Re: [PATCH net-next] net: ethernet: ti: davinci_emac: remove member phy_id from struct emac_platform_data

[Paolo Abeni <pabeni@redhat.com>]

On 2/6/26 10:53 PM, Heiner Kallweit wrote:
> b4aed01de486 ("ARM: davinci: Delete DM646x board files") removed the
> last user of member phy_id. So let's remove phy_id.
> 
> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
> ---
>  drivers/net/ethernet/ti/davinci_emac.c | 24 ++++++++++--------------
>  include/linux/davinci_emac.h           |  9 ---------
>  2 files changed, 10 insertions(+), 23 deletions(-)
> 
> diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c
> index ed8116fb05e..df51441a2b8 100644
> --- a/drivers/net/ethernet/ti/davinci_emac.c
> +++ b/drivers/net/ethernet/ti/davinci_emac.c
> @@ -329,7 +329,6 @@ struct emac_priv {
>  	u32 mac_hash2;
>  	u32 multicast_hash_cnt[EMAC_NUM_MULTICAST_BITS];
>  	u32 rx_addr_type;
> -	const char *phy_id;
>  	struct device_node *phy_node;
>  	spinlock_t lock;
>  	/*platform specific members*/
> @@ -1418,6 +1417,7 @@ static int emac_dev_open(struct net_device *ndev)
>  	struct emac_priv *priv = netdev_priv(ndev);
>  	struct phy_device *phydev = NULL;
>  	struct device *phy = NULL;
> +	const char *phy_id = NULL;
>  
>  	ret = pm_runtime_resume_and_get(&priv->pdev->dev);
>  	if (ret < 0) {
> @@ -1502,8 +1502,8 @@ static int emac_dev_open(struct net_device *ndev)
>  		}
>  	}
>  
> -	/* use the first phy on the bus if pdata did not give us a phy id */
> -	if (!phydev && !priv->phy_id) {
> +	/* use the first phy on the bus */
> +	if (!phydev) {

AFAICS emac_dev_open() is invoked also on resume, and priv->phy_id could
be not NULL at that point, the new code could lead to a different
behavior. I think it's worthy to at least mention it in the commit message.

/P