31 May
2009
31 May
'09
10:47 p.m.
Hiho,
I've finally tried to make a realtime kernel for Debian, as posted on my site
here: http://www.nescivi.nl/?p=111 (and also below, but without cross-
references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
I also found it a bit weird that the Debian package building did not make the
initrd image.
sincerely,
Marije
----------------------
I’ve finally sat down and tried to make a realtime kernel for Debian, for my
64bit machine.
I followed the instructions I found in this post on the Debian forums.
First, I got the Debian sources for 2.6.29, from my repository (so using
synaptic).
It puts the sources in /usr/src.
Then I got the latest realtime patches from
http://www.kernel.org/pub/linux/kernel/projects/rt/
Then I made a dir in my home directory
mkdir kernel
cd kernel
cp /usr/src/linux-source-2.6.29.tar.gz .
tar -xf linux-source-2.6.29.tar.gz
mv linux-source-2.6.29 linux-2.6.29-source-rt
cd linux-2.6.29-source-rt
bzcat ../patch-2.6.29.4-rt16.bz2 |patch -p1
I added this to the Makefile:
EXTRAVERSION = -rt16
To start from my current kernel configuration
make oldconfig
Then I got a whole bunch of questions, most of which I answered taking the
default options. Only for the realtime-preemption question I chose option 4,
the realtime option.
I looked at the options with menuconfig, but didn’t really change anything
more there. The timer frequency mentioned in the post above seems to be
preceded by the dynamic ticks option, which was turned on.
make menuconfig
I had to unselect “staging” and the “comedi” drivers as these caused errors
during building, and according to some messages on LAU I don’t really need it.
Build a kernel the Debian way:
make-kpkg clean
time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
To also build a source package:
time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
kernel_source
This will take a while.
Your new kernel & headers now live in ../kernel Use dpkg -i to install (as
root).
dpkg -i ../linux-headers-*.deb ../linux-image-*.deb
So good so far… but the kernel didn’t boot yet…
and the initrd file was missing
So with the initramfs-tools from the debian repository,
I did (as root)
mkinitramfs -o /boot/initrd.img-2.6.29-rt16 2.6.29-rt16
which created the initrd image.
Also editing it appropriately in grub (/boot/grub/menu.lst) so that the boot
process knows about it.
31 May
31 May
11:42 p.m.
Hi Marije
Did you make the necessary edits in /etc/security/limits.conf ?
As in...
Add the following lines to the bottom of the file: (Must do this as root, or with sudo)
@audio - rtprio 100
@audio - nice -10
@audio - memlock 400000
then save it.
Now add your users to the audio group on this machine.
HTH
Norv
Need a Holiday? Win a $10,000 Holiday of your choice. Enter
now.http://us.lrd.yahoo.com/_ylc=X3oDMTJxN2x2ZmNpBF9zAzIwMjM2MTY2MTMEdG1fZG…
1 Jun
1 Jun
12:59 p.m.
On Sunday 31 May 2009 21:42:03 Norval Watson wrote:
Hi Marije
Did you make the necessary edits in /etc/security/limits.conf ?
As in...
Add the following lines to the bottom of the file: (Must do this as root,
or with sudo) @audio - rtprio 100
@audio - nice -10
@audio - memlock 400000
then save it.
Now add your users to the audio group on this machine.
Yes, that's been there since I first installed stuff on this machine.
sincerely,
Marije
7:40 a.m.
(resent as it seems the ML is not fond of the attachment)
Hi Marije,
I also build an rt-kernel (2.6.29.2-rt10) for my Thinkpad T43 just a
couple of weeks ago and I am very pleased to find it working well.
On Sun, May 31, 2009 at 08:47:43PM -0400, nescivi wrote:
I've finally tried to make a realtime kernel for
Debian, as posted on my site
here: http://www.nescivi.nl/?p=111 (and also below, but without cross-
references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
Well, I don't know what might be missing in your config, but I put my
config here <http://www.jawebada.de/bucket/config-2.6.29.2-rt10-T43> as
a reference. The only problem is my Atheros WLAN controller, which
causes periodic xruns. That is why I configured a separate audio
runlevel without WLAN and a minimized process list. On the other hand,
this is the first time that I can use an rt-kernel for daily use as WLAN
and even suspend-to-ram work flawlessly.
I also found it a bit weird that the Debian package
building did not make the
initrd image.
