Re: [PATCH v10 01/11] usb: typec: Add notifier functions

[Chaoyi Chen <chaoyi.chen@rock-chips.com>]

On 11/26/2025 7:44 PM, Greg Kroah-Hartman wrote:
> On Wed, Nov 26, 2025 at 11:42:43AM +0200, Heikki Krogerus wrote:
>> Wed, Nov 26, 2025 at 09:46:19AM +0800, Chaoyi Chen kirjoitti:
>>> On 11/25/2025 7:49 PM, Greg Kroah-Hartman wrote:
>>>>> +static umode_t typec_is_visible(struct kobject *kobj, struct attribute *attr, int n)
>>>>> +{
>>>>> +	if (is_typec_port(kobj_to_dev(kobj)->parent))
>>>>
>>>> Why look at the parent?  Doesn't the device have a type that should show
>>>> this?
>>>>
>>>> Otherwise, looks good to me.
>>>
>>> They have same deivce type "typec_altmode_dev_type".
>>> The parent device has a different device type to distinguish between
>>> port device and partner device.
>>
>> I was already wondering would it make sense to provide separate device
>> types for the port, and also plug, alternate modes, but I'm not sure
>> if that's the right thing to do.
>>
>> There is a plan to register an "altmode" also for the USB4 mode,
>> which of course is not an alternate mode. So USB4 will definitely need a
>> separate device type.
>>
>> So if we supply separate device types for the port, plug and partner
>> alternate modes, we need to supply separate device types for port, plug
>> and partner USB4 mode as well.
>>
>> We certainly can still do that, but I'm just not sure if it makes
>> sense?
>>
>> I'll prepare a new version for this and include a separate patch where
>> instead of defining separate device types for the port and plug
>> alternate modes I'll just supply helpers is_port_alternate_mode() and
>> is_plug_alternate_mode().
> 
> That feels like it would be better in the long run as it would be
> easier to "match" on the device type.
>

It make sense. But now can we first use the current "match" device type
operation and then modify them later?

> thanks,
> 
> greg k-h
> 
> 

-- 
Best, 
Chaoyi