Re: [PATCH net v3 1/1] net: phy: nxp-c45-tja11xx: fix the PTP interrupt enablig/disabling

[Radu Pirea (OSS)]

Hi Jakub,

On 21.06.2023 06:31, Jakub Kicinski wrote:
> On Mon, 19 Jun 2023 16:28:51 +0300 Radu Pirea (NXP OSS) wrote:
> > Subject: [PATCH net v3 1/1] net: phy: nxp-c45-tja11xx: fix the PTP interrupt enablig/disabling
>
> typo: enablig -> enabling
>
> > .config_intr() handles only the link event interrupt and should
> > disable/enable the PTP interrupt also.
>
> I don't understand this sentence, TBH, could you rephrase? >
> > Fixes: 514def5dd339 ("phy: nxp-c45-tja11xx: add timestamping support")
>
> For a fix we really need to commit message to say what the problem is,
> in terms which will be understood by the user. User visible behavior.

If tools like ptp4l are killed, will leave egress timestamp interrupt 
enabled. And I would like to say that I was able to trigger any bug 
related to this, but I wasn't. So, I don't have any bad behaviour to 
describe :)

However, now I realize that disabling all the PTP IRQs is not a smart 
way to fix this virtual pseudo issue.

> > CC: stable@xxxxxxxxxxxxxxx # 5.15+
> > Signed-off-by: Radu Pirea (NXP OSS) <radu-nicolae.pirea@xxxxxxxxxxx>
> > ---
> >
> > Where is V1?
> > https://patchwork.kernel.org/project/netdevbpf/patch/20230410124856.287753-1-radu-nicolae.pirea@xxxxxxxxxxx/
> >
> > Where is V2?
> > https://patchwork.kernel.org/project/netdevbpf/patch/20230616135323
>
> This link looks cut off.

https://patchwork.kernel.org/project/netdevbpf/patch/20230616135323.98215-2-radu-nicolae.pirea@xxxxxxxxxxx/

> > +     /* 0x807A register is not present on SJA1110 PHYs. */
>
> Meaning? It's safe because the operation will be ignored?

Yes. The PHY will ignore the writes and will return 0 on reads.

> --
> pw-bot: cr

--
Radu P.