[3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler

Commit Message

claudiu beznea Aug. 26, 2024, 3:25 p.m. UTC

From: Claudiu Beznea <[e-mail]>

On RZ/G3S the watchdog can be part of a software-controlled PM domain. In
this case, the watchdog device need to be powered on in
struct watchdog_ops::restart API. This can be done though
pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog
device are marked as IRQ safe. We mark the watchdog PM domain as IRQ safe
with GENPD_FLAG_IRQ_SAFE when the watchdog PM domain is registered and the
watchdog device though pm_runtime_irq_safe().

Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
context'") pm_runtime_get_sync() was used in watchdog restart handler
(which is similar to pm_runtime_resume_and_get() except the later one
handles the runtime resume errors).

Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
context'") dropped the pm_runtime_get_sync() and replaced it with
clk_prepare_enable() to avoid invalid wait context due to genpd_lock()
in genpd_runtime_resume() being called from atomic context. But
clk_prepare_enable() doesn't fit for this either (as reported by
Ulf Hansson) as clk_prepare() can also sleep (it just not throw invalid
wait context warning as it is not written for this).

Because the watchdog device is marked now as IRQ safe (though this patch)
the irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume() returns
1 for devices not registering an IRQ safe PM domain for watchdog (as the
watchdog device is IRQ safe, PM domain is not and watchdog PM domain is
always-on), this being the case of RZ/G2 devices that uses this driver,
we can now drop also the clk_prepare_enable() calls in restart handler and
rely on pm_runtime_resume_and_get().

Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in
watchdog restart handler.

Signed-off-by: Claudiu Beznea <[e-mail]>
---
 drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

Comments

Guenter Roeck Aug. 27, 2024, 5:04 a.m. UTC | #1

On 8/26/24 08:25, Claudiu wrote:
> From: Claudiu Beznea <[e-mail]>
> 
> On RZ/G3S the watchdog can be part of a software-controlled PM domain. In
> this case, the watchdog device need to be powered on in
> struct watchdog_ops::restart API. This can be done though
> pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog
> device are marked as IRQ safe. We mark the watchdog PM domain as IRQ safe
> with GENPD_FLAG_IRQ_SAFE when the watchdog PM domain is registered and the
> watchdog device though pm_runtime_irq_safe().
> 
> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> context'") pm_runtime_get_sync() was used in watchdog restart handler
> (which is similar to pm_runtime_resume_and_get() except the later one
> handles the runtime resume errors).
> 
> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> context'") dropped the pm_runtime_get_sync() and replaced it with
> clk_prepare_enable() to avoid invalid wait context due to genpd_lock()
> in genpd_runtime_resume() being called from atomic context. But
> clk_prepare_enable() doesn't fit for this either (as reported by
> Ulf Hansson) as clk_prepare() can also sleep (it just not throw invalid
> wait context warning as it is not written for this).
> 
> Because the watchdog device is marked now as IRQ safe (though this patch)
> the irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume() returns
> 1 for devices not registering an IRQ safe PM domain for watchdog (as the
> watchdog device is IRQ safe, PM domain is not and watchdog PM domain is
> always-on), this being the case of RZ/G2 devices that uses this driver,
> we can now drop also the clk_prepare_enable() calls in restart handler and
> rely on pm_runtime_resume_and_get().
> 
> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in
> watchdog restart handler.
> 
> Signed-off-by: Claudiu Beznea <[e-mail]>

Acked-by: Guenter Roeck <[e-mail]>

Biju Das Aug. 27, 2024, 12:15 p.m. UTC | #2

Hi Claudiu,

Thanks for the feedback.

> -----Original Message-----
> From: Claudiu <[e-mail]>
> Sent: Monday, August 26, 2024 4:25 PM
> Subject: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler
> 
> From: Claudiu Beznea <[e-mail]>
> 
> On RZ/G3S the watchdog can be part of a software-controlled PM domain. In this case, the watchdog
> device need to be powered on in struct watchdog_ops::restart API. This can be done though
> pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog device are marked as IRQ safe.
> We mark the watchdog PM domain as IRQ safe with GENPD_FLAG_IRQ_SAFE when the watchdog PM domain is
> registered and the watchdog device though pm_runtime_irq_safe().
> 
> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> context'") pm_runtime_get_sync() was used in watchdog restart handler (which is similar to
> pm_runtime_resume_and_get() except the later one handles the runtime resume errors).
> 
> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> context'") dropped the pm_runtime_get_sync() and replaced it with
> clk_prepare_enable() to avoid invalid wait context due to genpd_lock() in genpd_runtime_resume() being
> called from atomic context. But
> clk_prepare_enable() doesn't fit for this either (as reported by Ulf Hansson) as clk_prepare() can
> also sleep (it just not throw invalid wait context warning as it is not written for this).
> 
> Because the watchdog device is marked now as IRQ safe (though this patch) the
> irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume() returns
> 1 for devices not registering an IRQ safe PM domain for watchdog (as the watchdog device is IRQ safe,
> PM domain is not and watchdog PM domain is always-on), this being the case of RZ/G2 devices that uses

RZ/G2L alike devices or be specific RZ/{G2L,G2LC,G2UL,V2L} as it is not applicable for
RZ/G2{H,M,N,E}devices.


