Re: [RFC PATCH net-next v2 7/7] net: mdio: en8811h: add the nested bus

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Using queue_delayed_work with 0 delay fails to execute an already-pending work immediately on resume.
--

commit e2ee026254e8a2afcaea891df04e388738201fc8
Author: Aleksei Sviridkin <f@lex.la>

net: mdio: en8811h: add the nested bus

This commit registers the PHY on a child bus so that the device tree can
describe it normally and the MCU can keep ownership of the reset line. It
introduces a two-phase download mechanism to allow retrying firmware loads.

> diff --git a/drivers/net/mdio/mdio-airoha-en8811h.c b/drivers/net/mdio/mdio-airoha-en8811h.c
> index d94d74d85f10d..7224558914727 100644
> --- a/drivers/net/mdio/mdio-airoha-en8811h.c
> +++ b/drivers/net/mdio/mdio-airoha-en8811h.c
[ ... ]
> @@ -150,23 +309,43 @@ static void en8811h_mcu_remove(struct mdio_device *mdiodev)
[ ... ]
>  static int en8811h_mcu_resume(struct device *dev)
>  {
>  	struct en8811h_mcu *mcu = dev_get_drvdata(dev);
[ ... ]
> +	if (ret < 0) {
> +		/* The reload restores this chip, not the PHY below it: its
> +		 * own resume has already failed by then and nothing calls
> +		 * phy_init_hw() twice.
> +		 */
> +		dev_err(dev, "firmware not restored, reloading: %pe\n",
> +			ERR_PTR(ret));
> +		mcu->fw_running = false;
> +		mcu->poll_ms = EN8811H_FW_POLL_MIN_MS;
> +		mcu->waited_ms = 0;
> +		mcu->warned = false;
> +		queue_delayed_work(system_freezable_wq, &mcu->fw_poll, 0);

[Severity: Medium]
Since queue_delayed_work() won't change the timeout of an already pending work,
does this leave the work scheduled with its previous timeout (which can be up to
30 seconds due to the exponential backoff) instead of executing immediately?

If the firmware load fails on resume while fw_poll was already pending (e.g.,
bus registration is deferring and in its backoff cycle), the retry poll might
be delayed significantly instead of executing immediately.

Should this use mod_delayed_work() instead to force the immediate execution of
a pending timer?

> +	}
>  
>  	return 0;
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260908155707.4164559-1-f@lex.la?part=7