netdev - Re: [PATCH net-next RFC V1 5/5] net: mdio: Add a driver for InES time stamping IP core.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 21 Mar 2018 23:16:52 +0100
From:   Andrew Lunn <andrew@...n.ch>
To:     Richard Cochran <richardcochran@...il.com>
Cc:     netdev@...r.kernel.org, devicetree@...r.kernel.org,
        David Miller <davem@...emloft.net>,
        Florian Fainelli <f.fainelli@...il.com>,
        Mark Rutland <mark.rutland@....com>,
        Miroslav Lichvar <mlichvar@...hat.com>,
        Rob Herring <robh+dt@...nel.org>,
        Willem de Bruijn <willemb@...gle.com>
Subject: Re: [PATCH net-next RFC V1 5/5] net: mdio: Add a driver for InES
 time stamping IP core.

On Wed, Mar 21, 2018 at 02:57:29PM -0700, Richard Cochran wrote:
> On Wed, Mar 21, 2018 at 10:44:36PM +0100, Andrew Lunn wrote:
> > O.K, so lets do the 20 questions approach.
> 
> :)
> 
> > As far as i can see, this is not an MDIO device. It is not connected
> > to the MDIO bus, it has no MDIO registers, you don't even pass a valid
> > MDIO address in device tree.
> 
> Right.

O.K, so i suggest we stop trying to model this thing as an MDIO
device. It is really an MMIO device.

> There might very well be other products out there that *do*
> use MDIO commands.  I know that there are MII time stamping asics and
> ip cores on the market, but I don't know all of their creative design
> details.

So i suggest we leave the design for those until we actual see one.
  
> > It it actually an MII bus snooper? Does it snoop, or is it actually in
> > the MII bus, and can modify packets, i.e. insert time stamps as frames
> > pass over the MII bus?
> 
> It acts like a "snooper" to provide out of band time stamps, but it
> also can modify packets when for the one-step functionality.
>  
> > When the driver talks about having three ports, does that mean it can
> > be on three different MII busses?

O.K, so here is how i think it should be done. It is a device which
offers services to other devices. It is not that different to an
interrupt controller, a GPIO controller, etc. Lets follow how they
work in device tree....

The device itself is just another MMIO mapped device in the SoC:

    	timestamper@...00000 {
		compatible = "ines,ptp-ctrl";
                reg = <0x60000000 0x80>;
		#address-cells = <1>;
		#size-cells = <0>;
	};

The MAC drivers are clients of this device. They then use a phandle
and specifier:

	eth0: ethernet-controller@...00 {
		compatible = "marvell,kirkwood-eth";
		#address-cells = <1>;
		#size-cells = <0>;
		reg = <0x72000 0x4000>;

		timerstamper = <&timestamper 2>
	}

The 2 indicates this MAC is using port 2.

The MAC driver can then do the standard device tree things to follow
the phandle to get access to the device and use the API it exports.

    Andrew