lkml.org 
[lkml]   [2002]   [Jul]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectO(1) batch scheduler
Hi Ingo,

I looked through your sched-2.5.25-A5 patch, and I'm confused by the
idle_count array. It calculates the idle average of the last 9 seconds -
but why not just use a weighted average. A weighted average is going to be
very close to the true average, and where it differs the weighted average
should be preferable.

Incremental patch (untested):

--- kernel/sched.c Tue Jul 9 22:06:38 2002
+++ ../linux-2.5.25-a/kernel/sched.c Tue Jul 9 22:23:41 2002
@@ -171,7 +171,7 @@
#define IDLE_TICKS (HZ)

int idle_ticks_left;
- int idle_count[IDLE_SLOTS];
+ int idle_count;
int idle_avg;

} ____cacheline_aligned;
@@ -886,17 +886,6 @@
#define BUSY_REBALANCE_TICK (HZ/4 ?: 1)
#define IDLE_REBALANCE_TICK (HZ/1000 ?: 1)

-static inline int recalc_idle_avg(runqueue_t *rq)
-{
- int i, count = 0, avg;
-
- for (i = 1; i < IDLE_SLOTS; i++)
- count += rq->idle_count[i];
-
- avg = count / (IDLE_SLOTS - 1);
- return avg;
-}
-
static inline void idle_tick(runqueue_t *rq)
{
if (jiffies % IDLE_REBALANCE_TICK)
@@ -938,17 +927,13 @@
* This code is rare, triggered only once per second:
*/
if (--rq->idle_ticks_left <= 0) {
- int i;
-
- rq->idle_ticks_left = IDLE_TICKS;
- for (i = IDLE_SLOTS-1; i > 0; i--)
- rq->idle_count[i] = rq->idle_count[i-1];
- rq->idle_count[0] = 0;
- rq->idle_avg = recalc_idle_avg(rq);
+ rq->idle_avg = (rq->idle_avg * (IDLE_SLOTS - 1)
+ + rq->idle_count) / IDLE_SLOTS;
+ rq->idle_count = 0;
}
}
if (p == rq->idle || p->policy == SCHED_BATCH)
- rq->idle_count[0]++;
+ rq->idle_count++;
#endif
if (p == rq->idle) {
if (local_bh_count(cpu) || local_irq_count(cpu) > 1)
--
------------------------------------------------------------------------
| Kevin O'Connor "BTW, IMHO we need a FAQ for |
| kevin@koconnor.net 'IMHO', 'FAQ', 'BTW', etc. !" |
------------------------------------------------------------------------
-
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/
\
 
 \ /
  Last update: 2005-03-22 13:27    [W:0.219 / U:0.048 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site