> this driver, we can now drop also the clk_prepare_enable() calls in restart handler and rely on
> pm_runtime_resume_and_get().
> 
> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in watchdog restart handler.

Can this patch be fix for Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> context'") on RZ/{G2L,G2LC,G2UL,V2L} SoC??

Cheers,
Biju

> 
> Signed-off-by: Claudiu Beznea <[e-mail]>
> ---
>  drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
>  1 file changed, 19 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/watchdog/rzg2l_wdt.c b/drivers/watchdog/rzg2l_wdt.c index
> 2a35f890a288..e9e0408c96f7 100644
> --- a/drivers/watchdog/rzg2l_wdt.c
> +++ b/drivers/watchdog/rzg2l_wdt.c
> @@ -12,6 +12,7 @@
>  #include <linux/module.h>
>  #include <linux/of.h>
>  #include <linux/platform_device.h>
> +#include <linux/pm_domain.h>
>  #include <linux/pm_runtime.h>
>  #include <linux/reset.h>
>  #include <linux/units.h>
> @@ -166,8 +167,23 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev,
>  	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
>  	int ret;
> 
> -	clk_prepare_enable(priv->pclk);
> -	clk_prepare_enable(priv->osc_clk);
> +	/*
> +	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
> +	 * domain that is currently powered off. In this case we need to power
> +	 * it on before accessing registers. Along with this the clocks will be
> +	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
> +	 * need to be on for the reboot to happen.
> +	 *
> +	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device trees
> +	 * where PM domains are registered like on RZ/G2 devices) it is safe
> +	 * to call pm_runtime_resume_and_get() as the
> +	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
> +	 * returns non zero value and the genpd_lock() is avoided, thus, there
> +	 * will be no invalid wait context reported by lockdep.
> +	 */
> +	ret = pm_runtime_resume_and_get(wdev->parent);
> +	if (ret)
> +		return ret;
> 
>  	if (priv->devtype == WDT_RZG2L) {
>  		ret = reset_control_deassert(priv->rstc);
> @@ -275,6 +291,7 @@ static int rzg2l_wdt_probe(struct platform_device *pdev)
> 
>  	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
> 
> +	pm_runtime_irq_safe(&pdev->dev);
>  	pm_runtime_enable(&pdev->dev);
> 
>  	priv->wdev.info = &rzg2l_wdt_ident;
> --
> 2.39.2
>

claudiu beznea Aug. 27, 2024, 12:32 p.m. UTC | #3

Hi, Biju,

On 27.08.2024 15:15, Biju Das wrote:
> Hi Claudiu,
> 
> Thanks for the feedback.
> 
>> -----Original Message-----
>> From: Claudiu <[e-mail]>
>> Sent: Monday, August 26, 2024 4:25 PM
>> Subject: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler
>>
>> From: Claudiu Beznea <[e-mail]>
>>
>> On RZ/G3S the watchdog can be part of a software-controlled PM domain. In this case, the watchdog
>> device need to be powered on in struct watchdog_ops::restart API. This can be done though
>> pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog device are marked as IRQ safe.
>> We mark the watchdog PM domain as IRQ safe with GENPD_FLAG_IRQ_SAFE when the watchdog PM domain is
>> registered and the watchdog device though pm_runtime_irq_safe().
>>
>> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
>> context'") pm_runtime_get_sync() was used in watchdog restart handler (which is similar to
>> pm_runtime_resume_and_get() except the later one handles the runtime resume errors).
>>
>> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
>> context'") dropped the pm_runtime_get_sync() and replaced it with
>> clk_prepare_enable() to avoid invalid wait context due to genpd_lock() in genpd_runtime_resume() being
>> called from atomic context. But
>> clk_prepare_enable() doesn't fit for this either (as reported by Ulf Hansson) as clk_prepare() can
>> also sleep (it just not throw invalid wait context warning as it is not written for this).
>>
>> Because the watchdog device is marked now as IRQ safe (though this patch) the
>> irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume() returns
>> 1 for devices not registering an IRQ safe PM domain for watchdog (as the watchdog device is IRQ safe,
>> PM domain is not and watchdog PM domain is always-on), this being the case of RZ/G2 devices that uses
> 
> RZ/G2L alike devices or be specific RZ/{G2L,G2LC,G2UL,V2L} as it is not applicable for
> RZ/G2{H,M,N,E}devices.

