On 10/20/26 11:48, Mattijs Korpershoek wrote:
> Hi Fabrice,
>
> Thank you for the patch.
>
> On Thu, Oct 08, 2026 at 17:37, Fabrice Gasnier <[email protected]>
> wrote:
>
>> When starting the dwc2 core, it may not be seen in the correct mode,
>> when the dwc2_core_reset() gets called. This is seen on stm32mp1 upon
>> first call to reconfig_usbd(), a warning can be seen:
>> - "Waiting for GINTSTS_CURMODE_HOST timeout"
>>
>> When the core gets initialized, wait for the debounce filters, so the
>> correct mode gets read and polled later in dwc2_core_reset().
>>
>> The routines are added to dwc2_core.c, and inspired by the linux dwc2
>> driver.
>
> It looks like this patch is based on linux commit
> 2938fc63e0c2 ("usb: dwc2: Properly account for the force mode delays")
Hi Mattijs,
Yes Indeed, and also...
>
> Can we add it to the commit message? Or please add which linux kernel
> version this change is based on.
>
> If you wish, I can edit the commit message when applying. Please let me know.
... there are routines picked e.g. dwc2_iddig_filter_enabled() and helpers.
This has been inspired by Linux v6.18. It looks like still the same on latest
v7.3.
Yes please feel free to fix that up while applying!
Thanks,
Best regards,
Fabrice
>
>>
>> Fixes: c5d685b8993c ("usb: dwc2: Unify flush and reset logic with v4.20a
>> support")
>> Signed-off-by: Fabrice Gasnier <[email protected]>
>
> The code looks good, so, with the additional details requested in the
> commit message:
>
> Reviewed-by: Mattijs Korpershoek <[email protected]>
>
