On Fri, Mar 27, 2026 at 09:42:05PM +0200, Dmitry Baryshkov wrote:
> On Thu, Mar 26, 2026 at 10:36:05AM +0530, Dikshita Agarwal wrote:
> > 
> > 
> > On 1/25/2026 5:00 PM, Dmitry Baryshkov wrote:
> > > Specifying UBWC data in each driver doesn't scale and is prone to
> > > errors. Request UBWC data from the central database in preparation to
> > > using it through the rest of the driver.
> > > 
> > > Reviewed-by: Bryan O'Donoghue <[email protected]>
> > > Reviewed-by: Konrad Dybcio <[email protected]>
> > > Reviewed-by: Dikshita Agarwal <[email protected]>
> > > Tested-by: Wangao Wang <[email protected]>
> > > Signed-off-by: Dmitry Baryshkov <[email protected]>
> > > ---
> > >  drivers/media/platform/qcom/iris/Kconfig      | 1 +
> > >  drivers/media/platform/qcom/iris/iris_core.h  | 4 ++++
> > >  drivers/media/platform/qcom/iris/iris_probe.c | 5 +++++
> > >  3 files changed, 10 insertions(+)
> > > 
> > > @@ -244,6 +245,10 @@ static int iris_probe(struct platform_device *pdev)
> > >  
> > >   core->iris_platform_data = of_device_get_match_data(core->dev);
> > >  
> > > + core->ubwc_cfg = qcom_ubwc_config_get_data();
> > > + if (IS_ERR(core->ubwc_cfg))
> > > +         return PTR_ERR(core->ubwc_cfg);
> > 
> > Afterthought: This change assumes that the presence of a UBWC config
> > implies Iris UBWC support. However, some platforms (e.g. qcm2290) do have
> > UBWC data defined at the SoC level but do not support UBWC in the video
> > firmware, which could potentially surface during SYS_INIT if UBWC is
> > advertised unconditionally, so this might be worth double‑checking.
> 
> No, this change implies that the lack of the UBWC config is an error for
> modern platforms (which it is anyway).

And that's not to mention that Iris doesn't support Agatti / QCM2290,
nor any other UBWC-less platform at this point. Those platforms will
need additional chagnes anyway, so you might as well include a patch to
skip sending UBWC-related packets in iris_hfi_gen2_packet_sys_init().

-- 
With best wishes
Dmitry

Reply via email to