Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-29 Thread Rafael J. Wysocki 
On Wednesday, October 29, 2014 09:54:43 AM Kevin Hilman wrote:
> Daniel Lezcano  writes:
> 
> > On 10/28/2014 08:01 AM, Preeti Murthy wrote:
> >> On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
> >>  wrote:
> >>> In the current code, the check to reflect or not the outcoming state is 
> >>> done
> >>> against the idle state which has been chosen and its value.
> >>>
> >>> Instead of doing a check in each of the reflect functions, just don't 
> >>> call reflect
> >>> if something went wrong in the idle path.
> >>>
> >>> Signed-off-by: Daniel Lezcano 
> >>> Acked-by: Nicolas Pitre 
> >>> ---
> >>>   drivers/cpuidle/governors/ladder.c | 3 +--
> >>>   drivers/cpuidle/governors/menu.c   | 4 +---
> >>>   kernel/sched/idle.c| 3 ++-
> >>>   3 files changed, 4 insertions(+), 6 deletions(-)
> >>>
> >>> diff --git a/drivers/cpuidle/governors/ladder.c 
> >>> b/drivers/cpuidle/governors/ladder.c
> >>> index fb396d6..c0b36a8 100644
> >>> --- a/drivers/cpuidle/governors/ladder.c
> >>> +++ b/drivers/cpuidle/governors/ladder.c
> >>> @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
> >>> *drv,
> >>>   static void ladder_reflect(struct cpuidle_device *dev, int index)
> >>>   {
> >>>  struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
> >>> -   if (index > 0)
> >>> -   ldev->last_state_idx = index;
> >>> +   ldev->last_state_idx = index;
> >>>   }
> >>>
> >>>   static struct cpuidle_governor ladder_governor = {
> >>> diff --git a/drivers/cpuidle/governors/menu.c 
> >>> b/drivers/cpuidle/governors/menu.c
> >>> index a17515f..3907301 100644
> >>> --- a/drivers/cpuidle/governors/menu.c
> >>> +++ b/drivers/cpuidle/governors/menu.c
> >>> @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, 
> >>> struct cpuidle_device *dev,
> >>>   static void menu_reflect(struct cpuidle_device *dev, int index)
> >>>   {
> >>>  struct menu_device *data = &__get_cpu_var(menu_devices);
> >>> -   data->last_state_idx = index;
> >>> -   if (index >= 0)
> >>> -   data->needs_update = 1;
> >>> +   data->needs_update = 1;
> >>
> >> Why is the last_state_idx not getting updated ?
> >
> > Oups, right. This is missing.
> >
> > Thanks for pointing this out.
> >
> > By the way, I don't think a back end driver is changing the selected
> > state currently and I am not sure this is desirable since we want to
> > trust the state we are going (as a best effort). 
> 
> FYI, the OMAP3 backend driver can does not always obey the selected
> state, and will return a different state than the one requested.

The ACPI driver can do that as well.

-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-29 Thread Kevin Hilman 
Daniel Lezcano  writes:

