Re: [PATCH 01/33] MAINTAINERS: Add Qualcomm Iris video accelerator driver

[Bryan O'Donoghue]

On 14/08/2023 19:44, Dikshita Agarwal wrote:
> On 7/29/2023 4:18 AM, Randy Dunlap wrote:
> > On 7/28/23 06:23, Vikash Garodia wrote:
> > > Add an entry for Iris video encoder/decoder accelerator driver.
> > >
> > > Signed-off-by: Dikshita Agarwal <quic_dikshita@xxxxxxxxxxx>
> > > Signed-off-by: Vikash Garodia <quic_vgarodia@xxxxxxxxxxx>
> > > ---
> > >   MAINTAINERS | 10 ++++++++++
> > >   1 file changed, 10 insertions(+)
> > >
> > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > index 3be1bdf..ea633b2 100644
> > > --- a/MAINTAINERS
> > > +++ b/MAINTAINERS
> > > @@ -17671,6 +17671,16 @@ T:	git git://linuxtv.org/media_tree.git
> > >   F:	Documentation/devicetree/bindings/media/*venus*
> > >   F:	drivers/media/platform/qcom/venus/
> > >   
> > > +QUALCOMM IRIS VIDEO ACCELERATOR DRIVER
> >
> > This entry should immediately follow:
> > QUALCOMM IPQ4019 VQMMC REGULATOR DRIVER
> >
> > to keep the file in alphabetical order.
> Sure, will fix this in next version.

I think TBH before we see a next version, there needs to be a conclusive 
argument on why a new driver - instead of an update to the existing 
venus - is the way to go.

We have an ongoing corpus of working code that people use. The attempt 
to at least _try_ to integrate 8550 and beyond to upstream venus should 
be made.

If it fails, then we can discuss a branched driver.

Its not up to me but, that's certainly my honest and unvarnished input.

Instead of investing time in V2 - please invest time in upstream venus 
or make the technical argument conclusively _prior_ to V2 as to why V2 
and beyond is the "only" way forward for 8550 and beyond.

---
bod