Re: L4_ThreadSwitch behaviour

Joshua Haberman wrote:
> On Feb 22, 2007, at 1:25 AM, Espen Skoglund wrote:
>> As you've already been told, doing scheduling purely based on
>> timeslice donation is not a very good idea.
> 
> What other options are there for implementing user-level scheduling  
> policy?

Hi Joshua,

I recently completed a project on a user-level reflective scheduler for
L4-embedded. It enables a real-time system to be aware of its own behaviour, and
manipulate it through a /model/ (representation) of itself.

However, my scheduler does not schedule multiple arbitrary threads, which is
possibly what you are looking for, but timed, atomic (non-preemptable) actions
in a single protection domain. Some applications may not fit this architecture,
but then the RS model can be evolved.

A decision on releasing the framework has not been made (yet), but you can read
more about it in this paper:

http://www.ertos.nicta.com.au/publications/papers/Ruocco_RTSS_06.abstract.pml

As usual, critical comments and feedback are welcome.

Cheers,

	Sergio

-- 

--

http://www.cse.unsw.edu.au/~sruocco/
ERTOS Researcher                                              Lecturer
National ICT Australia Ltd.              University of New South Wales

This email and any attachments are confidential. They may contain
legally privileged information or copyright material. You should not
read, copy, use or disclose them without authorisation.  If you are
not an intended recipient, please contact us at once by return email
and then delete both messages.  We do not accept liability in
connection with computer virus, data corruption, delay, interruption,
unauthorised access or unauthorised amendment.  This notice should not
be removed.

begin:vcard
fn:Sergio Ruocco
n:Ruocco;Sergio
email;internet:[email protected]
version:2.1
end:vcard