[PATCH 09/11] usb: dwc2: Skip clock gating on Broadcom SoCs

[Stefan Wahren]

Am 05.07.24 um 23:14 schrieb Lukas Wunner:
> On Fri, Jul 05, 2024 at 12:16:14PM -0500, Jeremy Linton wrote:
>>> Am 05.07.24 um 17:03 schrieb Lukas Wunner:
>>>> Careful there, the patch vaguely says...
>>>>
>>>>      With that added and identified as "BCM2848",
>>>>      an id in use by other OSs for this device, the dw2
>>>>      controller on the BCM2711 will work.
>>>>
>>>> ...which sounds like they copy-pasted the BCM2848 id from somewhere else.
>>>> I would assume that BCM2848 is really a different SoC and not just
>>>> a different name for the BCM2835, but hopefully BroadCom folks will
>>>> be able to confirm or deny this (and thus the necessity of the quirk
>>>> on BCM2848 and not just on BCM2835).
>> This id comes from the edk2-platforms ACPI tables and is currently used by
>> both the rpi3 and rpi4, and AFAIK nothing else as the rpi5-dev work is
>> currently only exposing XHCI.
>>
>> The ID is strictly the USB controller not the SoC. Its a bit confusingly
>> named, but something we inherited from the much older windows/edk2 port,
>> where it appears that the peripheral HID's were just picked in numerical
>> order.
>>
>> [0] https://github.com/tianocore/edk2-platforms/blob/12f68d29abdc9d703f67bd743fdec23ebb1e966e/Platform/RaspberryPi/AcpiTables/GpuDevs.asl#L15
> So BCM2848, BCM2849, BCM2850 and so on are just made-up IDs
> for a Windows/EDK2 port that got cargo-culted into the kernel?
> Yikes!
>
> Has anyone checked whether they collide with actual Broadcom products?
Using public available information like this [1], I wasn't able to find
any collision.

[1] - https://github.com/anholt/linux/wiki/Devices-with-Videocore-graphics