Re: [PATCH net-next v2 2/2] net: dsa: yt921x: Add DCB/priority support

[Oleksij Rempel <o.rempel@pengutronix.de>]

On Sat, Jan 24, 2026 at 01:47:59AM +0200, Vladimir Oltean wrote:
> > +static int
> > +yt921x_dsa_port_del_dscp_prio(struct dsa_switch *ds, int port, u8 dscp, u8 prio)
> > +{
> > +	struct yt921x_priv *priv = to_yt921x_priv(ds);
> > +	int res;
> > +
> > +	mutex_lock(&priv->reg_lock);
> > +	res = yt921x_reg_write(priv, YT921X_IPM_DSCPn(dscp),
> > +			       ipm_ctrl(IEEE8021Q_TT_BK));
> 
> Question regarding semantics, also for Oleksij: deleting the DSCP app
> table entry should cause that DSCP value to be mapped to bulk, as in
> your interpretation (or best effort for KSZ), or according to the next
> selector in line, as per trust order (for example port default priority,
> which may not be bulk)?

It maps to bulk.

My interpretation of this hardware is that it uses a direct lookup table
without a "valid/invalid" bit per entry. Once DSCP is enabled in the
trust order, the switch performs the lookup for every packet; it cannot
transparently skip a specific DSCP value to fall back to the next
selector.

Therefore, "deleting" an entry practically remaps it to the default
(Bulk/Best Effort) priority. This matches the logic we use in the KSZ
driver.

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |