> > > I liked the QNX-type scheduler, if it only were a bit improved to
> > > eliminate the deadlocks. Of course, that one _was_ a major rewrite.
> > That one is simple -- just let processes promote to
> > levels lower than their min priority level.
> > With the added trick that each level below their own
> > min level takes a timeout period longer...
>
> That solves starvation, but not deadlock resulting from priority
> inversion (high priority task waiting for low priority task), as can
Yes it does. If starvation is solved, you've automatically
cleared up the issue of priority inversion too. The only
problem is that it might take a little while, but hey,
nothing's perfect :)
Besides, I find that my patch exhibits the same improvement
in responsiveness as the QNXsched patch does -- and it was
all introduced with another goal in mind :)
Rik.
+-------------------------------------------------------------------+
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |
+-------------------------------------------------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/