82432  0
it87   29472  0
i2c_sensor  3712  1 it87
i2c_isa 2688  0
i2c_dev12288  0
powernow_k811088  0
freq_table  5192  1 powernow_k8
  ^ This column

If the count is 0, it means the module is not being used, either because
the device it supplies isn't currently being used, or because you don't
need it.


Hamish
-- 
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

remove unwanted modules

[Marc DM]

Stupid questions :

How can I find out which modules my system actually needs and disable 
the ones I don't need.


How can I know if a module I'm disabling at startup isn't needed for 
another module that I plan to load?


Thanks. And I won't ask anymore stupid questions for the rest of the week.

Regards,

Marc DM


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Correct modules list for MSI K8N Neo2 Platinum

[Rik Theys]

Hi,

I've been running kernel 2.6.11.12 (custom kernel) for a while. With this 
kernel, the amd74xx module loaded fine at boot so I had DMA for my PATA 
drives. I did this by adding the amd74xx module to /etc/mkinitrd/modules.

Since 2.6.12 (and the new udev) hit testing, I switched to Debian kernel 
2.6.12. With this kernel and "amd74xx" in /etc/mkinitrd/modules, the module 
is not correctly loaded at startup. I don't see the usual:

NFORCE3-250: IDE controller at PCI slot :00:08.0
NFORCE3-250: chipset revision 162
NFORCE3-250: not 100% native mode: will probe irqs later
NFORCE3-250: BIOS didn't set cable bits correctly. Enabling workaround.
NFORCE3-250: BIOS didn't set cable bits correctly. Enabling workaround.
NFORCE3-250: :00:08.0 (rev a2) UDMA133 controller
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:DMA
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:DMA, hdd:DMA


According to various usenet posts, it should work by 
keeping /etc/mkinitrd/modules empty and simple adding the following 
to /etc/modules:

amd74xx
ide-cd
ide-disk
ide-generic

By using this setting, the amd74xx module gets loaded OK, but then also the 
ide-scsi module gets loaded (although it's not mentioned in /etc/modules), 
causing my DVD drives to become scsi devices.

How can I prevent the ide-scsi module from loading?

Can somebody who also has the MSI K8N Neo2 Platinum motherboard and who is 
using 2.6.12 please post their /etc/modules and /etc/mkinitrd/modules?

Thanks

-- 
Rik Theys

-- 
Rik Theys


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Xavier Bestel]

Ok, I finally found a solution: I rebuilt the script/mod/modpost
executable and copied it over the one in the kernel-build directory, and
the installation went perfectly.

Thanks,

Xav



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Goswin von Brederlow]

Xavier Bestel <[EMAIL PROTECTED]> writes:

> On Fri, 2005-07-22 at 11:59, Jo Shields wrote:
>> I've personally had no difficulty whatsoever with VMware 5 natively on 
>> AMD64 Debian, including the building of kernel modules. Is your kernel 
>> built with the same version of GCC as "gcc -v" reports?
>
> Like the original poster, I'm using a 64bits kernel in a 32bits
> userspace environment. I have added -m64 to the kernel headers' Makefile
> CFLAGS to have gcc (which has the same version, as checked by
> vmware-config.pl) generate code for the correct architecture, but
> modpost fails in the same way.
>
>   Xav

The kernel-image-amd64 package overrides a few more things to get the
kernel building, like ld and as. You might have to do that too.

Personaly I recomend running cdebootstrap to build a 64bit chroot and
compile the stuff in there.

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Xavier Bestel]

On Fri, 2005-07-22 at 11:59, Jo Shields wrote:
> I've personally had no difficulty whatsoever with VMware 5 natively on 
> AMD64 Debian, including the building of kernel modules. Is your kernel 
> built with the same version of GCC as "gcc -v" reports?

Like the original poster, I'm using a 64bits kernel in a 32bits
userspace environment. I have added -m64 to the kernel headers' Makefile
CFLAGS to have gcc (which has the same version, as checked by
vmware-config.pl) generate code for the correct architecture, but
modpost fails in the same way.

Xav



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Jo Shields]

I've personally had no difficulty whatsoever with VMware 5 natively on 
AMD64 Debian, including the building of kernel modules. Is your kernel 
built with the same version of GCC as "gcc -v" reports?


