On Tue, 11 May 1999, Andrea Arcangeli wrote:

> >> 1) there are no idle cpu in the system
> >> 2) the prev task was in general the less priority one
> >
> >wrong. We actively _preempt_ processes on other CPUs, this means that a
> 
> If I understand well the only wrong thing I said is point (2). But if you
> would read better my email I stated clearly that point (2) it not the pure
> reality, but instead I said that it's a very good approximation of the
> pure reality.
> 
> >preempted process has all rights to try to replace even lesser priority
> >processes on other CPUs. Your problem might be that you are thinking in
> 
> You should rereadd my previous email. I never said it's useless, I have
> said that it's not worthwile.

it is always 'worthwile' to have a correct scheduler. This was the sole
purpose of all the 2.2.8 scheduler changes!

> You mean that a preemted process has all rights to preempt a even lesser
> priority CPU. But ask you _why_ such process is been preempted. Simply
> because in general we can see it as the _less_ priority one.

not necesserily, it might as well just be replaced by a RT process ... 

> I repeat that as global design I prefer to have such call in schedule_tail
> even if according to me it's only a performance _hit_.

it is not a performance hit at all because most processes reschedule
'voluntarily', ie. they get removed from the runqueue.

There is one inconsistency left though, if the previous process was
SCHED_YIELD then we should obviously not push it to other CPUs, because it
has just given up it's timeslice. (the attached untested patch fixes this) 

-- mingo

--- linux/kernel/sched.c.orig	Tue May 11 13:29:39 1999
+++ linux/kernel/sched.c	Tue May 11 13:38:32 1999
@@ -194,10 +194,8 @@
 static inline int prev_goodness (struct task_struct * prev,
 					struct task_struct * p, int this_cpu)
 {
-	if (p->policy & SCHED_YIELD) {
-		p->policy &= ~SCHED_YIELD;
+	if (p->policy & SCHED_YIELD)
 		return 0;
-	}
 	return goodness(prev, p, this_cpu);
 }
 
@@ -659,10 +657,16 @@
  */
 static inline void __schedule_tail (struct task_struct *prev)
 {
+	if (prev->policy & SCHED_YIELD)
+		prev->policy &= ~SCHED_YIELD;
+	else {
+#ifdef __SMP__
+		if ((prev->state == TASK_RUNNING) &&
+				(prev != idle_task(smp_processor_id())))
+			reschedule_idle(prev);
+#endif
+	}
 #ifdef __SMP__
-	if ((prev->state == TASK_RUNNING) &&
-			(prev != idle_task(smp_processor_id())))
-		reschedule_idle(prev);
 	wmb();
 	prev->has_cpu = 0;
 #endif /* __SMP__ */

-
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/