RE: [PATCH v12 net-next 0/8] dpll/ice: Add generic DPLL type and full TX reference clock control for E825

[Nitka, Grzegorz]

> -----Original Message-----
> From: Jakub Kicinski <kuba@xxxxxxxxxx>
> Sent: Saturday, June 6, 2026 1:26 AM
> To: Nitka, Grzegorz <grzegorz.nitka@xxxxxxxxx>
> Cc: netdev@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; intel-wired-
> lan@xxxxxxxxxxxxxxxx; Oros, Petr <poros@xxxxxxxxxx>;
> richardcochran@xxxxxxxxx; andrew+netdev@xxxxxxx; Kitszel, Przemyslaw
> <przemyslaw.kitszel@xxxxxxxxx>; Nguyen, Anthony L
> <anthony.l.nguyen@xxxxxxxxx>; Prathosh.Satish@xxxxxxxxxxxxx; Vecera,
> Ivan <ivecera@xxxxxxxxxx>; jiri@xxxxxxxxxxx; Kubalewski, Arkadiusz
> <arkadiusz.kubalewski@xxxxxxxxx>; vadim.fedorenko@xxxxxxxxx;
> donald.hunter@xxxxxxxxx; horms@xxxxxxxxxx; pabeni@xxxxxxxxxx;
> davem@xxxxxxxxxxxxx; edumazet@xxxxxxxxxx
> Subject: Re: [PATCH v12 net-next 0/8] dpll/ice: Add generic DPLL type and full
> TX reference clock control for E825
> 
> On Fri, 5 Jun 2026 16:10:19 +0000 Nitka, Grzegorz wrote:
> > Hi Kuba. Just submitted v13.
> >
> > It includes the following fixes for pre-existing issues:
> >   - dpll core fixes. Each AI review concern (3) is addressed in separate
> >     commit. If you think it's better to squash them, let me know (however
> >     it addresses issues from two different 'old' commits).
> >     Also, the hint form AI  to use different dpll xa_array (parent not pin) to
> >     address one of the issues, it simply does not work (WARNING flood
> observed,
> >     more details in patch 3/11)
> >   - ice driver fix for potential hung on flush_workqueue in error path for FW
> node pins
> >
> > As you noted, two other pre-existing issues are covered by ICE_FLAG_DPLL.
> > I left the code as it is for those concerns.
> > There is one more pre-existing one, related to ice_ptp_link_change. As the fix
> seems to be
> > rather simple one, I believe I need more time for more comprehensive
> testing.
> > So my preference is to go with standard fix-path on 'net'.
> 
> SG, thanks!
> 
> BTW either you or Arkadiusz should chime in on the NCO thread, please:
> https://lore.kernel.org/all/20260531194423.383366-2-
> ivecera@xxxxxxxxxx/
> Shouldn't take much time to express an opinion, I hope.

Sure, we will do on Monday (we had a long weekend this week, so limited
capacity).

Regarding this series. There are another pre-existing dpll-core findings
reported in v13. As I still had to address valid critical/high issues in 'ice',
I added 2 simple dpll-core fixes in v14 (just submitted, along with ice 
improvements).

Regards

Grzegorz