Xavier Bestel wrote:


Hi,

 


Hello,

try vmware-any-any-update found at http://ftp.cvut.cz/vmware/
   



I tried the any-any-update93 to install vmware build 13124, but it still
fails with:

make: Entering directory `/tmp/vmware-config6/vmmon-only'
make -C /lib/modules/2.6.11-9-amd64-k8/build/include/.. SUBDIRS=$PWD 
SRCROOT=$PWD/. modules
make[1]: Entering directory `/usr/src/kernel-headers-2.6.11-9-amd64-k8'
 CC [M]  /tmp/vmware-config6/vmmon-only/linux/driver.o
 CC [M]  /tmp/vmware-config6/vmmon-only/linux/hostif.o
 CC [M]  /tmp/vmware-config6/vmmon-only/common/cpuid.o
 CC [M]  /tmp/vmware-config6/vmmon-only/common/hash.o
 CC [M]  /tmp/vmware-config6/vmmon-only/common/memtrack.o
 CC [M]  /tmp/vmware-config6/vmmon-only/common/phystrack.o
 CC [M]  /tmp/vmware-config6/vmmon-only/common/task.o
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for 
Ada/C/ObjC but not for C++
cc1plus: warning: command line option "-ffreestanding" is valid for C/ObjC but 
not for C++
cc1plus: warning: command line option "-Wdeclaration-after-statement" is valid 
for C/ObjC but not for C++
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for 
Ada/C/ObjC but not for C++
 CC [M]  /tmp/vmware-config6/vmmon-only/common/vmx86.o
 CC [M]  /tmp/vmware-config6/vmmon-only/vmcore/compat.o
 CC [M]  /tmp/vmware-config6/vmmon-only/vmcore/moduleloop.o
 LD [M]  /tmp/vmware-config6/vmmon-only/vmmon.o
 Building modules, stage 2.
 MODPOST
modpost: /tmp/vmware-config6/vmmon-only/vmmon.o no symtab?
/bin/sh: line 1: 11235 Aborted scripts/mod/modpost -i 
/usr/src/kernel-headers-2.6.11-9-amd64-k8/Module.symvers 
/tmp/vmware-config6/vmmon-only/vmmon.o
make[2]: *** [__modpost] Error 134


Any hint ?

Thanks,

Xav

PS: please Cc: me, I'm off-list. Sorry for Cc:ing you, I'm not sure it
would have reached you otherwise.



 




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Xavier Bestel]

Hi,

> Hello,
> 
> try vmware-any-any-update found at http://ftp.cvut.cz/vmware/

I tried the any-any-update93 to install vmware build 13124, but it still
fails with:

make: Entering directory `/tmp/vmware-config6/vmmon-only'
make -C /lib/modules/2.6.11-9-amd64-k8/build/include/.. SUBDIRS=$PWD 
SRCROOT=$PWD/. modules
make[1]: Entering directory `/usr/src/kernel-headers-2.6.11-9-amd64-k8'
  CC [M]  /tmp/vmware-config6/vmmon-only/linux/driver.o
  CC [M]  /tmp/vmware-config6/vmmon-only/linux/hostif.o
  CC [M]  /tmp/vmware-config6/vmmon-only/common/cpuid.o
  CC [M]  /tmp/vmware-config6/vmmon-only/common/hash.o
  CC [M]  /tmp/vmware-config6/vmmon-only/common/memtrack.o
  CC [M]  /tmp/vmware-config6/vmmon-only/common/phystrack.o
  CC [M]  /tmp/vmware-config6/vmmon-only/common/task.o
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for 
Ada/C/ObjC but not for C++
cc1plus: warning: command line option "-ffreestanding" is valid for C/ObjC but 
not for C++
cc1plus: warning: command line option "-Wdeclaration-after-statement" is valid 
for C/ObjC but not for C++
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for 
Ada/C/ObjC but not for C++
  CC [M]  /tmp/vmware-config6/vmmon-only/common/vmx86.o
  CC [M]  /tmp/vmware-config6/vmmon-only/vmcore/compat.o
  CC [M]  /tmp/vmware-config6/vmmon-only/vmcore/moduleloop.o
  LD [M]  /tmp/vmware-config6/vmmon-only/vmmon.o
  Building modules, stage 2.
  MODPOST
modpost: /tmp/vmware-config6/vmmon-only/vmmon.o no symtab?
/bin/sh: line 1: 11235 Aborted scripts/mod/modpost -i 
/usr/src/kernel-headers-2.6.11-9-amd64-k8/Module.symvers 
/tmp/vmware-config6/vmmon-only/vmmon.o
make[2]: *** [__modpost] Error 134


Any hint ?

Thanks,

Xav

PS: please Cc: me, I'm off-list. Sorry for Cc:ing you, I'm not sure it
would have reached you otherwise.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Must load trivial modules in /etc/modules of Sarge-64Bit

[Dirk Salva]

Hi,

in Sarge-64Bit its a little bit nerving now:
In my /etc/modules I have to register
rtc
ppp_generic
pcspkr

and *floppy* to load the modules and to install the devices although I
use udev and hotplug. I dont have to do so in Sarge-32Bit. Whats
that!?
Apart from that my Sarge-64Bit and Sarge-32Bit are absolutely
identically.

