[PATCH] Revert "usb: cdns3: use VBUS Valid to determine role for dr_mode OTG"

[Marek Vasut]

On 4/2/26 1:03 PM, Siddharth Vadapalli wrote:

Hi,

>>>> Adding in the USB custodian as this revert is intended for master and
>>>> v2026.04 I assume as the commit being reverted is also in master.
>>>> Thanks.
>>> This may be related to
>>>
>>> [PATCH] usb: cdns3: fix cdns3_bind() to avoid returning error and 
>>> probe parent
>>>
>>> ?
>>>
>>> Can you coordinate with Siddharth Vadapalli ?
>> Please refer to my response at:
>> https://lore.kernel.org/r/17a341aa-4711-4389-8ea1-a8169616baeb@ti.com/
>> on the patch quoted above
>> ([PATCH] usb: cdns3: fix cdns3_bind() to avoid returning error and 
>> probe parent)
>>
>> A common .probe as a stub will not work and probing the parent in 
>> cdns3_bind() doesn't seem to be acceptable. So reverting the commit 
>> appears to be the only option at the moment to fix non-USB Boot Modes.
> Please note that this revert patch is required to fix non-USB Boot Mode. 
> If you have a suggestion for an alternative, please let me know.
I replied to [PATCH] usb: cdns3: fix cdns3_bind() to avoid returning 
error and probe parent now, we can continue there.

If this patch is a stop-gap measure to avoid breakage on TI platform for 
v2026.04 release, then let's pick it up.

And I'm sorry for my slow response, I am buried in email.