OK, but I said "RZ/G2 devices that uses this driver". Here are included
RZ/{G2L,G2LC,G2UL,V2L} AFAICT.

> 
> 
>> this driver, we can now drop also the clk_prepare_enable() calls in restart handler and rely on
>> pm_runtime_resume_and_get().
>>
>> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in watchdog restart handler.
> 
> Can this patch be fix for Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
>> context'") on RZ/{G2L,G2LC,G2UL,V2L} SoC??

Not sure... I thought about it, too. I chose to have it like this thinking
that:

1/ that may not apply cleanly as it depends on other cleanups done on this
   driver, e.g. commit d8997ed79ed7 ("watchdog: rzg2l_wdt: Rely on the
   reset driver for doing proper reset") so it may be worthless for
   backport machinery
2/ There is actually no seen bug reported by lockdep (as the clk_prepare()
   doesn't handle it)

Don't know, I can reply here and add it. Applying this patch with b4 will
take care of it. But not sure about it.

Thank you,
Claudiu Beznea

> 
> Cheers,
> Biju
> 
>>
>> Signed-off-by: Claudiu Beznea <[e-mail]>
>> ---
>>  drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
>>  1 file changed, 19 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/watchdog/rzg2l_wdt.c b/drivers/watchdog/rzg2l_wdt.c index
>> 2a35f890a288..e9e0408c96f7 100644
>> --- a/drivers/watchdog/rzg2l_wdt.c
>> +++ b/drivers/watchdog/rzg2l_wdt.c
>> @@ -12,6 +12,7 @@
>>  #include <linux/module.h>
>>  #include <linux/of.h>
>>  #include <linux/platform_device.h>
>> +#include <linux/pm_domain.h>
>>  #include <linux/pm_runtime.h>
>>  #include <linux/reset.h>
>>  #include <linux/units.h>
>> @@ -166,8 +167,23 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev,
>>  	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
>>  	int ret;
>>
>> -	clk_prepare_enable(priv->pclk);
>> -	clk_prepare_enable(priv->osc_clk);
>> +	/*
>> +	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
>> +	 * domain that is currently powered off. In this case we need to power
>> +	 * it on before accessing registers. Along with this the clocks will be
>> +	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
>> +	 * need to be on for the reboot to happen.
>> +	 *
>> +	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device trees
>> +	 * where PM domains are registered like on RZ/G2 devices) it is safe
>> +	 * to call pm_runtime_resume_and_get() as the
>> +	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
>> +	 * returns non zero value and the genpd_lock() is avoided, thus, there
>> +	 * will be no invalid wait context reported by lockdep.
>> +	 */
>> +	ret = pm_runtime_resume_and_get(wdev->parent);
>> +	if (ret)
>> +		return ret;
>>
>>  	if (priv->devtype == WDT_RZG2L) {
>>  		ret = reset_control_deassert(priv->rstc);
>> @@ -275,6 +291,7 @@ static int rzg2l_wdt_probe(struct platform_device *pdev)
>>
>>  	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
>>
>> +	pm_runtime_irq_safe(&pdev->dev);
>>  	pm_runtime_enable(&pdev->dev);
>>
>>  	priv->wdev.info = &rzg2l_wdt_ident;
>> --
>> 2.39.2
>>
>

