[PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Doug Anderson]

The original code for the exynos i2c controller registered for the
"noirq" variants.  However during review feedback it was moved to
SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
longer actually "noirq" (despite functions named
exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).

i2c controllers that might have wakeup sources on them seem to need to
resume at noirq time so that the individual drivers can actually read
the i2c bus to handle their wakeup.

NOTE: I took the original review feedback from Wolfram and added
poweroff, thaw, freeze, restore.

This patch has only been compile-tested since I don't have all the
patches needed to make my machine using this i2c driver actually
suspend/resume.

Signed-off-by: Doug Anderson <dianders@chromium.org>
---
Changes in v2:
- Added missing CONFIG_PM_SLEEP

 drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-exynos5.c
index 63d2292..348b1cd 100644
--- a/drivers/i2c/busses/i2c-exynos5.c
+++ b/drivers/i2c/busses/i2c-exynos5.c
@@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
 }
 #endif
 
-static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
-			 exynos5_i2c_resume_noirq);
+const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
+#ifdef CONFIG_PM_SLEEP
+	.suspend_noirq = exynos5_i2c_suspend_noirq,
+	.resume_noirq = exynos5_i2c_resume_noirq,
+	.freeze_noirq = exynos5_i2c_suspend_noirq,
+	.thaw_noirq = exynos5_i2c_resume_noirq,
+	.poweroff_noirq = exynos5_i2c_suspend_noirq,
+	.restore_noirq = exynos5_i2c_resume_noirq,
+#endif
+};
 
 static struct platform_driver exynos5_i2c_driver = {
 	.probe		= exynos5_i2c_probe,
-- 
2.0.0.526.g5318336

[PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Doug Anderson]

The original code for the exynos i2c controller registered for the
"noirq" variants.  However during review feedback it was moved to
SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
longer actually "noirq" (despite functions named
exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).

i2c controllers that might have wakeup sources on them seem to need to
resume at noirq time so that the individual drivers can actually read
the i2c bus to handle their wakeup.

NOTE: I took the original review feedback from Wolfram and added
poweroff, thaw, freeze, restore.

This patch has only been compile-tested since I don't have all the
patches needed to make my machine using this i2c driver actually
suspend/resume.

Signed-off-by: Doug Anderson <dianders@chromium.org>
---
Changes in v2:
- Added missing CONFIG_PM_SLEEP

 drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-exynos5.c
index 63d2292..348b1cd 100644
--- a/drivers/i2c/busses/i2c-exynos5.c
+++ b/drivers/i2c/busses/i2c-exynos5.c
@@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
 }
 #endif
 
-static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
-			 exynos5_i2c_resume_noirq);
+const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
+#ifdef CONFIG_PM_SLEEP
+	.suspend_noirq = exynos5_i2c_suspend_noirq,
+	.resume_noirq = exynos5_i2c_resume_noirq,
+	.freeze_noirq = exynos5_i2c_suspend_noirq,
+	.thaw_noirq = exynos5_i2c_resume_noirq,
+	.poweroff_noirq = exynos5_i2c_suspend_noirq,
+	.restore_noirq = exynos5_i2c_resume_noirq,
+#endif
+};
 
 static struct platform_driver exynos5_i2c_driver = {
 	.probe		= exynos5_i2c_probe,
-- 
2.0.0.526.g5318336

RE: [PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Kukjin Kim]

Doug Anderson wrote:
> 
> The original code for the exynos i2c controller registered for the
> "noirq" variants.  However during review feedback it was moved to
> SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
> longer actually "noirq" (despite functions named
> exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).
> 
> i2c controllers that might have wakeup sources on them seem to need to
> resume at noirq time so that the individual drivers can actually read
> the i2c bus to handle their wakeup.
> 
> NOTE: I took the original review feedback from Wolfram and added
> poweroff, thaw, freeze, restore.
> 
Yeah I'm not sure except .suspend_noirq and .resume_noirq but I'm fine if
Wolfram suggested ;-)

> This patch has only been compile-tested since I don't have all the
> patches needed to make my machine using this i2c driver actually
> suspend/resume.
> 
> Signed-off-by: Doug Anderson <dianders@chromium.org>
> ---
> Changes in v2:
> - Added missing CONFIG_PM_SLEEP
> 
>  drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-
> exynos5.c
> index 63d2292..348b1cd 100644
> --- a/drivers/i2c/busses/i2c-exynos5.c
> +++ b/drivers/i2c/busses/i2c-exynos5.c
> @@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
>  }
>  #endif
> 
> -static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
> -			 exynos5_i2c_resume_noirq);
> +const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {

Maybe static const struct...?