> On 10/28/2014 08:01 AM, Preeti Murthy wrote:
>> On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
>>  wrote:
>>> In the current code, the check to reflect or not the outcoming state is done
>>> against the idle state which has been chosen and its value.
>>>
>>> Instead of doing a check in each of the reflect functions, just don't call 
>>> reflect
>>> if something went wrong in the idle path.
>>>
>>> Signed-off-by: Daniel Lezcano 
>>> Acked-by: Nicolas Pitre 
>>> ---
>>>   drivers/cpuidle/governors/ladder.c | 3 +--
>>>   drivers/cpuidle/governors/menu.c   | 4 +---
>>>   kernel/sched/idle.c| 3 ++-
>>>   3 files changed, 4 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/drivers/cpuidle/governors/ladder.c 
>>> b/drivers/cpuidle/governors/ladder.c
>>> index fb396d6..c0b36a8 100644
>>> --- a/drivers/cpuidle/governors/ladder.c
>>> +++ b/drivers/cpuidle/governors/ladder.c
>>> @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
>>> *drv,
>>>   static void ladder_reflect(struct cpuidle_device *dev, int index)
>>>   {
>>>  struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
>>> -   if (index > 0)
>>> -   ldev->last_state_idx = index;
>>> +   ldev->last_state_idx = index;
>>>   }
>>>
>>>   static struct cpuidle_governor ladder_governor = {
>>> diff --git a/drivers/cpuidle/governors/menu.c 
>>> b/drivers/cpuidle/governors/menu.c
>>> index a17515f..3907301 100644
>>> --- a/drivers/cpuidle/governors/menu.c
>>> +++ b/drivers/cpuidle/governors/menu.c
>>> @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, 
>>> struct cpuidle_device *dev,
>>>   static void menu_reflect(struct cpuidle_device *dev, int index)
>>>   {
>>>  struct menu_device *data = &__get_cpu_var(menu_devices);
>>> -   data->last_state_idx = index;
>>> -   if (index >= 0)
>>> -   data->needs_update = 1;
>>> +   data->needs_update = 1;
>>
>> Why is the last_state_idx not getting updated ?
>
> Oups, right. This is missing.
>
> Thanks for pointing this out.
>
> By the way, I don't think a back end driver is changing the selected
> state currently and I am not sure this is desirable since we want to
> trust the state we are going (as a best effort). 

FYI, the OMAP3 backend driver can does not always obey the selected
state, and will return a different state than the one requested.

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-29 Thread Kevin Hilman 
Daniel Lezcano daniel.lezc...@linaro.org writes:

 On 10/28/2014 08:01 AM, Preeti Murthy wrote:
 On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
 daniel.lezc...@linaro.org wrote:
 In the current code, the check to reflect or not the outcoming state is done
 against the idle state which has been chosen and its value.

 Instead of doing a check in each of the reflect functions, just don't call 
 reflect
 if something went wrong in the idle path.

 Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
 Acked-by: Nicolas Pitre n...@linaro.org
 ---
   drivers/cpuidle/governors/ladder.c | 3 +--
   drivers/cpuidle/governors/menu.c   | 4 +---
   kernel/sched/idle.c| 3 ++-
   3 files changed, 4 insertions(+), 6 deletions(-)

 diff --git a/drivers/cpuidle/governors/ladder.c 
 b/drivers/cpuidle/governors/ladder.c
 index fb396d6..c0b36a8 100644
 --- a/drivers/cpuidle/governors/ladder.c
 +++ b/drivers/cpuidle/governors/ladder.c
 @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
 *drv,
   static void ladder_reflect(struct cpuidle_device *dev, int index)
   {
  struct ladder_device *ldev = __get_cpu_var(ladder_devices);
 -   if (index  0)
 -   ldev-last_state_idx = index;
 +   ldev-last_state_idx = index;
   }

   static struct cpuidle_governor ladder_governor = {
 diff --git a/drivers/cpuidle/governors/menu.c 
 b/drivers/cpuidle/governors/menu.c
 index a17515f..3907301 100644
 --- a/drivers/cpuidle/governors/menu.c
 +++ b/drivers/cpuidle/governors/menu.c
 @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, 
 struct cpuidle_device *dev,
   static void menu_reflect(struct cpuidle_device *dev, int index)
   {
  struct menu_device *data = __get_cpu_var(menu_devices);
 -   data-last_state_idx = index;
 -   if (index = 0)
 -   data-needs_update = 1;
 +   data-needs_update = 1;

 Why is the last_state_idx not getting updated ?

 Oups, right. This is missing.

 Thanks for pointing this out.

 By the way, I don't think a back end driver is changing the selected
 state currently and I am not sure this is desirable since we want to
 trust the state we are going (as a best effort). 

FYI, the OMAP3 backend driver can does not always obey the selected
state, and will return a different state than the one requested.