Biju Das Aug. 27, 2024, 12:51 p.m. UTC | #4

> -----Original Message-----
> From: claudiu beznea <[e-mail]>
> Sent: Tuesday, August 27, 2024 1:33 PM
> To: Biju Das <[e-mail]>; [e-mail]; [e-mail];
> [e-mail]; [e-mail]; [e-mail]; [e-mail]
> Cc: [e-mail]; [e-mail]; [e-mail]; linux-
> [e-mail]; [e-mail]; Claudiu Beznea <[e-mail]>
> Subject: Re: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler
> 
> Hi, Biju,
> 
> On 27.08.2024 15:15, Biju Das wrote:
> > Hi Claudiu,
> >
> > Thanks for the feedback.
> >
> >> -----Original Message-----
> >> From: Claudiu <[e-mail]>
> >> Sent: Monday, August 26, 2024 4:25 PM
> >> Subject: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog
> >> domain in the restart handler
> >>
> >> From: Claudiu Beznea <[e-mail]>
> >>
> >> On RZ/G3S the watchdog can be part of a software-controlled PM
> >> domain. In this case, the watchdog device need to be powered on in
> >> struct watchdog_ops::restart API. This can be done though
> >> pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog device are marked as IRQ
> safe.
> >> We mark the watchdog PM domain as IRQ safe with GENPD_FLAG_IRQ_SAFE
> >> when the watchdog PM domain is registered and the watchdog device though pm_runtime_irq_safe().
> >>
> >> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid
> >> wait
> >> context'") pm_runtime_get_sync() was used in watchdog restart handler
> >> (which is similar to
> >> pm_runtime_resume_and_get() except the later one handles the runtime resume errors).
> >>
> >> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> >> context'") dropped the pm_runtime_get_sync() and replaced it with
> >> clk_prepare_enable() to avoid invalid wait context due to
> >> genpd_lock() in genpd_runtime_resume() being called from atomic
> >> context. But
> >> clk_prepare_enable() doesn't fit for this either (as reported by Ulf
> >> Hansson) as clk_prepare() can also sleep (it just not throw invalid wait context warning as it is
> not written for this).
> >>
> >> Because the watchdog device is marked now as IRQ safe (though this
> >> patch) the
> >> irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume()
> >> returns
> >> 1 for devices not registering an IRQ safe PM domain for watchdog (as
> >> the watchdog device is IRQ safe, PM domain is not and watchdog PM
> >> domain is always-on), this being the case of RZ/G2 devices that uses
> >
> > RZ/G2L alike devices or be specific RZ/{G2L,G2LC,G2UL,V2L} as it is
> > not applicable for RZ/G2{H,M,N,E}devices.
> 
> OK, but I said "RZ/G2 devices that uses this driver". Here are included RZ/{G2L,G2LC,G2UL,V2L} AFAICT.

OK. Not sure you missed the same terminology on comment section, see below??

> 
> >
> >
> >> this driver, we can now drop also the clk_prepare_enable() calls in
> >> restart handler and rely on pm_runtime_resume_and_get().
> >>
> >> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in watchdog restart handler.
> >
> > Can this patch be fix for Commit e4cf89596c1f ("watchdog: rzg2l_wdt:
> > Fix 'BUG: Invalid wait
> >> context'") on RZ/{G2L,G2LC,G2UL,V2L} SoC??
> 
> Not sure... I thought about it, too. I chose to have it like this thinking
> that:
> 
> 1/ that may not apply cleanly as it depends on other cleanups done on this
>    driver, e.g. commit d8997ed79ed7 ("watchdog: rzg2l_wdt: Rely on the
>    reset driver for doing proper reset") so it may be worthless for
>    backport machinery
> 2/ There is actually no seen bug reported by lockdep (as the clk_prepare()
>    doesn't handle it)
> 
> Don't know, I can reply here and add it. Applying this patch with b4 will take care of it. But not
> sure about it.

Maybe leave it.

> >
> >>
> >> Signed-off-by: Claudiu Beznea <[e-mail]>
> >> ---
> >>  drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
> >>  1 file changed, 19 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/drivers/watchdog/rzg2l_wdt.c
> >> b/drivers/watchdog/rzg2l_wdt.c index
> >> 2a35f890a288..e9e0408c96f7 100644
> >> --- a/drivers/watchdog/rzg2l_wdt.c
> >> +++ b/drivers/watchdog/rzg2l_wdt.c
> >> @@ -12,6 +12,7 @@
> >>  #include <linux/module.h>
> >>  #include <linux/of.h>
> >>  #include <linux/platform_device.h>
> >> +#include <linux/pm_domain.h>
> >>  #include <linux/pm_runtime.h>
> >>  #include <linux/reset.h>
> >>  #include <linux/units.h>
> >> @@ -166,8 +167,23 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev,
> >>  	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
> >>  	int ret;
> >>
> >> -	clk_prepare_enable(priv->pclk);
> >> -	clk_prepare_enable(priv->osc_clk);
> >> +	/*
> >> +	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
> >> +	 * domain that is currently powered off. In this case we need to power
> >> +	 * it on before accessing registers. Along with this the clocks will be
> >> +	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
> >> +	 * need to be on for the reboot to happen.
> >> +	 *
> >> +	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device trees

NitPick: For the rest of RZ/G2 devices that uses this driver (This will make sure
It is not meant for RZ/G2{H,M,N,E} devices)

Cheers,
Biju



> >> +	 * where PM domains are registered like on RZ/G2 devices) it is safe
> >> +	 * to call pm_runtime_resume_and_get() as the
> >> +	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
> >> +	 * returns non zero value and the genpd_lock() is avoided, thus, there
> >> +	 * will be no invalid wait context reported by lockdep.
> >> +	 */
> >> +	ret = pm_runtime_resume_and_get(wdev->parent);
> >> +	if (ret)
> >> +		return ret;
> >>
> >>  	if (priv->devtype == WDT_RZG2L) {
> >>  		ret = reset_control_deassert(priv->rstc);
> >> @@ -275,6 +291,7 @@ static int rzg2l_wdt_probe(struct platform_device
> >> *pdev)
> >>
> >>  	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
> >>
> >> +	pm_runtime_irq_safe(&pdev->dev);
> >>  	pm_runtime_enable(&pdev->dev);
> >>
> >>  	priv->wdev.info = &rzg2l_wdt_ident;
> >> --
> >> 2.39.2
> >>
> >

claudiu beznea Aug. 27, 2024, 1:34 p.m. UTC | #5

On 27.08.2024 15:51, Biju Das wrote:
> 
> 
>> -----Original Message-----
>> From: claudiu beznea <[e-mail]>
>> Sent: Tuesday, August 27, 2024 1:33 PM
>> To: Biju Das <[e-mail]>; [e-mail]; [e-mail];
>> [e-mail]; [e-mail]; [e-mail]; [e-mail]
>> Cc: [e-mail]; [e-mail]; [e-mail]; linux-
>> [e-mail]; [e-mail]; Claudiu Beznea <[e-mail]>
>> Subject: Re: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler
>>
>> Hi, Biju,
>>
>> On 27.08.2024 15:15, Biju Das wrote:
>>> Hi Claudiu,
>>>
>>> Thanks for the feedback.
>>>
>>>> -----Original Message-----
>>>> From: Claudiu <[e-mail]>
>>>> Sent: Monday, August 26, 2024 4:25 PM
>>>> Subject: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog
>>>> domain in the restart handler
>>>>
>>>> From: Claudiu Beznea <[e-mail]>
>>>>
>>>> On RZ/G3S the watchdog can be part of a software-controlled PM
>>>> domain. In this case, the watchdog device need to be powered on in
>>>> struct watchdog_ops::restart API. This can be done though
>>>> pm_runtime_resume_and_get() API if the watchdog PM domain and watchdog device are marked as IRQ
>> safe.
>>>> We mark the watchdog PM domain as IRQ safe with GENPD_FLAG_IRQ_SAFE
>>>> when the watchdog PM domain is registered and the watchdog device though pm_runtime_irq_safe().
>>>>
>>>> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid
>>>> wait
>>>> context'") pm_runtime_get_sync() was used in watchdog restart handler
>>>> (which is similar to
>>>> pm_runtime_resume_and_get() except the later one handles the runtime resume errors).
>>>>
>>>> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
>>>> context'") dropped the pm_runtime_get_sync() and replaced it with
>>>> clk_prepare_enable() to avoid invalid wait context due to
>>>> genpd_lock() in genpd_runtime_resume() being called from atomic
>>>> context. But
>>>> clk_prepare_enable() doesn't fit for this either (as reported by Ulf
>>>> Hansson) as clk_prepare() can also sleep (it just not throw invalid wait context warning as it is
>> not written for this).
>>>>
>>>> Because the watchdog device is marked now as IRQ safe (though this
>>>> patch) the
>>>> irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume()
>>>> returns
>>>> 1 for devices not registering an IRQ safe PM domain for watchdog (as
>>>> the watchdog device is IRQ safe, PM domain is not and watchdog PM
>>>> domain is always-on), this being the case of RZ/G2 devices that uses
>>>
>>> RZ/G2L alike devices or be specific RZ/{G2L,G2LC,G2UL,V2L} as it is
>>> not applicable for RZ/G2{H,M,N,E}devices.
>>
>> OK, but I said "RZ/G2 devices that uses this driver". Here are included RZ/{G2L,G2LC,G2UL,V2L} AFAICT.
> 
> OK. Not sure you missed the same terminology on comment section, see below??
> 
>>
>>>
>>>
>>>> this driver, we can now drop also the clk_prepare_enable() calls in
>>>> restart handler and rely on pm_runtime_resume_and_get().
>>>>
>>>> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in watchdog restart handler.
>>>
>>> Can this patch be fix for Commit e4cf89596c1f ("watchdog: rzg2l_wdt:
>>> Fix 'BUG: Invalid wait
>>>> context'") on RZ/{G2L,G2LC,G2UL,V2L} SoC??
>>
>> Not sure... I thought about it, too. I chose to have it like this thinking
>> that:
>>
>> 1/ that may not apply cleanly as it depends on other cleanups done on this
>>    driver, e.g. commit d8997ed79ed7 ("watchdog: rzg2l_wdt: Rely on the
>>    reset driver for doing proper reset") so it may be worthless for
>>    backport machinery
>> 2/ There is actually no seen bug reported by lockdep (as the clk_prepare()
>>    doesn't handle it)
>>
>> Don't know, I can reply here and add it. Applying this patch with b4 will take care of it. But not
>> sure about it.
> 
> Maybe leave it.
> 
>>>
>>>>
>>>> Signed-off-by: Claudiu Beznea <[e-mail]>
>>>> ---
>>>>  drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
>>>>  1 file changed, 19 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/watchdog/rzg2l_wdt.c
>>>> b/drivers/watchdog/rzg2l_wdt.c index
>>>> 2a35f890a288..e9e0408c96f7 100644
>>>> --- a/drivers/watchdog/rzg2l_wdt.c
>>>> +++ b/drivers/watchdog/rzg2l_wdt.c
>>>> @@ -12,6 +12,7 @@
>>>>  #include <linux/module.h>
>>>>  #include <linux/of.h>
>>>>  #include <linux/platform_device.h>
>>>> +#include <linux/pm_domain.h>
>>>>  #include <linux/pm_runtime.h>
>>>>  #include <linux/reset.h>
>>>>  #include <linux/units.h>
>>>> @@ -166,8 +167,23 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev,
>>>>  	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
>>>>  	int ret;
>>>>
>>>> -	clk_prepare_enable(priv->pclk);
>>>> -	clk_prepare_enable(priv->osc_clk);
>>>> +	/*
>>>> +	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
>>>> +	 * domain that is currently powered off. In this case we need to power
>>>> +	 * it on before accessing registers. Along with this the clocks will be
>>>> +	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
>>>> +	 * need to be on for the reboot to happen.
>>>> +	 *
>>>> +	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device trees
> 
> NitPick: For the rest of RZ/G2 devices that uses this driver (This will make sure
> It is not meant for RZ/G2{H,M,N,E} devices)

