Realtime-preemption for 2.6.20-rc5 ?

Realtime-preemption for 2.6.20-rc5 ?

[Sunil Naidu]

Hi Ingo,

I would like to try with patch-2.6.20-rc5-rt7 for an experiment to
measure the latency.
Is there any documentation or help which talks about patching, issues,
and latency benchmarks?

~Akula2

Re: Realtime-preemption for 2.6.20-rc5 ?

[Ingo Molnar]

* Sunil Naidu <akula2.shark@gmail.com> wrote:

> Hi Ingo,
> 
> I would like to try with patch-2.6.20-rc5-rt7 for an experiment to 
> measure the latency. Is there any documentation or help which talks 
> about patching, issues, and latency benchmarks?

the best place to start is:

  http://rt.wiki.kernel.org

	Ingo

Re: Realtime-preemption for 2.6.20-rc5 ?

[Sunil Naidu]

On 1/18/07, Ingo Molnar <mingo@elte.hu> wrote:
> the best place to start is:
>
>   http://rt.wiki.kernel.org
>
>         Ingo

I did refer the same. Is it necessary to use only base kernel, say
2.6.19? Or, can I go ahead with 2.6.19 + 2.6.19.2 patch + 2.6.19-rt
patch?

If yes, any reason why we need to apply rt patch only to a base kernel?

Thanks,

~Akula2

Re: Realtime-preemption for 2.6.20-rc5 ?

[Pavel Pisa]

Hello Sunil and Ingo,

Date: 2007-01-20 02:56:40 GMT (20 hours and 26 minutes ago)
> 2007-01-20, Sunil Naidu <akula2.shark@gmail.com> wrote:
> I did refer the same. Is it necessary to use only base kernel, say
> 2.6.19? Or, can I go ahead with 2.6.19 + 2.6.19.2 patch + 2.6.19-rt
> patch?
>
> If yes, any reason why we need to apply rt patch only to a base kernel?

according to my observation 2.6.19-rt15 is based/includes 2.6.19.1 changes.

But there has been that nasty clear_page_dirty_for_io() bug causing
corruption of ext3. Even that I have tested more 2.6.20-rc + rt, I preffer
to stay on "stable" kernel on boxes which I use daily until next stable
appears. I have backported clear_page_dirty_for_io() to 2.6.19-rt15
and it worked fine. I have tried to update 2.6.19-rt15 to 2.6.19.2
base. There is result of my attempt

Unofficial incremental patch from 2.6.19-rt15 to 2.6.19.2 + rt
http://rtime.felk.cvut.cz/repos/ppisa-linux-devel/kernel-patches/current/patch-2.6.19.2-incr.patch

Kernel seems to work correctly. I have checked the patch contents
and I have not noticed any RT problematic changes in the code according
to my dumb knowledge.

I would be very happy, if Ingo would be so kind and could confirm my findings,
because I am not sure, if final 2.6.20+rt would be ready before we need
to prepare setup for our next semester classes at university.

Best wishes

                Pavel Pisa
        e-mail: pisa@cmp.felk.cvut.cz
        www:    http://cmp.felk.cvut.cz/~pisa
        work:   http://www.pikron.com

Re: Realtime-preemption for 2.6.20-rc5 ?

[Peter Zijlstra]

On Sun, 2007-01-21 at 00:39 +0100, Pavel Pisa wrote:
> Hello Sunil and Ingo,
> 
> Date: 2007-01-20 02:56:40 GMT (20 hours and 26 minutes ago)
> > 2007-01-20, Sunil Naidu <akula2.shark@gmail.com> wrote:
> > I did refer the same. Is it necessary to use only base kernel, say
> > 2.6.19? Or, can I go ahead with 2.6.19 + 2.6.19.2 patch + 2.6.19-rt
> > patch?
> >
> > If yes, any reason why we need to apply rt patch only to a base kernel?
> 
> according to my observation 2.6.19-rt15 is based/includes 2.6.19.1 changes.
> 
> But there has been that nasty clear_page_dirty_for_io() bug causing
> corruption of ext3. Even that I have tested more 2.6.20-rc + rt, 

> I preffer
> to stay on "stable" kernel on boxes which I use daily until next stable
> appears.

This is a very weird statement, the -rt kernel includes so much
experimental work it cannot be called 'stable' by a long shot.

Sure its not known unstable, but neither is .20-rc5.

If you want -rt, just run with the latest unless you have a very
specific need not to.

Re: Realtime-preemption for 2.6.20-rc5 ?

[Pavel Pisa]

On Sunday 21 January 2007 10:10, Peter Zijlstra wrote:
> > I preffer
> > to stay on "stable" kernel on boxes which I use daily until next stable
> > appears.
>
> This is a very weird statement, the -rt kernel includes so much
> experimental work it cannot be called 'stable' by a long shot.
>
> Sure its not known unstable, but neither is .20-rc5.

There are no security fixes for rc and our own experience
is, that RT patch has very low impact on base system stability.
The rc-s contains much more experimental stuff all over the
kernel which needs to be stabilized till next (hopefully) stable
release.

> If you want -rt, just run with the latest unless you have a very
> specific need not to.

We have run successfully 2.6.16.1-rt12 over last summer
semester on students diskless stations without much problems.
(Main problem has been some NFS FS problem with 1GB/s server, 100MB/s
stations and switches in between, but it has been same for non-RT kernel.
We solved that by switching NFS over TCP.)

We would like to upgrade to something which would not cause us much
troubles for next course run. We teach real time control in X35POS
course and we need fast responses and timing (100 usec) for direct
PWM and IO control.

On the other hand, I agree that for own experimentation and development
it is better to build on latest released version.

Best wishes

             Pavel