RE: [PATCH net-next 4/5] net: ravb: Do not apply RX checksum settings to hardware if the interface is down

[Biju Das]

> -----Original Message-----
> From: Biju Das <biju.das.jz@xxxxxxxxxxxxxx>
> Sent: Thursday, February 8, 2024 8:09 AM
> Subject: RE: [PATCH net-next 4/5] net: ravb: Do not apply RX checksum
> settings to hardware if the interface is down
> 
> Hi Sergey,
> 
> > -----Original Message-----
> > From: Sergey Shtylyov <s.shtylyov@xxxxxx>
> > Sent: Wednesday, February 7, 2024 8:50 PM
> > Subject: Re: [PATCH net-next 4/5] net: ravb: Do not apply RX checksum
> > settings to hardware if the interface is down
> >
> > On 2/7/24 3:07 PM, Claudiu wrote:
> >
> > > From: Claudiu Beznea <claudiu.beznea.uj@xxxxxxxxxxxxxx>
> > >
> > > Do not apply the RX checksum settings to hardware if the interface
> > > is
> > down.
> > > In case runtime PM is enabled, and while the interface is down, the
> > > IP will be in reset mode (as for some platforms disabling the clocks
> > > will switch the IP to reset mode, which will lead to losing
> > > registers
> > > content) and
> >
> >    The register contents? I thought I'd pointed out all of these...
> >
> > > applying settings in reset mode is not an option. Instead, cache the
> > > RX checksum settings and apply them in ravb_open() through
> > ravb_emac_init().
> > > This has been solved by introducing pm_runtime_active() check. The
> > > device runtime PM usage counter has been incremented to avoid
> > > disabling the device clocks while the check is in progress (if any).
> > >
> > > Commit prepares for the addition of runtime PM.
> > >
> > > Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@xxxxxxxxxxxxxx>
> >
> > Reviewed-by: Sergey Shtylyov <s.shtylyov@xxxxxx>
> >
> >    I'm afraid such check now needs to be added to
> > ravb_set_features_gbeth() that's populated by Biju Das' checksum
> > patches (which I've already ACKed)...
> 
> You mean this check to be moved to ravb_set_features_rcar() instead of
> ravb_set_rx_csum() as ravb_set_rx_csum() is called in receive path as well
> which is interface up case.
> ON reset mode, anyway we don't get any interrupts so there is no rx.
> Then possibility is through set_features??


Or are you suggesting to have a common code to avoid code duplication?

On interface down case, just cache the feature and return?

Active cases, call the family specific callback()?

Cheers,
Biju