Re: [PATCH v5 1/7] media: platform: amd: Introduce amd isp4 capture driver

["Du, Bin" <bin.du@amd.com>]

Hi Krzysztof,

We’ve set up the clang build and kernel static analysis tools and found 
additional warnings aside from those you noted. We will resolve these 
issues and submit version 6, including fixes for other version 5 review 
comments, by end of this week.
Many thanks.

On 10/28/2025 5:10 PM, Du, Bin wrote:
> 
> 
> On 10/28/2025 5:06 PM, Krzysztof Kozlowski wrote:
>> On 28/10/2025 10:00, Du, Bin wrote:
>>> Thanks Krzysztof.
>>>
>>> On 10/28/2025 4:41 PM, Krzysztof Kozlowski wrote:
>>>> On 28/10/2025 09:30, Du, Bin wrote:
>>>>>>> +    }
>>>>>>> +};
>>>>>>> +
>>>>>>> +module_platform_driver(isp4_capture_drv);
>>>>>>> +
>>>>>>> +MODULE_ALIAS("platform:" ISP4_DRV_NAME);
>>>>>>
>>>>>> You should not need MODULE_ALIAS() in normal cases. If you need it,
>>>>>> usually it means your device ID table is wrong (e.g. misses either
>>>>>> entries or MODULE_DEVICE_TABLE()). MODULE_ALIAS() is not a substitute
>>>>>> for incomplete ID table.
>>>>>>
>>>>>
>>>>> Thanks for the clarification, yes, MODULE_ALIAS() does not apply in 
>>>>> our
>>>>> case and should be removed.
>>>>
>>>>
>>>> You don't have ID table, so probably this should be fixed, unless it is
>>>> instantiated by some other driver, e.g. as MFD. That's the only typical
>>>> use case for alias in the drivers.
>>>>
>>>
>>> Yes, our device is a MFD, so alias works for it, as alternative, we can
>>> add module device table, it should work either, could you please
>>> indicate your preferred option?
>>>
>>
>>
>> So it is spawned by drivers/gpu/drm/amd/amdgpu/isp_v4_1_0.c or
>> drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c ?
>>
> 
> yes, it's spawned by drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c
> 
>> If so, then your code is fine, that's correct use of MODULE_ALIAS.
>>
> 
> Thanks for the quick confirmation, so, we'll keep it.
> 
>> Best regards,
>> Krzysztof
> 

-- 
Regards,
Bin