Re: [PATCH] [RFC] usb: dwc2: Run the core connect in dwc2_hcd_init()

[Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>]

Hi,

On 4/13/2020 9:00 PM, Marek Vasut wrote:
> On 4/13/20 5:27 PM, Minas Harutyunyan wrote:
>> Hi,
> 
> Hi,
> 
>> On 4/13/2020 6:54 PM, Marek Vasut wrote:
>>> On 4/13/20 4:45 PM, Minas Harutyunyan wrote:
>>>> Hi,
>>>
>>> Hi,
>>>
>>>> On 4/13/2020 6:31 PM, Marek Vasut wrote:
>>>>> On STM32MP1 without the STUSB1600 USB-C chip (i.e. with plain USB-OTG
>>>>
>>>> Could you please provide GSNPSID, GHWCFG1-4 register values for STM32MP1.
>>>
>>> GSNPSID = 0x4f54330a
>>> GHWCFG1 = 0x00000000
>>> GHWCFG2 = 0x228fe150
>>> GHWCFG3 = 0x03b892e8
>>> GHWCFG4 = 0xe3f00030
>>>
>> STM32MP1 core doesn't support ACG feature: GHWCFG4_ACG_SUPPORTED = 0.
>> So, in your patch for STM32MP1 at least these lines are no meaning:
>>
>> +	/* Enable ACG feature in device mode,if supported */
>> +	dwc2_enable_acg(hsotg);
> 
> The comment says it enables the ACG only if it's supported.
> 
>> Looks like you just copied this chirp from dwc2_conn_id_status_change()
>> function.
> 
> Yes, that's correct, verbatim.
> 
>> dwc2_hsotg_core_init_disconnected() function is device mode function,
>> not sure that it's safe to call it for host mode in dwc2_hcd_init().
> 
> OK well, if I don't use this patch and I plug in a USB cable from the
> host PC, then the STM32MP1 isn't detected by the host PC. Do you have
> any idea what the correct fix would be then ?
> 

Please provide without your patch:
1. debugfs: params and hw_params
2. debug log
3. regdump before and after connecting device to host where issue seen.

Thanks,
Minas