Re: [PATCH v1 3/4] power: supply: bq25890: Add input voltage limit

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] power/supply: transient VINDPM limit due to write sequence
--

commit 3c4a08cfdbd118c7f60e9223379cfa47c0168057
Author: Vincent Cloutier <vincent@cloutier.co>

power: supply: bq25890: Add input voltage limit

The BQ25890 family supports an absolute input voltage dynamic power
management threshold. Read input-voltage-limit-microvolt from firmware,
enable absolute VINDPM, and then program the requested threshold.

> diff --git a/drivers/power/supply/bq25890_charger.c b/drivers/power/supply/bq25890_charger.c
> index 741a45cc5f35d..71d6cf1796674 100644
> --- a/drivers/power/supply/bq25890_charger.c
> +++ b/drivers/power/supply/bq25890_charger.c

[ ... ]

> @@ -814,18 +818,67 @@ static int bq25890_get_chip_state(struct bq25890_device *bq,
>  	return 0;
>  }
>  
> +static int bq25890_set_vindpm(struct bq25890_device *bq)
> +{
> +	int ret;
> +
> +	if (!bq->init_data.force_vindpm)
> +		return 0;
> +
> +	ret = bq25890_field_write(bq, F_FORCE_VINDPM, 1);
> +	if (ret < 0)
> +		return ret;
> +
> +	return bq25890_field_write(bq, F_VINDPM, bq->init_data.vindpm);
> +}

[Severity: Medium]
Does writing F_FORCE_VINDPM before F_VINDPM create a transient VINDPM limit?

Because these fields are written via separate read-modify-write I2C
transactions, enabling F_FORCE_VINDPM before setting the actual F_VINDPM
threshold creates a brief window where the charger explicitly enforces
whatever stale limit was previously in the register.

If this stale limit is too high for the current input source, could it
cause momentary input current throttling during device initialization,
resume, or cable insertion?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260927145538.233548-1-vincent.cloutier@icloud.com?part=3