Message-ID: <e530d504-dfa3-4d39-b903-b2570bb8b014@suse.com>
Date: Fri, 23 Feb 2024 13:02:00 +0100
From: Matthias Brugger <mbrugger@...e.com>
To: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
 Ghennadi Procopciuc <ghennadi.procopciuc@....nxp.com>,
 Arnd Bergmann <arnd@...db.de>, Olof Johansson <olof@...om.net>,
 Chester Lin <chester62515@...il.com>, krzysztof.kozlowski+dt@...aro.org,
 Andreas Färber <afaerber@...e.de>, robh+dt@...nel.org
Cc: linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
 soc@...nel.org, NXP S32 Linux Team <s32@....com>,
 Ghennadi Procopciuc <ghennadi.procopciuc@....com>
Subject: Re: [PATCH 1/1] MAINTAINERS: Add maintainer for NXP S32G boards



On 22/02/2024 12:13, Krzysztof Kozlowski wrote:
> On 21/02/2024 18:00, Ghennadi Procopciuc wrote:
>> On 2/21/24 17:43, Krzysztof Kozlowski wrote:
>>> On 21/02/2024 16:19, Ghennadi Procopciuc wrote:
>>>> On 2/21/24 16:45, Krzysztof Kozlowski wrote:
>>>>> On 21/02/2024 15:42, Krzysztof Kozlowski wrote:
>>>>>> On 21/02/2024 13:01, Ghennadi Procopciuc wrote:
>>>>>>> From: Ghennadi Procopciuc <ghennadi.procopciuc@....com>
>>>>>>>
>>>>>>> Add myself as a maintainer of the NXP S32G DT files.
>>>>>>
>>>>>> No need for cover letters for single patches. OTOH, this commit msg is
>>>>>> empty...
>>>> Thank you, I can correct that.
>>>>
>>>>>> Plus your patch looks corrupted (wrong encoding): F??rber
>>>>
>>>> Indeed, it is due to 'Content-Type: text/plain; charset="us-ascii"'.
>>>> I can fix this as part of v2.
>>>>
>>>>>> BTW, did you contribute anything to the upstream Linux kernel? Do you
>>>>>> know the process? Downstream does not really matter.
>>>>>
>>>>> I found the answer:
>>>>>
>>>>> From: Ghennadi Procopciuc <ghennadi.procopciuc@....com>
>>>>> Signed-off-by: Ghennadi Procopciuc <ghennadi.procopciuc@....nxp.com>
>>>>>
>>>>> Does not look like. Please get some upstream work experience first.
>>>>>
>>>>> https://lore.kernel.org/all/?q=f%3Aghennadi.procopciuc%40oss.nxp.com
>>>>
>>>> Although I am new to upstream communities and may make mistakes, I am
>>>> eager to learn and improve.
>>>>
>>>> After leaving SuSe[0], the current maintainer of the NXP S32G DT files
>>>> became inactive[1]. As a result, this area is not currently being
>>>> maintained. This is the actual reason why I attempted to add myself as a
>>>> maintainer of that area. Although I acknowledge that I may not have
>>>> enough experience to become a maintainer, I am concerned that the NXP
>>>> S32G DT patch submission may be blocked for everyone due to the current
>>>> situation.
>>>
>>> I would be just happy to see first actual contributions or any sort of
>>> activity, like reviewing, before taking over something.
>>>
>>> You do not need to become maintainer to provide reviews. By reviewing
>>> patches you already reduce burden/work from the maintainers.
>>>
>>> Best regards,
>>> Krzysztof
>>>
>>
>> I fully understand and agree with your perspective on this matter and
>> accept the fact that I will not be a maintainer as initially intended.
>> Furthermore, I am very willing to participate in any reviews related to
>> S32G SoCs.
> 
> Just give it some time...
> 
>>
>> Patches, including those I have created ([0], [1]), will likely be
>> submitted for this area. This is because each enabled driver for S32G
>> SoCs is expected to have at least one node in the device tree. These
>> patches will undergo review and receive feedback. However, the upstream
>> process will come to a halt at this point since there are no maintainers
>> to apply and integrate them.
> 
> Indeed that's a problem.
> 
>>
>> I do not know how this situation should be handled, given my lack of
>> experience in upstreaming maintenance. The documentation for the Linux
>> kernel is insufficient in providing guidance [2] on how to handle
>> inactive maintainers and it is unclear who should take over their
>> responsibilities. This is likely not the first time this has happened in
>> the kernel's history.
>>
>> Could you please guide me on how these patches should be integrated into
>> a maintainer's  tree and by whom?
> 
> Chester left Suse, so maybe this also means less interest in maintaining
> it? Stepping up in such case, so your proposal, is reasonable, so in
> general I agree and thank you for trying to do something here.
> 
> Andreas or Matthias,
> Maybe you could change your R: into M: and take s32g patches?
> 
> If not, then I will ack this patch making Ghennadi the maintainer.
> 

Normal process would be that Arnd would contact Chester to see if he is not able 
to do the maintainer work any more. In that case maybe Arnd could take over.

I agree with you that some one should get maintainer for a SoC because he is 
involved in the Linux kernel community and not because he is working on 
downstream patches for the same silicon. Especially being paid by the company 
that produces the silicon can quickly get you into dificult situation. Think for 
example that NXP, which pays you, wants something to be added in the upstream 
kernel, but the code is not in a shape to be part of Linux kernel. That can 
generate conflict and for the upstream community it is important that the only 
criteria to accept something upstream is the fact that it is in a good shape for 
that, not any comercial roadmap by a company.

I'm not saying that Ghennadi won't be able to defend this, if it ever occurs. 
Basically because I don't have a good track record of him due to missing 
upstream collaboration.

I would prefer that Arnd (or Andreas) step up to take the maintainer role. I 
already have one and it's difficult for me to find the time to do it in an 
acceptable way.

Regards,
Matthias

Powered by blists - more mailing lists