This might be related to a recent change in Debian's kernel-package (are
we talking about the unstable branch?). Check
/usr/share/doc/kernel-package/NEWS.Debian.gz and
/usr/share/doc/kernel-package/README.gz in particular:
"Note that you will have to arrange for the actual initrd creation to
take place by installting a script like
/usr/share/kerne-package/examples/etc/kernel/post{inst,rm}.d/yaird or,
alternately,
/usr/share/kerne-package/examples/etc/kernel/post{inst,rm}.d/initramfs
into the correspondung directories /etc/kernel/post{inst,rm}.d, since
the kernel-postinst does not arrange for the initramfs creatorto be
called. You can thuse select your own; initramfs-tools or yaird."
Cheers
Jan
1:01 p.m.
Hiho,
On Monday 01 June 2009 05:40:24 Jan Weil wrote:
Ah, ok.
I would have expected the option --init-rd to the packaging then to complain
at least...
sincerely,
Marije
(resent as it seems the ML is not fond of the
attachment)
Hi Marije,
I also build an rt-kernel (2.6.29.2-rt10) for my Thinkpad T43 just a
couple of weeks ago and I am very pleased to find it working well.
ok.
On Sun, May 31, 2009 at 08:47:43PM -0400, nescivi
wrote:
Hmm...
I'll try turning of the WLan...
I've finally tried to make a realtime kernel
for Debian, as posted on my
site here: http://www.nescivi.nl/?p=111 (and also below, but without
cross- references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
Well, I don't know what might be missing in your config, but I put my
config here <http://www.jawebada.de/bucket/config-2.6.29.2-rt10-T43> as
a reference. The only problem is my Atheros WLAN controller, which
causes periodic xruns. That is why I configured a separate audio
runlevel without WLAN and a minimized process list. On the other hand,
this is the first time that I can use an rt-kernel for daily use as WLAN
and even suspend-to-ram work flawlessly. I also found
it a bit weird that the Debian package building did not make
the initrd image.
This might be related to a recent change in Debian's kernel-package (are
we talking about the unstable branch?). Check
/usr/share/doc/kernel-package/NEWS.Debian.gz and
/usr/share/doc/kernel-package/README.gz in particular:
"Note that you will have to arrange for the actual initrd creation to
take place by installting a script like
/usr/share/kerne-package/examples/etc/kernel/post{inst,rm}.d/yaird or,
alternately,
/usr/share/kerne-package/examples/etc/kernel/post{inst,rm}.d/initramfs
into the correspondung directories /etc/kernel/post{inst,rm}.d, since
the kernel-postinst does not arrange for the initramfs creatorto be
called. You can thuse select your own; initramfs-tools or yaird."
1:45 p.m.
Hi again,
On Monday 01 June 2009 11:01:48 nescivi wrote:
Actually, haven't compiled that module at all yet, as the madwifi module has
to be compiled separately.
sincerely,
Marije
On Sun, May
31, 2009 at 08:47:43PM -0400, nescivi wrote:
Hmm...
I'll try turning of the WLan... I've finally tried to make a realtime kernel
for Debian, as posted on
my site here: http://www.nescivi.nl/?p=111 (and also below, but without
cross- references).
However, I'm still getting xruns in jackd, at rather uncritical
settings (large period sizes...).
So I'm wondering what else could be going wrong...
Well, I don't know what might be missing in your config, but I put my
config here <http://www.jawebada.de/bucket/config-2.6.29.2-rt10-T43> as
a reference. The only problem is my Atheros WLAN controller, which
causes periodic xruns. That is why I configured a separate audio
runlevel without WLAN and a minimized process list. On the other hand,
this is the first time that I can use an rt-kernel for daily use as WLAN
and even suspend-to-ram work flawlessly.
1:03 p.m.
Hiho,
some more on this...
I had a problem using my RME Multiface with this kernel... It saw the PCMCIA
card, but not the device. Tried my basic 2.6.29 kernel, that one had the same
problem.
2.6.26 had no problem with it.
sincerely,
Marije
On Sunday 31 May 2009 20:47:43 nescivi wrote:
Hiho,
I've finally tried to make a realtime kernel for Debian, as posted on my
site here: http://www.nescivi.nl/?p=111 (and also below, but without cross-
references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
I also found it a bit weird that the Debian package building did not make
the initrd image.
sincerely,
Marije
----------------------
I’ve finally sat down and tried to make a realtime kernel for Debian, for
my 64bit machine.
I followed the instructions I found in this post on the Debian forums.
First, I got the Debian sources for 2.6.29, from my repository (so using
synaptic).
It puts the sources in /usr/src.