> +#ifdef CONFIG_PM_SLEEP
> +	.suspend_noirq = exynos5_i2c_suspend_noirq,
> +	.resume_noirq = exynos5_i2c_resume_noirq,
> +	.freeze_noirq = exynos5_i2c_suspend_noirq,
> +	.thaw_noirq = exynos5_i2c_resume_noirq,
> +	.poweroff_noirq = exynos5_i2c_suspend_noirq,
> +	.restore_noirq = exynos5_i2c_resume_noirq,
> +#endif
> +};
> 
>  static struct platform_driver exynos5_i2c_driver = {
>  	.probe		= exynos5_i2c_probe,
> --
> 2.0.0.526.g5318336

Others look good to me,

Acked-by: Kukjin Kim <kgene.kim@samsung.com>

Thanks,
Kukjin

[PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Kukjin Kim]

Doug Anderson wrote:
> 
> The original code for the exynos i2c controller registered for the
> "noirq" variants.  However during review feedback it was moved to
> SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
> longer actually "noirq" (despite functions named
> exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).
> 
> i2c controllers that might have wakeup sources on them seem to need to
> resume at noirq time so that the individual drivers can actually read
> the i2c bus to handle their wakeup.
> 
> NOTE: I took the original review feedback from Wolfram and added
> poweroff, thaw, freeze, restore.
> 
Yeah I'm not sure except .suspend_noirq and .resume_noirq but I'm fine if
Wolfram suggested ;-)

> This patch has only been compile-tested since I don't have all the
> patches needed to make my machine using this i2c driver actually
> suspend/resume.
> 
> Signed-off-by: Doug Anderson <dianders@chromium.org>
> ---
> Changes in v2:
> - Added missing CONFIG_PM_SLEEP
> 
>  drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-
> exynos5.c
> index 63d2292..348b1cd 100644
> --- a/drivers/i2c/busses/i2c-exynos5.c
> +++ b/drivers/i2c/busses/i2c-exynos5.c
> @@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
>  }
>  #endif
> 
> -static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
> -			 exynos5_i2c_resume_noirq);
> +const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {

Maybe static const struct...?

> +#ifdef CONFIG_PM_SLEEP
> +	.suspend_noirq = exynos5_i2c_suspend_noirq,
> +	.resume_noirq = exynos5_i2c_resume_noirq,
> +	.freeze_noirq = exynos5_i2c_suspend_noirq,
> +	.thaw_noirq = exynos5_i2c_resume_noirq,
> +	.poweroff_noirq = exynos5_i2c_suspend_noirq,
> +	.restore_noirq = exynos5_i2c_resume_noirq,
> +#endif
> +};
> 
>  static struct platform_driver exynos5_i2c_driver = {
>  	.probe		= exynos5_i2c_probe,
> --
> 2.0.0.526.g5318336

Others look good to me,

Acked-by: Kukjin Kim <kgene.kim@samsung.com>

Thanks,
Kukjin

Re: [PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Doug Anderson]

Kukjin,

On Wed, Jun 25, 2014 at 4:13 AM, Kukjin Kim <kgene.kim@samsung.com> wrote:
> Doug Anderson wrote:
>>
>> The original code for the exynos i2c controller registered for the
>> "noirq" variants.  However during review feedback it was moved to
>> SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
>> longer actually "noirq" (despite functions named
>> exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).
>>
>> i2c controllers that might have wakeup sources on them seem to need to
>> resume at noirq time so that the individual drivers can actually read
>> the i2c bus to handle their wakeup.
>>
>> NOTE: I took the original review feedback from Wolfram and added
>> poweroff, thaw, freeze, restore.
>>
> Yeah I'm not sure except .suspend_noirq and .resume_noirq but I'm fine if
> Wolfram suggested ;-)

Sorry, I didn't mean to imply that Wolfram suggested the "noirq" versions.

Specifically in <https://lkml.org/lkml/2013/9/8/133> Naveen had:

> +static const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
> + .suspend_noirq = exynos5_i2c_suspend_noirq,
> + .resume_noirq = exynos5_i2c_resume_noirq,
> +};
> +
> +#define EXYNOS5_DEV_PM_OPS (&exynos5_i2c_dev_pm_ops)
> +#else
> +#define EXYNOS5_DEV_PM_OPS NULL
> +#endif

And Wolfram said:

> Isn't there a macro for this? SIMPLE_DEV_PM_OPS*? Not sure, I always mix
> them up...

That had the side effect of getting freeze, restore, ...

