On 19.04.2019 01:30, David Miller wrote:
> From: Heiner Kallweit <hkallwe...@gmail.com>
> Date: Wed, 17 Apr 2019 00:06:28 +0200
>
>> This driver is the only user of PHY_10GBIT_FEC_FEATURES. So we may be
>> able to remove this predefined feature constant later. Setting
>> phydev->advertising to what is supported is done by phy_probe(),
>> therefore we don't have to do it in the config_init callback.
>> Resetting phydev->autoneg if autoneg isn't supported is also done by
>> phy_probe() now, and manually setting the phylib state machine to
>> NOLINK is neither needed nor advisable.
>> In addition propagate the return value of bcm87xx_of_reg_init.
>>
>> Signed-off-by: Heiner Kallweit <hkallwe...@gmail.com>
>
> I read the discussion on this patch, there doesn't seem to be outright
> NACK but rather some talk about future enhancements and something about
> Florian looking at a few datasheets next week.
>
> What's the story?
>
Apart from this smaller improvement there may be more to improve in this
driver, but it needs the datasheet to find out.
I think it should also be ok to apply the current patch now, and further
improvements later.
