Re: [PATCH v3 RESEND 2/2] drm/bridge: imx: Add i.MX93 parallel display format configuration support

[Krzysztof Kozlowski <krzk@kernel.org>]

On 16/08/2024 11:44, Liu Ying wrote:
> On 08/16/2024, Krzysztof Kozlowski wrote:
>> On 16/08/2024 10:09, Liu Ying wrote:
>>> NXP i.MX93 mediamix blk-ctrl contains one DISPLAY_MUX register which
>>> configures parallel display format by using the "PARALLEL_DISP_FORMAT"
>>> field. Add a DRM bridge driver to support the display format configuration.
>>>
>>> Signed-off-by: Liu Ying <victor.liu@nxp.com>
>>> ---
>>
>> ...
>>
>>> +
>>> +static int imx93_pdfc_bridge_probe(struct platform_device *pdev)
>>> +{
>>> +	struct device *dev = &pdev->dev;
>>> +	struct imx93_pdfc *pdfc;
>>> +	int ret;
>>> +
>>> +	pdfc = devm_kzalloc(dev, sizeof(*pdfc), GFP_KERNEL);
>>> +	if (!pdfc)
>>> +		return -ENOMEM;
>>> +
>>> +	pdfc->regmap = syscon_node_to_regmap(dev->of_node->parent);
>>> +	if (IS_ERR(pdfc->regmap)) {
>>> +		ret = PTR_ERR(pdfc->regmap);
>>> +		if (ret != -EPROBE_DEFER)
>>> +			DRM_DEV_ERROR(dev, "failed to get regmap: %d\n", ret);
>>> +		return ret;
>>
>> Nope, you just open-coded dev_err_probe. Syntax is - return
>> dev_err_probe(). if you need wrapper for DRM, add such.
> 
> Will use dev_err_probe().
> 
>>
>>> +	}
>>> +
>>> +	pdfc->next_bridge = devm_drm_of_get_bridge(dev, dev->of_node, 1, 0);
>>> +	if (IS_ERR(pdfc->next_bridge)) {
>>> +		ret = PTR_ERR(pdfc->next_bridge);
>>> +		if (ret != -EPROBE_DEFER)
>>> +			DRM_DEV_ERROR(dev, "failed to get next bridge: %d\n", ret);
>>> +		return ret;
>>
>> Ditto
> 
> Will use dev_err_probe().
> 
>>
>>
>>> +	}
>>> +
>>
>> ...
>>
>>> +MODULE_DESCRIPTION("NXP i.MX93 parallel display format configuration driver");
>>> +MODULE_AUTHOR("Liu Ying <victor.liu@nxp.com>");
>>> +MODULE_LICENSE("GPL v2");
>>> +MODULE_ALIAS("platform:" DRIVER_NAME);
>>
>> Which other driver needs this platform alias?
> 
> Quote include/linux/module.h:
> 
> "
> /* For userspace: you can also call me... */                                     
> #define MODULE_ALIAS(_alias) MODULE_INFO(alias, _alias)   
> "
> 
> Anything wrong with using MODULE_ALIAS() here?

Yes, it redundant. Do not answer with question to the question.

Are you adding random code that you cannot justify? It looks like this
if you cannot answer why do you need it.

Form letter:

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.

Best regards,
Krzysztof