Ah, I also see that Felipe Balbi was the one that gave earlier
feedback about this also at <https://lkml.org/lkml/2012/11/27/262>.
He said "you need to define poweroff, thaw, freeze, restore."


>> This patch has only been compile-tested since I don't have all the
>> patches needed to make my machine using this i2c driver actually
>> suspend/resume.
>>
>> Signed-off-by: Doug Anderson <dianders@chromium.org>
>> ---
>> Changes in v2:
>> - Added missing CONFIG_PM_SLEEP
>>
>>  drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
>>  1 file changed, 10 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-
>> exynos5.c
>> index 63d2292..348b1cd 100644
>> --- a/drivers/i2c/busses/i2c-exynos5.c
>> +++ b/drivers/i2c/busses/i2c-exynos5.c
>> @@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
>>  }
>>  #endif
>>
>> -static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
>> -                      exynos5_i2c_resume_noirq);
>> +const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
>
> Maybe static const struct...?

Duh, right.  Fixing and will spin.


>> +#ifdef CONFIG_PM_SLEEP
>> +     .suspend_noirq = exynos5_i2c_suspend_noirq,
>> +     .resume_noirq = exynos5_i2c_resume_noirq,
>> +     .freeze_noirq = exynos5_i2c_suspend_noirq,
>> +     .thaw_noirq = exynos5_i2c_resume_noirq,
>> +     .poweroff_noirq = exynos5_i2c_suspend_noirq,
>> +     .restore_noirq = exynos5_i2c_resume_noirq,
>> +#endif
>> +};
>>
>>  static struct platform_driver exynos5_i2c_driver = {
>>       .probe          = exynos5_i2c_probe,
>> --
>> 2.0.0.526.g5318336
>
> Others look good to me,
>
> Acked-by: Kukjin Kim <kgene.kim@samsung.com>
>
> Thanks,
> Kukjin
>

Re: [PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Doug Anderson]

Kukjin,

On Wed, Jun 25, 2014 at 4:13 AM, Kukjin Kim <kgene.kim-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org> wrote:
> Doug Anderson wrote:
>>
>> The original code for the exynos i2c controller registered for the
>> "noirq" variants.  However during review feedback it was moved to
>> SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
>> longer actually "noirq" (despite functions named
>> exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).
>>
>> i2c controllers that might have wakeup sources on them seem to need to
>> resume at noirq time so that the individual drivers can actually read
>> the i2c bus to handle their wakeup.
>>
>> NOTE: I took the original review feedback from Wolfram and added
>> poweroff, thaw, freeze, restore.
>>
> Yeah I'm not sure except .suspend_noirq and .resume_noirq but I'm fine if
> Wolfram suggested ;-)

Sorry, I didn't mean to imply that Wolfram suggested the "noirq" versions.

Specifically in <https://lkml.org/lkml/2013/9/8/133> Naveen had:

> +static const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
> + .suspend_noirq = exynos5_i2c_suspend_noirq,
> + .resume_noirq = exynos5_i2c_resume_noirq,
> +};
> +
> +#define EXYNOS5_DEV_PM_OPS (&exynos5_i2c_dev_pm_ops)
> +#else
> +#define EXYNOS5_DEV_PM_OPS NULL
> +#endif

And Wolfram said:

> Isn't there a macro for this? SIMPLE_DEV_PM_OPS*? Not sure, I always mix
> them up...

That had the side effect of getting freeze, restore, ...

Ah, I also see that Felipe Balbi was the one that gave earlier
feedback about this also at <https://lkml.org/lkml/2012/11/27/262>.
He said "you need to define poweroff, thaw, freeze, restore."


>> This patch has only been compile-tested since I don't have all the
>> patches needed to make my machine using this i2c driver actually
>> suspend/resume.
>>
>> Signed-off-by: Doug Anderson <dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
>> ---
>> Changes in v2:
>> - Added missing CONFIG_PM_SLEEP
>>
>>  drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
>>  1 file changed, 10 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-
>> exynos5.c
>> index 63d2292..348b1cd 100644
>> --- a/drivers/i2c/busses/i2c-exynos5.c
>> +++ b/drivers/i2c/busses/i2c-exynos5.c
>> @@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
>>  }
>>  #endif
>>
>> -static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
>> -                      exynos5_i2c_resume_noirq);
>> +const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
>
> Maybe static const struct...?

Duh, right.  Fixing and will spin.