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-29 Thread Rafael J. Wysocki 
On Wednesday, October 29, 2014 09:54:43 AM Kevin Hilman wrote:
 Daniel Lezcano daniel.lezc...@linaro.org writes:
 
  On 10/28/2014 08:01 AM, Preeti Murthy wrote:
  On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
  daniel.lezc...@linaro.org wrote:
  In the current code, the check to reflect or not the outcoming state is 
  done
  against the idle state which has been chosen and its value.
 
  Instead of doing a check in each of the reflect functions, just don't 
  call reflect
  if something went wrong in the idle path.
 
  Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
  Acked-by: Nicolas Pitre n...@linaro.org
  ---
drivers/cpuidle/governors/ladder.c | 3 +--
drivers/cpuidle/governors/menu.c   | 4 +---
kernel/sched/idle.c| 3 ++-
3 files changed, 4 insertions(+), 6 deletions(-)
 
  diff --git a/drivers/cpuidle/governors/ladder.c 
  b/drivers/cpuidle/governors/ladder.c
  index fb396d6..c0b36a8 100644
  --- a/drivers/cpuidle/governors/ladder.c
  +++ b/drivers/cpuidle/governors/ladder.c
  @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
  *drv,
static void ladder_reflect(struct cpuidle_device *dev, int index)
{
   struct ladder_device *ldev = __get_cpu_var(ladder_devices);
  -   if (index  0)
  -   ldev-last_state_idx = index;
  +   ldev-last_state_idx = index;
}
 
static struct cpuidle_governor ladder_governor = {
  diff --git a/drivers/cpuidle/governors/menu.c 
  b/drivers/cpuidle/governors/menu.c
  index a17515f..3907301 100644
  --- a/drivers/cpuidle/governors/menu.c
  +++ b/drivers/cpuidle/governors/menu.c
  @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, 
  struct cpuidle_device *dev,
static void menu_reflect(struct cpuidle_device *dev, int index)
{
   struct menu_device *data = __get_cpu_var(menu_devices);
  -   data-last_state_idx = index;
  -   if (index = 0)
  -   data-needs_update = 1;
  +   data-needs_update = 1;
 
  Why is the last_state_idx not getting updated ?
 
  Oups, right. This is missing.
 
  Thanks for pointing this out.
 
  By the way, I don't think a back end driver is changing the selected
  state currently and I am not sure this is desirable since we want to
  trust the state we are going (as a best effort). 
 
 FYI, the OMAP3 backend driver can does not always obey the selected
 state, and will return a different state than the one requested.

The ACPI driver can do that as well.

-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Preeti U Murthy 
On 10/28/2014 11:58 PM, Daniel Lezcano wrote:
> On 10/28/2014 08:01 AM, Preeti Murthy wrote:
>> On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
>>  wrote:
>>> In the current code, the check to reflect or not the outcoming state
>>> is done
>>> against the idle state which has been chosen and its value.
>>>
>>> Instead of doing a check in each of the reflect functions, just don't
>>> call reflect
>>> if something went wrong in the idle path.
>>>
>>> Signed-off-by: Daniel Lezcano 
>>> Acked-by: Nicolas Pitre 
>>> ---
>>>   drivers/cpuidle/governors/ladder.c | 3 +--
>>>   drivers/cpuidle/governors/menu.c   | 4 +---
>>>   kernel/sched/idle.c| 3 ++-
>>>   3 files changed, 4 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/drivers/cpuidle/governors/ladder.c
>>> b/drivers/cpuidle/governors/ladder.c
>>> index fb396d6..c0b36a8 100644
>>> --- a/drivers/cpuidle/governors/ladder.c
>>> +++ b/drivers/cpuidle/governors/ladder.c
>>> @@ -165,8 +165,7 @@ static int ladder_enable_device(struct
>>> cpuidle_driver *drv,
>>>   static void ladder_reflect(struct cpuidle_device *dev, int index)
>>>   {
>>>  struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
>>> -   if (index > 0)
>>> -   ldev->last_state_idx = index;
>>> +   ldev->last_state_idx = index;
>>>   }
>>>
>>>   static struct cpuidle_governor ladder_governor = {
>>> diff --git a/drivers/cpuidle/governors/menu.c
>>> b/drivers/cpuidle/governors/menu.c
>>> index a17515f..3907301 100644
>>> --- a/drivers/cpuidle/governors/menu.c
>>> +++ b/drivers/cpuidle/governors/menu.c
>>> @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver
>>> *drv, struct cpuidle_device *dev,
>>>   static void menu_reflect(struct cpuidle_device *dev, int index)
>>>   {
>>>  struct menu_device *data = &__get_cpu_var(menu_devices);
>>> -   data->last_state_idx = index;
>>> -   if (index >= 0)
>>> -   data->needs_update = 1;
>>> +   data->needs_update = 1;
>>
>> Why is the last_state_idx not getting updated ?
> 
> Oups, right. This is missing.
> 
> Thanks for pointing this out.
> 
> By the way, I don't think a back end driver is changing the selected
> state currently and I am not sure this is desirable since we want to
> trust the state we are going (as a best effort). So if the 'enter'
> function does not change the index, that means the last_state_idx has
> not to be changed since it has been assigned in the 'select' function.

