Em Tue, 5 May 2020 22:22:20 +0200
Thierry Reding <thierry.reding@xxxxxxxxx> escreveu:
> On Tue, May 05, 2020 at 04:26:04PM +0200, Hans Verkuil wrote:
> > On 05/05/2020 16:15, Mauro Carvalho Chehab wrote:
> > > Em Tue, 5 May 2020 14:24:22 +0200
> > > Hans Verkuil <hverkuil@xxxxxxxxx> escreveu:
> > >
> > >> On 05/05/2020 14:17, Mauro Carvalho Chehab wrote:
> > >>> Em Tue, 5 May 2020 10:33:00 +0200
> > >>> Hans Verkuil <hverkuil@xxxxxxxxx> escreveu:
> > >>>
> > >>>> The following changes since commit e51759f56d314d28c25be7606b03791f048e44c7:
> > >>>>
> > >>>> media: v4l2-ctrls.h: clarify the p_def argument of v4l2_ctrl_new_std_compound (2020-04-29 12:04:58 +0200)
> > >>>>
> > >>>> are available in the Git repository at:
> > >>>>
> > >>>> git://linuxtv.org/hverkuil/media_tree.git tags/br-tegra
> > >>>>
> > >>>> for you to fetch changes up to 0bba0154ded02babccd59e21c6a27c5ad7d50faf:
> > >>>>
> > >>>> MAINTAINERS: Add Tegra Video driver section (2020-05-05 10:26:05 +0200)
> > >>>>
> > >>>> ----------------------------------------------------------------
> > >>>> Tag branch
> > >>>>
> > >>>> ----------------------------------------------------------------
> > >>>> Sowjanya Komatineni (2):
> > >>>> media: tegra: Add Tegra210 Video input driver
> > >>>> MAINTAINERS: Add Tegra Video driver section
> > >>>
> > >>>
> > >>> As pointed by checkpatch:
> > >>>
> > >>> WARNING: DT compatible string "nvidia,tegra210-csi" appears un-documented -- check ./Documentation/devicetree/bindings/
> > >>> #628: FILE: drivers/staging/media/tegra-video/csi.c:521:
> > >>> + { .compatible = "nvidia,tegra210-csi", .data = &tegra210_csi_soc },
> > >>>
> > >>> WARNING: DT compatible string "nvidia,tegra210-csi" appears un-documented -- check ./Documentation/devicetree/bindings/
> > >>> #3238: FILE: drivers/staging/media/tegra-video/video.c:106:
> > >>> + { .compatible = "nvidia,tegra210-csi", },
> > >>>
> > >>> and double-checked with:
> > >>>
> > >>> $ git grep "nvidia,tegra210-csi"
> > >>> drivers/staging/media/tegra-video/csi.c: { .compatible = "nvidia,tegra210-csi", .data = &tegra210_csi_soc },
> > >>> drivers/staging/media/tegra-video/video.c: { .compatible = "nvidia,tegra210-csi", },
> > >>>
> > >>> The DT specs for this driver are missing. I even checked at linux-next:
> > >>> there's no DT bindings for this.
> > >>
> > >> Sorry, I should have mentioned this in the cover letter: the binding file
> > >> is here: https://patchwork.linuxtv.org/patch/63576/
> > >>
> > >> But it will be merged via the tegra subsystem (Thierry Reding) since it is
> > >> part of the display bindings, not media bindings. Also, Thierry was planning
> > >> to convert it to yaml, so he offered to merge it instead.
> > >
> > > Ok. Please re-send the pull request (or just mark it as New on patchwork)
> > > once the bindings arrive linux-next.
> >
> > It's still marked as New in patchwork :-) (https://patchwork.linuxtv.org/patch/63596/)
> >
> > >
> > > Alternatively, I don't mind if those patches would also be merged via
> > > Tegra's tree, together with the bindings. On such case, feel free to
> > > add, for both patches:
> > >
> > > Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@xxxxxxxxxx>
>
> Mauro, do you mind if I turn that into an Acked-by? I think the ARM SoC
> maintainers may prefer that as being more of an explicit statement than
> Reviewed-by: that it's fine to take this through ARM SoC.
Sure, go ahead.
>
> >
> > Thierry, let us know what you want to do. It might not be a bad idea to merge
> > the whole series through your tree, as Mauro suggested. It keeps everything
> > together.
>
> I can do that. For ARM SoC I typically create a for-X.Y/dt-bindings
> branch where I collect device tree bindings patches that don't have
> a better place anywhere else, so I'll probably stick the DT binding
> update into that branch and then I can base a for-5.8/media branch
> on top of that to resolve the checkpatch warnings.
>
> Actually, once I have that I could even just send a PR to Mauro to
> take this through the media tree since the dependencies should all
> be resolved. But with an Acked-by I'm sure the ARM SoC maintainers
> won't mind picking this up.
Whatever works best. I don't mind to either pick it together with
the DT, or to merge it via some other tree with either my
acked-by or reviewed-by.
Thanks,
Mauro
