alter kthread prio

alter kthread prio

Timeslice proportion has been increased substantially for -niced tasks. As a result of this kernel threads have much larger timeslices than they previously had.

Change kernel threads' nice value to -5 to bring their timeslice back in line with previous behaviour. This means kernel threads will be less likely to cause large latencies under periods of system stress for normal nice 0 tasks.

Index: linux-2.6.10-rc1-mm2/kernel/workqueue.c
===================================================================
@@ -188,7 +188,7 @@ static int worker_thread(void *__cwq)
 
 	current->flags |= PF_NOFREEZE;
 
-	set_user_nice(current, -10);
+	set_user_nice(current, -5);
 
 	/* Block and flush all signals */
 	sigfillset(&blocked);

ack - but this too needs testing in -mm. Could be 2.6.11 material.

Ingo