Re: [PATCH] Bluetooth: BT Driver: mediatek: add gpio pin to reset bt

[Krzysztof Kozlowski <krzk@kernel.org>]

On 25/06/2025 05:42, Chris Lu (陸稚泓) wrote:
> Hi Krzysztof,
> 
> Sorry for late reply. Some problem with Zhangchao's envrionment that he
> can't reply the mail. So I'll reply on his behalf.
> 
> Before submitting the next version, we'd like to discuss and confirm
> the questions you mentioned in v1 according to the suggestion from
> Bluetooth Maintainer Luiz.
> 
> On Wed, 2025-05-28 at 14:07 +0200, Krzysztof Kozlowski wrote:
>>
>> External email : Please do not click links or open attachments until
>> you have verified the sender or the content.
>>
>>
>> On 28/05/2025 09:03, Zhangchao Zhang wrote:
>>> +
>>> +static int btmtk_reset_by_gpio(struct hci_dev *hdev)
>>> +{
>>> +     struct btmtk_data *data = hci_get_priv(hdev);
>>> +     struct btmtk_reset_gpio *reset_gpio_data;
>>> +     struct device_node *node;
>>> +     int reset_gpio_number;
>>> +
>>> +     node = of_find_compatible_node(NULL, NULL, "mediatek,usb-
>>> bluetooth");
>>
>> There is no such compatible. Just git grep for it.
>>
>>> +     if (node) {
>>> +             reset_gpio_number = of_get_named_gpio(node, "reset-
>>> gpios", 0);
>>
>> Where is the ABI documented? Anyway, you should not pick GPIOs from
>> random devices.
>>
> In v1, we haven't prepared yaml file about dts setting yet. Does your
> question above referring to the same thing?
> 
> Zhangchao will add it in the next submission and run it through
> "dt_binding_check" according to the Kernel's document.
> 
No, because this does not explain what you send in v2 and why now you
sent again v1!

Nothing here makes sense, not only mediatek,usb-bluetooth which does not
exist and has 1% chances of acceptance anyway.

Best regards,
Krzysztof