netdev - Re: [PATCH] net/phy: micrel: Follow attached

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]

Date:	Sat, 17 Oct 2015 10:38:43 -0700
From:	Florian Fainelli <f.fainelli@...il.com>
To:	Dinh Nguyen <dinguyen@...nsource.altera.com>
Cc:	Andrew Lunn <andrew@...n.ch>, netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH] net/phy: micrel: Follow attached_dev to get to the MAC device

2015-10-17 8:06 GMT-07:00 Dinh Nguyen <dinguyen@...nsource.altera.com>:
> On Sat, 17 Oct 2015, Dinh Nguyen wrote:
>
>> On Fri, 16 Oct 2015, Andrew Lunn wrote:
>>
>> > On Fri, Oct 16, 2015 at 05:58:41PM -0500, Dinh Nguyen wrote:
>> > > On Fri, 16 Oct 2015, Andrew Lunn wrote:
>> > >
>> > > > Commit "8b63ec1837fa phylib: Make PHYs children of their MDIO bus, not
>> > > > the bus' parent." broke finding PHY properties in the MAC device tree
>> > > > node. The parent device is now the MDIO bus, not the MAC. Use
>> > > > attached_dev towards the MAC device tree node.
>> > > >
>> > > > Fixes: 8b63ec1837fa ("phylib: Make PHYs children of their MDIO bus, not the bus' parent.")
>> > > > Signed-off-by: Andrew Lunn <andrew@...n.ch>
>> > > > ---
>> > > >
>> > > > Compile tested only.
>> > > >
>> > > > Dinh, please could you test it and report back if it works or not.
>> > > >
>> > >
>> > > This patch did not seem to fix the problem. The following code did seem to
>> > > fix the problem:
>> > >
>> > >   if (!of_node && dev->parent->of_node)
>> > > -               of_node = dev->parent->of_node;
>> > > +       do {
>> > > +               of_node = dev->of_node;
>> > > +               dev = dev->parent;
>> > > +               i++;
>> > > +       } while (!of_node && dev);
>> >
>> > This might fix the issue, but it has disadvantages. As i said before,
>> > it allows people to place phy properties into the mdio device node. We
>> > want to be reducing placing you can add phy properties, not adding
>> > more.
>> >
>
> I've also tried creating a separate phy node in the DTS and have the EMAC
> point the PHY with a 'phy = <&phy0>;', but that also didn't seem to work with
> your patch.

If you intend this to be a real phandle to a phy, it should be something like:

phy-handle = <&phy0>;

'phy' is not a valid phandle property that stmmac recognizes

Looking at stmmac and how the various dwmac-* probe the driver,
SET_NETDEV_DEV() is what assigns the network device's device pointer,
and that seems to be done correctly with the proper device argument.
-- 
Florian
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html