On 11/11/20 8:05 PM, Rob Herring wrote:

quoted

after v5.10-rc1 the flexcan bindings were converted to yaml. This causes
several unneeded regressions on i.MX53 based boards and/or SoC specifying the
fsl,stop-mode property in their flexcan node.

This series fixes these problems by first updating the affected i.MX SoC dtsi
files and then fixing the flexcan yaml binding.

After I got the OK from the DT and fsl people, the plan is to upstream this via
net/master. If this is not an option, I'll send individual patches.

There's no need for dts changes to go into 5.10. dtbs_check is nowhere
near warning free yet. They should go via the soc tree. The schema
fixes do need to go in and I can take them. However, all the issues
still aren't fixed:

Right, this shows up with the dt_binding_check...

Documentation/devicetree/bindings/clock/imx5-clock.example.dt.yaml:
can@53fc8000: compatible: 'oneOf' conditional failed, one must be
fixed:
        ['fsl,imx53-flexcan', 'fsl,p1010-flexcan'] is too long
        Additional items are not allowed ('fsl,p1010-flexcan' was unexpected)
        'fsl,imx53-flexcan' is not one of ['fsl,imx7d-flexcan',
'fsl,imx6ul-flexcan', 'fsl,imx6sx-flexcan']
        'fsl,imx53-flexcan' is not one of ['fsl,ls1028ar1-flexcan']
        'fsl,imx25-flexcan' was expected
        'fsl,imx6q-flexcan' was expected
        'fsl,lx2160ar1-flexcan' was expected
        From schema:
/home/rob/proj/git/linux-dt/Documentation/devicetree/bindings/net/can/fsl,flexcan.yaml

Either the imx5-clock.yaml example needs changing or the schema does.
I'm guessing it's the former. 

Ack. I've created a patch to fix this:

http://lore.kernel.org/r/20201111213548.1621094-1-mkl@pengutronix.de (local)

I've applied the 2 schema patches here.

Thanks

Marc

-- 
Pengutronix e.K.                 | Marc Kleine-Budde           |
Embedded Linux                   | https://www.pengutronix.de  |
Vertretung West/Dortmund         | Phone: +49-231-2826-924     |
Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-5555 |