If one considers this driver is used by RZ/G2{H,M,N,E} when reaching this
point then surely is in the wrong place.

RZ/Five is also uses this driver. Later, maybe devices compatible with this
driver will be added and this comment will not fit. Later, will we be
updating the comment for that? I'm not a fan of it.

Thank you,
Claudiu Beznea

> 
> Cheers,
> Biju
> 
> 
> 
>>>> +	 * where PM domains are registered like on RZ/G2 devices) it is safe
>>>> +	 * to call pm_runtime_resume_and_get() as the
>>>> +	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
>>>> +	 * returns non zero value and the genpd_lock() is avoided, thus, there
>>>> +	 * will be no invalid wait context reported by lockdep.
>>>> +	 */
>>>> +	ret = pm_runtime_resume_and_get(wdev->parent);
>>>> +	if (ret)
>>>> +		return ret;
>>>>
>>>>  	if (priv->devtype == WDT_RZG2L) {
>>>>  		ret = reset_control_deassert(priv->rstc);
>>>> @@ -275,6 +291,7 @@ static int rzg2l_wdt_probe(struct platform_device
>>>> *pdev)
>>>>
>>>>  	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
>>>>
>>>> +	pm_runtime_irq_safe(&pdev->dev);
>>>>  	pm_runtime_enable(&pdev->dev);
>>>>
>>>>  	priv->wdev.info = &rzg2l_wdt_ident;
>>>> --
>>>> 2.39.2
>>>>
>>>