ciao, Dirk
-- 
|  Akkuschrauber Kaufberatung and AEG GSM stuff   |
|   Visit my homepage:   http://www.nutrimatic.ping.de/   |
| FIDO: Dirk Salva 2:244/6305.10 Internet: dsalvaATgmx.de |
|The "Ruhrgebiet", best place to live in Germany! |


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: can't load/unload modules

[Cyril Thévenet]

Le dimanche 01 mai 2005 à 12:32 +0200, Goswin von Brederlow a écrit :
> Cyril Thévenet <[EMAIL PROTECTED]> writes:
> 
> > Hi,
> >
> > I'm new on amd64 and I have some problems with modules. When I want to
> > load a module (for example nfsd) I get this message :
> >
> > FATAL: Error inserting nfsd
> > (/lib/modules/2.6.11-inde.14/kernel/fs/nfsd/nfsd.ko): Unknown symbol in
> > module, or unknown parameter (see dmesg)
> >
> > And the problem seems to the same for all the modules. The problem is
> > also the same with modprobe or modconf. And the dmesg only says :
> >
> > nfsd: Unknown parameter `nfsd'
> >
> > I'm using debian/sid with a self-compiled kernel from debian
> > kernel-source-2.6.8 .
> >
> > Thanks for your help
> > Cyril
> 
> Does it work with Debians kernel-image packages?

The problem is the same with pre-compiled kernel (I tried
kernel-image-2.6.11-9-amd64-k8).

> How did you build the kernel? make-kpkg?

$ fakeroot make-kpkg --initrd --append_to_version -inde.14 kernel_image




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: can't load/unload modules

[Goswin von Brederlow]

Cyril Thévenet <[EMAIL PROTECTED]> writes:

> Hi,
>
> I'm new on amd64 and I have some problems with modules. When I want to
> load a module (for example nfsd) I get this message :
>
> FATAL: Error inserting nfsd
> (/lib/modules/2.6.11-inde.14/kernel/fs/nfsd/nfsd.ko): Unknown symbol in
> module, or unknown parameter (see dmesg)
>
> And the problem seems to the same for all the modules. The problem is
> also the same with modprobe or modconf. And the dmesg only says :
>
> nfsd: Unknown parameter `nfsd'
>
> I'm using debian/sid with a self-compiled kernel from debian
> kernel-source-2.6.8 .
>
> Thanks for your help
> Cyril

Does it work with Debians kernel-image packages?
How did you build the kernel? make-kpkg?

MfG
Goswin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

can't load/unload modules

[Cyril Thévenet]

Hi,

I'm new on amd64 and I have some problems with modules. When I want to
load a module (for example nfsd) I get this message :

FATAL: Error inserting nfsd
(/lib/modules/2.6.11-inde.14/kernel/fs/nfsd/nfsd.ko): Unknown symbol in
module, or unknown parameter (see dmesg)

And the problem seems to the same for all the modules. The problem is
also the same with modprobe or modconf. And the dmesg only says :

nfsd: Unknown parameter `nfsd'

I'm using debian/sid with a self-compiled kernel from debian
kernel-source-2.6.8 .

Thanks for your help
Cyril



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Problem in the /lib/modules/ directory...

[Sønke Lorenzen]

On Sun, 2005-04-17 at 12:53 +0200, N-0-X (http://noxistes.org) wrote:
> Hello.
> It has been I while that I have tried to install debian on my amd64
> box. The installation seems very good, while 'on the boot cd', my
> ethernet connection is reconnized and activated by debian.
> But, for the first boot, it isn't reconized anymore.
> In fact, the problem lies in the /lib/modules/ directory.
> On my box, I also have a Mandrake 10.1 Community for amd64 which works
> farelly well, but, I would like to swicht to debian.
> Anyways, it has been a month now that i've had this problem. Some
> people suggested me to make an other installation of debian, which an
> other cd. So, that's what I did yesterday, but I have the exact same
> problem...
> 
> Here's what I have in my /lib/modules/ directory => "
> 2.6.8-10-amd64-k8 " (normal).
> But when I do a modprobe, here's the output:
> """
> 3Com:~# modprobe forcedeth
> FATAL: Could not load /lib/modules/2.6.8.1-12.5mdk/modules.dep: No
> such file or directory

This is probaply because you are running a mandrake kernel.
Does uname -r say 2.6.8.1-12.5mdk ?
Check your lilo/grub settings if that is the case.

Regards, Sønke


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Problem in the /lib/modules/ directory...

[N-0-X (http://noxistes.org)]

Hello.
It has been I while that I have tried to install debian on my amd64
box. The installation seems very good, while 'on the boot cd', my
ethernet connection is reconnized and activated by debian.
But, for the first boot, it isn't reconized anymore.
In fact, the problem lies in the /lib/modules/ directory.
On my box, I also have a Mandrake 10.1 Community for amd64 which works
farelly well, but, I would like to swicht to debian.
Anyways, it has been a month now that i've had this problem. Some
people suggested me to make an other installation of debian, which an
other cd. So, that's what I did yesterday, but I have the exact same
problem...

