Possible problem with malloc_free_space using ss-20021007?

Joel Sherrill joel.sherrill at OARcorp.com
Wed Oct 30 02:20:42 UTC 2002



Till Straumann wrote:
> 
> BTW - talking about malloc_free_space()
> 
> Joel - has the performance problem been fixed?
> (remember: malloc_free_space() ran for a long
> time with thread dispatching disabled)

It can still run for a long time but it does so with
a mutex locked.  So I consider the context switch
latency problem resolved.  The algorithm could still
be improved by keeping track of things as part of
alloc/dealloc but I didn't want more memory for a 
heap or region control block.

FWIW based upon Bob's data in private email, it looks
like the amount of free space is about 2x too much.
Does this match for anyone else?

> -- Till

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel at OARcorp.com                 On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985



More information about the users mailing list