Re: [PATCH v8 0/3] media: venus: enable venus on qcs615

[Dmitry Baryshkov]

On 06/06/2025 08:54, Krzysztof Kozlowski wrote:
> On 06/06/2025 09:51, Renjiang Han wrote:
> > > > Reading this thread, I don't think that is the case.
> > > >
> > > > I don't see how patches 2/3 or 3/3 depend on 1/3.
> > > >
> > > > The frequency table is a fallback in the driver and the DT changes
> > > > are completely straight forward.
> > > >
> > > > TBH, I think we are hitting an email comms/social barrier here, not a
> > > > technical one.
> > > >
> > > > @Renjiang can you please confirm that freq_table is a fallback,
> > > > qcs615 will work without OPP table and the DTS stuff doesn't depend
> > > > on the driver.
> > > yes, freq_table is a fallback. driver will use freq_table without OPP
> > > table. the DTS doesn't depend on the driver.
> > To correct my previous response:
> > For this project, the driver patch needs to be merged first. Since it
> > falls back to SC7180, but their frequencies are different which is
> > reflected in the OPP table defined in the DTS.
> >
> > The DTS patch is intended to enable video function on the QCS615 platform.
> >
> > The driver patch is to switch from using the driver freq_table to the
> > OPP table from the DTS. Without it, the driver will continue to use the
> > internal freq_table as before.
> >
> > Therefore, if the DTS patch is applied without the driver patch, the video
> > function will be enabled on the QCS615 platform, but the driver will use
> > the SC7180 frequency table, which can lead to an overclocking issue.
>
> This means devices are not really compatible and DTS is not correct.

The devices are compatible: all programming interface is the same. 
Having the freq table in the driver is an error.


--
With best wishes
Dmitry