On Tue, 5 Mar 2024 08:36:54 +0000 "Russell King (Oracle)" <linux@xxxxxxxxxxxxxxx> wrote: > On Mon, Mar 04, 2024 at 06:57:34PM -0800, Jakub Kicinski wrote: > > On Mon, 26 Feb 2024 14:39:59 +0100 Kory Maincent wrote: > > > Prepare for future hardware timestamp selection by adding source and > > > corresponding pointers to ptp_clock structure. Additionally, introduce > > > helpers for registering specific phydev or netdev PTP clocks, retrieving > > > PTP clock information such as hwtstamp source or phydev/netdev pointers, > > > and obtaining the ptp_clock structure from the phc index. > > > > Can we assume there's one PHC per netdev? > > We both store the netdev/phydev info in the ptp clock > > and ptp clock in the netdev. Is there a reason for that? > > No. In the case of mvpp2 + marvell PHY, the two PTP implementations are > entirely separate. Yes the PTP clock can be independent from the netdev. We need to know which software layer register the PHC to be able to call its callbacks. My commit log is a bit small here. I will enhance it in the next version. Regards, -- Köry Maincent, Bootlin Embedded Linux and kernel engineering https://bootlin.com
