4 Jul
2008
4 Jul
'08
12:05 p.m.
Hi,
I'm very much pist off about those real time kernels!!!
Why is there no decent debian realtime kernel? If they can build
thousands of kernels for Debian, why not one decent real time kernel?!?!?!?
I'm using now one of musix, but I can't build my nvidia driver on it.
Thats a little problem, but with my normal debian kernel I have wireless
internet connection with one click... but 6 of the 10 times on that
rt-kernel is failed :/ Why???
Maybe linuxaudio.org should put some 'pressure' on those distro
builders to include a decent realtime kernel!
...
Does someone have a good kernel for me (debian testing/ unstable)
/d
4 Jul
4 Jul
4:08 p.m.
I'm using now one of musix, but I can't build my nvidia driver on it.
In my experience, nvidia drivers are much worse and harder to get working
than the realtime kernel -- even the free software ones. I'd imagine that
specific problem has a lot more to do with those drivers than the rt kernel.
5:02 p.m.
On Fri, Jul 4, 2008 at 11:08 AM, Peter Lutz <peter(a)gotsomeideas.com> wrote:
rt-sources, as supported by Ingo, Thomas, Steven, Daniel and others,
has *never* supported non-open source video drivers. You can use them
but real-time performance has never been guaranteed, they say it never
will be guaranteed, and don't bother reporting rt-problems if you use
one. The problem with the closed source drivers is there is no way to
look at the code to guarantee it's written in a real-time safe manner.
Therefore you use them at your own risk in terms of real-time
operation.
- Mark
I'm using
now one of musix, but I can't build my nvidia driver on it.
In my experience, nvidia drivers are much worse and harder to get working
than the realtime kernel -- even the free software ones. I'd imagine that
specific problem has a lot more to do with those drivers than the rt kernel. 
6:53 p.m.
I think your frustration is due to a little misunderstanding. If the
inability to install the nvidia driver is cause for this dissatisfaction,
then rest assured that it's not the packager's fault. Installing kernel
modules means placing the files in the correct directory.
A real-time kernel, or any kernel on a distro _will_ have some sort of
"extraver" or "localver". It's the extra characters after the
actual linux
version, sometimes/some distros including the -rc or upstream patch version.
Confusing? No worries. Examples galore:
$ uname -r
2.6.25-rt-sidux
$uname -r
2.6.25.4-rt21
Whatever it is, you have to make sure the nvidia stuff go into whatever the
correct -rt kernel directory is in /lib/modules ;) The following are the
steps similar to what a particular distro buildscript takes:
bash NVIDIA-Linux*pkg0.run --extract-only
cd NVIDIA-Linux*pkg0
cd usr/src/nv/
ln -s Makefile.kbuild Makefile
make SYSSRC=/lib/modules/$thelibmoduledirasshownabove/build module
install -m644 nvidia.ko /lib/modules/$asabove/kernel/drivers/video/
depmod -v $asbove
rmmod nvidia
CTRL+ALT+BACKSPACE
Maybe there's a command you can use try run a help on the proprietary
executable. Then, there are other packages like ndiswrapper - and oh yeah
catalyst who can forget that one - that may need a little hack or two on the
rt-patched kernel. A checklist:
1) Upstream kernel
2) Upstream kernel patchset
3) RT kernel patchset
4) Apply whatever hacks needed (so far only catalyst/fglrx needs one)
5) Install proprietary module to correct directory
6) Good luck and enjoy!
As for performance, I've always been having better with older kernels as
newer ones roll out. I'm currently on 2.6.23.1-rt11 :) Realtime performance
actually is no longer a problem. It used to be a problem sometime ago, but
after migration to pam security and all that it's become a breeze. On
Windows recording with my laptop's Intel HDA is a no-go @ > 500ms latency +
ASIO-4-All, but on Linux I get 20+-.
6 Jul
6 Jul
10:45 a.m.
Hi,
Since over three years jacklab try to find out what is the best
supported rt-kernel on opensuse. The early self compiled 2.6.x
rt-kernels broke a lot of drivers and it was not easy for users to
extend them with kernel modules or drivers (for nvidia etc) and often
the audio priority was a showstopper (xruns!). We decided to rebuild an
own rt-kernel (incl all suse patches) for JAD 1.0. (based on oS 10.2).
Since 10.3 opensuse offers a kernel-rt, but without the corresponding
kernel modules nor kernel-rt-source for build up a proprietary
nividia driver (needed for dualhead support) and not at least the "rtc
tick is to low" for Rosegarden.
The opensuse rt-kernel is not for musicians, it is for SLERT, an
enterprise realtime linux. (btw: One of the former LAC founder and
synthgeek Matthias Nargoni is the Novell Manager for SLERT)
The rt-kernel for oS 11 also doesn't supported well by Suse, still no
sources. But the experienced Suser (free suse package builder)
Jan Engelhard (jengelh) now offers in his repository a good (mature and
fast) rt kernel with nivida-gfx and some kernel modules. He have an
open ear for the needs of musicians. With the 2.6.25.8 (patched with
rt6) opensuse 11 has a better audio realtime then ever. I can use half
of the buffers that I needed with the rt-kernels from the past. The rt
audio performance is comparable to 64studio (my personal reference
audio linux).
Packman offers most of the pro audio software we need, since
jacklab and packman joined forces by merging their repositories, and if
there is something missing (like wineasio or dssi-vst), you will find it
in the jacklab repo or in the 100 other repos around the opensuse build
server. http://forum.jacklab.net/viewtopic.php?f=25&t=794
The situation on opensuse Linux is pretty good for musicians at this
time :) The next time the jacklab devel Olli Bengs (Applelonkel) has
planed to release an installable live dvd named JAD 2.0, a remastered oS
11 for musicians. The jacklab community is very enthusiastic about this
release, because it looks very good, also the 64-bit version (oS
11 x86_64) works well on my test system.
have a nice day!
Michael
Am Fri, 04 Jul 2008 16:05:30 +0200
schrieb schoappied <schoappied(a)gmail.com>om>:
Hi,
I'm very much pist off about those real time kernels!!!
Why is there no decent debian realtime kernel? If they can build
thousands of kernels for Debian, why not one decent real time
kernel?!?!?!?
I'm using now one of musix, but I can't build my nvidia driver on it.
Thats a little problem, but with my normal debian kernel I have
wireless internet connection with one click... but 6 of the 10
times on that rt-kernel is failed :/ Why???
Maybe linuxaudio.org should put some 'pressure' on those distro
builders to include a decent realtime kernel!
...
Does someone have a good kernel for me (debian testing/ unstable)
/d
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
--
Michael Bohle
Vorsitzender
jacklab e.V.
D-27572 Bremerhaven | Am Baggerloch 20
Phone/Fax: +49-471-7004937
Mail: opendaw(a)jacklab.org
www: http://jacklab.org
Bürozeiten: Mo. - Fr. 12:00 - 18:00 Uhr
und nach Vereinbarung
12:17 p.m.
Michael Bohle wrote:
The situation on opensuse Linux is pretty good for
musicians at this
time... [snip]
Hi Michael,
I'm very happy to read this message. I've been pleased with JAD 1.0, but
I haven't had much time lately to meet with the crew on IRC and I've
dropped out of touch with what's happening there.
If anyone else has been thinking about trying JAD, give it a go. It's my
primary 32-bit system, I get excellent performance from it, and the
devel team has been responsive to problems users have encountered. It is
a fine system for musicians.
Best,
dp
7:54 p.m.
Dave Phillips wrote:
Michael Bohle wrote:
I keep having these errors:
when start booting:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
please type in the full path name to try again
or press ENTER to boot the system
later:
fsck.ext3 superblok could not write or read (hda3/ hda2...)
corrupt?
if it is a ext3 file, then maybe it is corrupt...
check /var/log/fsck/checkfs
the normal kernel 25.2 runs just nicely....
:(
The situation on opensuse Linux is pretty good
for musicians at this
time... [snip]
Hi Michael,
I'm very happy to read this message. I've been pleased with JAD 1.0, but
I haven't had much time lately to meet with the crew on IRC and I've
dropped out of touch with what's happening there.
If anyone else has been thinking about trying JAD, give it a go. It's my
primary 32-bit system, I get excellent performance from it, and the
devel team has been responsive to problems users have encountered. It is
a fine system for musicians.
Best,
dp
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
8:24 p.m.
2008/7/6 schoappied <schoappied(a)gmail.com>om>:
I keep having these errors:
when start booting:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
fsck.ext3 superblok could not write or read (hda3/ hda2...)
have you compiled the filesystem into the kernel (not as module)?
with
make menuconfig
it is somewhere in
..../filesystems/ext2 and ext3
8:47 p.m.
Emanuel Rumpf wrote:
2008/7/6 schoappied <schoappied(a)gmail.com>om>:
Thanks for your help.
I have/had Ext3 journalling file system support (EXT3_FS) as module...
http://img377.imageshack.us/img377/47/snapshot3qa8.png
this is my .config
http://pastebin.com/m1b07a30f
I keep having these errors:
when start booting:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
fsck.ext3 superblok could not write or read (hda3/ hda2...)
have you compiled the filesystem into the kernel (not as module)?
with
make menuconfig
it is somewhere in
..../filesystems/ext2 and ext3
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
7 Jul
7 Jul
1:50 p.m.