Here's what I have in my /lib/modules/ directory => "
2.6.8-10-amd64-k8 " (normal).
But when I do a modprobe, here's the output:
"""
3Com:~# modprobe forcedeth
FATAL: Could not load /lib/modules/2.6.8.1-12.5mdk/modules.dep: No
such file or directory
3Com:~# 
"""
So I don't get it... I used the forcedeth because that is what someone
(sorry, forgot your name...) in this mailing list who has the same
motherboard as mine, said that he had to use forcedeth so that his
ethernet connection is reconnized.
I need help beacause i'm impacient to have more then 667 commands
while being root... lol.

Thanks ;)

-- 
N-0-X, webmaster de http://www.noxistes.org

Re: vmware-modules on AMD64

[Markus Hirschmann]

Hi Seb
[EMAIL PROTECTED] wrote:
On Mon, Mar 07, 2005 at 02:48:53PM +0100, Markus Hirschmann wrote:
System is sarge with kernel-image 2.6.8-11 (-amd64-k8), the right
kernel-headers are installed.
There is also gcc-3.4 (which was used to compile the kernel) and symlink 
from /usr/bin/gcc.

I get the following error:
 CC [M]  /usr/src/modules/vmware/vmmon-only/linux/driver.o
/usr/src/modules/vmware/vmmon-only/linux/driver.c:1: error: code model
`kernel' not supported in the 32 bit mode
Try to increase the verbosity of the build by putting V=1 in the
make command that builds the modules.
I did, see below.
Thank you.
Markus
Output:
indik:/usr/src/modules/vmware# LC_ALL=C debian/rules binary-modules 
KSRC=/usr/src/kernel-headers-2.6.8-10-amd64-k8/
export DH_OPTIONS='-pvmware-modules-2.6.8-10-amd64-k8'
# Is this needed for setting up a Depends?
#echo "kpkg:Package-Version=4.5.2-8848-0fbriere.3+2.6.8-10-amd64-k8" \
#>> debian/vmware.substvars
echo "kpkg:Kernel-Version=2.6.8-10-amd64-k8" > \
debian/vmware-modules-2.6.8-10-amd64-k8.substvars
# The substvars mechanism seems slightly broken, hack to get around it
# stolen from the qce-ga package. Y!
rm -f debian/control
sed -e 's/${kpkg\:Kernel\-Version}/2.6.8-10-amd64-k8/' \
debian/control.in > debian/control
dh_testdir
dh_testroot
dh_clean -k
# Build the module
/usr/bin/make KSRC=/usr/src/kernel-headers-2.6.8-10-amd64-k8/ 
KVERS=2.6.8-10-amd64-k8
make[1]: Entering directory `/usr/src/modules/vmware'
rm -rf *.o *.ko
/usr/bin/make V=1 VM_UNAME=2.6.8-10-amd64-k8 
HEADER_DIR=/usr/src/kernel-headers-2.6.8-10-amd64-k8//include -C 
vmmon-only auto-build
Unable to find VMware installation database. Using 'vmware'.
Building for VMware Workstation 4.5.2 or VMware GSX Server 3.1.0.
Using 2.6.x kernel build system.
make[2]: Entering directory `/usr/src/modules/vmware/vmmon-only'
make -C /usr/src/kernel-headers-2.6.8-10-amd64-k8//include/.. 
SUBDIRS=$PWD SRCROOT=$PWD/. modules
make[3]: Entering directory `/usr/src/kernel-headers-2.6.8-10-amd64-k8'
mkdir -p /usr/src/modules/vmware/vmmon-only/.tmp_versions
make -f scripts/Makefile.build obj=/usr/src/modules/vmware/vmmon-only
  gcc -Wp,-MD,/usr/src/modules/vmware/vmmon-only/linux/.driver.o.d 
-nostdinc -iwithprefix include -D__KERNEL__ -Iinclude  -Wall 
-Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common 
-march=k8 -mno-red-zone -mcmodel=kernel -pipe -fno-reorder-blocks 
-Wno-sign-compare -fno-asynchronous-unwind-tables 
-O2-fomit-frame-pointer   -Wall -Wstrict-prototypes 
-DVME_DEFAULT=VME_V452 -DVMW_USING_KBUILD 
-I/usr/src/modules/vmware/vmmon-only/./include 
-I/usr/src/modules/vmware/vmmon-only/./common 
-I/usr/src/modules/vmware/vmmon-only/./linux 
-I/usr/src/modules/vmware/vmmon-only/./vmcore 
-DVMWARE__FIX_IO_APIC_BASE=FIX_IO_APIC_BASE-DVMW_HAVE_COMPILE_H 
-DMODULE -DKBUILD_BASENAME=driver -DKBUILD_MODNAME=vmmon -c -o 
/usr/src/modules/vmware/vmmon-only/linux/.tmp_driver.o 
/usr/src/modules/vmware/vmmon-only/linux/driver.c
/usr/src/modules/vmware/vmmon-only/linux/driver.c:1: error: code model 
`kernel'not supported in the 32 bit mode
make[4]: *** [/usr/src/modules/vmware/vmmon-only/linux/driver.o] Error 1
make[3]: *** [_module_/usr/src/modules/vmware/vmmon-only] Error 2
make[3]: Leaving directory `/usr/src/kernel-headers-2.6.8-10-amd64-k8'
make[2]: *** [vmmon.ko] Error 2
make[2]: Leaving directory `/usr/src/modules/vmware/vmmon-only'
make[1]: *** [build] Error 2
make[1]: Leaving directory `/usr/src/modules/vmware'
make: *** [binary-modules] Error 2
indik:/usr/src/modules/vmware#

