[PATCH v7 2/7] mtk-mdp: add driver to probe mdp components

[Ezequiel Garcia]

Hi everyone,

On Tue, 7 Dec 2021 at 09:37, Hans Verkuil <hverkuil at xs4all.nl> wrote:
>
> On 9/5/21 6:23 PM, houlong wei wrote:
> > Hi Ezequiel,
> >
> > Thank you for your attention to this series of patches. I answer partial of your questions below.
> > Regards,
> > Houlong
> >
> > On Sat, 2021-09-04 at 20:34 +0800, Ezequiel Garcia wrote:
> >> Hi Eizan,
> >>
> >> Sorry for seeing this series so late.
> >>
> >> On Wed, 25 Aug 2021 at 03:35, Eizan Miyamoto <eizan at chromium.org>
> >> wrote:
> >>>
> >>> Broadly, this patch (1) adds a driver for various MTK MDP
> >>> components to
> >>> go alongside the main MTK MDP driver, and (2) hooks them all
> >>> together
> >>> using the component framework.
> >>>
> >>> (1) Up until now, the MTK MDP driver controls 8 devices in the
> >>> device
> >>> tree on its own. When running tests for the hardware video decoder,
> >>> we
> >>> found that the iommus and LARBs were not being properly configured.
> >>
> >> Why were not being properly configured? What was the problem?
> >> Why not fixing that instead?
> >>
> >> Does this mean the driver is currently broken and unusable?
> >
> > This series of patches are supplements to another series, please refer
> > to
> > https://patchwork.kernel.org/project/linux-mediatek/list/?series=515129c
> > , which add device link between the mtk-iommu consumer and the mtk-larb
> > devices. Without that series of patches, the mtk-mdp driver can work
> > well so far.
> > But with that series, it seems the device link only can be established
> > for the device which is registered as a platform driver. That's why
> > Eizan adds this series of patches to make all mdp components to be
> > registered as platform drivers.
>
> Hold on, so this means that if that iommu device-link series is merged,
> then the mtk-mdp driver breaks? I posted a PR for that iommu series, but
> I've just withdrawn that PR until this issue is clarified.
>
> Is it only mtk-mdp that is affected by this iommu device-link series, or
> others as well?
>

Like I said before, I think it's a mistake to introduce the component
framework in V4L2. This whole idea looks like a hack to me.

If we merge this, then we validate using the component framework
as a way to avoid fixing things properly.

Regards,
Ezequiel