On Sun, Jul 6, 2008 at 3:46 PM, schoappied <schoappied(a)gmail.com> wrote:
Emanuel Rumpf wrote:
So if I'm understanding this correctly, and likely I'm not, then you
get through grub and then the kernel fails to understand your hard
drive. If you are using ext2/ext3 drive formatting then why are you
building ext3 support as a module? It's not large and is useful or
even needed during boot. I think you should consider building ext2/3
support into the kernel.
I've never seen the libgcrypt messages myself. Did you use some sort
of encrypted file system? Or is this because you're including some
encryption support in the kernel?
- Mark
2008/7/6 schoappied
<schoappied(a)gmail.com>om>:
Thanks for your help.
I have/had Ext3 journalling file system support (EXT3_FS) as module...
I keep having these errors:
when start booting:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
fsck.ext3 superblok could not write or read (hda3/ hda2...)
have you compiled the filesystem into the kernel (not as module)?
with
make menuconfig
it is somewhere in
..../filesystems/ext2 and ext3
_______________________________________________
6 Jul
6 Jul
9:04 p.m.
Emanuel Rumpf wrote:
2008/7/6 schoappied <schoappied(a)gmail.com>om>:
should I have this enabled? I seems weird to set it to No_HZ if further
on Hz is set to 1000 ... or do i misunderstand this item?
Tickless System (Dynamic Ticks) (NO_HZ)
This option enables a tickless system: timer interrupts will
only trigger on an as-needed basis both when the system is
busy and when the system is idle.
And how do I enable my rt2x00 ralink wireless?
I keep having these errors:
when start booting:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
fsck.ext3 superblok could not write or read (hda3/ hda2...)
have you compiled the filesystem into the kernel (not as module)?
with
make menuconfig
it is somewhere in
..../filesystems/ext2 and ext3
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
9:25 p.m.
Back on Sunday 06 July 2008, schoappied was like:
should I have this enabled? I seems weird to set it to No_HZ if further
on Hz is set to 1000 ... or do i misunderstand this item?
Tickless System (Dynamic Ticks) (NO_HZ)
This option enables a tickless system: timer interrupts will
only trigger on an as-needed basis both when the system is
busy and when the system is idle.
It's sort of like using a callback rather than polling to deal with
interrupts. (not literally, but the analogy works)
It's great for laptops because it helps preserve battery power. It _sounds_
like it might be a cool option for real time, but I've tried it and it caused
all kinds of problems with real time performance. (It also totally hosed
compiz)
-Reuben
7 Jul
7 Jul
1:34 a.m.
2008/7/6 schoappied <schoappied(a)gmail.com>om>:
resume: libgcrypt version: 1.4.1
resume: couldn't not stat the resume device file '/dev/hda3'
please type in the full path name to try again
or press ENTER to boot the system
I fixed this by recreating the initrd image in the /boot directory
with mkinitramfs like:
mkinitramfs initrd.img-2.6.25.8.rt7 2.6.25.8.rt7
(do: apt-get install initramfs-tools fist)
7:19 a.m.
Emanuel Rumpf wrote:
Thanks:
I did:
# mkinitramfs -o initrd.img-2.6.25.8-rt7 2.6.25.8-rt7
But the errors still remains:
resume: libgcrypt version: 1.4.1
mkinitramfs
initrd.img-2.6.25.8.rt7 2.6.25.8.rt7
sorry, I forgot the -o option:
cd /boot
mkinitramfs -o initrd.img-2.6.25.8.rt7 2.6.25.8.rt7
(replace the version strings with yours!)
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
resume: couldn't not stat the resume device file
'/dev/hda3'
please type in the full path name to try again
or press ENTER to boot the system
when I press enter, I get after a few seconds booting:
fsck.ext3: no such file or directory while trying to open /dev/hda2
/dev/hda2
The superblock could not be read or does not describe a correct ext2
filesystem. If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else) then the superblock
is corrupt, and you might try running e2fsk with an alternate superblock:
e2fsck -b 8193 <device>
fsck died with exit status 8
failed (code 8)
When login in kdm: Could not start kstartupconfig: check your installation.
1:31 p.m.
2008/7/7 schoappied <schoappied(a)gmail.com>om>:
I did:
# mkinitramfs -o initrd.img-2.6.25.8-rt7 2.6.25.8-rt7
But the errors still remains
hm .. maybe it was the change to fstab then, that
had an effect on it.
