Re: [PATCH v3 2/2] hwmon: acpi_power_meter: convert to hwmon_device_register_with_info

[LABBE Corentin <clabbe@baylibre.com>]

Le Sat, May 21, 2022 at 06:52:15AM -0700, Guenter Roeck a écrit :
> On 5/16/22 05:21, Guenter Roeck wrote:
> > On 5/15/22 23:21, LABBE Corentin wrote:
> >> Le Sun, May 15, 2022 at 05:29:54PM -0700, Guenter Roeck a écrit :
> >>> On 5/15/22 12:36, LABBE Corentin wrote:
> >>>> Le Wed, May 11, 2022 at 07:10:29PM -0700, Guenter Roeck a écrit :
> >>>>> Corentin,
> >>>>>
> >>>>> On 5/8/22 23:30, Corentin Labbe wrote:
> >>>>>> Booting lead to a hwmon_device_register() is deprecated. Please convert the driver to use hwmon_device_register_with_info().
> >>>>>> So let's convert the driver to use hwmon_device_register_with_info().
> >>>>>>
> >>>>>> Signed-off-by: Corentin Labbe <clabbe@baylibre.com>
> >>>>>> ---
> >>>>> [ ... ]
> >>>>>
> >>>>>> @@ -836,20 +740,20 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event)
> >>>>>>             if (res)
> >>>>>>                 break;
> >>>>>> -        remove_attrs(resource);
> >>>>>> +        remove_domain_devices(resource);
> >>>>>>             setup_attrs(resource);
> >>>>>
> >>>>> Zhang Rui found an interesting problem with this code:
> >>>>> It needs a call to sysfs_update_groups(hwmon_dev->groups)
> >>>>> to update sysfs attribute visibility, probably between
> >>>>> remove_domain_devices() and setup_attrs().
> >>>>>
> >>>>>>             break;
> >>>>>>         case METER_NOTIFY_TRIP:
> >>>>>> -        sysfs_notify(&device->dev.kobj, NULL, POWER_AVERAGE_NAME);
> >>>>>> +        hwmon_notify_event(&device->dev, hwmon_power, hwmon_power_average, 0);
> >>>>>
> >>>>> ... which makes realize: The notification device should be the hwmon device.
> >>>>> That would be resource->hwmon_dev, not the acpi device.
> >>>>>
> >>>>
> >>>> Hello
> >>>>
> >>>> Since my hardware lacks capabilities testing this, I have emulated it on qemu:
> >>>> https://github.com/montjoie/qemu/commit/320f2ddacb954ab308ef699f66fca6313f75bc2b
> >>>>
> >>>> I have added a custom ACPI _DBX method for triggering some ACPI state change. (like config change, like enabling CAP).
> >>>>
> >>>> For testing config change I have tried lot of way:
> >>>>                   res = read_capabilities(resource);
> >>>> @@ -742,18 +758,22 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event)
> >>>>                   remove_domain_devices(resource);
> >>>>                   setup_attrs(resource);
> >>>> +               res = sysfs_update_groups(&resource->hwmon_dev->kobj, acpi_power_groups);
> >>>> +               res = sysfs_update_groups(&resource->acpi_dev->dev.kobj, acpi_power_groups);
> >>>> +               res = hwmon_notify_event(resource->hwmon_dev, hwmon_power, hwmon_power_cap, 0);
> >>>> +               res = hwmon_notify_event(resource->hwmon_dev, hwmon_power, hwmon_power_average, 0);
> >>>
> >>> Did you add a debug log here ?
> >>
> >> Yes I added debug log to check what is called.
> >>
> >>>
> >>> acpi_power_groups would be the wrong parameter for sysfs_update_groups().
> >>> It would have to be resource->hwmon_dev->groups.
> >>>
> >>
> >> Even with that, no call to is_visible:
> >> @@ -742,18 +758,22 @@ static void acpi_power_meter_notify(struct acpi_device *device, u32 event)
> >>                  remove_domain_devices(resource);
> >>                  setup_attrs(resource);
> >> +               res = sysfs_update_groups(&resource->hwmon_dev->kobj, resource->hwmon_dev->groups);
> >> +               res = sysfs_update_groups(&resource->acpi_dev->dev.kobj, resource->hwmon_dev->groups);
> >> +               res = hwmon_notify_event(resource->hwmon_dev, hwmon_power, hwmon_power_cap, 0);
> >> +               res = hwmon_notify_event(resource->hwmon_dev, hwmon_power, hwmon_power_average, 0);
> >>                  break;
> >>
> >> I checked drivers/hwmon/hwmon.c is seems that is_visible is only called by gen_attr/gen_attrs which is only called by __hwmon_create_attrs and then by registers functions.
> >> So perhaps it explain why it is never called.
> > 
> > Ah yes, you are correct. Sorry, it has been too long ago that I wrote that code.
> > Effectively that means we'll have to rework the hwmon core to generate attributes
> > anyway and leave it up to the driver core to call the is_visible function.
> > 
> 
> Attached is an outline of what would be needed in the hwmon core.
> Completely untested. I wonder if it may be easier to always
> create all attributes and have them return -ENODATA if not
> supported.
> 
> Guenter


Hello

Do you plan to send your change as patch ?
My patch serie is stuck since now it depend on it.

Or can I send a new iteration of my serie with partial support for notify.

Regards