linux-kernel - Re: [PATCH v5 2/3] dt-bindings: media: imx274: Add optional input clock and supplies

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <f38bb328-b282-783b-3ac5-5441001d10b6@nvidia.com>
Date:   Thu, 3 Sep 2020 09:40:57 -0700
From:   Sowjanya Komatineni <skomatineni@...dia.com>
To:     Jacopo Mondi <jacopo@...ndi.org>
CC:     <thierry.reding@...il.com>, <jonathanh@...dia.com>,
        <sakari.ailus@....fi>, <hverkuil@...all.nl>,
        <jacopo+renesas@...ndi.org>, <luca@...aceresoli.net>,
        <leonl@...pardimaging.com>, <robh+dt@...nel.org>,
        <lgirdwood@...il.com>, <broonie@...nel.org>,
        <linux-media@...r.kernel.org>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v5 2/3] dt-bindings: media: imx274: Add optional input
 clock and supplies


On 9/3/20 9:35 AM, Jacopo Mondi wrote:
> Hi Sowjanya,
>
> On Thu, Sep 03, 2020 at 09:05:27AM -0700, Sowjanya Komatineni wrote:
>> On 9/3/20 5:55 AM, Jacopo Mondi wrote:
>>> Hello Sowjanya,
>>>
>>> On Tue, Sep 01, 2020 at 07:04:37PM -0700, Sowjanya Komatineni wrote:
>>>> This patch adds IMX274 optional external clock input and voltage
>>>> supplies to device tree bindings.
>>>>
>>>> Reviewed-by: Luca Ceresoli <luca@...aceresoli.net>
>>>> Signed-off-by: Sowjanya Komatineni <skomatineni@...dia.com>
>>>> ---
>>>>    .../devicetree/bindings/media/i2c/sony,imx274.yaml  | 21 +++++++++++++++++++++
>>>>    1 file changed, 21 insertions(+)
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/media/i2c/sony,imx274.yaml b/Documentation/devicetree/bindings/media/i2c/sony,imx274.yaml
>>>> index 7ae47a6..57e7176 100644
>>>> --- a/Documentation/devicetree/bindings/media/i2c/sony,imx274.yaml
>>>> +++ b/Documentation/devicetree/bindings/media/i2c/sony,imx274.yaml
>>>> @@ -25,6 +25,27 @@ properties:
>>>>      reset-gpios:
>>>>        maxItems: 1
>>>>
>>> I just sent an update to my json-schema conversion of this bindings
>>> document (not yet on patchwork, sorry) and Sakari pointed me to the
>>> fact in between my v2 and my v4 this patch from you went in:
>>> 4ea3273d24b ("dt-bindings: media: imx274: Add optional input clock and supplies")
>>>
>>> I should probably now update my bindings conversion patch, basically
>>> taking in what you've done here, but I would have one question.
>>>
>>>> +  clocks:
>>>> +    maxItems: 1
>>>> +    description: Reference to the sensor input clock
>>>> +
>>>> +  clock-names:
>>>> +    maxItems: 1
>>>> +    items:
>>>> +      - const: inck
>>>> +
>>>> +  vana-supply:
>>>> +    description:
>>>> +      Analog voltage supply, 2.8 volts
>>>> +
>>>> +  vdig-supply:
>>>> +    description:
>>>> +      Digital IO voltage supply, 1.8 volts
>>>> +
>>>> +  vddl-supply:
>>>> +    description:
>>>> +      Digital core voltage supply, 1.2 volts
>>> 4ea3273d24b introduced these regulators as VANA-supply, VDIG-supply
>>> and VDDL-supply (please note the upper-case names). This version uses
>>> lower-case ones instead. Is this intentional ? The driver currently
>>> does not parse any of these if I'm not mistaken, but as the bindings
>>> in textual form defines an ABI which should be preserved during the
>>> conversion to json-schema, should these be kept in upper-case ?
>>>
>>> Thanks
>>>      j
>> Yes, based on feedback lower case was recommended. So, changed to use
>> lower-case names.
>>
>> These properties were not used by driver currently and from my prior series
>> only dt-binding got merged as  no feedback was received on it for all prior
>> versions.
>>
>> So, should be ok to change to lower-case as there properties are introduced
>> now and driver update using these properties is under review
>>
> Well, I see that patch went in v5.9-rc1, so it will be part of v5.9.
>
> If the bindings update goes in in v5.10 (or whatever comes after v5.9)
> then we have a problem, as the DTB created for v5.9 won't work anymore
> on any later version, and that should not happen. Alternatively, a fix
> for the next -rc release could be fast-tracked, but you would
> need to synchronize with the dt maintainers for that and make a patch
> for the existing .txt bindings file.
>
> If the name change happens in the yaml file and one release is made
> with the old names, then we're stuck with those forever and ever, if I
> got the situation right.
>
> Please check with the dt and media maintainers, or they can comment
> here if they glance through these lines.
>
> Thanks
>    j

Hi Leon Luo,

I used upper case for regulator supply names in all prior 4 versions of 
IMX274 patch series as I see some other media i2c drivers doing it and 
dt-binding patch from v3 got merged in 5.9-rc1 which was using upper-case.

Later received feedback from Sakari requesting to use lower-case names 
so updated to use lower case name now in v5.

Not sure if we have timeline to squeeze in patch to change names to 
lower-case before they get into 5.10.

Can you please comment?

Sakari,

Can you also help understand why can't we keep upper case for regulator 
supplies?

I see some other media i2c drivers using upper case as well.

Thanks

Sowjanya




>>>> +
>>>>      port:
>>>>        type: object
>>>>        description: |
>>>> --
>>>> 2.7.4
>>>>