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: vmware-modules on AMD64

[seb]

On Mon, Mar 07, 2005 at 02:48:53PM +0100, Markus Hirschmann wrote:
> Hi
> 
> I have a big Problem I cannot resolve. I use vmware from
> 
> deb http://www.fbriere.net/debian/dists/testing vmware/
> 
> System is sarge with kernel-image 2.6.8-11 (-amd64-k8), the right
> kernel-headers are installed.
> 
> There is also gcc-3.4 (which was used to compile the kernel) and symlink 
> from /usr/bin/gcc.
> 
> I get the following error:
> 
>   CC [M]  /usr/src/modules/vmware/vmmon-only/linux/driver.o
> /usr/src/modules/vmware/vmmon-only/linux/driver.c:1: error: code model
> `kernel' not supported in the 32 bit mode
> 
> vmware-any-any-patches tries, no success.
> 
> Any suggestions out there? Can't I compile modules for this kernel on 
> official sarge snapshot? Has anybody module packages oder modules for me?
> 
> Thanks
> 

Try to increase the verbosity of the build by putting V=1 in the
make command that builds the modules.

Seb


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

vmware-modules on AMD64

[Markus Hirschmann]

Hi
I have a big Problem I cannot resolve. I use vmware from
deb http://www.fbriere.net/debian/dists/testing vmware/
System is sarge with kernel-image 2.6.8-11 (-amd64-k8), the right
kernel-headers are installed.
There is also gcc-3.4 (which was used to compile the kernel) and symlink 
from /usr/bin/gcc.

I get the following error:
  CC [M]  /usr/src/modules/vmware/vmmon-only/linux/driver.o
/usr/src/modules/vmware/vmmon-only/linux/driver.c:1: error: code model
`kernel' not supported in the 32 bit mode
vmware-any-any-patches tries, no success.
Any suggestions out there? Can't I compile modules for this kernel on 
official sarge snapshot? Has anybody module packages oder modules for me?

Thanks
Greetings
Markus
indik:/usr/src/modules/vmware# LC_ALL=C debian/rules binary-modules
KSRC=/usr/src/kernel-headers-2.6.8-10-amd64-k8/
export DH_OPTIONS='-pvmware-modules-2.6.8-10-amd64-k8'
# Is this needed for setting up a Depends?
#echo "kpkg:Package-Version=4.5.2-8848-0fbriere.3+2.6.8-10-amd64-k8" \
#>> debian/vmware.substvars
echo "kpkg:Kernel-Version=2.6.8-10-amd64-k8" > \
debian/vmware-modules-2.6.8-10-amd64-k8.substvars
# The substvars mechanism seems slightly broken, hack to get around it
# stolen from the qce-ga package. Y!
rm -f debian/control
sed -e 's/${kpkg\:Kernel\-Version}/2.6.8-10-amd64-k8/' \
debian/control.in > debian/control
dh_testdir
dh_testroot
dh_clean -k
# Build the module
/usr/bin/make KSRC=/usr/src/kernel-headers-2.6.8-10-amd64-k8/
KVERS=2.6.8-10-amd64-k8
make[1]: Entering directory `/usr/src/modules/vmware'
rm -rf *.o *.ko
/usr/bin/make VM_UNAME=2.6.8-10-amd64-k8
HEADER_DIR=/usr/src/kernel-headers-2.6.8-10-amd64-k8//include -C
vmmon-only auto-build
Unable to find VMware installation database. Using 'vmware'.
Building for VMware Workstation 4.5.2 or VMware GSX Server 3.1.0.
Using 2.6.x kernel build system.
make[2]: Entering directory `/usr/src/modules/vmware/vmmon-only'
make -C /usr/src/kernel-headers-2.6.8-10-amd64-k8//include/..
SUBDIRS=$PWD SRCROOT=$PWD/. modules
make[3]: Entering directory `/usr/src/kernel-headers-2.6.8-10-amd64-k8'
  CC [M]  /usr/src/modules/vmware/vmmon-only/linux/driver.o
/usr/src/modules/vmware/vmmon-only/linux/driver.c:1: error: code model
`kernel' not supported in the 32 bit mode
make[4]: *** [/usr/src/modules/vmware/vmmon-only/linux/driver.o] Error 1
make[3]: *** [_module_/usr/src/modules/vmware/vmmon-only] Error 2
make[3]: Leaving directory `/usr/src/kernel-headers-2.6.8-10-amd64-k8'
make[2]: *** [vmmon.ko] Error 2
make[2]: Leaving directory `/usr/src/modules/vmware/vmmon-only'
make[1]: *** [build] Error 2
make[1]: Leaving directory `/usr/src/modules/vmware'
make: *** [binary-modules] Error 2
indik:/usr/src/modules/vmware#
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Missing pam_userdb.so from libpam-modules