Then I got the latest realtime patches from
http://www.kernel.org/pub/linux/kernel/projects/rt/
Then I made a dir in my home directory
mkdir kernel
cd kernel
cp /usr/src/linux-source-2.6.29.tar.gz .
tar -xf linux-source-2.6.29.tar.gz
mv linux-source-2.6.29 linux-2.6.29-source-rt
cd linux-2.6.29-source-rt
bzcat ../patch-2.6.29.4-rt16.bz2 |patch -p1
I added this to the Makefile:
EXTRAVERSION = -rt16
To start from my current kernel configuration
make oldconfig
Then I got a whole bunch of questions, most of which I answered taking the
default options. Only for the realtime-preemption question I chose option
4, the realtime option.
I looked at the options with menuconfig, but didn’t really change anything
more there. The timer frequency mentioned in the post above seems to be
preceded by the dynamic ticks option, which was turned on.
make menuconfig
I had to unselect “staging” and the “comedi” drivers as these caused errors
during building, and according to some messages on LAU I don’t really need
it. Build a kernel the Debian way:
make-kpkg clean
time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
To also build a source package:
time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
kernel_source
This will take a while.
Your new kernel & headers now live in ../kernel Use dpkg -i to install (as
root).
dpkg -i ../linux-headers-*.deb ../linux-image-*.deb
So good so far… but the kernel didn’t boot yet…
and the initrd file was missing
So with the initramfs-tools from the debian repository,
I did (as root)
mkinitramfs -o /boot/initrd.img-2.6.29-rt16 2.6.29-rt16
which created the initrd image.
Also editing it appropriately in grub (/boot/grub/menu.lst) so that the
boot process knows about it.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
1:16 p.m.
On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi wrote:
some more on this...
I had a problem using my RME Multiface with this kernel... It saw the PCMCIA
card, but not the device. Tried my basic 2.6.29 kernel, that one had the same
problem.
2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td22280851.html>
<http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
Jan
3:26 p.m.
On Monday 01 June 2009 11:16:35 Jan Weil wrote:
On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi
wrote:
Probably :)
I don't need to rebuild the whole kernel for this, right?
sincerely,
Marije
some more on this...
I had a problem using my RME Multiface with this kernel... It saw the
PCMCIA card, but not the device. Tried my basic 2.6.29 kernel, that one
had the same problem.
2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td22280
851.html> <http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
3:43 p.m.
On Mon, Jun 01, 2009 at 01:26:21PM -0400, nescivi wrote:
On Monday 01 June 2009 11:16:35 Jan Weil wrote:
Yes, if you built snd-hdsp as a module. Until recently, you would have
had to call 'make-kpkg clean' after patching, i.e., recompile the entire
kernel, which was a real PITA. Fortunately, this is no longer needed. A
long overdue change in my opinion.
On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi
wrote:
Probably :)
I don't need to rebuild the whole kernel for this, right? some more on this...
I had a problem using my RME Multiface with this kernel... It saw the
PCMCIA card, but not the device. Tried my basic 2.6.29 kernel, that one
had the same problem.
2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td22280
851.html> <http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
3:52 p.m.
On Monday 01 June 2009 13:43:22 Jan Weil wrote:
On Mon, Jun 01, 2009 at 01:26:21PM -0400, nescivi
wrote:
So which command do I use?
sincerely,
Marije
On Monday 01 June 2009 11:16:35 Jan Weil wrote:
Yes, if you built snd-hdsp as a module. Until recently, you would have
had to call 'make-kpkg clean' after patching, i.e., recompile the entire
kernel, which was a real PITA. Fortunately, this is no longer needed. A
long overdue change in my opinion. On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi
wrote:
Probably :)
I don't need to rebuild the whole kernel for this, right? some more on this...
I had a problem using my RME Multiface with this kernel... It saw the
PCMCIA card, but not the device. Tried my basic 2.6.29 kernel, that
one had the same problem.
2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td2
2280 851.html> <http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
4:04 p.m.
On Mon, Jun 01, 2009 at 01:51:41PM -0400, nescivi wrote:
On Monday 01 June 2009 13:43:22 Jan Weil wrote:
Just call make-kpkg again as before, something like
fakeroot make-kpkg --initrd --revision XYZ kernel_image
The kernel Makefile will catch your changes and only recompile what is
needed. If I understand correctly, the debian sub-directory will be
recreated on the fly. It should be safe to remove it, though, if the
above does not work out. dpkg will yell out some warnings when you
install the newly built version as it is not able to determine new
module dependencies. In this case, just recompiling a single module,
there shouldn't be a problem.