Hmm Right. So you might want to remove the last_state_idx update in
ladder_reflect() also?

Regards
Preeti U Murthy
> 
> 
> 

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Daniel Lezcano 

On 10/28/2014 03:01 AM, Len Brown wrote:

On Thu, Oct 23, 2014 at 5:01 AM, Daniel Lezcano
 wrote:

In the current code, the check to reflect or not the outcoming state is done
against the idle state which has been chosen and its value.

Instead of doing a check in each of the reflect functions, just don't call 
reflect
if something went wrong in the idle path.

Signed-off-by: Daniel Lezcano 
Acked-by: Nicolas Pitre 
---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/cpuidle/governors/ladder.c 
b/drivers/cpuidle/governors/ladder.c
index fb396d6..c0b36a8 100644
--- a/drivers/cpuidle/governors/ladder.c
+++ b/drivers/cpuidle/governors/ladder.c
@@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
-   if (index > 0)
-   ldev->last_state_idx = index;


Before this patch, last_state_idx was never set to 0 here.
After this patch, last_state_idx will be set to 0 when entered_state is 0.
Is that okay?


Yes, I think so because the state zero won't be never selected on x86 
but on the other arch it will. So before this patch, on the other archs, 
the state 0 was never reflected as it should do.


This is resulting from the CPUIDLE_DRIVER_STATE_START macro (I hope I 
can kill this macro in a couple of patchset after this one).



+   ldev->last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index a17515f..3907301 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = &__get_cpu_var(menu_devices);
-   data->last_state_idx = index;
-   if (index >= 0)
-   data->needs_update = 1;
+   data->needs_update = 1;
  }

  /**
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 58c7522..49dcc7d 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -162,7 +162,8 @@ use_default:
 /*
  * Give the governor an opportunity to reflect on the outcome
  */
-   cpuidle_reflect(dev, entered_state);
+   if (entered_state >= 0)
+   cpuidle_reflect(dev, entered_state);

  exit_idle:
 __current_set_polling();
--
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html







--
  Linaro.org │ Open source software for ARM SoCs

Follow Linaro:   Facebook |
 Twitter |
 Blog

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Daniel Lezcano 

On 10/28/2014 08:01 AM, Preeti Murthy wrote:

On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
 wrote:

In the current code, the check to reflect or not the outcoming state is done
against the idle state which has been chosen and its value.

Instead of doing a check in each of the reflect functions, just don't call 
reflect
if something went wrong in the idle path.

Signed-off-by: Daniel Lezcano 
Acked-by: Nicolas Pitre 
---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/cpuidle/governors/ladder.c 
b/drivers/cpuidle/governors/ladder.c
index fb396d6..c0b36a8 100644
--- a/drivers/cpuidle/governors/ladder.c
+++ b/drivers/cpuidle/governors/ladder.c
@@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
-   if (index > 0)
-   ldev->last_state_idx = index;
+   ldev->last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index a17515f..3907301 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = &__get_cpu_var(menu_devices);
-   data->last_state_idx = index;
-   if (index >= 0)
-   data->needs_update = 1;
+   data->needs_update = 1;


Why is the last_state_idx not getting updated ?