[Kurt Roeckx]

On Fri, Jan 21, 2005 at 06:39:01PM +0100, Kurt Roeckx wrote:
> Andreas Bath is working on all db versions to get them working
> properly.  I expect him to upload any day now.  I'd just wait for
> him.

That's Andreas Barth of course.


Kurt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Missing pam_userdb.so from libpam-modules

[Kurt Roeckx]

On Fri, Jan 21, 2005 at 11:47:27AM +0100, Andreas Jochens wrote:
> 
> did you try to rebuild 'db3' in pure64 recently with the
> 'export LD_ASSUME_KERNEL=2.4.24' line uncommented? 
> In the gcc-3.4 archive this does not work,
> which is the reason why I commented it out.
> 
> I would be interested if 'db3' compiles in pure64 without this change.

It does not build without a patch.  See my latest patch for db4.1
for instance to see what needs to be changed.  It basicly comes
down to adding the fast mutexes for x86_64 and disabling
detection of the posix mutexes.

Andreas Bath is working on all db versions to get them working
properly.  I expect him to upload any day now.  I'd just wait for
him.


Kurt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Missing pam_userdb.so from libpam-modules

[Andreas Jochens]

On 05-Jan-20 20:59, Kurt Roeckx wrote:
> This seems to bug in the libdb3 package on gcc-3.4 archive.  The
> libdb3 maintainers used a trick to prevent it from linking to
> libpthread which the gcc-3.4 archive undid.
> 
> > The same test works fine on i386 Sid.
> 
> It also seems to work on pure64.
> 
> > I noticed that in db3's debian/rules there is a commented-out line:
> > #export LD_ASSUME_KERNEL=2.4.24
> 
> Yes, that's the trick I'm talking about above.

Hello Kurt,

did you try to rebuild 'db3' in pure64 recently with the
'export LD_ASSUME_KERNEL=2.4.24' line uncommented? 
In the gcc-3.4 archive this does not work,
which is the reason why I commented it out.

I would be interested if 'db3' compiles in pure64 without this change.

Regards
Andreas Jochens


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Missing pam_userdb.so from libpam-modules

[Kurt Roeckx]

On Thu, Jan 20, 2005 at 08:42:45PM +0100, Michal Schmidt wrote:
> 
> $ gcc -ldb3 db3test.c  #this is what scares the pam ./configure script:
> /usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
> reference to `pthread_condattr_setpshared'

This seems to bug in the libdb3 package on gcc-3.4 archive.  The
libdb3 maintainers used a trick to prevent it from linking to
libpthread which the gcc-3.4 archive undid.

> The same test works fine on i386 Sid.

It also seems to work on pure64.

> I noticed that in db3's debian/rules there is a commented-out line:
> #export LD_ASSUME_KERNEL=2.4.24

Yes, that's the trick I'm talking about above.

The libdb people are working on fixing this properly for all
there versions (db2, db3, db4.1, db4.2), db4.0 is going to get
removed and db4.3 should be fine.


Kurt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Missing pam_userdb.so from libpam-modules

[Michal Schmidt]

Hello,
The package libpam-modules (version 0.76-22.0.0.1.amd64 from gcc-3.4) 
doesn't contain pam_userdb.so. This PAM plugin is present in the i386 
Sid version of the package.

Compiling pam from the source package shows:
...
make[3]: Entering directory 
`/home/michich/src/pam-0.76/Linux-PAM/modules/pam_userdb'
This module will not be compiled on this system
...

The reason for this is the ./configure script decides that there is no 
Berkeley database available although I have libdb3-dev installed.