>> +#ifdef CONFIG_PM_SLEEP
>> +     .suspend_noirq = exynos5_i2c_suspend_noirq,
>> +     .resume_noirq = exynos5_i2c_resume_noirq,
>> +     .freeze_noirq = exynos5_i2c_suspend_noirq,
>> +     .thaw_noirq = exynos5_i2c_resume_noirq,
>> +     .poweroff_noirq = exynos5_i2c_suspend_noirq,
>> +     .restore_noirq = exynos5_i2c_resume_noirq,
>> +#endif
>> +};
>>
>>  static struct platform_driver exynos5_i2c_driver = {
>>       .probe          = exynos5_i2c_probe,
>> --
>> 2.0.0.526.g5318336
>
> Others look good to me,
>
> Acked-by: Kukjin Kim <kgene.kim-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>
>
> Thanks,
> Kukjin
>

[PATCH v2] i2c: exynos5: Properly use the "noirq" variants of suspend/resume

[Doug Anderson]

Kukjin,

On Wed, Jun 25, 2014 at 4:13 AM, Kukjin Kim <kgene.kim@samsung.com> wrote:
> Doug Anderson wrote:
>>
>> The original code for the exynos i2c controller registered for the
>> "noirq" variants.  However during review feedback it was moved to
>> SIMPLE_DEV_PM_OPS without anyone noticing that it meant we were no
>> longer actually "noirq" (despite functions named
>> exynos5_i2c_suspend_noirq and exynos5_i2c_resume_noirq).
>>
>> i2c controllers that might have wakeup sources on them seem to need to
>> resume at noirq time so that the individual drivers can actually read
>> the i2c bus to handle their wakeup.
>>
>> NOTE: I took the original review feedback from Wolfram and added
>> poweroff, thaw, freeze, restore.
>>
> Yeah I'm not sure except .suspend_noirq and .resume_noirq but I'm fine if
> Wolfram suggested ;-)

Sorry, I didn't mean to imply that Wolfram suggested the "noirq" versions.

Specifically in <https://lkml.org/lkml/2013/9/8/133> Naveen had:

> +static const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
> + .suspend_noirq = exynos5_i2c_suspend_noirq,
> + .resume_noirq = exynos5_i2c_resume_noirq,
> +};
> +
> +#define EXYNOS5_DEV_PM_OPS (&exynos5_i2c_dev_pm_ops)
> +#else
> +#define EXYNOS5_DEV_PM_OPS NULL
> +#endif

And Wolfram said:

> Isn't there a macro for this? SIMPLE_DEV_PM_OPS*? Not sure, I always mix
> them up...

That had the side effect of getting freeze, restore, ...

Ah, I also see that Felipe Balbi was the one that gave earlier
feedback about this also at <https://lkml.org/lkml/2012/11/27/262>.
He said "you need to define poweroff, thaw, freeze, restore."


>> This patch has only been compile-tested since I don't have all the
>> patches needed to make my machine using this i2c driver actually
>> suspend/resume.
>>
>> Signed-off-by: Doug Anderson <dianders@chromium.org>
>> ---
>> Changes in v2:
>> - Added missing CONFIG_PM_SLEEP
>>
>>  drivers/i2c/busses/i2c-exynos5.c | 12 ++++++++++--
>>  1 file changed, 10 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-
>> exynos5.c
>> index 63d2292..348b1cd 100644
>> --- a/drivers/i2c/busses/i2c-exynos5.c
>> +++ b/drivers/i2c/busses/i2c-exynos5.c
>> @@ -789,8 +789,16 @@ static int exynos5_i2c_resume_noirq(struct device *dev)
>>  }
>>  #endif
>>
>> -static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
>> -                      exynos5_i2c_resume_noirq);
>> +const struct dev_pm_ops exynos5_i2c_dev_pm_ops = {
>
> Maybe static const struct...?

Duh, right.  Fixing and will spin.


>> +#ifdef CONFIG_PM_SLEEP
>> +     .suspend_noirq = exynos5_i2c_suspend_noirq,
>> +     .resume_noirq = exynos5_i2c_resume_noirq,
>> +     .freeze_noirq = exynos5_i2c_suspend_noirq,
>> +     .thaw_noirq = exynos5_i2c_resume_noirq,
>> +     .poweroff_noirq = exynos5_i2c_suspend_noirq,
>> +     .restore_noirq = exynos5_i2c_resume_noirq,
>> +#endif
>> +};
>>
>>  static struct platform_driver exynos5_i2c_driver = {
>>       .probe          = exynos5_i2c_probe,
>> --
>> 2.0.0.526.g5318336
>
> Others look good to me,
>
> Acked-by: Kukjin Kim <kgene.kim@samsung.com>
>
> Thanks,
> Kukjin
>