Re: [LAU] Realtime Kernel

From: Arda Eden <ardaeden@email-addr-hidden>
Date: Wed Mar 19 2008 - 00:36:15 EET

Hey guys,
Have a look at this:
bowman@email-addr-hidden ~ $ uname -a
Linux melinda 2.6.24_rc8-100-rt #1 SMP PREEMPT RT Sun Jan 20 18:23:34 EET
2008 i686 mobile AMD Athlon(tm) XP-M 2200+ GNU/Linux

I'm really happy now. I did it.
Thanks to all.

On Tue, Mar 18, 2008 at 9:49 PM, Mark Knecht <markknecht@email-addr-hidden> wrote:

> There is really no reason that I know of that you cannot build this
> kernel yourself. It's not horribly difficult once you've done it a few
> times. It can be quite daunting when you're up to bat for the first
> time though. the trick is understanding what driver modules you are
> using and making sure all of those are included in the configuration
> of your new kernel. Once you get past that step it's just a matter of
> adding a new option in the grub config file and trying it out.
>
> From the RTWiki these are the commands to download and patch a kernel:
>
>
> # wget ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.23.1.tar.bz2
> # wget
> http://www.kernel.org/pub/linux/kernel/projects/rt/patch-2.6.23.1-rt11.bz2
> # tar xfj linux-2.6.23.1.tar.bz2
> # cd linux-2.6.23.1
> # bzcat2 ../patch-2.6.23.1-rt11.bz2 | patch -p1
>
> The above is for 2.6.23. If you were going to do it you might as well
> go for 2.6.24.
>
> Changing the kernel itself should not cause any changes in the
> stability of your applications. They really exist at a higher level
> and just talk to the kernel underneath it. If a new minor revision
> kernel upgrade breaks an application I think it would generally be
> considered a regression in the kernel and would get fixed. I have
> kernels ranging back to 2.6.14. they all work.
>
> Hope this helps,
> Mark
>
> On Tue, Mar 18, 2008 at 12:41 PM, Arda Eden <ardaeden@email-addr-hidden> wrote:
> > I'm a PARDUS (a Turkish GNU/Linux distro) user so that the latest kernel
> > provided by my distro developers is 2.6.18.8-86. I can't know why they
> don't
> > complile a new one but i think that it's about the stability of most
> > applications provided by thet distro.
> >
> > Which distro of Linux is most suitable for most linux audio applications
> ?
> >
> >
> >
> > On Tue, Mar 18, 2008 at 9:36 PM, Mark Knecht <markknecht@email-addr-hidden>
> wrote:
> >
> > > That kernel is getting pretty old. I'm running 2.6.24-rt1 here. you
> > > might want to check out the RT-Wiki or some of the sites for using RT
> > > kernels with audio apps.
> > >
> > >
> > > http://rt.wiki.kernel.org/index.php/Main_Page
> > >
> > > The above link gives instructions for patching a kernel to get to an
> RT
> > kernel.
> > >
> > > Hope this helps,
> > > Mark
> > >
> > >
> > >
> > >
> > > On Tue, Mar 18, 2008 at 12:02 PM, Arda Eden <ardaeden@email-addr-hidden>
> wrote:
> > > > Hmm,
> > > > So this is all about my kernel:
> > > > Linux melinda 2.6.18.8-86 #2 SMP Mon Feb 11 00:50:59 EET 2008 i686
> > mobile
> > > > AMD Athlon(tm) XP-M 2200+ GNU/Linux
> > > >
> > > > Is it possible to patch it in order to make it a RT kernel ?
> > > >
> > > >
> > > > On Tue, Mar 18, 2008 at 8:59 PM, Mark Knecht <markknecht@email-addr-hidden>
> > wrote:
> > > > >
> > > > > On Tue, Mar 18, 2008 at 11:41 AM, Arda Eden <ardaeden@email-addr-hidden>
> > wrote:
> > > > > > So what does a realtime kernel change ? Makes my latency 1 ms ?
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Tue, Mar 18, 2008 at 8:38 PM, Arnold Krille
> > <arnold@email-addr-hidden>
> > > > wrote:
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Am Dienstag, 18. März 2008 schrieb Mark Knecht:
> > > > > > >
> > > > > > > > On Tue, Mar 18, 2008 at 11:25 AM, Arda Eden <
> ardaeden@email-addr-hidden>
> > > > wrote:
> > > > > > > > > So is it really necessary to have a realtime kernel in
> order
> > to
> > > > use
> > > > > > linux
> > > > > > > > > audio applications without any latencies or xruns ?
> > > > > > > > It is not possible to use ANY kernel with NO latencies.
> EVERY
> > kernel
> > > > > > > > has latencies. The question is how low do you want the
> latencies
> > to
> > > > > > > > be? If you can exist with 50mS or higher you might get away
> with
> > a
> > > > > > > > standard kernel. If you want to run with 1.2mS latency then
> you
> > will
> > > > > > > > absolutely have to have a real-time enabled kernel.
> > > > > > >
> > > > > > > Still it has to be noted that you can achieve 5ms latency
> without
> > a
> > > > > > > RT-kernel... Even with an el-cheapo builtin soundcard...
> > > > > > >
> > > > > > > Arnold
> > > > >
> > > > > Absolutely. The most recent kernels are often quite good. However
> I
> > > > > would say that the difference between a standard kernel running
> 5mS
> > > > > and a RT kernel running 5mS is that the RT kernel is far more
> likely
> > > > > to meet the 5mS requirement under heavy system loads while the
> > > > > standard kernel may, at times, give a bit too much time to non-rt
> > > > > threads and not get back to your audio thread as fast as someone
> might
> > > > > want.
> > > > >
> > > > > If you want the highest reliability means of meeting your RT
> > > > > requirements all the time then use the RT kernel. If you are just
> > > > > doing home recording and can accept an occasional xrun (i.e. - a
> > > > > glitch in the recorded data) then stick with the standard kernel
> for
> > > > > as long as it works for you.
> > > > >
> > > > > Just my view,
> > > > > Mark
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Arda EDEN
> > > > Cumhuriyet University
> > > > Faculty of Fine Arts
> > > > Department of Music Technology
> > > > Sivas/TURKEY
> > >
> >
> >
> >
> > --
> >
> > Arda EDEN
> > Cumhuriyet University
> > Faculty of Fine Arts
> > Department of Music Technology
> > Sivas/TURKEY
>

-- 
Arda EDEN
Cumhuriyet University
Faculty of Fine Arts
Department of Music Technology
Sivas/TURKEY

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
Received on Wed, 19 Mar 2008 00:36:15 +0200

This archive was generated by hypermail 2.1.8 : Wed Mar 19 2008 - 04:15:07 EET