The problem can be shown on compilation of a simple test program:
/* db3test.c */
int main()
{ return 0; }
$ gcc -ldb3 db3test.c  #this is what scares the pam ./configure script:
/usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
reference to `pthread_condattr_setpshared'
/usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
reference to `pthread_mutexattr_destroy'
/usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
reference to `pthread_mutexattr_setpshared'
/usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
reference to `pthread_mutexattr_init'
/usr/lib/gcc/x86_64-linux/3.4.4/../../../../lib/libdb3.so: undefined 
reference to `pthread_mutex_trylock'
collect2: ld returned 1 exit status

The same test works fine on i386 Sid.
It can be made to compile on my amd64 system with:
$ gcc -ldb3 -lpthread db3test.c
So either the pam's ./configure needs to add "-lpthread" too or
(more likely) db3 needs to be fixed somehow.
I noticed that in db3's debian/rules there is a commented-out line:
#export LD_ASSUME_KERNEL=2.4.24
In the i386 Sid pam source package this export is present.
I think this export was there for a reason and it is related to the problem:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=223142
Regards
Michal
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Frederik Schueler]

Hi,

On Sat, Oct 16, 2004 at 10:39:25AM +0200, Frederik Schueler wrote:
> http://athlon.lowpingbastards.de/~freddy/debian-amd64-howto.html#id258694

Sorry, this must be

https://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html#id258694

Greetings
Frederik Schueler

-- 
ENOSIG


pgpI4oWhk5JSz.pgp
Description: PGP signature

Re: Compiling third-party kernel modules for amd64 kernel on Debian/386

[Frederik Schueler]

Hello,

try vmware-any-any-update found at http://ftp.cvut.cz/vmware/
or see the faq: 

http://athlon.lowpingbastards.de/~freddy/debian-amd64-howto.html#id258694

Greetings
Frederik Schueler

-- 
ENOSIG


pgpMhd8wISp8r.pgp
Description: PGP signature

Compiling third-party kernel modules for amd64 kernel on Debian/386

[Dave Carrigan]

Hi,

I just installed kernel-image-2.6.8-9-amd64-k8 and
kernel-headers-2.6.8-9-amd64-k8 from Debian/unstable. I also have gcc
3.4 installed and made sure my /usr/bin/gcc was pointing to gcc-3.4 (and
the same for g++).

Now I want to compile the VMWare modules:


# vmware-config.pl
  ... snip ...
Building for VMware Workstation 4.5.2 or VMware GSX Server 3.1.0.
Using 2.6.x kernel build system.
make: Entering directory `/tmp/vmware-config7/vmmon-only'
make -C /usr/src/kernel-headers-2.6.8-9-amd64-k8/include/.. SUBDIRS=$PWD 
SRCROOT=$PWD/. modules
make[1]: Entering directory `/usr/src/kernel-headers-2.6.8-9-amd64-k8'
  CC [M]  /tmp/vmware-config7/vmmon-only/linux/driver.o
/tmp/vmware-config7/vmmon-only/linux/driver.c:1: error: code model `kernel' not 
supportedin the 32 bit mode


So I edited /usr/src/kernel-headers-2.6.8-9-amd64-k8/Makefile and added
-m64 to the CFLAGS and retried the build:


# vmware-config.pl
  ... snip ...
make[1]: Entering directory `/usr/src/kernel-headers-2.6.8-9-amd64-k8'
  CC [M]  /tmp/vmware-config8/vmmon-only/linux/driver.o
  CC [M]  /tmp/vmware-config8/vmmon-only/linux/hostif.o
  CC [M]  /tmp/vmware-config8/vmmon-only/common/cpuid.o
  ... snip ...
  LD [M]  /tmp/vmware-config8/vmmon-only/vmmon.o
  Building modules, stage 2.
  MODPOST
modpost: /tmp/vmware-config8/vmmon-only/vmmon.o no symtab?
/bin/sh: line 1: 17532 Aborted (core dumped) 
scripts/mod/modpost -i /usr/src/kernel-headers-2.6.8-9-amd64-k8/Module.symvers 
/tmp/vmware-config8/vmmon-only/vmmon.o
make[2]: *** [__modpost] Error 134
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/usr/src/kernel-headers-2.6.8-9-amd64-k8'
make: *** [vmmon.ko] Error 2
make: Leaving directory `/tmp/vmware-config8/vmmon-only'
Unable to build the vmmon module.


I am having similar problems building the NVIDIA drivers, and I am at a
loss how to proceed. Any suggestions?

-- 
Dave Carrigan
Seattle, WA, USA
[EMAIL PROTECTED] | http://www.rudedog.org/ | ICQ:161669680
UNIX-Apache-Perl-Linux-Firewalls-LDAP-C-C++-DNS-PalmOS-PostgreSQL-MySQL

Dave is currently listening to Andy White - All Of The Thing I Can Bring 
(Teenage)


signature.asc
Description: Digital signature

Re: modules

[Peter Nelson]

was was wrote:
Hi, I have a problem with the modutils, when I use
modconf or update-modules the shell show:
Architecture-specific modutils configuration not
found, using defaults.
What´s wrong?
 