Biju Das Aug. 27, 2024, 1:51 p.m. UTC | #6

Hi Claudiu,

> -----Original Message-----
> From: claudiu beznea <[e-mail]>
> Sent: Tuesday, August 27, 2024 2:35 PM
> Subject: Re: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog domain in the restart handler
> 
> 
> 
> On 27.08.2024 15:51, Biju Das wrote:
> >
> >
> >> -----Original Message-----
> >> From: claudiu beznea <[e-mail]>
> >> Sent: Tuesday, August 27, 2024 1:33 PM
> >> To: Biju Das <[e-mail]>; [e-mail];
> >> [e-mail]; [e-mail]; [e-mail];
> >> [e-mail]; [e-mail]
> >> Cc: [e-mail]; [e-mail];
> >> [e-mail]; linux- [e-mail];
> >> [e-mail]; Claudiu Beznea
> >> <[e-mail]>
> >> Subject: Re: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog
> >> domain in the restart handler
> >>
> >> Hi, Biju,
> >>
> >> On 27.08.2024 15:15, Biju Das wrote:
> >>> Hi Claudiu,
> >>>
> >>> Thanks for the feedback.
> >>>
> >>>> -----Original Message-----
> >>>> From: Claudiu <[e-mail]>
> >>>> Sent: Monday, August 26, 2024 4:25 PM
> >>>> Subject: [PATCH 3/3] watchdog: rzg2l_wdt: Power on the watchdog
> >>>> domain in the restart handler
> >>>>
> >>>> From: Claudiu Beznea <[e-mail]>
> >>>>
> >>>> On RZ/G3S the watchdog can be part of a software-controlled PM
> >>>> domain. In this case, the watchdog device need to be powered on in
> >>>> struct watchdog_ops::restart API. This can be done though
> >>>> pm_runtime_resume_and_get() API if the watchdog PM domain and
> >>>> watchdog device are marked as IRQ
> >> safe.
> >>>> We mark the watchdog PM domain as IRQ safe with GENPD_FLAG_IRQ_SAFE
> >>>> when the watchdog PM domain is registered and the watchdog device though pm_runtime_irq_safe().
> >>>>
> >>>> Before commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid
> >>>> wait
> >>>> context'") pm_runtime_get_sync() was used in watchdog restart
> >>>> handler (which is similar to
> >>>> pm_runtime_resume_and_get() except the later one handles the runtime resume errors).
> >>>>
> >>>> Commit e4cf89596c1f ("watchdog: rzg2l_wdt: Fix 'BUG: Invalid wait
> >>>> context'") dropped the pm_runtime_get_sync() and replaced it with
> >>>> clk_prepare_enable() to avoid invalid wait context due to
> >>>> genpd_lock() in genpd_runtime_resume() being called from atomic
> >>>> context. But
> >>>> clk_prepare_enable() doesn't fit for this either (as reported by
> >>>> Ulf
> >>>> Hansson) as clk_prepare() can also sleep (it just not throw invalid
> >>>> wait context warning as it is
> >> not written for this).
> >>>>
> >>>> Because the watchdog device is marked now as IRQ safe (though this
> >>>> patch) the
> >>>> irq_safe_dev_in_sleep_domain() call from genpd_runtime_resume()
> >>>> returns
> >>>> 1 for devices not registering an IRQ safe PM domain for watchdog
> >>>> (as the watchdog device is IRQ safe, PM domain is not and watchdog
> >>>> PM domain is always-on), this being the case of RZ/G2 devices that
> >>>> uses
> >>>
> >>> RZ/G2L alike devices or be specific RZ/{G2L,G2LC,G2UL,V2L} as it is
> >>> not applicable for RZ/G2{H,M,N,E}devices.
> >>
> >> OK, but I said "RZ/G2 devices that uses this driver". Here are included RZ/{G2L,G2LC,G2UL,V2L}
> AFAICT.
> >
> > OK. Not sure you missed the same terminology on comment section, see below??
> >
> >>
> >>>
> >>>
> >>>> this driver, we can now drop also the clk_prepare_enable() calls in
> >>>> restart handler and rely on pm_runtime_resume_and_get().
> >>>>
> >>>> Thus, drop clk_prepare_enable() and use pm_runtime_resume_and_get() in watchdog restart handler.
> >>>
> >>> Can this patch be fix for Commit e4cf89596c1f ("watchdog: rzg2l_wdt:
> >>> Fix 'BUG: Invalid wait
> >>>> context'") on RZ/{G2L,G2LC,G2UL,V2L} SoC??
> >>
> >> Not sure... I thought about it, too. I chose to have it like this
> >> thinking
> >> that:
> >>
> >> 1/ that may not apply cleanly as it depends on other cleanups done on this
> >>    driver, e.g. commit d8997ed79ed7 ("watchdog: rzg2l_wdt: Rely on the
> >>    reset driver for doing proper reset") so it may be worthless for
> >>    backport machinery
> >> 2/ There is actually no seen bug reported by lockdep (as the clk_prepare()
> >>    doesn't handle it)
> >>
> >> Don't know, I can reply here and add it. Applying this patch with b4
> >> will take care of it. But not sure about it.
> >
> > Maybe leave it.
> >
> >>>
> >>>>
> >>>> Signed-off-by: Claudiu Beznea <[e-mail]>
> >>>> ---
> >>>>  drivers/watchdog/rzg2l_wdt.c | 21 +++++++++++++++++++--
> >>>>  1 file changed, 19 insertions(+), 2 deletions(-)
> >>>>
> >>>> diff --git a/drivers/watchdog/rzg2l_wdt.c
> >>>> b/drivers/watchdog/rzg2l_wdt.c index
> >>>> 2a35f890a288..e9e0408c96f7 100644
> >>>> --- a/drivers/watchdog/rzg2l_wdt.c
> >>>> +++ b/drivers/watchdog/rzg2l_wdt.c
> >>>> @@ -12,6 +12,7 @@
> >>>>  #include <linux/module.h>
> >>>>  #include <linux/of.h>
> >>>>  #include <linux/platform_device.h>
> >>>> +#include <linux/pm_domain.h>
> >>>>  #include <linux/pm_runtime.h>
> >>>>  #include <linux/reset.h>
> >>>>  #include <linux/units.h>
> >>>> @@ -166,8 +167,23 @@ static int rzg2l_wdt_restart(struct watchdog_device *wdev,
> >>>>  	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
> >>>>  	int ret;
> >>>>
> >>>> -	clk_prepare_enable(priv->pclk);
> >>>> -	clk_prepare_enable(priv->osc_clk);
> >>>> +	/*
> >>>> +	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
> >>>> +	 * domain that is currently powered off. In this case we need to power
> >>>> +	 * it on before accessing registers. Along with this the clocks will be
> >>>> +	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
> >>>> +	 * need to be on for the reboot to happen.
> >>>> +	 *
> >>>> +	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device
> >>>> +trees
> >
> > NitPick: For the rest of RZ/G2 devices that uses this driver (This
> > will make sure It is not meant for RZ/G2{H,M,N,E} devices)
> 
> If one considers this driver is used by RZ/G2{H,M,N,E} when reaching this point then surely is in the
> wrong place.

