Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1226565 > unrolled thread

[PATCH] nohz: make the idle_entrytime/idle_exittime correct

Back to article view | Back to linux.kernel

Contents

  [PATCH] nohz: make the idle_entrytime/idle_exittime correct Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-09-17 01:30 +0200
    Re: [PATCH] nohz: make the idle_entrytime/idle_exittime correct Thomas Gleixner <tglx@linutronix.de> - 2015-09-17 23:50 +0200
      Re: [PATCH] nohz: make the idle_entrytime/idle_exittime correct Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-09-18 20:50 +0200

#1226565 — [PATCH] nohz: make the idle_entrytime/idle_exittime correct

Currently the idle_entrytime is updated on idle_entry, idle_exit and
get_cpu_idle/iowait_time_us() with non-NULL parameter. This makes the
idle_entrytime not the real entrytime anymore. The idle_exittiime is
updated when exit tickless state, which can be idle tickless or full
tickless.

Change the idle_entrytime to be updated only when enter idle, the
idle_exittime to be updated only when exit idle. Update
the get_cpu_idle_iowait_time_us() accordingly.

Signed-off-by: Yunhong Jiang <yunhong.jiang@linux.intel.com>
---
 kernel/time/tick-sched.c | 44 ++++++++++++++++----------------------------
 1 file changed, 16 insertions(+), 28 deletions(-)

diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index c792429e98c6..f8470ff20f77 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -448,7 +448,7 @@ static void tick_nohz_update_jiffies(ktime_t now)
  * Updates the per cpu time idle statistics counters
  */
 static void
-update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now, u64 *last_update_time)
+update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now)
 {
 	ktime_t delta;
 
@@ -458,17 +458,13 @@ update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now, u64 *last_upda
 			ts->iowait_sleeptime = ktime_add(ts->iowait_sleeptime, delta);
 		else
 			ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-		ts->idle_entrytime = now;
 	}
-
-	if (last_update_time)
-		*last_update_time = ktime_to_us(now);
-
 }
 
 static void tick_nohz_stop_idle(struct tick_sched *ts, ktime_t now)
 {
-	update_ts_time_stats(smp_processor_id(), ts, now, NULL);
+	update_ts_time_stats(smp_processor_id(), ts, now);
+	ts->idle_exittime = now;
 	ts->idle_active = 0;
 
 	sched_clock_idle_wakeup_event(0);
@@ -507,21 +503,16 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
 		return -1;
 
 	now = ktime_get();
-	if (last_update_time) {
-		update_ts_time_stats(cpu, ts, now, last_update_time);
-		idle = ts->idle_sleeptime;
-	} else {
-		if (ts->idle_active && !nr_iowait_cpu(cpu)) {
-			ktime_t delta = ktime_sub(now, ts->idle_entrytime);
+	idle = ts->idle_sleeptime;
+	if (ts->idle_active && !nr_iowait_cpu(cpu)) {
+		ktime_t delta = ktime_sub(now, ts->idle_entrytime);
 
-			idle = ktime_add(ts->idle_sleeptime, delta);
-		} else {
-			idle = ts->idle_sleeptime;
-		}
+		idle = ktime_add(ts->idle_sleeptime, delta);
 	}
 
+	if (last_update_time)
+		*last_update_time = ktime_to_us(now);
 	return ktime_to_us(idle);
-
 }
 EXPORT_SYMBOL_GPL(get_cpu_idle_time_us);
 
@@ -548,19 +539,16 @@ u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
 		return -1;
 
 	now = ktime_get();
-	if (last_update_time) {
-		update_ts_time_stats(cpu, ts, now, last_update_time);
-		iowait = ts->iowait_sleeptime;
-	} else {
-		if (ts->idle_active && nr_iowait_cpu(cpu) > 0) {
-			ktime_t delta = ktime_sub(now, ts->idle_entrytime);
+	iowait = ts->iowait_sleeptime;
+	if (ts->idle_active && nr_iowait_cpu(cpu) > 0) {
+		ktime_t delta = ktime_sub(now, ts->idle_entrytime);
 
-			iowait = ktime_add(ts->iowait_sleeptime, delta);
-		} else {
-			iowait = ts->iowait_sleeptime;
-		}
+		iowait = ktime_add(ts->iowait_sleeptime, delta);
 	}
 
+	if (last_update_time)
+		*last_update_time = ktime_to_us(now);
+
 	return ktime_to_us(iowait);
 }
 EXPORT_SYMBOL_GPL(get_cpu_iowait_time_us);
-- 
1.8.3.1

--
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/

[toc] | [next] | [standalone]

#1227427

On Wed, 16 Sep 2015, Yunhong Jiang wrote:

> Currently the idle_entrytime is updated on idle_entry, idle_exit and
> get_cpu_idle/iowait_time_us() with non-NULL parameter. This makes the
> idle_entrytime not the real entrytime anymore. The idle_exittiime is
> updated when exit tickless state, which can be idle tickless or full
> tickless.

If something calls one of the functions then the cpu is not idle. It's
running code.
 
> Change the idle_entrytime to be updated only when enter idle, the
> idle_exittime to be updated only when exit idle. Update
> the get_cpu_idle_iowait_time_us() accordingly.

And the rationale for this change is?

Thanks,

	tglx
--
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/

[toc] | [prev] | [next] | [standalone]

#1228226

On Thu, Sep 17, 2015 at 11:45:37PM +0200, Thomas Gleixner wrote:
> On Wed, 16 Sep 2015, Yunhong Jiang wrote:
> 
> > Currently the idle_entrytime is updated on idle_entry, idle_exit and
> > get_cpu_idle/iowait_time_us() with non-NULL parameter. This makes the
> > idle_entrytime not the real entrytime anymore. The idle_exittiime is
> > updated when exit tickless state, which can be idle tickless or full
> > tickless.
> 
> If something calls one of the functions then the cpu is not idle. It's
> running code.

I think the get_cpu_idle/iowait_time_us() can be called from other CPUs.

Hmm, I should state this patch as "no idle_entrytime update on
get_cpu_idle/iowait_time_us()". If you think this statement is correct,
I will re-submit a patch for it.

>  
> > Change the idle_entrytime to be updated only when enter idle, the
> > idle_exittime to be updated only when exit idle. Update
> > the get_cpu_idle_iowait_time_us() accordingly.
> 
> And the rationale for this change is?

I tried to check /proc/timer_list for idle_entrytime/idle_exittime, and
noticed this when reading the corresponding code. So no special reason for
this change, just correctness.

Thanks
--jyh

> 
> Thanks,
> 
> 	tglx
> --
> 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/
--
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/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web