[PATCH] dt-bindings: mailbox: Convert omap-mailbox.txt binding to YAML

[Suman Anna]

On 5/20/21 2:21 PM, Rob Herring wrote:
> On Thu, May 20, 2021 at 11:45 AM Suman Anna <s-anna at ti.com> wrote:
>>
>> On 5/19/21 7:08 PM, Rob Herring wrote:
>>> On Wed, May 19, 2021 at 10:14:14AM -0500, Suman Anna wrote:
>>>> On 5/18/21 12:20 PM, Suman Anna wrote:
>>>>> Convert the current OMAP Mailbox binding from text format to YAML
>>>>> format/DT schema, and delete the legacy text binding file.
>>>>>
>>>>> The new YAML binding conversion is an updated version compared to
>>>>> the original. The descriptions for certain properties have been
>>>>> improved to provide more clarity. Constraints are added to the
>>>>> properties 'ti,mbox-num-users', 'ti,mbox-num-fifos' and 'interrupts'.
>>>>> The 'ti,hwmods' is a legacy property and is retained only to reflect
>>>>> the existing usage on some older OMAP2 and OMAP3 platforms.
>>>>>
>>>>> All the existing examples have also been updated to reflect the
>>>>> latest dts nodes (ti,hwmods removed from OMAP4 and AM33xx examples,
>>>>> and interrupts value updated for AM65x SoCs).
>>>>>
>>>>> Signed-off-by: Suman Anna <s-anna at ti.com>
>>>>> ---
>>>>> Hi,
>>>>>
>>>>> This patch does fix a number of dtbs_check warnings seen around OMAP Mailbox
>>>>> nodes with the latest kernel. There are few left-over warnings when just
>>>>> this patch is used on v5.13-rc1 or next-20210518. I have posted a separate
>>>>> fix for a warning on TI K3 SoCs [1], and will be posting a separate cleanup
>>>>> series for OMAP2+ SoCs. The dts patches can be picked up independently
>>>>> of this patch.
>>>>
>>>> FYI, All the dtbs_check warnings will be gone with [1] and the OMAP2+ cleanup
>>>> series [2].
>>>
>>> Nice, though it is a moving target. :) Is that still true with the
>>> undocumented compatible checks that's been added?
>>
>> [1] is acked, so will definitely get picked up for the next merge window. Should
>> hit next sometime in the next couple of days.
>>
>> I didn't exactly understand your second comment, but no new compatibles were
>> added. The existing nodes are already in compliance with the constraints I added
>> (so that's strictly binding enforcements). These constraints are almost all on
>> legacy SoCs, so these do not pose any issues.
> 
> When you say 'All the dtbs_check warnings', you mean *all* or just all
> the mailbox ones. I read as the former, but I guess from the prior
> paragraph and the links you are talking about, it's just mailbox
> warnings.

Yeah, sorry about the confusion. It is indeed just the mailbox ones, whatever is
showing up in the build log for this patch on the devicetree patchworks.

https://gitlab.com/robherring/linux-dt-review/-/jobs/1274246349/artifacts/file/build-dtbs-1480365.log

> 
> Overall, this is what TI 32-bit warnings look like:
> 
> Processing (am3|am4|am5|dra|keystone|omap|compulab|logicpd|elpida|motorola-cpcap|twl|da|dm):
> warnings: 60358
> undocumented compat: 3195

Yeah, we have many derivative boards on each of the SoC families.

> 
> This comes from my CI job[1]. The warnings aren't deduplicated, so 1
> SoC warning expands to N board warnings. Only 3879 unique warnings. :)
> 
> The top warnings are:
> 
> arch/arm/boot/dts/(am3|am4|am5|dra|keystone|omap|compulab|logicpd|elpida|motorola-cpcap|twl|da|dm):60358:3879:3195
>     253  segment at 0: 'anyOf' conditional failed, one must be fixed:
>     253  segment at 0: $nodename:0: 'segment at 0' does not match '^bus(@[0-9a-f]+)?$'
>     140  segment at 200000: 'anyOf' conditional failed, one must be fixed:
>     140  segment at 200000: $nodename:0: 'segment at 200000' does not match
> '^bus(@[0-9a-f]+)?$'
>     128 0:0: /ocp/target-module at 56000000: failed to match any schema
> with compatible: ['ti,sysc-omap4', 'ti,sysc']
>     128 0:0: /ocp/target-module at 50000000: failed to match any schema
> with compatible: ['ti,sysc-omap2', 'ti,sysc']
>     128 0:0: /ocp/interconnect at 48000000/segment at 0/target-module at 60000:
> failed to match any schema with compatible: ['ti,sysc-omap2',
> 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 49032000: failed to match any
> schema with compatible: ['ti,sysc-omap2-timer', 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 48320000: failed to match any
> schema with compatible: ['ti,sysc-omap2', 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 48318000: failed to match any
> schema with compatible: ['ti,sysc-omap2-timer', 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 48304000: failed to match any
> schema with compatible: ['ti,sysc-omap2-timer', 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 480a0000: failed to match any
> schema with compatible: ['ti,sysc-omap2', 'ti,sysc']
>     114 0:0: /ocp at 68000000/target-module at 48056000: failed to match any
> schema with compatible: ['ti,sysc-omap2', 'ti,sysc']
>     114 0:0: /ocp at 68000000/l4 at 48000000/scm at 2000/scm_conf at 270/pbias_regulator at 2b0:
> failed to match any schema with compatible: ['ti,pbias-omap3',
> 'ti,pbias-omap']
>     114 0:0: /ocp at 68000000/l4 at 48000000/scm at 2000/pinmux at a00: failed to
> match any schema with compatible: ['ti,omap3-padconf',
> 'pinctrl-single']
>     114 0:0: /ocp at 68000000/l4 at 48000000/scm at 2000/pinmux at 30: failed to
> match any schema with compatible: ['ti,omap3-padconf',
> 'pinctrl-single']
>     112 0:0: /opp-table: failed to match any schema with compatible:
> ['operating-points-v2-ti-cpu']
>     111  segment at 100000: 'anyOf' conditional failed, one must be fixed:
>     111  segment at 100000: $nodename:0: 'segment at 100000' does not match
> '^bus(@[0-9a-f]+)?$'
>     108 0:0: /soc: failed to match any schema with compatible: ['ti,omap-infra']

Tony,

Any chance you can add a 'ti-sysc' binding and address the 'segment' warnings
for 5.14? Those alone should bring the warning numbers down drastically.

regards
Suman

> 
> Rob
> 
> [1] https://gitlab.com/robherring/linux-dt/-/jobs/1261951514
>