Re: [PATCH 1/1 RFC] net/phy: Add Lantiq PHY driver

[John Crispin]

[ changing haukes mail addr to the intel one ]

On 19/05/2016 08:57, Alexander Stein wrote:
> Hi John,
> 
> On Thursday 19 May 2016 06:50:56, John Crispin wrote:
>> On 18/05/2016 18:24, Florian Fainelli wrote:
>>> CC'ing Andrew, John,
>>
>> also CC'ing Matthias and Hauke. we have had a driver in OpenWrt/LEDE for
>> several years that seems a little more complete than this one.
>>
>> https://git.lede-project.org/?p=source.git;a=blob;f=target/linux/lantiq/patc
>> hes-4.4/0023-NET-PHY-adds-driver-for-lantiq-PHY11G.patch;h=93bb4275ec1d261f3
>> 98afb8fdc879c1dd973f997;hb=HEAD
> 
> Thanks for the link, I wasn't aware of that patch. I like it in general, but 
> there are some things I'd like to get addressed first:
> * vr9_gphy_of_reg_init() writes uncoditionally to led3h and led3l even on
>   PEf7071 which does not have this register at all

we use this driver mainly on the 11g and 22f version. mathias recently
added the led3 handling.

@Mathias, can you have a look at this and fix it inside the lede tree ?

> * Why is PHY_HAS_INTERRUPT commented out everywhere?

legacy code, the old mips silicon had a bug and the internal phys irq
lines worked unreliably so we used polling instead. rather than remove
the code i just disabled that part. code is not cleaned up yet for
upstream submission as you can tell :-)

> * ltq_phy_init and ltq_phy_exit can be simplified using phy_drivers_register
>   and phy_drivers_unregister

yes, this driver is based on a version made by daniel from spharion
about 5 years ago. that api probably did not exist at the time

> * A mdio_device_id table is missing

same as previous answer, old api version

	John