TLSF dynamic storage allocator - Was: Anyone know this paper?

Pavel Pisa ppisa4lists at pikron.com
Thu Dec 16 03:13:52 UTC 2004


>  > I believe that there is no perfect algorithm for all real-time
>  > embedded systems.  It would be great to be able to choose from
>  > a variety of heap, CPU scheduling, disk scheduling and disk
>  > caching algorithms based upon your application requirements.
>  >
>  > It seems that no algorithm is perfect for every situation.
>
> I agree.  That said, I'm a big fan of the slab concepts, I suppose
> their downside is complexity & subtlety- its great for the user but
> you have to put your software weenie hat on to understand what its
> doing inside.
>
> Gregm

Hello All,

I am not expert for dynamic allocators research and I believe,
that you right, that there is no perfect solution.

I participate in the OCERA project. The partners from UPV university
has designed relatively nice O(1) allocator. We have used it and even
little contributed to its interfacing to RT-Linux GPL framework.

You can find sources and papers with description at

  http://rtportal.upv.es/apps/malloc/index.shtml
  http://www.ocera.org/download/components/WP5/dynmem-1.3.2.html
  http://cvs.sourceforge.net/viewcvs.py/ocera/ocera/components/sched/dynmem/

Authors are Miguel Masmano and Ismael Ripoll from UPV.

If you find, that this code can be of some use for RTEMS, 
OCERA people would be happy, with that and I can try to test
some day this allocator in RTEMS environment.

The license should not be problem, it is GPL/LGPL and if required,
it should not be possible to explicitly add some RTEMS GPL extension.


Yours sincerely

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



More information about the users mailing list