Re: [PATCH net-next v7 01/12] net: phylink: keep and use MAC supported_interfaces in phylink struct

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The patch makes a one-time copy of `supported_interfaces` during `phylink_create()`, which breaks drivers like `mtk_eth_soc` that modify `phylink_config.supported_interfaces` after calling `phylink_create()`.
--

commit c809e12abb9cfbf31d18f5e71f4e3d15bb4f6a01
Author: Christian Marangi <ansuelsmth@gmail.com>

net: phylink: keep and use MAC supported_interfaces in phylink struct

This patch introduces a cached copy of the supported_interfaces mask in the
phylink struct. It initializes this copy during phylink_create() and
transitions the phylink core to validate against this internal mask rather
than the original phylink_config.

> diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
> index 087ac63f9193d..4d59c0dd78dbf 100644
> --- a/drivers/net/phy/phylink.c
> +++ b/drivers/net/phy/phylink.c
[ ... ]
> @@ -1868,6 +1873,9 @@ struct phylink *phylink_create(struct phylink_config *config,
>  	mutex_init(&pl->state_mutex);
>  	INIT_WORK(&pl->resolve, phylink_resolve);
>  
> +	phy_interface_copy(pl->supported_interfaces,
> +			   config->supported_interfaces);
> +

[Severity: High]
Does this one-time copy in phylink_create() break drivers that modify their
supported_interfaces after initialization?

For example, in drivers/net/ethernet/mediatek/mtk_eth_soc.c:mtk_add_mac(),
the driver appends PHY_INTERFACE_MODE_INTERNAL to
mac->phylink_config.supported_interfaces after phylink_create() has already
been called:

drivers/net/ethernet/mediatek/mtk_eth_soc.c:mtk_add_mac() {
        ...
        mac->phylink = phylink_create(&mac->phylink_config, ...);
        ...
        if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_2P5GPHY) &&
            id == MTK_GMAC2_ID)
                __set_bit(PHY_INTERFACE_MODE_INTERNAL,
                          mac->phylink_config.supported_interfaces);
        ...
}

Since phylink now validates against the cached pl->supported_interfaces mask,
will it reject PHY_INTERFACE_MODE_INTERNAL as unsupported and prevent the
internal 2.5G PHY from linking successfully on affected Mediatek platforms?

>  	pl->config = config;
>  	if (config->type == PHYLINK_NETDEV) {
>  		pl->netdev = to_net_dev(config->dev);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260615122950.22281-1-ansuelsmth@gmail.com?part=1