netdev - Re: [PATCH] mdio-bitbang: add support for lowlevel mdio read/write

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] [day] [month] [year] [list]

Message-ID: <20200421143156.GD2338@pengutronix.de>
Date:   Tue, 21 Apr 2020 16:31:56 +0200
From:   Michael Grzeschik <mgr@...gutronix.de>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     netdev@...r.kernel.org, f.fainelli@...il.com, davem@...emloft.net,
        kernel@...gutronix.de
Subject: Re: [PATCH] mdio-bitbang: add support for lowlevel mdio read/write

Hi Andrew,

I want to refresh this thread.

On Wed, Jan 29, 2020 at 10:48:05PM +0100, Michael Grzeschik wrote:
>On Wed, Jan 29, 2020 at 04:53:46PM +0100, Andrew Lunn wrote:
>> On Wed, Jan 29, 2020 at 04:42:01PM +0100, Michael Grzeschik wrote:
>> > Hi Andrew!
>> >
>> > I tested your patch. But it works only partially. For the case that
>> > the upper driver is directly communicating in SMI mode with the phy,
>> > this works fine. But the regular MDIO connection does not work anymore
>> > afterwards.
>> >
>> > The normals MDIO communication still needs to work, as mdio-gpio is
>> > calling of_mdiobus_register that on the other end calls get_phy_device
>> > and tries to communicate via regular MDIO to the device.
>>
>> Do you mean you have a mix of devices on the bus, some standards
>> comformant, and others using this hacked up SMI0 mode?
>
>Actually it is the same device used in both modes. The SMI0
>mode is used by the switch driver to address the extended switch
>functions. But on the same bus we have the fec connected to
>the cpu bound fixed-phy (microchip,ks8863) via MDIO.
>
>> You need to specify per device if SMI0 should be used?
>
>Yes, we have to use the same bus fot both modes SMI0 and MDIO.

In fact I for now used the cpu bound port with phy-handle to the fec.
This way it still used mdio for the initial probe.

But it should also work to use fixed-phy for it don't run into mdio
communication on the same bus. This way your patch should work.

In case you did not think of anything else, I will send the series
including your patch after I tested it with master.

Regards,
Michael

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)