Oups, right. This is missing.

Thanks for pointing this out.

By the way, I don't think a back end driver is changing the selected 
state currently and I am not sure this is desirable since we want to 
trust the state we are going (as a best effort). So if the 'enter' 
function does not change the index, that means the last_state_idx has 
not to be changed since it has been assigned in the 'select' function.




--
  Linaro.org │ Open source software for ARM SoCs

Follow Linaro:   Facebook |
 Twitter |
 Blog

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Preeti Murthy 
On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
 wrote:
> In the current code, the check to reflect or not the outcoming state is done
> against the idle state which has been chosen and its value.
>
> Instead of doing a check in each of the reflect functions, just don't call 
> reflect
> if something went wrong in the idle path.
>
> Signed-off-by: Daniel Lezcano 
> Acked-by: Nicolas Pitre 
> ---
>  drivers/cpuidle/governors/ladder.c | 3 +--
>  drivers/cpuidle/governors/menu.c   | 4 +---
>  kernel/sched/idle.c| 3 ++-
>  3 files changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/cpuidle/governors/ladder.c 
> b/drivers/cpuidle/governors/ladder.c
> index fb396d6..c0b36a8 100644
> --- a/drivers/cpuidle/governors/ladder.c
> +++ b/drivers/cpuidle/governors/ladder.c
> @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
> *drv,
>  static void ladder_reflect(struct cpuidle_device *dev, int index)
>  {
> struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
> -   if (index > 0)
> -   ldev->last_state_idx = index;
> +   ldev->last_state_idx = index;
>  }
>
>  static struct cpuidle_governor ladder_governor = {
> diff --git a/drivers/cpuidle/governors/menu.c 
> b/drivers/cpuidle/governors/menu.c
> index a17515f..3907301 100644
> --- a/drivers/cpuidle/governors/menu.c
> +++ b/drivers/cpuidle/governors/menu.c
> @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
> cpuidle_device *dev,
>  static void menu_reflect(struct cpuidle_device *dev, int index)
>  {
> struct menu_device *data = &__get_cpu_var(menu_devices);
> -   data->last_state_idx = index;
> -   if (index >= 0)
> -   data->needs_update = 1;
> +   data->needs_update = 1;

Why is the last_state_idx not getting updated ?

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Preeti Murthy 
On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
daniel.lezc...@linaro.org wrote:
 In the current code, the check to reflect or not the outcoming state is done
 against the idle state which has been chosen and its value.

 Instead of doing a check in each of the reflect functions, just don't call 
 reflect
 if something went wrong in the idle path.

 Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
 Acked-by: Nicolas Pitre n...@linaro.org
 ---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

 diff --git a/drivers/cpuidle/governors/ladder.c 
 b/drivers/cpuidle/governors/ladder.c
 index fb396d6..c0b36a8 100644
 --- a/drivers/cpuidle/governors/ladder.c
 +++ b/drivers/cpuidle/governors/ladder.c
 @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
 *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = __get_cpu_var(ladder_devices);
 -   if (index  0)
 -   ldev-last_state_idx = index;
 +   ldev-last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
 diff --git a/drivers/cpuidle/governors/menu.c 
 b/drivers/cpuidle/governors/menu.c
 index a17515f..3907301 100644
 --- a/drivers/cpuidle/governors/menu.c
 +++ b/drivers/cpuidle/governors/menu.c
 @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
 cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = __get_cpu_var(menu_devices);
 -   data-last_state_idx = index;
 -   if (index = 0)
 -   data-needs_update = 1;
 +   data-needs_update = 1;

Why is the last_state_idx not getting updated ?