On Mon, Jun 01, 2009 at 01:26:21PM -0400, nescivi
wrote:
So which command do I use? On Monday 01 June 2009 11:16:35 Jan Weil wrote:
Yes, if you built snd-hdsp as a module. Until recently, you would have
had to call 'make-kpkg clean' after patching, i.e., recompile the entire
kernel, which was a real PITA. Fortunately, this is no longer needed. A
long overdue change in my opinion. On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi
wrote:
> some more on this...
> I had a problem using my RME Multiface with this kernel... It saw the
> PCMCIA card, but not the device. Tried my basic 2.6.29 kernel, that
> one had the same problem.
> 2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td2
2280 851.html> <http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
Probably :)
I don't need to rebuild the whole kernel for this, right?
4:52 p.m.
On Monday 01 June 2009 14:04:33 Jan Weil wrote:
On Mon, Jun 01, 2009 at 01:51:41PM -0400, nescivi
wrote:
ah, ok.
got it all built...
so now it's just seeing if it all works when I reboot...
got to go now though, so more later!
I updated my website/blog entry with the comments from this thread for future
reference.
sincerely,
Marije
On Monday 01 June 2009 13:43:22 Jan Weil wrote:
Just call make-kpkg again as before, something like
fakeroot make-kpkg --initrd --revision XYZ kernel_image
The kernel Makefile will catch your changes and only recompile what is
needed. If I understand correctly, the debian sub-directory will be
recreated on the fly. It should be safe to remove it, though, if the
above does not work out. dpkg will yell out some warnings when you
install the newly built version as it is not able to determine new
module dependencies. In this case, just recompiling a single module,
there shouldn't be a problem. On Mon, Jun 01, 2009 at 01:26:21PM -0400, nescivi
wrote:
So which command do I use? On Monday 01 June 2009 11:16:35 Jan Weil wrote:
> On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi wrote:
> > some more on this...
> > I had a problem using my RME Multiface with this kernel... It saw
> > the PCMCIA card, but not the device. Tried my basic 2.6.29
> > kernel, that one had the same problem.
> > 2.6.26 had no problem with it.
>
> yup, been there. Does this
> <http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1
>-td2 2280 851.html>
> <http://article.gmane.org/gmane.linux.alsa.devel/60411> help?
Probably :)
I don't need to rebuild the whole kernel for this, right?
Yes, if you built snd-hdsp as a module. Until recently, you would have
had to call 'make-kpkg clean' after patching, i.e., recompile the
entire kernel, which was a real PITA. Fortunately, this is no longer
needed. A long overdue change in my opinion. 
16 Jun
16 Jun
11:24 a.m.
hey,
just trying this method of compiling rt kernel....
i seem to be getting up to compile time
( ie time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
kernel_source )
and it all looks good til this error....
could someone help?
CC arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.o
LD arch/x86/kernel/cpu/cpufreq/built-in.o
CC arch/x86/kernel/cpu/mtrr/main.o
CC arch/x86/kernel/cpu/mtrr/if.o
CC arch/x86/kernel/cpu/mtrr/generic.o
arch/x86/kernel/cpu/mtrr/generic.c:311: error: redefinition of ‘mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:38: error: previous definition of
‘mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_str_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_str_mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_mtrr_debug’ was here
make[4]: *** [arch/x86/kernel/cpu/mtrr/generic.o] Error 1
make[3]: *** [arch/x86/kernel/cpu/mtrr] Error 2
make[2]: *** [arch/x86/kernel/cpu] Error 2
make[1]: *** [arch/x86/kernel] Error 2
make[1]: Leaving directory `/home/tom/kernel/linux-2.6.29-source-rt'
make: *** [debian/stamp/build/kernel] Error 2
real 1m19.333s
user 0m50.659s
sys 0m12.893s
On Tue, Jun 2, 2009 at 4:51 AM, nescivi <nescivi(a)gmail.com> wrote:
On Monday 01 June 2009 14:04:33 Jan Weil wrote:
Just call make-kpkg again as before, something like
fakeroot make-kpkg --initrd --revision XYZ kernel_image
The kernel Makefile will catch your changes and only recompile what is
needed. If I understand correctly, the debian sub-directory will be
recreated on the fly. It should be safe to remove it, though, if the
above does not work out. dpkg will yell out some warnings when you
install the newly built version as it is not able to determine new
module dependencies. In this case, just recompiling a single module,
there shouldn't be a problem.
ah, ok.
got it all built...
so now it's just seeing if it all works when I reboot...
got to go now though, so more later!
