Re: [PATCH v3 0/6] mips: dts: ralink: update system controller nodes and its consumers

[Thomas Bogendoerfer <tsbogend@alpha.franken.de>]

On Fri, Feb 21, 2025 at 05:40:34PM +0100, Sergio Paracuellos wrote:
> On Fri, Feb 21, 2025 at 4:37 PM Thomas Bogendoerfer
> <tsbogend@alpha.franken.de> wrote:
> >
> > On Fri, Feb 21, 2025 at 03:50:09PM +0100, Sergio Paracuellos wrote:
> > > Hi Thomas,
> > >
> > > On Fri, Feb 21, 2025 at 3:05 PM Thomas Bogendoerfer
> > > <tsbogend@alpha.franken.de> wrote:
> > > >
> > > > On Fri, Feb 21, 2025 at 11:48:34AM +0100, Sergio Paracuellos wrote:
> > > > > Hi Thomas,
> > > > >
> > > > > El El lun, 20 ene 2025 a las 10:21, Sergio Paracuellos <
> > > > > sergio.paracuellos@gmail.com> escribió:
> > > > >
> > > > > > Hi all!
> > > > > >
> > > > > > Ralinks SoCs have a system controller node which serves as clock and reset
> > > > > > providers for the rest of the world. This patch series introduces clock
> > > > > > definitions for these SoCs. The clocks are registered in the driver using
> > > > > > a bunch of arrays in specific order so these definitions represent the
> > > > > > assigned
> > > > > > identifier that is used when this happens so client nodes can easily use it
> > > > > > to specify the clock which they consume without the need of checking
> > > > > > driver code.
> > > > > >
> > > > > > DTS files which are currently on tree are not matching system controller
> > > > > > bindings. So all of them are updated to properly match them.
> > > > > >
> > > > > > I'd like this series to go through kernel mips git tree if possible.
> > > > > >
> > > > > > Thanks in advance for your time.
> > > > > >
> > > > > > Changes in v3:
> > > > > > - Address Krzysztof comments in v2 (Thanks!):
> > > > > >   + Drop reset include file since what it was defined there were hardware
> > > > > >     constants and no binding related indexes at all.
> > > > > >   + Update patches for not referring to this reset removed file.
> > > > >
> > > > >
> > > > > I was expecting this series going through the mips tree.
> > > >
> > > >   DTC     arch/mips/boot/dts/ralink/rt3883_eval.dtb
> > > > Error: /local/tbogendoerfer/korg/linux/arch/mips/boot/dts/ralink/rt3883.dtsi:2.1-9 syntax error
> > > > FATAL ERROR: Unable to parse input tree
> > >
> > > Weird, it looks like dtc is not happy with the "include" line with new
> > > definitions? Are you getting this only with rt3883? Since all the
> > > patches are almost the same and I compile tested this before sending..
> > > Something got corrupted? I don't have my laptop now to check but I
> > > will recheck again on monday.
> >
> > rt2880_eval.dts:/include/ "rt2880.dtsi"
> > rt3052_eval.dts:#include "rt3050.dtsi"
> > rt3883_eval.dts:/include/ "rt3883.dtsi"
> >
> > rt3052 works, rt2880 and rt3883 don't.
> >
> > changing the /include/ to #include makes them compile.
> 
> Mmmm...does this mean that this was broken before my patches? Since I
> have not touched the files that need the replacement.

no, without your patches everything compiles. I guess dtc (?) doesn't
allow #include in files, which were included via / include /. But that's
just guesswork

 So I probably
> checked in the openwrt tree and missed this totally. Sorry for that.
> How do you want to handle this? Should I send v4 including these
> replacements? Or do you prefer to handle them directly?

I'll fix the includes while applying.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]