* [PATCH] tick: make sleep length calculation more accurate
@ 2013-09-27 10:52 Daniel Lezcano
2013-10-01 21:31 ` Stephen Boyd
0 siblings, 1 reply; 3+ messages in thread
From: Daniel Lezcano @ 2013-09-27 10:52 UTC (permalink / raw)
To: tglx; +Cc: linux-kernel, patches, linaro-kernel
The sleep_length is computed in the tick_nohz_stop_sched_tick function but it
is used later in the code with in between the local irq enabled.
cpu_idle_loop
tick_nohz_idle_enter [ exits with local irq enabled ]
__tick_nohz_idle_enter
tick_nohz_stop_sched_tick
...
arch_cpu_idle
menu_select [ uses here 'sleep_length' ]
...
Between the computation of the sleep length and its usage, some interrupts
can occur, making the sleep length shorter than actually it is.
This patch fixes that by moving the sleep_length computation in the
tick_nohz_get_sleep_length function and store the next_event for the device
instead of the sleep_length.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
---
include/linux/tick.h | 2 +-
kernel/time/tick-sched.c | 5 +++--
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/include/linux/tick.h b/include/linux/tick.h
index 5128d33..4932004 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -67,7 +67,7 @@ struct tick_sched {
ktime_t idle_exittime;
ktime_t idle_sleeptime;
ktime_t iowait_sleeptime;
- ktime_t sleep_length;
+ ktime_t next_event;
unsigned long last_jiffies;
unsigned long next_jiffies;
ktime_t idle_expires;
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 3612fc7..2007a7f 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -673,7 +673,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
out:
ts->next_jiffies = next_jiffies;
ts->last_jiffies = last_jiffies;
- ts->sleep_length = ktime_sub(dev->next_event, now);
+ ts->next_event = dev->next_event;
return ret;
}
@@ -837,8 +837,9 @@ void tick_nohz_irq_exit(void)
ktime_t tick_nohz_get_sleep_length(void)
{
struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched);
+ ktime_t now = ktime_get();
- return ts->sleep_length;
+ return ktime_sub(ts->next_event, now);
}
static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
--
1.7.9.5
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [PATCH] tick: make sleep length calculation more accurate
2013-09-27 10:52 [PATCH] tick: make sleep length calculation more accurate Daniel Lezcano
@ 2013-10-01 21:31 ` Stephen Boyd
2013-10-02 9:55 ` Daniel Lezcano
0 siblings, 1 reply; 3+ messages in thread
From: Stephen Boyd @ 2013-10-01 21:31 UTC (permalink / raw)
To: Daniel Lezcano; +Cc: tglx, linux-kernel, patches, linaro-kernel
On 09/27/13 03:52, Daniel Lezcano wrote:
> The sleep_length is computed in the tick_nohz_stop_sched_tick function but it
> is used later in the code with in between the local irq enabled.
>
> cpu_idle_loop
> tick_nohz_idle_enter [ exits with local irq enabled ]
> __tick_nohz_idle_enter
> tick_nohz_stop_sched_tick
> ...
>
> arch_cpu_idle
> menu_select [ uses here 'sleep_length' ]
> ...
>
> Between the computation of the sleep length and its usage, some interrupts
> can occur, making the sleep length shorter than actually it is.
>
> This patch fixes that by moving the sleep_length computation in the
> tick_nohz_get_sleep_length function and store the next_event for the device
> instead of the sleep_length.
>
> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> ---
> include/linux/tick.h | 2 +-
> kernel/time/tick-sched.c | 5 +++--
> 2 files changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/include/linux/tick.h b/include/linux/tick.h
> index 5128d33..4932004 100644
> --- a/include/linux/tick.h
> +++ b/include/linux/tick.h
> @@ -67,7 +67,7 @@ struct tick_sched {
> ktime_t idle_exittime;
> ktime_t idle_sleeptime;
> ktime_t iowait_sleeptime;
> - ktime_t sleep_length;
> + ktime_t next_event;
> unsigned long last_jiffies;
> unsigned long next_jiffies;
> ktime_t idle_expires;
Documentation update?
> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
> index 3612fc7..2007a7f 100644
> --- a/kernel/time/tick-sched.c
> +++ b/kernel/time/tick-sched.c
> @@ -673,7 +673,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
> out:
> ts->next_jiffies = next_jiffies;
> ts->last_jiffies = last_jiffies;
> - ts->sleep_length = ktime_sub(dev->next_event, now);
> + ts->next_event = dev->next_event;
>
> return ret;
> }
> @@ -837,8 +837,9 @@ void tick_nohz_irq_exit(void)
> ktime_t tick_nohz_get_sleep_length(void)
> {
> struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched);
> + ktime_t now = ktime_get();
>
> - return ts->sleep_length;
> + return ktime_sub(ts->next_event, now);
> }
>
> static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
What happens if the idling CPU's next_event is updated via that
interrupt? Say if the interrupt handler schedules a timer to fire before
the next timer on the CPU? It looks like we won't notice that.
Perhaps it's better to do this instead?
ktime_t tick_nohz_get_sleep_length(void)
{
struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched);
+ ktime_t now = ktime_get();
+ struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
- return ts->sleep_length;
+ return ktime_sub(dev->next_event, now);
}
--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH] tick: make sleep length calculation more accurate
2013-10-01 21:31 ` Stephen Boyd
@ 2013-10-02 9:55 ` Daniel Lezcano
0 siblings, 0 replies; 3+ messages in thread
From: Daniel Lezcano @ 2013-10-02 9:55 UTC (permalink / raw)
To: Stephen Boyd; +Cc: tglx, linux-kernel, patches, linaro-kernel
On 10/01/2013 11:31 PM, Stephen Boyd wrote:
> On 09/27/13 03:52, Daniel Lezcano wrote:
>> The sleep_length is computed in the tick_nohz_stop_sched_tick function but it
>> is used later in the code with in between the local irq enabled.
>>
>> cpu_idle_loop
>> tick_nohz_idle_enter [ exits with local irq enabled ]
>> __tick_nohz_idle_enter
>> tick_nohz_stop_sched_tick
>> ...
>>
>> arch_cpu_idle
>> menu_select [ uses here 'sleep_length' ]
>> ...
>>
>> Between the computation of the sleep length and its usage, some interrupts
>> can occur, making the sleep length shorter than actually it is.
>>
>> This patch fixes that by moving the sleep_length computation in the
>> tick_nohz_get_sleep_length function and store the next_event for the device
>> instead of the sleep_length.
>>
>> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
>> ---
>> include/linux/tick.h | 2 +-
>> kernel/time/tick-sched.c | 5 +++--
>> 2 files changed, 4 insertions(+), 3 deletions(-)
>>
>> diff --git a/include/linux/tick.h b/include/linux/tick.h
>> index 5128d33..4932004 100644
>> --- a/include/linux/tick.h
>> +++ b/include/linux/tick.h
>> @@ -67,7 +67,7 @@ struct tick_sched {
>> ktime_t idle_exittime;
>> ktime_t idle_sleeptime;
>> ktime_t iowait_sleeptime;
>> - ktime_t sleep_length;
>> + ktime_t next_event;
>> unsigned long last_jiffies;
>> unsigned long next_jiffies;
>> ktime_t idle_expires;
>
> Documentation update?
>
>> diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
>> index 3612fc7..2007a7f 100644
>> --- a/kernel/time/tick-sched.c
>> +++ b/kernel/time/tick-sched.c
>> @@ -673,7 +673,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
>> out:
>> ts->next_jiffies = next_jiffies;
>> ts->last_jiffies = last_jiffies;
>> - ts->sleep_length = ktime_sub(dev->next_event, now);
>> + ts->next_event = dev->next_event;
>>
>> return ret;
>> }
>> @@ -837,8 +837,9 @@ void tick_nohz_irq_exit(void)
>> ktime_t tick_nohz_get_sleep_length(void)
>> {
>> struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched);
>> + ktime_t now = ktime_get();
>>
>> - return ts->sleep_length;
>> + return ktime_sub(ts->next_event, now);
>> }
>>
>> static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
>
> What happens if the idling CPU's next_event is updated via that
> interrupt? Say if the interrupt handler schedules a timer to fire before
> the next timer on the CPU? It looks like we won't notice that.
Yes, or after.
It sounds like this issue also occurs with the current code, no ?
> Perhaps it's better to do this instead?
>
> ktime_t tick_nohz_get_sleep_length(void)
> {
> struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched);
> + ktime_t now = ktime_get();
> + struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
>
> - return ts->sleep_length;
> + return ktime_sub(dev->next_event, now);
> }
Yes, I agree.
Thanks for the review.
-- Daniel
--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2013-10-02 9:55 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-09-27 10:52 [PATCH] tick: make sleep length calculation more accurate Daniel Lezcano
2013-10-01 21:31 ` Stephen Boyd
2013-10-02 9:55 ` Daniel Lezcano
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox