Use virtual cpu accounting if available for guest times.

Use virtual cpu accounting if available for guest times.

[Christian Borntraeger]

Avi,

ppc and s390 offer the possibility to track process times precisely
by looking at cpu timer on every context switch, irq, softirq etc.
We can use that infrastructure as well for guest time accounting.
We need to account the used time before we change the state. 
This patch adds a call to account_system_vtime to kvm_guest_enter
and kvm_guest exit. If CONFIG_VIRT_CPU_ACCOUNTING is not set,
account_system_vtime is defined in hardirq.h as an empty function,
which means this patch does not change the behaviour on other
platforms.

I compile tested this patch on x86 and function tested the patch on
s390.

Avi, please apply.


Signed-off-by: Christian Borntraeger <borntraeger-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org>

---
 drivers/kvm/kvm.h |    3 +++
 1 file changed, 3 insertions(+)

Index: kvm/drivers/kvm/kvm.h
===================================================================
--- kvm.orig/drivers/kvm/kvm.h
+++ kvm/drivers/kvm/kvm.h
@@ -7,6 +7,7 @@
  */
 
 #include <linux/types.h>
+#include <linux/hardirq.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
 #include <linux/spinlock.h>
@@ -669,11 +670,13 @@ __init void kvm_arch_init(void);
 
 static inline void kvm_guest_enter(void)
 {
+	account_system_vtime(current);
 	current->flags |= PF_VCPU;
 }
 
 static inline void kvm_guest_exit(void)
 {
+	account_system_vtime(current);
 	current->flags &= ~PF_VCPU;
 }
 

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: Use virtual cpu accounting if available for guest times.

[Avi Kivity]

Christian Borntraeger wrote:
> ppc and s390 offer the possibility to track process times precisely
> by looking at cpu timer on every context switch, irq, softirq etc.
> We can use that infrastructure as well for guest time accounting.
> We need to account the used time before we change the state. 
> This patch adds a call to account_system_vtime to kvm_guest_enter
> and kvm_guest exit. If CONFIG_VIRT_CPU_ACCOUNTING is not set,
> account_system_vtime is defined in hardirq.h as an empty function,
> which means this patch does not change the behaviour on other
> platforms.
>
> I compile tested this patch on x86 and function tested the patch on
> s390.
>
> Avi, please apply.
>
>   

Done.  Thanks!

-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [kvm-devel] Use virtual cpu accounting if available for guest times.

[Hollis Blanchard]

On Thu, 2007-10-18 at 14:39 +0200, Christian Borntraeger wrote:
> Avi,
> 
> ppc and s390 offer the possibility to track process times precisely
> by looking at cpu timer on every context switch, irq, softirq etc.
> We can use that infrastructure as well for guest time accounting.
> We need to account the used time before we change the state. 
> This patch adds a call to account_system_vtime to kvm_guest_enter
> and kvm_guest exit. If CONFIG_VIRT_CPU_ACCOUNTING is not set,
> account_system_vtime is defined in hardirq.h as an empty function,
> which means this patch does not change the behaviour on other
> platforms.
> 
> I compile tested this patch on x86 and function tested the patch on
> s390.
> 
> Avi, please apply.
> 
> 
> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
> 
> ---
>  drivers/kvm/kvm.h |    3 +++
>  1 file changed, 3 insertions(+)
> 
> Index: kvm/drivers/kvm/kvm.h
> ===================================================================
> --- kvm.orig/drivers/kvm/kvm.h
> +++ kvm/drivers/kvm/kvm.h
> @@ -7,6 +7,7 @@
>   */
> 
>  #include <linux/types.h>
> +#include <linux/hardirq.h>
>  #include <linux/list.h>
>  #include <linux/mutex.h>
>  #include <linux/spinlock.h>
> @@ -669,11 +670,13 @@ __init void kvm_arch_init(void);
> 
>  static inline void kvm_guest_enter(void)
>  {
> +	account_system_vtime(current);
>  	current->flags |= PF_VCPU;
>  }
> 
>  static inline void kvm_guest_exit(void)
>  {
> +	account_system_vtime(current);
>  	current->flags &= ~PF_VCPU;
>  }

I don't understand. Should kvm_guest_exit() be calling
account_user_vtime() (instead of account_system_vtime())?

-- 
Hollis Blanchard
IBM Linux Technology Center

Re: Use virtual cpu accounting if available for guest times.

[Hollis Blanchard]

On Fri, 2007-10-19 at 11:57 -0500, Hollis Blanchard wrote:
> > @@ -669,11 +670,13 @@ __init void kvm_arch_init(void);
> > 
> >  static inline void kvm_guest_enter(void)
> >  {
> > +	account_system_vtime(current);
> >  	current->flags |= PF_VCPU;
> >  }
> > 
> >  static inline void kvm_guest_exit(void)
> >  {
> > +	account_system_vtime(current);
> >  	current->flags &= ~PF_VCPU;
> >  }
> 
> I don't understand. Should kvm_guest_exit() be calling
> account_user_vtime() (instead of account_system_vtime())?

Never mind; the tree I was looking at didn't have the
account_guest_time() stuff in account_system_time().

-- 
Hollis Blanchard
IBM Linux Technology Center


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

[PATCH] move kvm_guest_exit() after local_irq_enable()

[Laurent Vivier]

According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if we move
it after local_irq_enable().

http://lkml.org/lkml/2007/10/15/114

To simplify s390 port, we don't clear it in account_system_time().

http://lkml.org/lkml/2007/10/15/183

---
 drivers/kvm/kvm_main.c |    5 ++++-
 kernel/sched.c         |    1 -
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 87275be..b9cd1f0 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -2194,12 +2194,15 @@ again:
 
 	kvm_x86_ops->run(vcpu, kvm_run);
 
-	kvm_guest_exit();
 	vcpu->guest_mode = 0;
 	local_irq_enable();
 
 	++vcpu->stat.exits;
 
+	barrier();
+
+	kvm_guest_exit();
+
 	preempt_enable();
 
 	/*
diff --git a/kernel/sched.c b/kernel/sched.c
index b27ab3e..57fac22 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3315,7 +3315,6 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 #ifdef CONFIG_GUEST_ACCOUNTING
 	if (p->flags & PF_VCPU) {
 		account_guest_time(p, cputime);
-		p->flags &= ~PF_VCPU;
 		return;
 	}
 #endif
-- 
1.5.2.4

Re: [PATCH] move kvm_guest_exit() after local_irq_enable()

[Avi Kivity]

Laurent Vivier wrote:
> According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if we move
> it after local_irq_enable().
>
> http://lkml.org/lkml/2007/10/15/114
>
> To simplify s390 port, we don't clear it in account_system_time().
>
> http://lkml.org/lkml/2007/10/15/183
>   

Applied (the kvm part), and added a fat comment on the barrier.  Can you 
send a signed-off-by: line?

-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [kvm-devel] [PATCH] move kvm_guest_exit() after local_irq_enable()

[Laurent Vivier]

Avi Kivity a écrit :
> Laurent Vivier wrote:
>> According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if we move
>> it after local_irq_enable().
>>
>> http://lkml.org/lkml/2007/10/15/114
>>
>> To simplify s390 port, we don't clear it in account_system_time().
>>
>> http://lkml.org/lkml/2007/10/15/183
>>   
> 
> Applied (the kvm part), and added a fat comment on the barrier.  Can you 
> send a signed-off-by: line?
> 

Sorry, I missed it...

Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>

Re: [PATCH] move kvm_guest_exit() after local_irq_enable()

[Ingo Molnar]

* Laurent Vivier <Laurent.Vivier-6ktuUTfB/bM@public.gmane.org> wrote:

> According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if 
> we move it after local_irq_enable().
> 
> http://lkml.org/lkml/2007/10/15/114
> 
> To simplify s390 port, we don't clear it in account_system_time().
> 
> http://lkml.org/lkml/2007/10/15/183

thanks - i have picked your patch up.

	Ingo

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [PATCH] move kvm_guest_exit() after local_irq_enable()

[Avi Kivity]

Ingo Molnar wrote:
> * Laurent Vivier <Laurent.Vivier-6ktuUTfB/bM@public.gmane.org> wrote:
>
>   
>> According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if 
>> we move it after local_irq_enable().
>>
>> http://lkml.org/lkml/2007/10/15/114
>>
>> To simplify s390 port, we don't clear it in account_system_time().
>>
>> http://lkml.org/lkml/2007/10/15/183
>>     
>
> thanks - i have picked your patch up.
>
>   

Note the kvm part is already in my 2.6.24 queue.


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [PATCH] move kvm_guest_exit() after local_irq_enable()

[Ingo Molnar]

* Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org> wrote:

> Ingo Molnar wrote:
> >* Laurent Vivier <Laurent.Vivier-6ktuUTfB/bM@public.gmane.org> wrote:
> >
> >  
> >>According comments from Avi, we can clear PF_VCPU in kvm_guest_exit if 
> >>we move it after local_irq_enable().
> >>
> >>http://lkml.org/lkml/2007/10/15/114
> >>
> >>To simplify s390 port, we don't clear it in account_system_time().
> >>
> >>http://lkml.org/lkml/2007/10/15/183
> >>    
> >
> >thanks - i have picked your patch up.
> >
> >  
> 
> Note the kvm part is already in my 2.6.24 queue.

ok, then please push the whole patch to Linus, the scheduler bits are:

  Acked-by: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>

	Ingo

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/

Re: [PATCH] move kvm_guest_exit() after local_irq_enable()

[Avi Kivity]

Ingo Molnar wrote:
>>
>> Note the kvm part is already in my 2.6.24 queue.
>>     
>
> ok, then please push the whole patch to Linus, the scheduler bits are:
>
>   Acked-by: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
>   

Sure, thanks.

-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/