[PATCH v3] sched/cputime: add steal time support to full dynticks CPU time accounting

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

[PATCH v3] sched/cputime: add steal time support to full dynticks CPU time accounting

Wanpeng Li-4
From: Wanpeng Li <[hidden email]>

This patch adds steal guest time support to full dynticks CPU
time accounting. After 'commit ff9a9b4c4334 ("sched, time: Switch
VIRT_CPU_ACCOUNTING_GEN to jiffy granularity")', time is jiffy
based sampling even if it's still listened to ring boundaries, so
steal_account_process_tick() is reused to account how much 'ticks'
are steal time after the last accumulation.

Suggested-by: Rik van Riel <[hidden email]>
Cc: Ingo Molnar <[hidden email]>
Cc: Peter Zijlstra (Intel) <[hidden email]>
Cc: Rik van Riel <[hidden email]>
Cc: Thomas Gleixner <[hidden email]>
Cc: Frederic Weisbecker <[hidden email]>
Cc: Paolo Bonzini <[hidden email]>
Cc: Radim <[hidden email]>
Signed-off-by: Wanpeng Li <[hidden email]>
---
v2 -> v3:
 * convert steal time jiffies to cputime
v1 -> v2:
 * fix divide zero bug, thanks Rik

 kernel/sched/cputime.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index 75f98c5..f51c98c 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -257,7 +257,7 @@ void account_idle_time(cputime_t cputime)
  cpustat[CPUTIME_IDLE] += (__force u64) cputime;
 }
 
-static __always_inline bool steal_account_process_tick(void)
+static __always_inline unsigned long steal_account_process_tick(void)
 {
 #ifdef CONFIG_PARAVIRT
  if (static_key_false(&paravirt_steal_enabled)) {
@@ -279,7 +279,7 @@ static __always_inline bool steal_account_process_tick(void)
  return steal_jiffies;
  }
 #endif
- return false;
+ return 0;
 }
 
 /*
@@ -691,8 +691,14 @@ static cputime_t get_vtime_delta(struct task_struct *tsk)
 
 static void __vtime_account_system(struct task_struct *tsk)
 {
+ cputime_t steal_time;
  cputime_t delta_cpu = get_vtime_delta(tsk);
+ unsigned long delta_st = steal_account_process_tick();
+ steal_time = jiffies_to_cputime(delta_st);
 
+ if (steal_time >= delta_cpu)
+ return;
+ delta_cpu -= steal_time;
  account_system_time(tsk, irq_count(), delta_cpu, cputime_to_scaled(delta_cpu));
 }
 
@@ -723,7 +729,16 @@ void vtime_account_user(struct task_struct *tsk)
  write_seqcount_begin(&tsk->vtime_seqcount);
  tsk->vtime_snap_whence = VTIME_SYS;
  if (vtime_delta(tsk)) {
+ cputime_t steal_time;
+ unsigned long delta_st = steal_account_process_tick();
  delta_cpu = get_vtime_delta(tsk);
+ steal_time = jiffies_to_cputime(delta_st);
+
+ if (steal_time >= delta_cpu) {
+ write_seqcount_end(&tsk->vtime_seqcount);
+ return;
+ }
+ delta_cpu -= steal_time;
  account_user_time(tsk, delta_cpu, cputime_to_scaled(delta_cpu));
  }
  write_seqcount_end(&tsk->vtime_seqcount);
--
1.9.1

Reply | Threaded
Open this post in threaded view
|

Re: [PATCH v3] sched/cputime: add steal time support to full dynticks CPU time accounting

Rik van Riel
On Wed, 2016-05-18 at 20:27 +0800, Wanpeng Li wrote:

> From: Wanpeng Li <[hidden email]>
>
> This patch adds steal guest time support to full dynticks CPU 
> time accounting. After 'commit ff9a9b4c4334 ("sched, time: Switch 
> VIRT_CPU_ACCOUNTING_GEN to jiffy granularity")', time is jiffy 
> based sampling even if it's still listened to ring boundaries, so 
> steal_account_process_tick() is reused to account how much 'ticks' 
> are steal time after the last accumulation. 
>
> Suggested-by: Rik van Riel <[hidden email]>
> Cc: Ingo Molnar <[hidden email]>
> Cc: Peter Zijlstra (Intel) <[hidden email]>
> Cc: Rik van Riel <[hidden email]>
> Cc: Thomas Gleixner <[hidden email]>
> Cc: Frederic Weisbecker <[hidden email]>
> Cc: Paolo Bonzini <[hidden email]>
> Cc: Radim <[hidden email]>
> Signed-off-by: Wanpeng Li <[hidden email]>
This also nicely fixes up f9c904b7613b ("sched/cputime: 
Fix steal_account_process_tick() to always return jiffies"),
which relies on a bool function returning a certain number
of jiffies :)

