Skip to content

Commit

Permalink
sched: fix load average accounting vs. cpu hotplug
Browse files Browse the repository at this point in the history
The new load average code clears rq->calc_load_active on
CPU_ONLINE. That's wrong as the new onlined CPU might have got a
scheduler tick already and accounted the delta to the stale value of
the time we offlined the CPU.

Clear the value when we cleanup the dead CPU instead. 

Also move the update of the calc_load_update time for the newly online
CPU to CPU_UP_PREPARE to avoid that the CPU plays catch up with the
stale update time value.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
  • Loading branch information
Thomas Gleixner committed Jul 18, 2009
1 parent 54fdc58 commit a468d38
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions kernel/sched.c
Original file line number Diff line number Diff line change
Expand Up @@ -7289,6 +7289,7 @@ static void migrate_dead_tasks(unsigned int dead_cpu)
static void calc_global_load_remove(struct rq *rq)
{
atomic_long_sub(rq->calc_load_active, &calc_load_tasks);
rq->calc_load_active = 0;
}
#endif /* CONFIG_HOTPLUG_CPU */

Expand Down Expand Up @@ -7515,6 +7516,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
task_rq_unlock(rq, &flags);
get_task_struct(p);
cpu_rq(cpu)->migration_thread = p;
rq->calc_load_update = calc_load_update;
break;

case CPU_ONLINE:
Expand All @@ -7525,8 +7527,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
/* Update our root-domain */
rq = cpu_rq(cpu);
spin_lock_irqsave(&rq->lock, flags);
rq->calc_load_update = calc_load_update;
rq->calc_load_active = 0;
if (rq->rd) {
BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));

Expand Down

0 comments on commit a468d38

Please sign in to comment.