[RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[Jon Hunter]

The function pm_genpd_remove_subdomain() removes a subdomain from a
generic PM domain, however, it does not check if the subdomain has any
slave domains or device attached before doing so. Therefore, add a test
to verify that the subdomain does not have any slave domains associated
or any device attached before removing.

Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
---
 drivers/base/power/domain.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
index 27888a90ca98..a3ed71d8129a 100644
--- a/drivers/base/power/domain.c
+++ b/drivers/base/power/domain.c
@@ -1469,6 +1469,11 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
 
 	mutex_lock(&genpd->lock);
 
+	if (!list_empty(&subdomain->slave_links) || subdomain->device_count) {
+		ret = -EBUSY;
+		goto out;
+	}
+
 	list_for_each_entry(link, &genpd->master_links, master_node) {
 		if (link->slave != subdomain)
 			continue;
@@ -1487,6 +1492,7 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
 		break;
 	}
 
+out:
 	mutex_unlock(&genpd->lock);
 
 	return ret;
-- 
2.1.4

Re: [RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[Geert Uytterhoeven]

Hi Jon,

On Thu, Aug 27, 2015 at 11:21 AM, Jon Hunter <jonathanh@nvidia.com> wrote:
> The function pm_genpd_remove_subdomain() removes a subdomain from a
> generic PM domain, however, it does not check if the subdomain has any
> slave domains or device attached before doing so. Therefore, add a test
> to verify that the subdomain does not have any slave domains associated
> or any device attached before removing.

So far no-one ran into this, as there are no callers of
pm_genpd_remove_subdomain() in mainline or -next.

Do you plan to add a caller?

> Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
> ---
>  drivers/base/power/domain.c | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
> index 27888a90ca98..a3ed71d8129a 100644
> --- a/drivers/base/power/domain.c
> +++ b/drivers/base/power/domain.c
> @@ -1469,6 +1469,11 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
>
>         mutex_lock(&genpd->lock);
>
> +       if (!list_empty(&subdomain->slave_links) || subdomain->device_count) {
> +               ret = -EBUSY;
> +               goto out;
> +       }
> +
>         list_for_each_entry(link, &genpd->master_links, master_node) {
>                 if (link->slave != subdomain)
>                         continue;
> @@ -1487,6 +1492,7 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
>                 break;
>         }
>
> +out:
>         mutex_unlock(&genpd->lock);
>
>         return ret;
> --
> 2.1.4

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[Jon Hunter]

Hi Geert,

On 27/08/15 12:37, Geert Uytterhoeven wrote:
> Hi Jon,
> 
> On Thu, Aug 27, 2015 at 11:21 AM, Jon Hunter <jonathanh@nvidia.com> wrote:
>> The function pm_genpd_remove_subdomain() removes a subdomain from a
>> generic PM domain, however, it does not check if the subdomain has any
>> slave domains or device attached before doing so. Therefore, add a test
>> to verify that the subdomain does not have any slave domains associated
>> or any device attached before removing.
> 
> So far no-one ran into this, as there are no callers of
> pm_genpd_remove_subdomain() in mainline or -next.
> 
> Do you plan to add a caller?

Yes exactly. However, given that the API was present I thought why not
fix it now or at least get some feedback on this.

I have also noticed that there is no function to remove a pm-domain (ie.
reverse of pm_genpd_init()). I assume that currently people add
pm-domains and never remove them, which makes sense, however the
scenario I am trying to solve is allowing the registering of pm-domains
to be deferred if something is not yet present (ie. a regulator).

Cheers
Jon

-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------

Re: [RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[Kevin Hilman]

Jon Hunter <jonathanh@nvidia.com> writes:

> The function pm_genpd_remove_subdomain() removes a subdomain from a
> generic PM domain, however, it does not check if the subdomain has any
> slave domains or device attached before doing so. Therefore, add a test
> to verify that the subdomain does not have any slave domains associated
> or any device attached before removing.
>
> Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
> ---
>  drivers/base/power/domain.c | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
> index 27888a90ca98..a3ed71d8129a 100644
> --- a/drivers/base/power/domain.c
> +++ b/drivers/base/power/domain.c
> @@ -1469,6 +1469,11 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
>  
>  	mutex_lock(&genpd->lock);
>  
> +	if (!list_empty(&subdomain->slave_links) || subdomain->device_count) {

maybe a pr_warn() or similar is appropriate here?

> +		ret = -EBUSY;
> +		goto out;
> +	}
> +

... otherwise looks like a good check to add.

Acked-by: Kevin Hilman <khilman@linaro.org>

Re: [RFC PATCH] PM / Domains: Ensure subdomain is not in use before removing

[Jon Hunter]

On 28/08/15 18:38, Kevin Hilman wrote:
> Jon Hunter <jonathanh@nvidia.com> writes:
> 
>> The function pm_genpd_remove_subdomain() removes a subdomain from a
>> generic PM domain, however, it does not check if the subdomain has any
>> slave domains or device attached before doing so. Therefore, add a test
>> to verify that the subdomain does not have any slave domains associated
>> or any device attached before removing.
>>
>> Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
>> ---
>>  drivers/base/power/domain.c | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
>> index 27888a90ca98..a3ed71d8129a 100644
>> --- a/drivers/base/power/domain.c
>> +++ b/drivers/base/power/domain.c
>> @@ -1469,6 +1469,11 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd,
>>  
>>  	mutex_lock(&genpd->lock);
>>  
>> +	if (!list_empty(&subdomain->slave_links) || subdomain->device_count) {
> 
> maybe a pr_warn() or similar is appropriate here?

Thanks, good suggestion. Will add.

>> +		ret = -EBUSY;
>> +		goto out;
>> +	}
>> +
> 
> ... otherwise looks like a good check to add.
> 
> Acked-by: Kevin Hilman <khilman@linaro.org>

Cheers
Jon


-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------