Reviewed-by: Rik van Riel <[hidden email]>

--
All rights reversed

signature.asc (484 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [PATCH v3] sched/cputime: add steal time support to full dynticks CPU time accounting

Wanpeng Li-4
Ping Paolo or Peterz.
2016-05-25 3:22 GMT+08:00 Rik van Riel <[hidden email]>:

> On Wed, 2016-05-18 at 20:27 +0800, Wanpeng Li wrote:
>> From: Wanpeng Li <[hidden email]>
>>
>> This patch adds steal guest time support to full dynticks CPU
>> time accounting. After 'commit ff9a9b4c4334 ("sched, time: Switch
>> VIRT_CPU_ACCOUNTING_GEN to jiffy granularity")', time is jiffy
>> based sampling even if it's still listened to ring boundaries, so
>> steal_account_process_tick() is reused to account how much 'ticks'
>> are steal time after the last accumulation.
>>
>> Suggested-by: Rik van Riel <[hidden email]>
>> Cc: Ingo Molnar <[hidden email]>
>> Cc: Peter Zijlstra (Intel) <[hidden email]>
>> Cc: Rik van Riel <[hidden email]>
>> Cc: Thomas Gleixner <[hidden email]>
>> Cc: Frederic Weisbecker <[hidden email]>
>> Cc: Paolo Bonzini <[hidden email]>
>> Cc: Radim <[hidden email]>
>> Signed-off-by: Wanpeng Li <[hidden email]>
>
> This also nicely fixes up f9c904b7613b ("sched/cputime:
> Fix steal_account_process_tick() to always return jiffies"),
> which relies on a bool function returning a certain number
> of jiffies :)
>
> Reviewed-by: Rik van Riel <[hidden email]>
>
> --
> All rights reversed



--
Regards,
Wanpeng Li
Reply | Threaded
Open this post in threaded view
|

Re: [PATCH v3] sched/cputime: add steal time support to full dynticks CPU time accounting

Paolo Bonzini


On 25/05/2016 04:16, Wanpeng Li wrote:
> Ping Paolo or Peterz.

No need to ping, since Rik reviewed it 7 hours ago so the thread has
gotten a bump in our mailboxes.

And anyway this is the merge window, which is the most annoying time to
get pings and one-patch changes.  I don't mind at all getting large
series during the merge window, but the small ones definitely can wait a
week or two.

Thanks,

Paolo

> 2016-05-25 3:22 GMT+08:00 Rik van Riel <[hidden email]>:
>> On Wed, 2016-05-18 at 20:27 +0800, Wanpeng Li wrote:
>>> From: Wanpeng Li <[hidden email]>
>>>
>>> This patch adds steal guest time support to full dynticks CPU
>>> time accounting. After 'commit ff9a9b4c4334 ("sched, time: Switch
>>> VIRT_CPU_ACCOUNTING_GEN to jiffy granularity")', time is jiffy
>>> based sampling even if it's still listened to ring boundaries, so
>>> steal_account_process_tick() is reused to account how much 'ticks'
>>> are steal time after the last accumulation.
>>>
>>> Suggested-by: Rik van Riel <[hidden email]>
>>> Cc: Ingo Molnar <[hidden email]>
>>> Cc: Peter Zijlstra (Intel) <[hidden email]>
>>> Cc: Rik van Riel <[hidden email]>
>>> Cc: Thomas Gleixner <[hidden email]>
>>> Cc: Frederic Weisbecker <[hidden email]>
>>> Cc: Paolo Bonzini <[hidden email]>
>>> Cc: Radim <[hidden email]>
>>> Signed-off-by: Wanpeng Li <[hidden email]>
>>
>> This also nicely fixes up f9c904b7613b ("sched/cputime:
>> Fix steal_account_process_tick() to always return jiffies"),
>> which relies on a bool function returning a certain number
>> of jiffies :)
>>
>> Reviewed-by: Rik van Riel <[hidden email]>
>>
>> --
>> All rights reversed
>
>
>