* Re: iss: camera interface omap5
[not found] ` <EDF29A32-77BB-4346-BBCC-C12F2BB4745E@goldelico.com>
@ 2019-10-07 16:35 ` Tony Lindgren
0 siblings, 0 replies; only message in thread
From: Tony Lindgren @ 2019-10-07 16:35 UTC (permalink / raw)
To: H. Nikolaus Schaller
Cc: Linux-OMAP, Michael Allwright, Laurent Pinchart,
Mauro Carvalho Chehab, Sakari Ailus, linux-media
* H. Nikolaus Schaller <hns@goldelico.com> [191004 16:27]:
> Hi Tony,
>
> > Am 04.10.2019 um 18:11 schrieb Tony Lindgren <tony@atomide.com>:
> >
> > * H. Nikolaus Schaller <hns@goldelico.com> [191004 07:25]:
> >> Hi Tony,
> >> is there a similar node for omap5 comparable to this for omap4:
> >>
> >> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm/boot/dts/omap4.dtsi?h=v5.4-rc1#n176
>
> (I just noticed there seems to be a missing empty line between 216 and 217)
>
> >
> > Not sure if it's the same interface as on omap4, but at
> > least TRM has "8.1.3.1 ISS Instance Summary" at 0x52000000.
>
> Ok.
>
> IMHO there was a patch where someone got it working a while ago so it is likely the same:
>
> https://e2e.ti.com/support/interface/f/138/t/647460
>
> Now I wonder how it can be updated to current target-module style.
>
> Is it correct to do it similar to omap4 and replace
>
> /* No child device binding, driver in staging */
>
> by a child node that is compatible to "ti,omap4-iss".
> And there define a reg record like in the example on e2e?
> I.e. split into sysc registers for the target-module@52000000
> and driver specific registers in the child node?
Yeah something like that. Michael Alwright has a patch at [0]
with improvments for the iss staging driver to get it working.
Have you seen that one?
> Regarding clocks they likely have to stay with the target-module@52000000
> node, right?
I think it should be just something like this for
the target-module (on top of Michael's changes):
clocks = <&iss_clkctrl OMAP4_ISS_CLKCTRL 0>;
clock-names = "fck";
Then in the child node:
clocks = <&iss_clkctrl OMAP4_ISS_CLKCTRL 8>;
clock-names = "ctrlclk";
> And what about the dma nodes? Also keep in the target-module@52000000?
Those should be all in the child iss driver(s).
In general, it sounds like the iss is a private interconnect with
a control module for clocks etc, so doing it as multiple device
drivers probably makes sense.
Regards,
Tony
[0] https://github.com/allsey87/meta-builderbot/blob/master/recipes-kernel/linux/linux-stable-4.16/0008-omap4iss-Fix-multiple-bugs-and-use-device-tree.patch
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2019-10-07 16:35 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <0D08B352-F0DF-45A4-8279-51B07D366AD0@goldelico.com>
[not found] ` <20191004161117.GO5610@atomide.com>
[not found] ` <EDF29A32-77BB-4346-BBCC-C12F2BB4745E@goldelico.com>
2019-10-07 16:35 ` iss: camera interface omap5 Tony Lindgren
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).