It's because /etc/modutils/arch only has i386, no x86_64 in it.  As far 
as I can tell the error is harmless since 2.6 (which I assume you're 
using) uses /etc/modprobe.d anyway.  If you really care though you can 
copy (or symlink) i386 over to x86_64.

-Peter

modules

[was was]

Hi, I have a problem with the modutils, when I use
modconf or update-modules the shell show:

Architecture-specific modutils configuration not
found, using defaults.

What´s wrong?



__
Renovamos el Correo Yahoo!: ¡100 MB GRATIS!
Nuevos servicios, más seguridad
http://correo.yahoo.es

Re: Sound modules problème with kde !

[Raul Miller]

On Wed, Sep 15, 2004 at 01:11:28PM +0200, Ludovic watteaux wrote:
> modprobe snd-via82xx

Try
snd-via82xx dxs_support=1

-- 
Raul

Sound modules problème with kde !

[Ludovic watteaux]

Hello !

I've some trouble to use correctly my via82xx sound chip !
I use the module snd-via82xx but i 'm force to make a script launch in /etc/init.d/snd-via82xx  -->  /etc/rc2.d/S90snd-via82xx 
Otherwise juste with a line with snd-via82xx in /etc/modules : Sound is with crackle noise  !
__
#!/bin/sh
set -e
PATH=/usr/local/sbin:/sbin:/usr/sbin:/usr/local/bin:/bin:/usr/bin
modprobe snd-via82xx
modprobe -r snd-via82xx
modprobe snd-via82xx
___
My dmesg :

via82xx: Assuming DXS channels with 48k fixed sample rate.
 Please try dxs_support=1 or dxs_support=4 option
 and report if it works on your machine.
ACPI: PCI interrupt :00:11.5[C] -> GSI 22 (level, low) -> IRQ 22


This work fine with xmms in xfce Desktop : The sound is good.

But if a launch kde with or without kde audio server : 
snd_seq & snd_pcm_oss   are launch. And again the sound is crackle with oss and alsa output option in xmms. 
___
My /etc/modutils/alsa-base :

above snd-pcm snd-pcm-oss
above snd-mixer snd-mixer-oss
above snd-seq snd-seq-oss


With Pure Data i try oss and alsa i've this :
error: OSS: SNDCTL_DSP_CHANNELS failed /dev/dsp
ALSA lib pcm_hw.c:1155:(snd_pcm_hw_open) open /dev/snd/pcmC0D0c failed: Device or resource busy

I am dispair ... 

I use my own kernels 2.6.7 and 2.6.8-1 and it's the same.

Please help me!
Thanks !

Greetings
Ludovic

Re: Prob with modules 2.6.7

[Jérôme Warnier]

Le lun 09/08/2004 à 13:07, Ludovic watteaux a écrit :
> Hi Frederik,
> 
> > do you have hotplug installed? this might load all those modules for
> > you.
> 
> Yes Hotplug is install and launch the modules. 
> For now i turn it off and put the good modules in the /etc/modules
> It's ok now evrything work fine.
Why didn't you just blacklist them by putting something in
/etc/hotplug/blacklist.d (or /etc/hotplug/blacklist, which I would not
recommend)?
I think using hotplug is now standard way of doing things.

> Thanks a lot. 
> 
> Greetings.
> 
> Ludovic Watteaux
-- 
Jérôme Warnier
Consultant
BeezNest
http://beeznest.net

Re: Prob with modules 2.6.7

[Ludovic watteaux]

Hi Frederik,

> do you have hotplug installed? this might load all those modules for
> you.

Yes Hotplug is install and launch the modules. 
For now i turn it off and put the good modules in the /etc/modules
It's ok now evrything work fine.

Thanks a lot. 

Greetings.

Ludovic Watteaux

Re: Prob with modules 2.6.7

[Frederik Schueler]

Hi Lud,

On Sun, Aug 08, 2004 at 03:18:18PM +0200, Lud Professor wrote:
> discover and discover1 is not install , 

do you have hotplug installed? this might load all those modules for
you.

Greetings
Frederik Schueler

-- 
ENOSIG

Prob with modules 2.6.7

[Lud Professor]

Hi evrybody !

I'm looking for to know why i've modules charged that is not on 
my /etc/modules /etc/modules.conf  , /etc/modutils/.. , /etc/modprobe.d/.. ,  
/etc/modprobe.conf 

A exemple for my sound card with snd-via82xx modules
i must to unload module and reload it for it work fine if not it craking.
For my network with the b44 modules for my network card and prism54 for the 
wireless card. I can't load b44 in first to be eth0 and prism54 to by eth1.

discover and discover1 is not install , 

in /etc/rcS.d/ i've S20module-init-tools and S20modutils

I can unload modules b44, snd-via82xx and prism54 by modprobe -r 
b44 ...compiled in module.

What's happening ? i don't know ! 
If anybody have any idea ...

Thanks.