Message-ID: <20210408203348.vafestu55hpw3zeq@bluff>
Date:   Thu, 8 Apr 2021 15:33:48 -0500
From:   Nishanth Menon <nm@...com>
To:     Arnd Bergmann <arnd@...nel.org>
CC:     SoC Team <soc@...nel.org>, Olof Johansson <olof@...om.net>,
        arm-soc <arm@...nel.org>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Tero Kristo <kristo@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] arm64: dts: TI K3 updates for v5.13

On 22:13-20210408, Arnd Bergmann wrote:
> On Thu, Apr 8, 2021 at 6:34 PM Nishanth Menon <nm@...com> wrote:
> > On 17:24-20210408, Arnd Bergmann wrote:
> > >
> > > ti/k3-am64-main.dtsi:376.40-385.4: Warning (simple_bus_reg): /bus@...00/interrupt-controller0: missing or empty reg/ranges property
> > > ti/k3-am64-main.dtsi:45.13-135.4: Warning (simple_bus_reg): /bus@...00/dmss: missing or empty reg/ranges property
> > > ti/k3-am64-mcu.dtsi:77.39-86.4: Warning (simple_bus_reg): /bus@...00/bus@...0000/interrupt-controller1: missing or empty reg/ranges property
> > >
> > > ti/k3-am642-evm.dt.yaml: bus@...0000: interrupt-controller1: {'type': 'object'} is not allowed for {'compatible': ['ti,sci-intr'], 'ti,intr-trigger-type': [[1]], 'interrupt-controller': True, 'interrupt-parent': [[1]], '#interrupt-cells': [[1]], 'ti,sci': [[4]], 'ti,sci-dev-id': [[5]], 'ti,interrupt-ranges': [[0, 104, 4]], 'phandle': [[5]]}
> > > ti/k3-am642-evm.dt.yaml: bus@...00: dmss: {'type': 'object'} is not allowed for {'compatible': ['simple-mfd'], '#address-cells': [[2]], '#size-cells': [[2]], 'dma-ranges': True, 'ranges':  'phandle': [[7]]}}
> > > ti/k3-am642-evm.dt.yaml: bus@...00: interrupt-controller0: {'type': 'object'} is not allowed for {'compatible': ['ti,sci-intr'], 'ti,intr-trigger-type': [[1]], 'interrupt-controller': True, 'interrupt-parent': [[1]], '#interrupt-cells': [[1]], 'ti,sci': [[4]], 'ti,sci-dev-id': [[3]], 'ti,interrupt-ranges': [[0, 32, 16]], 'phandle': [[15]]}
> > > ti/k3-am642-evm.dt.yaml: flash@0: 'cdns,read-delay', 'cdns,tchsh-ns', 'cdns,tsd2d-ns', 'cdns,tshsl-ns', 'cdns,tslch-ns' do not match any of the regexes: '^partition@', 'pinctrl-[0-9]+'
> > > ti/k3-am642-sk.dt.yaml: bus@...0000: interrupt-controller1: {'type': 'object'} is not allowed for {'compatible': ['ti,sci-intr'], 'ti,intr-trigger-type': [[1]], ...
> > > ti/k3-am642-sk.dt.yaml: bus@...00: dmss: {'type': 'object'} is not allowed for {'compatible': ['simple-mfd'], ...
> > > ti/k3-am642-sk.dt.yaml: bus@...00: interrupt-controller0: {'type': 'object'} is not allowed for {'compatible': ['ti,sci-intr'], 'ti,intr-trigger-type': [[1]], 'interrupt-controller': True, 'interrupt-parent': [[1]], '#interrupt-cells': [[1]], 'ti,sci': [[4]], 'ti,sci-dev-id': [[3]], 'ti,interrupt-ranges': [[0, 32, 16]], 'phandle': [[11]]}
> > > ti/k3-am642-sk.dt.yaml: flash@0: 'cdns,read-delay', 'cdns,tchsh-ns', 'cdns,tsd2d-ns', 'cdns,tshsl-ns', 'cdns,tslch-ns' do not match any of the regexes: '^partition@', 'pinctrl-[0-9]+'
> > > ti/k3-j7200-common-proc-board.dt.yaml: flash@0: 'cdns,read-delay', 'cdns,tchsh-ns', 'cdns,tsd2d-ns', 'cdns,tshsl-ns', 'cdns,tslch-ns' do not match any of the regexes: '^partition@', 'pinctrl-[0-9]+'
> > >
> > >
> > > merge commit: 4b8cf90637f295ff1cb1f4bdfde134dab1196296
> >
> >
> > We have a bunch of problems with interrupt controller definitions -
> > attempts in [1] [2] did'nt seem to come to any form of conclusion yet.
> >
> > Cdns is a convert from txt to yaml and is queued on Marks' tree[3]
> 
> Ok, thanks for taking a look
> 

Apologies on not flagging in my PR itself.

> > These issues pre-existed in existing definitions and/or convertion
> > from txt and we are already working to make sure we try and reach a
> > conclusion soon[4]. I am guessing from your note that it is high time we
> > put our foot down on txt bindings and clean yamls, I will ensure we
> > follow up on that.
> 
> I'm still trying to figure out how to best script it, my intention was to
> only list issues that got introduced in the latest patches rather than
> the existing problems.

I had done a bit of scripting around this already in my checker:
https://github.com/nmenon/kernel_patch_verify/blob/master/kernel_patch_verify#L212
Though it never works quite right all the time.. So, i do run explicit
checks afterwards to try and catch these.. and containerizing them
helped deployment aspects(developers having different schema package
versions) in control as well.

but yeah, a clean way of doing things still eludes me.

> 
> On the other hand, it would indeed help to address any issues that
> are going to come up in every newly added board first, and then
> eventually get to all the remaining warnings for issues with the
> yaml bindings.

yep, some of the troubles I had has relation to schema checks getting
stricter after I had queued things up, and then having to make a
choice :(. But internally, I am tracking those in our tracking system
and team has been given a headsup that we cleanup yamls before we add
any further nodes.

> 

The following also need to be converted to yaml (internal
JIRA tracking already done).

Documentation/devicetree/bindings/arm/keystone/ti,sci.txt
Documentation/devicetree/bindings/gpio/gpio-davinci.txt
Documentation/devicetree/bindings/i2c/i2c-omap.txt
Documentation/devicetree/bindings/input/touchscreen/ti-tsc-adc.txt
Documentation/devicetree/bindings/mailbox/omap-mailbox.txt
Documentation/devicetree/bindings/mailbox/ti,secure-proxy.txt
Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt
Documentation/devicetree/bindings/pci/pci-keystone.txt
Documentation/devicetree/bindings/phy/ti,phy-am654-serdes.txt
Documentation/devicetree/bindings/pwm/pwm-tiecap.txt
Documentation/devicetree/bindings/pwm/pwm-tiehrpwm.txt
Documentation/devicetree/bindings/serial/omap_serial.txt
Documentation/devicetree/bindings/spi/omap-spi.txt

-- 
Regards,
Nishanth Menon
Key (0xDDB5849D1736249D) / Fingerprint: F8A2 8693 54EB 8232 17A3  1A34 DDB5 849D 1736 249D

Powered by blists - more mailing lists