Re: [PATCH net-next] net: dsa: convert to ndo_hwtstamp_get() and ndo_hwtstamp_set()

[Vladimir Oltean <vladimir.oltean@nxp.com>]

On Thu, May 08, 2025 at 09:48:40PM +0100, Vadim Fedorenko wrote:
> On 08/05/2025 21:40, Vladimir Oltean wrote:
> > On Thu, May 08, 2025 at 09:25:14PM +0100, Vadim Fedorenko wrote:
> > > The new interface also supports providing error explanation via extack,
> > > it would be great to add some error messages in case when setter fails.
> > > For example, HIRSCHMANN HellCreek switch doesn't support disabling
> > > of timestamps, it's not obvious from general -ERANGE error code, but can
> > > be explained by the text in extack message.
> > 
> > I wanted to keep the patches spartan and not lose track of the conversion
> > subtleties in embelishments like extack messages which can be added later
> > and do not require nearly as much attention to the flow before and after.
> > I'm afraid if I say "yes" here to the request to add extack to hellcreek
> > I'm opening the door to further requests to do that for other DSA drivers,
> > and sadly I do not have infinite time to fulfill them. Plus, I would
> > like to finalize the conversion tree-wide by the end of this development
> > cycle.
> > 
> > Even if I were to follow through with your request, I would do so in a
> > separate patch. I've self-reviewed this patch prior to posting it, and I
> > was already of the impression that it is pretty busy as it is.
> 
> I agree that the patch is pretty busy, and the extack additions should
> go into separate patch. The only thing which bothers me is that it may never
> happen if it's not done with this patch.

That may well be. But look at it another way, I wrote this patch in July
2023 and never got to upstream it, then Kory pinged me because it's
necessary to get rid of the old API. I don't want to go back and spend
time on extack messages when there's still a long way to go, and the
priority is obviously somewhere else. I've added this request to my
to-do list, and if I still have time at the end of all conversions, I'll
go through DSA drivers and see what can be improved.

> Anyway, the conversion code looks good, so
> 
> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

Thanks for the review, here and elsewhere. Do you want me to copy you on
the remaining conversions?