Re: [PATCH] net: lan743x: disconnect the PHY when PTP setup fails

[Simon Horman <horms@kernel.org>]

On Tue, Nov 03, 2026 at 10:28:52AM +0000, Wentao Liang wrote:
> lan743x_netdev_open() connects the PHY before opening PTP, but if
> lan743x_ptp_open() fails it jumps to close_mac, which skips
> lan743x_phylink_disconnect(). The reference taken by
> phylink_connect_phy() is then never released, and the PHY stays
> attached to the phylink instance.
> 
> Add a close_phylink label in front of the phylink disconnect, as the
> pre-phylink conversion code did with close_phy, and use it on the
> lan743x_ptp_open() failure path.
> 
> Fixes: a5f199a8d8a0 ("net: lan743x: Migrate phylib to phylink")
> Cc: stable@vger.kernel.org
> Assisted-by: DeepSeek-V4.1-Flash and OpenAI Codex
> Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
> ---
>  drivers/net/ethernet/microchip/lan743x_main.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
> index 82d3ec20ba25..390aec639c71 100644
> --- a/drivers/net/ethernet/microchip/lan743x_main.c
> +++ b/drivers/net/ethernet/microchip/lan743x_main.c
> @@ -3331,7 +3331,7 @@ static int lan743x_netdev_open(struct net_device *netdev)
>  
>  	ret = lan743x_ptp_open(adapter);
>  	if (ret)
> -		goto close_mac;
> +		goto close_phylink;
>  
>  	lan743x_rfe_open(adapter);
>  
> @@ -3374,6 +3374,8 @@ static int lan743x_netdev_open(struct net_device *netdev)
>  			lan743x_rx_close(&adapter->rx[index]);
>  	}
>  	lan743x_ptp_close(adapter);
> +
> +close_phylink:

It seems odd that basically this unwind logic existed before the
cited patch - even if the internals of the unwind differed - but
was removed by it. And yet, the unwind happens for other cases.

>  	if (adapter->phylink)

As an aside, I don't believe adapter->phylink can ever be NULL here.

>  		lan743x_phylink_disconnect(adapter);

Please do answer the email from netdev-bot+sinfo requesting,
elsewhere in this email thread, information about the providence of
this patch.

The above not withstanding, this looks good to me.

Reviewed-by: Simon Horman <horms@kernel.org>