| Message ID | 20210318141550.646383-3-tobias@waldekranz.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | net: dsa: mv88e6xxx: Offload bridge port flags | expand |
On Thu, Mar 18, 2021 at 03:15:44PM +0100, Tobias Waldekranz wrote: > When a port is a part of a LAG, the ATU will create dynamic entries > belonging to the LAG ID when learning is enabled. So trying to > fast-age those out using the constituent port will have no > effect. Unfortunately the hardware does not support move operations on > LAGs so there is no obvious way to transform the request to target the > LAG instead. > > Instead we document this known limitation and at least avoid wasting > any time on it. > > Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> > --- Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
On 3/18/2021 7:15 AM, Tobias Waldekranz wrote: > When a port is a part of a LAG, the ATU will create dynamic entries > belonging to the LAG ID when learning is enabled. So trying to > fast-age those out using the constituent port will have no > effect. Unfortunately the hardware does not support move operations on > LAGs so there is no obvious way to transform the request to target the > LAG instead. > > Instead we document this known limitation and at least avoid wasting > any time on it. > > Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index f0a9423af85d..ed38b4431d74 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -1479,6 +1479,13 @@ static void mv88e6xxx_port_fast_age(struct dsa_switch *ds, int port) struct mv88e6xxx_chip *chip = ds->priv; int err; + if (dsa_to_port(ds, port)->lag_dev) + /* Hardware is incapable of fast-aging a LAG through a + * regular ATU move operation. Until we have something + * more fancy in place this is a no-op. + */ + return; + mv88e6xxx_reg_lock(chip); err = mv88e6xxx_g1_atu_remove(chip, 0, port, false); mv88e6xxx_reg_unlock(chip);
When a port is a part of a LAG, the ATU will create dynamic entries belonging to the LAG ID when learning is enabled. So trying to fast-age those out using the constituent port will have no effect. Unfortunately the hardware does not support move operations on LAGs so there is no obvious way to transform the request to target the LAG instead. Instead we document this known limitation and at least avoid wasting any time on it. Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> --- drivers/net/dsa/mv88e6xxx/chip.c | 7 +++++++ 1 file changed, 7 insertions(+)