[uClinux-dev] realtime scheduling...

Lars Segerlund lars.segerlund at comsys.se
Thu Mar 13 04:53:57 EST 2003


  I am hacking away at my small realtime scheduler which works with 
uclinux, and wanted some opinions as for if the functionality would be 
sufficient for most users that need this kind of functionality.

  Basicly it runs some stuff on the NMI and the 'heavy' tasks on linux 
regular interrupts, it can hit a 15 ns window for the code to be run ( 
first instruction ) on the MC5246C3 developement board, and most of it 
should be easily portable to other architectures and constraints, ( 
single timer, multiple timers ). ( also I have some work left on 
verifying the behaviour of sti cli in the drivers and kernel, ie. 
estimating worst case for the linux interrupt parts ).

  The main restriction is that it collects taskt to be scheduled and 
schedules these in a 'batch', after this batch has been proccessed it 
cannot schedule anything which should run before the last starting point 
of the batch, so my question is if this is acceptable ?

  Also it doesn't handle periodic scheduling of tasks, ( it's easily 
implemented though but I don't need it for now ) since I am using it for 
dynamic scheduling of prioritized realtime routines.

  does this sound ok ? or is there any part which sould be included 
before I do a first release of the code ?

  I was thinking of calling it uCRT and post it here first, if this 
would be ok with everybody ?

  Also I don't think I can commit to supporting it forever for reasons 
about employement and such, so if nothing else it would be a nice 
example if someone needs this.

  All comments welcome, and I hope this is not 'out of topic'.

  / regards, Lars Segerlund.




More information about the uClinux-dev mailing list