I updated my website/blog entry with the comments from this thread for
future
reference.
sincerely,
Marije
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
On Mon, Jun 01, 2009 at 01:51:41PM -0400, nescivi
wrote:
> On Monday 01 June 2009 13:43:22 Jan Weil wrote:
> > On Mon, Jun 01, 2009 at 01:26:21PM -0400, nescivi wrote:
> > > On Monday 01 June 2009 11:16:35 Jan Weil wrote:
> > > > On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi wrote:
> > > > > some more on this...
> > > > > I had a problem using my RME Multiface with this kernel... It
saw
> > > > > the PCMCIA card, but not
the device. Tried my basic 2.6.29
> > > > > kernel, that one had the same problem.
> > > > > 2.6.26 had no problem with it.
> > > >
> > > > yup, been there. Does this
> > > > <
http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1
> > > >-td2 2280 851.html>
> > > > <http://article.gmane.org/gmane.linux.alsa.devel/60411> help?
> > >
> > > Probably :)
> > > I don't need to rebuild the whole kernel for this, right?
> >
> > Yes, if you built snd-hdsp as a module. Until recently, you would
have
had to call 'make-kpkg clean' after patching,
i.e., recompile the
entire kernel, which was a real PITA. Fortunately, this is no longer
needed. A long overdue change in my opinion.
So which command do I use? 
11:36 a.m.
2009/6/16 Tom Dunstan <gateswideopen(a)gmail.com>om>:
hey,
just trying this method of compiling rt kernel....
i seem to be getting up to compile time
( ie time fakeroot make-kpkg --initrd -rev mz1 kernel_image kernel_headers
kernel_source )
and it all looks good til this error....
could someone help?
CC arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.o
LD arch/x86/kernel/cpu/cpufreq/built-in.o
CC arch/x86/kernel/cpu/mtrr/main.o
CC arch/x86/kernel/cpu/mtrr/if.o
CC arch/x86/kernel/cpu/mtrr/generic.o
arch/x86/kernel/cpu/mtrr/generic.c:311: error: redefinition of ‘mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:38: error: previous definition of
‘mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_str_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_str_mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_mtrr_debug’ was here
make[4]: *** [arch/x86/kernel/cpu/mtrr/generic.o] Error 1
make[3]: *** [arch/x86/kernel/cpu/mtrr] Error 2
make[2]: *** [arch/x86/kernel/cpu] Error 2
make[1]: *** [arch/x86/kernel] Error 2
make[1]: Leaving directory `/home/tom/kernel/linux-2.6.29-source-rt'
which distro (it seems ubuntu), sources, rt-patch are you using?
2.6.29.4-rt18 here
-r
11:43 a.m.
hi
Ubuntu jaunty
linux 2.6.29 from http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.29/
patch-2.6.29.4-rt19.bz2
<http://www.kernel.org/pub/linux/kernel/projects/rt/patch-2.6.29.4-rt19.bz2>
15-Jun-2009 09:39 949K
On Tue, Jun 16, 2009 at 11:36 PM, Raffaele Morelli <
raffaele.morelli(a)gmail.com> wrote:
2009/6/16 Tom Dunstan
<gateswideopen(a)gmail.com>om>:
hey,
just trying this method of compiling rt kernel....
i seem to be getting up to compile time
( ie time fakeroot make-kpkg --initrd -rev mz1 kernel_image
kernel_headers
kernel_source )
and it all looks good til this error....
could someone help?
CC arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.o
LD arch/x86/kernel/cpu/cpufreq/built-in.o
CC arch/x86/kernel/cpu/mtrr/main.o
CC arch/x86/kernel/cpu/mtrr/if.o
CC arch/x86/kernel/cpu/mtrr/generic.o
arch/x86/kernel/cpu/mtrr/generic.c:311: error: redefinition of
‘mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:38: error:
previous definition of
‘mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_str_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_str_mtrr_debug’ was here
arch/x86/kernel/cpu/mtrr/generic.c:316: error: redefinition of
‘__setup_mtrr_debug’
arch/x86/kernel/cpu/mtrr/generic.c:42: error: previous definition of
‘__setup_mtrr_debug’ was here
make[4]: *** [arch/x86/kernel/cpu/mtrr/generic.o] Error 1
make[3]: *** [arch/x86/kernel/cpu/mtrr] Error 2
make[2]: *** [arch/x86/kernel/cpu] Error 2
make[1]: *** [arch/x86/kernel] Error 2
make[1]: Leaving directory `/home/tom/kernel/linux-2.6.29-source-rt'
which distro (it seems ubuntu), sources, rt-patch are you using?
