[RFC PATCH 3/4] linsched: avoid invoke tick_nohz_idle_enter() multipletimes in idle

From: Michael Wang
Date: Sun Sep 02 2012 - 23:58:13 EST


From: Michael Wang <wangyun@xxxxxxxxxxxxxxxxxx>

In real world, tick_nohz_idle_enter() will be invoked by idle thread
when the cpu change from active to idle, and will only be invoked
again after tick_nohz_idle_exit() was invoked by idle thread when
cpu is going to recover, invoke it multiple times in one idle may
cause unexpected troubles.

The patch add a check to avoid invoke tick_nohz_idle_enter() again
after idle.

Signed-off-by: Michael Wang <wangyun@xxxxxxxxxxxxxxxxxx>
---
tools/linsched/hrtimer.c | 11 ++++++++++-
1 files changed, 10 insertions(+), 1 deletions(-)

diff --git a/tools/linsched/hrtimer.c b/tools/linsched/hrtimer.c
index de88b25..3203253 100644
--- a/tools/linsched/hrtimer.c
+++ b/tools/linsched/hrtimer.c
@@ -149,6 +149,15 @@ void linsched_check_idle_cpu(void)
}
}

+void linsched_enter_idle(void)
+{
+ int cpu = smp_processor_id();
+ struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+
+ if (!ts->inidle && idle_cpu(cpu))
+ tick_nohz_idle_enter();
+}
+
/* Run a simulation for some number of ticks. Each tick,
* scheduling and load balancing decisions are made. Obviously, we
* could create tasks, change priorities, etc., at certain ticks
@@ -210,7 +219,7 @@ void linsched_run_sim(int sim_ticks)

BUG_ON(irqs_disabled());
if (idle_cpu(active_cpu) && !need_resched()) {
- tick_nohz_idle_enter();
+ linsched_enter_idle();
} else {
linsched_current_handler();
}
--
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/