Re: [PATCH v8 7/7] power: supply: Add charger driver for Asus Transformers

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Strict equality check on `action` bitmask may cause dropped events.
--

commit 55ed27a4bfa91a64727ac41721ebdb9163ebef2a
Author: Michał Mirosław <mirq-linux@rere.qmqm.pl>

power: supply: Add charger driver for Asus Transformers

This commit adds a new driver to support charger detection capabilities
provided by the embedded controller on ASUS Transformer devices. It
reports properties like USB type, charging behaviour, and online status.

> diff --git a/drivers/power/supply/asus-transformer-ec-charger.c b/drivers/power/supply/asus-transformer-ec-charger.c
> new file mode 100644
> index 0000000000000..c7a6bd2ba533f
> --- /dev/null
> +++ b/drivers/power/supply/asus-transformer-ec-charger.c
[ ... ]
> +static int asus_ec_charger_notify(struct notifier_block *nb,
> +				  unsigned long action, void *data)
> +{
> +	struct asus_ec_charger_data *priv =
> +		container_of(nb, struct asus_ec_charger_data, nb);
> +
> +	switch (action) {
> +	case ASUSEC_SMI_ACTION(POWER_NOTIFY):
> +	case ASUSEC_SMI_ACTION(ADAPTER_EVENT):

[Severity: Medium]
Will this strict equality check cause charger events to be dropped if multiple
interrupt flags are set simultaneously?

The action parameter appears to be constructed by the parent MFD driver by
taking the raw ASUSEC_IRQ_STATUS byte and OR'ing it with the SMI code shifted
by 8 bits. Since ASUSEC_IRQ_STATUS is a bitmask, if an SMI event occurs
concurrently with another event (such as a keyboard press), the action bitmask
will contain additional bits.

Could this strict switch check be replaced by checking the ASUSEC_SMI_MASK with
a bitwise AND operation, and then evaluating the SMI code separately?

> +		power_supply_changed(priv->psy);
> +		break;
> +	}
> +
> +	return NOTIFY_DONE;
> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260528053203.9339-1-clamor95@gmail.com?part=7