2.6.29.4-rt18 here
-r
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
11:51 a.m.
2009/6/16 Tom Dunstan <gateswideopen(a)gmail.com>om>:
hi
Ubuntu jaunty
linux 2.6.29 from http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.29/
patch-2.6.29.4-rt19.bz2 15-Jun-2009 09:39 949K
2.6.29 != 2.6.29.4
You must use vanilla sources 2.6.29.4
ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.29.4.tar.bz2
-r
1 Jun
1 Jun
3:54 p.m.
nescivi wrote:
On Monday 01 June 2009 11:16:35 Jan Weil wrote:
If the interface card is there, you don't have to recompile anything.
Look into the kernel log, what does it say about recognizing the io box?
In 2.6.28 they introduced a new power managment which might power up the
cardbus slot too late for the interface card to initialize in time. If
you run 'hdsploader' in user space after boot, and the card works, that
was the problem. And that is what Tim's patch fixes by increasing the
number of polling loops.
Flo
--
Machines can do the work, so people have time to think.
public key DA43FEF4 x-hkp://wwwkeys.eu.pgp.net
On Mon, Jun 01, 2009 at 11:03:49AM -0400, nescivi
wrote:
Probably :)
I don't need to rebuild the whole kernel for this, right? some more on this...
I had a problem using my RME Multiface with this kernel... It saw the
PCMCIA card, but not the device. Tried my basic 2.6.29 kernel, that one
had the same problem.
2.6.26 had no problem with it.
yup, been there. Does this
<http://www.nabble.com/kernel-2.6.29-rc6-no-working-rme-multiface-1-td22280
851.html> <http://article.gmane.org/gmane.linux.alsa.devel/60411>
help?
2:12 p.m.
On Sun, May 31, 2009 at 08:47:43PM -0400, nescivi wrote:
I've finally tried to make a realtime kernel for
Debian, as posted on my site
here: http://www.nescivi.nl/?p=111 (and also below, but without cross-
references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
Did you tune your interrupt priorities yet? I use Rui's rtirq
<http://www.rncbc.org/jack/>
along with this /etc/rtirq.conf
<http://www.jawebada.de/bucket/rtirq.conf>.
Jan
2:58 p.m.
On Monday 01 June 2009 12:12:50 Jan Weil wrote:
On Sun, May 31, 2009 at 08:47:43PM -0400, nescivi
wrote:
No, I didn't yet.
the script goes in /etc/init.d/
the conf into:
/etc/sysconfig/rtirq ?
there's no readme with the package :)
sincerely,
Marije
PS,
/usr/bin/jackd -R -p128 -t200 -dalsa -dhw:0 -r44100 -p1024 -n3
seems to be fine right now, but still have to do some actual audio with it...
I've finally tried to make a realtime kernel
for Debian, as posted on my
site here: http://www.nescivi.nl/?p=111 (and also below, but without
cross- references).
However, I'm still getting xruns in jackd, at rather uncritical settings
(large period sizes...).
So I'm wondering what else could be going wrong...
Did you tune your interrupt priorities yet? I use Rui's rtirq
<http://www.rncbc.org/jack/>
along with this /etc/rtirq.conf
<http://www.jawebada.de/bucket/rtirq.conf>.
3:18 p.m.
On Mon, Jun 01, 2009 at 12:58:14PM -0400, nescivi wrote:
Yes, assuming you are using standard init.d scripts, put it into
/etc/init.d and a symbolic link to your audio runlevel, e.g.,
/etc/rc3.d/S99rtirq -> ../init.d/rtirq
Did you tune
your interrupt priorities yet? I use Rui's rtirq
<http://www.rncbc.org/jack/>
along with this /etc/rtirq.conf
<http://www.jawebada.de/bucket/rtirq.conf>.
No, I didn't yet.
the script goes in /etc/init.d/ the conf into:
/etc/sysconfig/rtirq ?
/etc/rtirq.conf is more debian-ish, but both are fine.
5640
days inactive
5656
days old
linux-audio-user@lists.linuxaudio.org
20 comments
6 participants
participants (6)
-
Florian Faber -
Jan Weil -
nescivi -
Norval Watson -
Raffaele Morelli -
Tom Dunstan