Re: [PATCH net-next v7 3/3] net: dsa: yt921x: Add support for Motorcomm YT921x

[Andrew Lunn <andrew@lunn.ch>]

> > The DSA framework makes no such guarantees. The DSA framework is also
> > not the only entry point into the driver, phylink will directly call
> > into the driver, and if you implement things like LEDs, they will have
> > direct access to the driver.
> >
> > So i suggest only having a high level lock, acquired on entry,
> > released on exit, e.g. as mv88e6xxx does. KISS.

> So you mean holding bus->mdio_lock during any operations instead of
> implementing driver's own lock? Wouldn't other bus participants starve
> if I want to poll a register for like 100ms?

No, leave the mdio lock alone.

The mv88e6xxx driver has a lock in the 'priv' structure:

https://elixir.bootlin.com/linux/v6.16.5/source/drivers/net/dsa/mv88e6xxx/chip.h#L354

Function which are exposed in dsa_switch_ops take the lock at the
beginning, and release it at the end:

static int mv88e6xxx_port_mdb_add(struct dsa_switch *ds, int port,
				  const struct switchdev_obj_port_mdb *mdb,
				  struct dsa_db db)
{
	struct mv88e6xxx_chip *chip = ds->priv;
	int err;

	mv88e6xxx_reg_lock(chip);
	err = mv88e6xxx_port_db_load_purge(chip, port, mdb->addr, mdb->vid,
					   MV88E6XXX_G1_ATU_DATA_STATE_MC_STATIC);
	if (err)
		goto out;

	if (!mv88e6xxx_port_db_find(chip, mdb->addr, mdb->vid))
		err = -ENOSPC;

out:
	mv88e6xxx_reg_unlock(chip);

	return err;
}

There are functions which wait for the device, polling:

https://elixir.bootlin.com/linux/v6.16.5/source/drivers/net/dsa/mv88e6xxx/chip.c#L87

and we do this while holding the lock. We have a timeout of 50ms, but
it never takes that long, on average it is just a couple of MDIO bus
transactions.

	Andrew