Re: [PATCH net-next v2 2/9] phy: add hwtstamp_get callback to phy drivers

[Vadim Fedorenko <vadim.fedorenko@linux.dev>]

On 13/11/2025 17:15, Russell King (Oracle) wrote:
> On Thu, Nov 13, 2025 at 05:05:18PM +0000, Vadim Fedorenko wrote:
>> On 13/11/2025 16:57, Russell King (Oracle) wrote:
>>> On Thu, Nov 13, 2025 at 04:48:00PM +0000, Vadim Fedorenko wrote:
>>>> If the above is correct, then yes, there is no reason to implement
>>>> SIOCGHWTSTAMP, and even more, SIOCSHWTSTAMP can be technically removed
>>>> as a dead code.
>>>
>>> I think you're missing the clarification in this sentence "... to
>>> implement SIOCGHWTSTAMP in phy_mii_ioctl(), and even more,
>>> SIOCSHWTSTAMP can be removed from this function as dead code.""
>>
>> Ok, it's better to "there is no reason to have SIOCGHWTSTAMP chunk,
>> provided in patch 2 of this series"
>>
>> Or are you asking for the clarification of SIOCSHWTSTAMP removal?
>> I don't plan to remove it, at least not in this series. I just wanted
>> to mention that there will be no way to reach SIOCSHWTSTAMP case in
>> phy_mii_ioctl() from user-space ABI. Does it make sense?
> 
> I'm not asking. I'm trying to work out what you're trying to say, and
> suggesting a clarification to your last paragraph that would clarify
> what I thought you mean. Now I'm even more confused about what you're
> proposing.

The original idea is to have SIOCGHWTSTAMP and netlink hwtstamp_get for
PHY devices. But as ioctl is deprecated interface, I'm thinking of
removing SIOCSHWTSTAMP in favor of netlink hwtstamp_set.