netdev - Re: [RFC PATCH 3/3] net: phy: at803x: add device tree binding

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <9de3e32f081c9b2d1227514b4e9f166e@walle.cc>
Date:   Sat, 02 Nov 2019 02:18:27 +0100
From:   Michael Walle <michael@...le.cc>
To:     Florian Fainelli <f.fainelli@...il.com>
Cc:     linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        netdev@...r.kernel.org, Andrew Lunn <andrew@...n.ch>,
        Heiner Kallweit <hkallweit1@...il.com>
Subject: Re: [RFC PATCH 3/3] net: phy: at803x: add device tree binding

Am 2019-10-31 18:35, schrieb Florian Fainelli:
> On 10/31/19 10:22 AM, Michael Walle wrote:
>> Am 2019-10-31 00:59, schrieb Michael Walle:
>>>>> +
>>>>> +    if (of_property_read_bool(node, "atheros,keep-pll-enabled"))
>>>>> +        priv->flags |= AT803X_KEEP_PLL_ENABLED;
>>>> 
>>>> This should probably be a PHY tunable rather than a Device Tree
>>>> property
>>>> as this delves more into the policy than the pure hardware 
>>>> description.
>>> 
>>> To be frank. I'll first need to look into PHY tunables before
>>> answering ;)
>>> But keep in mind that this clock output might be used anywhere on the
>>> board. It must not have something to do with networking. The PHY has 
>>> a
>>> crystal and it can generate these couple of frequencies regardless of
>>> its network operation.
>> 
>> Although it could be used to provide any clock on the board, I don't 
>> know
>> if that is possible at the moment, because the PHY is configured in
>> config_init() which is only called when someone brings the interface 
>> up,
>> correct?
>> 
>> Anyway, I don't know if that is worth the hassle because in almost all
>> cases the use case is to provide a fixed clock to the MAC for an RGMII
>> interface. I don't know if that really fits a PHY tunable, because in
>> the worst case the link won't work at all if the SoC expects an
>> always-on clock.
> 
> Well, that was my question really, is the clock output being controlled
> the actual RXC that will feed back to the MAC or is this is another
> clock output pin (sorry if you indicated that before and I missed it)?

No it is not the RX clock. The PHY has three clock pins, RX clock, TX
clock and a general purpose CLK_25M pin.

> If this is the PHY's RXC, then does the configuration (DSP, PLL, XTAL)
> matter at all on the generated output frequency, or is this just a
> choice for the board designer, and whether the PHY is configured for
> MII/RGMII, it outputs the appropriate clock at 25/125Mhz?

The RXC changes the frequency according to the speed.

-- 
-michael