Re: [PATCH net v2] net: dsa: mv88e6xxx: 88E6191X and 88E6193X have no PTP

[Andrew Lunn <andrew@lunn.ch>]

On Mon, Sep 21, 2026 at 05:56:29AM +0200, Nicolo Giuliani via B4 Relay wrote:
> From: Nicolo Giuliani <nicolo.giuliani6@studio.unibo.it>
> 
> The 88E6191X and 88E6193X are 6393 family devices that share
> mv88e6393x_ops with the 88E6393X and are marked as ptp_support. Marvell's
> UMSD driver describes both as parts without AVB (88E6193X: "BGA package -
> No AVB, No Routing, No Cut-through"), and the register access confirms it
> on an 88E6193X: the whole indirect AVB register space behind Global 2
> registers 0x16 and 0x17 reads zero, for every port, block and address,
> with the 6390 and with the 6352 command encoding. Writes to the TAI
> registers, including the clock period register and the TAI global
> configuration register, read back as zero.
> 
> Since commit 7e3c18097a70 ("net: dsa: mv88e6xxx: read cycle counter
> period from hardware") the PTP setup reads the TAI clock period, so the
> switch fails to probe:
> 
> mv88e6xxx ...: unexpected cycle counter period of 0 ps
> 
> Add mv88e6191x_ops, a copy of mv88e6393x_ops without avb_ops and ptp_ops,
> use it for the 88E6191X and the 88E6193X and stop setting ptp_support for
> them. The 88E6393X is unchanged.
> 
> Tested on an 88E6193X (Sophos XGS 107w): the switch probes and the ports
> work. I do not have an 88E6191X, it is changed because UMSD describes it
> the same way.
> 
> Fixes: de776d0d316f ("net: dsa: mv88e6xxx: add support for mv88e6393x family")
> Suggested-by: Andrew Lunn <andrew@lunn.ch>
> Assisted-by: LLM
> Signed-off-by: Nicolo Giuliani <nicolo.giuliani6@studio.unibo.it>

Reviewed-by: Andrew Lunn <andrew@lunn.ch>

    Andrew