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Daniel Lezcano 

On 10/28/2014 08:01 AM, Preeti Murthy wrote:

On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
daniel.lezc...@linaro.org wrote:

In the current code, the check to reflect or not the outcoming state is done
against the idle state which has been chosen and its value.

Instead of doing a check in each of the reflect functions, just don't call 
reflect
if something went wrong in the idle path.

Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
Acked-by: Nicolas Pitre n...@linaro.org
---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/cpuidle/governors/ladder.c 
b/drivers/cpuidle/governors/ladder.c
index fb396d6..c0b36a8 100644
--- a/drivers/cpuidle/governors/ladder.c
+++ b/drivers/cpuidle/governors/ladder.c
@@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = __get_cpu_var(ladder_devices);
-   if (index  0)
-   ldev-last_state_idx = index;
+   ldev-last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index a17515f..3907301 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = __get_cpu_var(menu_devices);
-   data-last_state_idx = index;
-   if (index = 0)
-   data-needs_update = 1;
+   data-needs_update = 1;


Why is the last_state_idx not getting updated ?


Oups, right. This is missing.

Thanks for pointing this out.

By the way, I don't think a back end driver is changing the selected 
state currently and I am not sure this is desirable since we want to 
trust the state we are going (as a best effort). So if the 'enter' 
function does not change the index, that means the last_state_idx has 
not to be changed since it has been assigned in the 'select' function.




--
 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

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Daniel Lezcano 

On 10/28/2014 03:01 AM, Len Brown wrote:

On Thu, Oct 23, 2014 at 5:01 AM, Daniel Lezcano
daniel.lezc...@linaro.org wrote:

In the current code, the check to reflect or not the outcoming state is done
against the idle state which has been chosen and its value.

Instead of doing a check in each of the reflect functions, just don't call 
reflect
if something went wrong in the idle path.

Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
Acked-by: Nicolas Pitre n...@linaro.org
---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/cpuidle/governors/ladder.c 
b/drivers/cpuidle/governors/ladder.c
index fb396d6..c0b36a8 100644
--- a/drivers/cpuidle/governors/ladder.c
+++ b/drivers/cpuidle/governors/ladder.c
@@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = __get_cpu_var(ladder_devices);
-   if (index  0)
-   ldev-last_state_idx = index;


Before this patch, last_state_idx was never set to 0 here.
After this patch, last_state_idx will be set to 0 when entered_state is 0.
Is that okay?


Yes, I think so because the state zero won't be never selected on x86 
but on the other arch it will. So before this patch, on the other archs, 
the state 0 was never reflected as it should do.


This is resulting from the CPUIDLE_DRIVER_STATE_START macro (I hope I 
can kill this macro in a couple of patchset after this one).



+   ldev-last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index a17515f..3907301 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = __get_cpu_var(menu_devices);
-   data-last_state_idx = index;
-   if (index = 0)
-   data-needs_update = 1;
+   data-needs_update = 1;
  }

  /**
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 58c7522..49dcc7d 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -162,7 +162,8 @@ use_default:
 /*
  * Give the governor an opportunity to reflect on the outcome
  */
-   cpuidle_reflect(dev, entered_state);
+   if (entered_state = 0)
+   cpuidle_reflect(dev, entered_state);

  exit_idle:
 __current_set_polling();
--
1.9.1

--
To unsubscribe from this list: send the line unsubscribe linux-pm in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html







--
 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

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-28 Thread Preeti U Murthy 
On 10/28/2014 11:58 PM, Daniel Lezcano wrote:
 On 10/28/2014 08:01 AM, Preeti Murthy wrote:
 On Thu, Oct 23, 2014 at 2:31 PM, Daniel Lezcano
 daniel.lezc...@linaro.org wrote:
 In the current code, the check to reflect or not the outcoming state
 is done
 against the idle state which has been chosen and its value.

 Instead of doing a check in each of the reflect functions, just don't
 call reflect
 if something went wrong in the idle path.

 Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
 Acked-by: Nicolas Pitre n...@linaro.org
 ---
   drivers/cpuidle/governors/ladder.c | 3 +--
   drivers/cpuidle/governors/menu.c   | 4 +---
   kernel/sched/idle.c| 3 ++-
   3 files changed, 4 insertions(+), 6 deletions(-)

 diff --git a/drivers/cpuidle/governors/ladder.c
 b/drivers/cpuidle/governors/ladder.c
 index fb396d6..c0b36a8 100644
 --- a/drivers/cpuidle/governors/ladder.c
 +++ b/drivers/cpuidle/governors/ladder.c
 @@ -165,8 +165,7 @@ static int ladder_enable_device(struct
 cpuidle_driver *drv,
   static void ladder_reflect(struct cpuidle_device *dev, int index)
   {
  struct ladder_device *ldev = __get_cpu_var(ladder_devices);
 -   if (index  0)
 -   ldev-last_state_idx = index;
 +   ldev-last_state_idx = index;
   }

   static struct cpuidle_governor ladder_governor = {
 diff --git a/drivers/cpuidle/governors/menu.c
 b/drivers/cpuidle/governors/menu.c
 index a17515f..3907301 100644
 --- a/drivers/cpuidle/governors/menu.c
 +++ b/drivers/cpuidle/governors/menu.c
 @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver
 *drv, struct cpuidle_device *dev,
   static void menu_reflect(struct cpuidle_device *dev, int index)
   {
  struct menu_device *data = __get_cpu_var(menu_devices);
 -   data-last_state_idx = index;
 -   if (index = 0)
 -   data-needs_update = 1;
 +   data-needs_update = 1;

 Why is the last_state_idx not getting updated ?
 
 Oups, right. This is missing.
 
 Thanks for pointing this out.
 
 By the way, I don't think a back end driver is changing the selected
 state currently and I am not sure this is desirable since we want to
 trust the state we are going (as a best effort). So if the 'enter'
 function does not change the index, that means the last_state_idx has
 not to be changed since it has been assigned in the 'select' function.

Hmm Right. So you might want to remove the last_state_idx update in
ladder_reflect() also?

Regards
Preeti U Murthy
 
 
 

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

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-27 Thread Len Brown 
On Thu, Oct 23, 2014 at 5:01 AM, Daniel Lezcano
 wrote:
> In the current code, the check to reflect or not the outcoming state is done
> against the idle state which has been chosen and its value.
>
> Instead of doing a check in each of the reflect functions, just don't call 
> reflect
> if something went wrong in the idle path.
>
> Signed-off-by: Daniel Lezcano 
> Acked-by: Nicolas Pitre 
> ---
>  drivers/cpuidle/governors/ladder.c | 3 +--
>  drivers/cpuidle/governors/menu.c   | 4 +---
>  kernel/sched/idle.c| 3 ++-
>  3 files changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/cpuidle/governors/ladder.c 
> b/drivers/cpuidle/governors/ladder.c
> index fb396d6..c0b36a8 100644
> --- a/drivers/cpuidle/governors/ladder.c
> +++ b/drivers/cpuidle/governors/ladder.c
> @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
> *drv,
>  static void ladder_reflect(struct cpuidle_device *dev, int index)
>  {
> struct ladder_device *ldev = &__get_cpu_var(ladder_devices);
> -   if (index > 0)
> -   ldev->last_state_idx = index;

Before this patch, last_state_idx was never set to 0 here.
After this patch, last_state_idx will be set to 0 when entered_state is 0.
Is that okay?

thanks,
-Len

> +   ldev->last_state_idx = index;
>  }
>
>  static struct cpuidle_governor ladder_governor = {
> diff --git a/drivers/cpuidle/governors/menu.c 
> b/drivers/cpuidle/governors/menu.c
> index a17515f..3907301 100644
> --- a/drivers/cpuidle/governors/menu.c
> +++ b/drivers/cpuidle/governors/menu.c
> @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
> cpuidle_device *dev,
>  static void menu_reflect(struct cpuidle_device *dev, int index)
>  {
> struct menu_device *data = &__get_cpu_var(menu_devices);
> -   data->last_state_idx = index;
> -   if (index >= 0)
> -   data->needs_update = 1;
> +   data->needs_update = 1;
>  }
>
>  /**
> diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
> index 58c7522..49dcc7d 100644
> --- a/kernel/sched/idle.c
> +++ b/kernel/sched/idle.c
> @@ -162,7 +162,8 @@ use_default:
> /*
>  * Give the governor an opportunity to reflect on the outcome
>  */
> -   cpuidle_reflect(dev, entered_state);
> +   if (entered_state >= 0)
> +   cpuidle_reflect(dev, entered_state);
>
>  exit_idle:
> __current_set_polling();
> --
> 1.9.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pm" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Len Brown, Intel Open Source Technology Center
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH V2 3/5] cpuidle: idle: menu: Don't reflect when a state selection failed

2014-10-27 Thread Len Brown 
On Thu, Oct 23, 2014 at 5:01 AM, Daniel Lezcano
daniel.lezc...@linaro.org wrote:
 In the current code, the check to reflect or not the outcoming state is done
 against the idle state which has been chosen and its value.

 Instead of doing a check in each of the reflect functions, just don't call 
 reflect
 if something went wrong in the idle path.

 Signed-off-by: Daniel Lezcano daniel.lezc...@linaro.org
 Acked-by: Nicolas Pitre n...@linaro.org
 ---
  drivers/cpuidle/governors/ladder.c | 3 +--
  drivers/cpuidle/governors/menu.c   | 4 +---
  kernel/sched/idle.c| 3 ++-
  3 files changed, 4 insertions(+), 6 deletions(-)

 diff --git a/drivers/cpuidle/governors/ladder.c 
 b/drivers/cpuidle/governors/ladder.c
 index fb396d6..c0b36a8 100644
 --- a/drivers/cpuidle/governors/ladder.c
 +++ b/drivers/cpuidle/governors/ladder.c
 @@ -165,8 +165,7 @@ static int ladder_enable_device(struct cpuidle_driver 
 *drv,
  static void ladder_reflect(struct cpuidle_device *dev, int index)
  {
 struct ladder_device *ldev = __get_cpu_var(ladder_devices);
 -   if (index  0)
 -   ldev-last_state_idx = index;

Before this patch, last_state_idx was never set to 0 here.
After this patch, last_state_idx will be set to 0 when entered_state is 0.
Is that okay?

thanks,
-Len

 +   ldev-last_state_idx = index;
  }

  static struct cpuidle_governor ladder_governor = {
 diff --git a/drivers/cpuidle/governors/menu.c 
 b/drivers/cpuidle/governors/menu.c
 index a17515f..3907301 100644
 --- a/drivers/cpuidle/governors/menu.c
 +++ b/drivers/cpuidle/governors/menu.c
 @@ -365,9 +365,7 @@ static int menu_select(struct cpuidle_driver *drv, struct 
 cpuidle_device *dev,
  static void menu_reflect(struct cpuidle_device *dev, int index)
  {
 struct menu_device *data = __get_cpu_var(menu_devices);
 -   data-last_state_idx = index;
 -   if (index = 0)
 -   data-needs_update = 1;
 +   data-needs_update = 1;
  }

  /**
 diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
 index 58c7522..49dcc7d 100644
 --- a/kernel/sched/idle.c
 +++ b/kernel/sched/idle.c
 @@ -162,7 +162,8 @@ use_default:
 /*
  * Give the governor an opportunity to reflect on the outcome
  */
 -   cpuidle_reflect(dev, entered_state);
 +   if (entered_state = 0)
 +   cpuidle_reflect(dev, entered_state);

  exit_idle:
 __current_set_polling();
 --
 1.9.1

 --
 To unsubscribe from this list: send the line unsubscribe linux-pm in
 the body of a message to majord...@vger.kernel.org
 More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Len Brown, Intel Open Source Technology Center
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/