netdev - Re: [RFC PATCH 2/3] dt-bindings: net: phy: Add support for AT803X

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:   Thu, 31 Oct 2019 09:45:24 -0700
From:   Florian Fainelli <f.fainelli@...il.com>
To:     Michael Walle <michael@...le.cc>, Andrew Lunn <andrew@...n.ch>
Cc:     linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        netdev@...r.kernel.org
Subject: Re: [RFC PATCH 2/3] dt-bindings: net: phy: Add support for AT803X

On 10/30/19 5:14 PM, Michael Walle wrote:
>> So we can later add atheros,rgmii-io-2v5. That might need a regulator
>> as well. Maybe add that 2.5V is currently not supported.
> 
> There is no special setting for the 2.5V mode. This is how it works: there is one voltage pad for the RGMII interface. Either you connect this pad to a 2.5V voltage or you leave it open (well you would connect some decoupling Cs). If you leave it open the internal LDO, which seems to be enabled in any case takes over, supplying 1.5V. then there is a bit in the debug register which can switch the internal LDO to 1.8V. So if you'll use 2.5V the bit is irrelevant. 
> 
> Like I said maybe a "rgmii-io-microvolts" is a better property and only in the 1800000 setting would turn on this bit. but then both other setting would be a noop. 

That would align with the regulator subsystem units, but maybe you
should have the PHY driver be a regulator provider for itself so you can
chose wether you want to operate at 1.5V or 1.8V, or you have an
external regulator providing I/O supplies. That would make the whole
thing consistent from the driver's perspective and would not necessarily
be too far fetched from a HW description perspective?
--
Florian