So strictly speaking, then it is RZ/G3S and rest of the devices.
as RZ/G2 is not applicable here as we have RZ/V2L and RZ/Five apart
from RZ/{G2L,G2LC,G2UL}.

> 
> RZ/Five is also uses this driver. Later, maybe devices compatible with this driver will be added and
> this comment will not fit. Later, will we be updating the comment for that? I'm not a fan of it.

The comment RZ/G2 made confusion as the driver supports RZ/V2L and RZ/Five SoCs as well.
Also, it gives an impression about RZ/G2{H,M,N,E} devices.

Maybe replace "For the rest of RZ/G2 devices"->"For the rest of devices" should be sufficient.
as it covers RZ/{G2L,G2LC,G2UL}, RZ/V2L and RZ/Five SoCs.

Cheers.
Biju




> >
> >
> >
> >>>> +	 * where PM domains are registered like on RZ/G2 devices) it is safe
> >>>> +	 * to call pm_runtime_resume_and_get() as the
> >>>> +	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
> >>>> +	 * returns non zero value and the genpd_lock() is avoided, thus, there
> >>>> +	 * will be no invalid wait context reported by lockdep.
> >>>> +	 */
> >>>> +	ret = pm_runtime_resume_and_get(wdev->parent);
> >>>> +	if (ret)
> >>>> +		return ret;
> >>>>
> >>>>  	if (priv->devtype == WDT_RZG2L) {
> >>>>  		ret = reset_control_deassert(priv->rstc);
> >>>> @@ -275,6 +291,7 @@ static int rzg2l_wdt_probe(struct
> >>>> platform_device
> >>>> *pdev)
> >>>>
> >>>>  	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
> >>>>
> >>>> +	pm_runtime_irq_safe(&pdev->dev);
> >>>>  	pm_runtime_enable(&pdev->dev);
> >>>>
> >>>>  	priv->wdev.info = &rzg2l_wdt_ident;
> >>>> --
> >>>> 2.39.2
> >>>>
> >>>

Patch

diff --git a/drivers/watchdog/rzg2l_wdt.c b/drivers/watchdog/rzg2l_wdt.c
index 2a35f890a288..e9e0408c96f7 100644
--- a/drivers/watchdog/rzg2l_wdt.c
+++ b/drivers/watchdog/rzg2l_wdt.c
@@ -12,6 +12,7 @@ 
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/platform_device.h>
+#include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
 #include <linux/reset.h>
 #include <linux/units.h>
@@ -166,8 +167,23 @@  static int rzg2l_wdt_restart(struct watchdog_device *wdev,
 	struct rzg2l_wdt_priv *priv = watchdog_get_drvdata(wdev);
 	int ret;
 
-	clk_prepare_enable(priv->pclk);
-	clk_prepare_enable(priv->osc_clk);
+	/*
+	 * In case of RZ/G3S the watchdog device may be part of an IRQ safe power
+	 * domain that is currently powered off. In this case we need to power
+	 * it on before accessing registers. Along with this the clocks will be
+	 * enabled. We don't undo the pm_runtime_resume_and_get() as the device
+	 * need to be on for the reboot to happen.
+	 *
+	 * For the rest of RZ/G2 devices (and for RZ/G3S with old device trees
+	 * where PM domains are registered like on RZ/G2 devices) it is safe
+	 * to call pm_runtime_resume_and_get() as the
+	 * irq_safe_dev_in_sleep_domain() call in genpd_runtime_resume()
+	 * returns non zero value and the genpd_lock() is avoided, thus, there
+	 * will be no invalid wait context reported by lockdep.
+	 */
+	ret = pm_runtime_resume_and_get(wdev->parent);
+	if (ret)
+		return ret;
 
 	if (priv->devtype == WDT_RZG2L) {
 		ret = reset_control_deassert(priv->rstc);
@@ -275,6 +291,7 @@  static int rzg2l_wdt_probe(struct platform_device *pdev)
 
 	priv->devtype = (uintptr_t)of_device_get_match_data(dev);
 
+	pm_runtime_irq_safe(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
 
 	priv->wdev.info = &rzg2l_wdt_ident;