Re: [PATCH RFC v5 4/6] drm/sprd: add Unisoc's drm display controller driver

[Kevin Tang]

Hi Emil,
Sorry, I left for a long time because of other things. Now i'm back~

V6 fix access registers via readl/writel, check unsupported KMS
properties (format, rotation, blend mode, etc) by plane_check callback
ops
and remove always true checks for dpu core ops

Add changelog within the corresponding patch from now on, thanks for
your advice.

BR,
Kevin


Emil Velikov <emil.l.velikov@xxxxxxxxx> ä2020å3æ20æåä äå2:07åéï
>
> Hey Kevin,
>
> On Sun, 15 Mar 2020 at 23:19, Kevin Tang <kevin3.tang@xxxxxxxxx> wrote:
> >
> > Adds DPU(Display Processor Unit) support for the Unisoc's display subsystem.
> > It's support multi planes, scaler, rotation, PQ(Picture Quality) and more.
> >
> > Cc: Orson Zhai <orsonzhai@xxxxxxxxx>
> > Cc: Baolin Wang <baolin.wang@xxxxxxxxxx>
> > Cc: Chunyan Zhang <zhang.lyra@xxxxxxxxx>
> > Signed-off-by: Kevin Tang <kevin.tang@xxxxxxxxxx>
> > ---
> >  drivers/gpu/drm/sprd/Makefile       |   5 +-
> >  drivers/gpu/drm/sprd/dpu/Makefile   |   7 +
> >  drivers/gpu/drm/sprd/dpu/dpu_r2p0.c | 750 ++++++++++++++++++++++++++++++++++++
> >  drivers/gpu/drm/sprd/sprd_dpu.c     | 501 ++++++++++++++++++++++++
> >  drivers/gpu/drm/sprd/sprd_dpu.h     | 170 ++++++++
> >  drivers/gpu/drm/sprd/sprd_drm.c     |   1 +
> >  drivers/gpu/drm/sprd/sprd_drm.h     |   2 +
>
> Seems like v5 (patch at least, I did not look at the rest) does not
> address a handful of the feedback.
>  - Access registers via readl/writel, instead of current approach
>  - With ^^ you can drop the compiler workaround
>  - Checking for unsupported KMS properties (format, rotation, etc)
> should happen in the XXX_check callbacks
>  - Remove always true checks, around the (rather moot) abstraction layer
>
> For the future, please keep the revision/changelog within the
> corresponding patch. This way reviewers can see, from the onset, what
> is addressed.
> Look at `git log -- drivers/gpu/drm` for examples.
>
>
> -Emil