Hi!
I've been poking around and found the following link on O(1) scheduler
starvation problems:
http://www.hpl.hp.com/research/linux/kernel/o1-starve.php
The web page contains a small test program which supposedly is able to
make two processes starvate. However, I've been unable to reproduce what
is described in the above link. In fact, the CPU usage ratio ranges
between 60-40% or 70-30% in the worst cases.
I'm running 2.5.72-mm1 with Mike Galbraith's scheduler patches and a
small patch I made myself to improve interactivity (mainly, to stop XMMS
from skipping by adjusting some scheduler parameters).
What's going on here? Is the previous article outdated, or have the
starvation problems been definitively fixed?
Thanks!
--- old/kernel/sched.c 2003-06-17 21:04:21.240902000 +0200
+++ new/kernel/sched.c 2003-06-17 20:58:54.840902000 +0200
@@ -66,14 +66,14 @@
* they expire.
*/
#define MIN_TIMESLICE ( 10 * HZ / 1000)
-#define MAX_TIMESLICE (200 * HZ / 1000)
+#define MAX_TIMESLICE ( 20 * HZ / 1000)
#define CHILD_PENALTY 50
#define PARENT_PENALTY 100
#define EXIT_WEIGHT 3
-#define PRIO_BONUS_RATIO 25
+#define PRIO_BONUS_RATIO 20
#define INTERACTIVE_DELTA 2
-#define MAX_SLEEP_AVG (10*HZ)
-#define STARVATION_LIMIT (10*HZ)
+#define MAX_SLEEP_AVG (2*HZ)
+#define STARVATION_LIMIT (2*HZ)
#define NODE_THRESHOLD 125
#define SCHED_NANOSECOND 1
#define SCHED_SECOND (1000000000 * SCHED_NANOSECOND)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jun 23 2003 - 22:00:23 EST