within hours of the 2.1.128 release (within 30 minutes
of the patch reaching my mailbox) I have put out the
2.1.128 version of my scheduler bigpatch.
features:
- adds a SCHED_IDLE class for ultra-low priority
processing (can this be used to lock the system?
can anyone find a place within the kernel where
schedule() is called within a critical region?)
- greatly improved interactive response, the system
remain useable under triple-digit loads
- proper/better/different reporting of process priorities
so you have a way of recognizing RT/idle processes
from top:
- idle: nice = prio = 20
- rt: nice = (-20 - p->rt_priority), prio = 0
- lower latency RT scheduling, through clever and low cost
use of multiple runqueues (thanks Richard)
- lower overall scheduling overhead, no longer any need
to recalculate the priorities of _all_ tasks when the
timeslices have run out --> better rc5des performance,
up to 3% measured by one brave tester
- sysctl tuneable SCHED_IDLE timeslice length
You can get the patch from my home page, as usual.
Have fun,
Rik -- slowly getting used to dvorak kbd layout...
+-------------------------------------------------------------------+
| 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/