linux-kernel - Re: [PATCH] mfd: axp20x: Add supplied-from property to axp288_fuel

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <a9fbf785-0dab-2e55-5983-ea279f8dc3a2@redhat.com>
Date:   Tue, 27 Jul 2021 21:24:38 +0200
From:   Hans de Goede <hdegoede@...hat.com>
To:     Lee Jones <lee.jones@...aro.org>
Cc:     Chen-Yu Tsai <wens@...e.org>, C++ / GCC <cpp@....lt>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mfd: axp20x: Add supplied-from property to
 axp288_fuel_gauge cell

Hi Lee,

On 7/19/21 12:48 PM, Lee Jones wrote:
> On Sat, 17 Jul 2021, Hans de Goede wrote:
> 
>> The power-supply framework has the notion of one power-supply device
>> being supplied by another. A typical example of this is a charger
>> charging a battery.
>>
>> A tablet getting plugged in to charge (or plugged out) only results in
>> events seen by the axp288_charger device / MFD cell. Which means that
>> a change udev-event only gets send for the charger power-supply class
>> device, not for the battery (the axp288_fuel_gauge device).
>>
>> The axp288_fuel_gauge does have an external_power_change'd callback
>> which will generate a change udev-event when called. But before this
>> commit this never got called because the power-supply core only calls
>> this when a power-supply class device's supplier changes and the
>> supplier link from axp288_charger to axp288_fuel_gauge was missing.
>>
>> Add a "supplied-from" property to axp288_fuel_gauge cell, pointing
>> to the "axp288_charger" power-supply class device, so that the
>> axp288_fuel_gauge's external_power_change'd callback gets called on
>> axp288_charger state changes.
>>
>> Signed-off-by: Hans de Goede <hdegoede@...hat.com>
>> ---
>>  drivers/mfd/axp20x.c | 13 +++++++++++++
>>  1 file changed, 13 insertions(+)
>>
>> diff --git a/drivers/mfd/axp20x.c b/drivers/mfd/axp20x.c
>> index d0ac019850d1..8161a5dc68e8 100644
>> --- a/drivers/mfd/axp20x.c
>> +++ b/drivers/mfd/axp20x.c
>> @@ -700,6 +700,18 @@ static const struct resource axp288_charger_resources[] = {
>>  	DEFINE_RES_IRQ(AXP288_IRQ_CBTO),
>>  };
>>  
>> +static const char * const axp288_fuel_gauge_suppliers[] = { "axp288_charger" };
>> +
>> +static const struct property_entry axp288_fuel_gauge_properties[] = {
>> +	PROPERTY_ENTRY_STRING_ARRAY("supplied-from", axp288_fuel_gauge_suppliers),
>> +	{ }
>> +};
>> +
>> +static const struct software_node axp288_fuel_gauge_sw_node = {
>> +	.name = "axp288_fuel_gauge",
>> +	.properties = axp288_fuel_gauge_properties,
>> +};
>> +
>>  static const struct mfd_cell axp288_cells[] = {
>>  	{
>>  		.name		= "axp288_adc",
>> @@ -717,6 +729,7 @@ static const struct mfd_cell axp288_cells[] = {
>>  		.name		= "axp288_fuel_gauge",
>>  		.num_resources	= ARRAY_SIZE(axp288_fuel_gauge_resources),
>>  		.resources	= axp288_fuel_gauge_resources,
>> +		.swnode		= &axp288_fuel_gauge_sw_node,
>>  	}, {
>>  		.name		= "axp221-pek",
>>  		.num_resources	= ARRAY_SIZE(axp288_power_button_resources),
> 
> That's a lot of code to pass a string.
> 
> Is this really the most efficient method?

Currently there is no other method to provide information about the supplier
consumer relation for non device-tree platforms (in device-tree platforms
this info is present inside the device-tree), so yes this is the most
efficient method available.

Regards,

Hans