[v4,2/8] ACPI: scan: Add function to fetch dependent of acpi device

In some ACPI tables we encounter, devices use the _DEP method to assert
a dependence on other ACPI devices as opposed to the OpRegions that the
specification intends. We need to be able to find those devices "from"
the dependee, so add a callback and a wrapper to walk over the
acpi_dep_list and return the dependent ACPI device.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Daniel Scally <djrscally@gmail.com>
---
Changes since v3:

	Both new functions were renamed.

 drivers/acpi/scan.c     | 38 ++++++++++++++++++++++++++++++++++++++
 include/acpi/acpi_bus.h |  1 +
 2 files changed, 39 insertions(+)

On Thu, May 20, 2021 at 4:11 PM Daniel Scally <djrscally@gmail.com> wrote:
>
> In some ACPI tables we encounter, devices use the _DEP method to assert
> a dependence on other ACPI devices as opposed to the OpRegions that the
> specification intends. We need to be able to find those devices "from"
> the dependee, so add a callback and a wrapper to walk over the
> acpi_dep_list and return the dependent ACPI device.
>
> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
> Signed-off-by: Daniel Scally <djrscally@gmail.com>
> ---
> Changes since v3:
>
>         Both new functions were renamed.
>
>  drivers/acpi/scan.c     | 38 ++++++++++++++++++++++++++++++++++++++
>  include/acpi/acpi_bus.h |  1 +
>  2 files changed, 39 insertions(+)
>
> diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
> index 195635c3462b..1a76fbdfa669 100644
> --- a/drivers/acpi/scan.c
> +++ b/drivers/acpi/scan.c
> @@ -2105,6 +2105,21 @@ static void acpi_bus_attach(struct acpi_device *device, bool first_pass)
>                 device->handler->hotplug.notify_online(device);
>  }
>
> +static int acpi_return_dep_dev(struct acpi_dep_data *dep, void *data)

What about calling this acpi_get_first_consumer_cb()?

> +{
> +       struct acpi_device *adev;
> +       int ret;
> +
> +       ret = acpi_bus_get_device(dep->consumer, &adev);
> +       if (ret)
> +               /* If we don't find an adev then we want to continue parsing */
> +               return 0;
> +
> +       *(struct acpi_device **)data = adev;

And it can do the get_device() here, can't it?

So maybe use acpi_bus_get_acpi_device() instead of
acpi_bus_get_device()?  Would be simpler.

> +
> +       return 1;
> +}
> +
>  static int acpi_scan_clear_dep(struct acpi_dep_data *dep, void *data)
>  {
>         struct acpi_device *adev;
> @@ -2168,6 +2183,29 @@ void acpi_dev_clear_dependencies(struct acpi_device *supplier)
>  }
>  EXPORT_SYMBOL_GPL(acpi_dev_clear_dependencies);
>
> +/**
> + * acpi_dev_get_dependent_dev - Return ACPI device dependent on @supplier

And what about calling this acpi_get_first_consumer() ?

> + * @supplier: Pointer to the dependee device
> + *
> + * Returns the first &struct acpi_device which declares itself dependent on
> + * @supplier via the _DEP buffer, parsed from the acpi_dep_list.
> + *
> + * The caller is responsible for putting the reference to adev when it is no
> + * longer needed.
> + */
> +struct acpi_device *acpi_dev_get_dependent_dev(struct acpi_device *supplier)
> +{
> +       struct acpi_device *adev = NULL;
> +
> +       acpi_walk_dep_device_list(supplier->handle, acpi_return_dep_dev, &adev);
> +
> +       if (adev)
> +               get_device(&adev->dev);
> +
> +       return adev;
> +}
> +EXPORT_SYMBOL_GPL(acpi_dev_get_dependent_dev);
> +
>  /**
>   * acpi_bus_scan - Add ACPI device node objects in a given namespace scope.
>   * @handle: Root of the namespace scope to scan.
> diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
> index 0b2c4f170f4d..68d378207704 100644
> --- a/include/acpi/acpi_bus.h
> +++ b/include/acpi/acpi_bus.h
> @@ -692,6 +692,7 @@ static inline bool acpi_device_can_poweroff(struct acpi_device *adev)
>  bool acpi_dev_hid_uid_match(struct acpi_device *adev, const char *hid2, const char *uid2);
>
>  void acpi_dev_clear_dependencies(struct acpi_device *supplier);
> +struct acpi_device *acpi_dev_get_dependent_dev(struct acpi_device *supplier);
>  struct acpi_device *
>  acpi_dev_get_next_match_dev(struct acpi_device *adev, const char *hid, const char *uid, s64 hrv);
>  struct acpi_device *
> --
> 2.25.1
>

Hi Rafael

On 20/05/2021 19:55, Rafael J. Wysocki wrote:
> On Thu, May 20, 2021 at 8:33 PM Rafael J. Wysocki <rafael@kernel.org> wrote:

>> On Thu, May 20, 2021 at 4:11 PM Daniel Scally <djrscally@gmail.com> wrote:

>>> In some ACPI tables we encounter, devices use the _DEP method to assert

>>> a dependence on other ACPI devices as opposed to the OpRegions that the

>>> specification intends. We need to be able to find those devices "from"

>>> the dependee, so add a callback and a wrapper to walk over the

>>> acpi_dep_list and return the dependent ACPI device.

>>>

>>> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>

>>> Signed-off-by: Daniel Scally <djrscally@gmail.com>

>>> ---

>>> Changes since v3:

>>>

>>>         Both new functions were renamed.

>>>

>>>  drivers/acpi/scan.c     | 38 ++++++++++++++++++++++++++++++++++++++

>>>  include/acpi/acpi_bus.h |  1 +

>>>  2 files changed, 39 insertions(+)

>>>

>>> diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c

>>> index 195635c3462b..1a76fbdfa669 100644

>>> --- a/drivers/acpi/scan.c

>>> +++ b/drivers/acpi/scan.c

>>> @@ -2105,6 +2105,21 @@ static void acpi_bus_attach(struct acpi_device *device, bool first_pass)

>>>                 device->handler->hotplug.notify_online(device);

>>>  }

>>>

>>> +static int acpi_return_dep_dev(struct acpi_dep_data *dep, void *data)

>> What about calling this acpi_get_first_consumer_cb()?

> Or acpi_dev_get_first_consumer_dev_cb() if you want to be super-precise?



Sure; fine by me, and same for the other function