I discovered some problems with lines like this in /etc/fstab :
UUID=13495293458246835258345 / reiserfs ............etc.
things went better, when replaced with:
/dev/disk/by-uuid/13495293458246835258345 / reiserfs ............etc.
(directory names instead of "UUID=" (probably more compatible - who knows..))
2:26 p.m.
Emanuel Rumpf wrote:
2008/7/7 schoappied <schoappied(a)gmail.com>om>:
I'm glad I can use the 64studio kernel now and have wireless internet
working.
When looking at my fstab file, its looking fine to my, but I question if
the mount option for /hda3 is fine.... <none>
# /etc/fstab: static file system information.
#
# <file system> <mount point> <type> <options>
<dump> <pass>
proc /proc proc defaults 0 0
/dev/hda1 / ext3 errors=remount-ro 0 1
/dev/hda2 /home ext3 defaults 0 2
/dev/hda3 none swap sw 0 0
/dev/hdc /media/cdrom0 udf,iso9660 user,noauto 0 0
/dev/fd0 /media/floppy0 auto rw,user,noauto 0 0
I did:
# mkinitramfs -o initrd.img-2.6.25.8-rt7 2.6.25.8-rt7
But the errors still remains
hm .. maybe it was the change to fstab then, that
had an effect on it.
I discovered some problems with lines like this in /etc/fstab :
UUID=13495293458246835258345 / reiserfs ............etc.
things went better, when replaced with:
/dev/disk/by-uuid/13495293458246835258345 / reiserfs ............etc.
(directory names instead of "UUID=" (probably more compatible - who knows..))
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
3:05 p.m.
schoappied wrote:
Emanuel Rumpf wrote:
But how can it be that the normal kernel just works fine with this fstab?
I think the problem is more likely related to the .config file or
initrd.img-2.6.25.8-rt7 stuff or can it be that there's something wrong
in the settings/ .config for ATA drives... ??? But I must say, I'm not
very experienced in this kind of problem solving...
And why doesn't the system complain about the /hda1??
2008/7/7 schoappied
<schoappied(a)gmail.com>om>:
I'm glad I can use the 64studio kernel now and have wireless internet
working.
When looking at my fstab file, its looking fine to my, but I question if
the mount option for /hda3 is fine.... <none>
# /etc/fstab: static file system information.
#
# <file system> <mount point> <type> <options>
<dump> <pass>
proc /proc proc defaults 0 0
/dev/hda1 / ext3 errors=remount-ro 0 1
/dev/hda2 /home ext3 defaults 0 2
/dev/hda3 none swap sw 0 0
/dev/hdc /media/cdrom0 udf,iso9660 user,noauto 0 0
/dev/fd0 /media/floppy0 auto rw,user,noauto 0 0
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
I did:
# mkinitramfs -o initrd.img-2.6.25.8-rt7 2.6.25.8-rt7
But the errors still remains
hm .. maybe it was the change to fstab then, that
had an effect on it.
I discovered some problems with lines like this in /etc/fstab :
UUID=13495293458246835258345 / reiserfs ............etc.
things went better, when replaced with:
/dev/disk/by-uuid/13495293458246835258345 / reiserfs ............etc.
(directory names instead of "UUID=" (probably more compatible - who knows..))
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
3:29 p.m.
2008/7/7 schoappied <schoappied(a)gmail.com>om>:
schoappied wrote:
>
> Emanuel Rumpf wrote:
>
>>
>> 2008/7/7 schoappied <schoappied(a)gmail.com>om>:
>>
>>>
>>> I did:
>>> # mkinitramfs -o initrd.img-2.6.25.8-rt7 2.6.25.8-rt7
>>>
>>> But the errors still remains
>>>
>>
>> hm .. maybe it was the change to fstab then, that
But how can it be that the normal kernel just works fine with this fstab?
I think the problem is more likely related to the .config file or
initrd.img-2.6.25.8-rt7 stuff or can it be that there's something wrong in
the settings/ .config for ATA drives... ??? But I must say, I'm not very
experienced in this kind of problem solving...
your fstab looks ok. using uuids is just a newly recommended option.
I also remember, that I updated the initramfs-tools package before
creating the initrd.
maybe it's a bug, that has been resolved in the most recent version.
( I got the update from debian sid repository)
My version is 0.92d
download:
http://packages.debian.org/sid/all/initramfs-tools/download
5985
days inactive
5988
days old
linux-audio-user@lists.linuxaudio.org
18 comments
8 participants
participants (8)
-
Dave Phillips -
Emanuel Rumpf -
Mark Knecht -
Michael Bohle -
Peter Lutz -
Ray Rashif -
Reuben Martin -
schoappied