Re: [PATCH net v2 4/5] net: dsa: mt7530: clear flood flags on bridge leave

[Paolo Abeni <pabeni@redhat.com>]

On 5/14/26 4:05 PM, Daniel Golle wrote:
> Flood flags set by port_bridge_flags persist after a port leaves the
> bridge, causing unknown unicast to be forwarded to standalone ports.
> 
> Clear UNU_FFP, UNM_FFP and BC_FFP in port_bridge_leave so that the
> port returns to its initial state without flooding.
> 
> Fixes: 5a30833b9a16 ("net: dsa: mt7530: support MDB and bridge flag operations")
> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
> ---
> v2: no changes
> 
>  drivers/net/dsa/mt7530.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
> index 752ba92b0851..5b58e42bfda9 100644
> --- a/drivers/net/dsa/mt7530.c
> +++ b/drivers/net/dsa/mt7530.c
> @@ -1767,6 +1767,11 @@ mt7530_port_bridge_leave(struct dsa_switch *ds, int port,
>  	mt7530_rmw(priv, MT7530_PCR_P(port), PCR_PORT_VLAN_MASK,
>  		   MT7530_PORT_MATRIX_MODE);
>  
> +	/* Clear flood flags so they don't persist across bridge leave */
> +	mt7530_clear(priv, MT753X_MFC,
> +		     UNU_FFP(BIT(port)) | UNM_FFP(BIT(port)) |
> +		     BC_FFP(BIT(port)));

I think sashiko has a point here:

---
Will these cleared flags be immediately overwritten by the DSA core?
When a port leaves a bridge, the DSA core's dsa_port_bridge_leave()
broadcasts the DSA_NOTIFIER_BRIDGE_LEAVE event, which triggers this
callback and clears the bits.
However, immediately after the broadcast returns,
dsa_port_bridge_leave() unconditionally calls
dsa_port_switchdev_unsync_attrs(), which calls
dsa_port_clear_brport_flags().
This explicitly configures standalone ports to flood everything by
setting the BR_FLOOD, BR_MCAST_FLOOD, and BR_BCAST_FLOOD flags to 1, and
then invokes the driver's port_bridge_flags callback.
As a result, mt7530_port_bridge_flags() processes this call and executes
mt7530_rmw() to turn the UNU_FFP, UNM_FFP, and BC_FFP bits right back to
1 for the port.
Does this leave the isolation bug unfixed since the register clearing is
reverted before the DSA bridge leave sequence fully completes?
---

Since the other fixes are independent, I'm applying them.

/P