[PATCH v5 00/19] Add USB on SAM9X60, SAMA7G5 and SAMA5D2 boards

[Marek Vasut]

On 12/22/22 18:32, Eugen.Hristev at microchip.com wrote:
> On 12/22/22 18:11, Marek Vasut wrote:
>> On 12/22/22 16:54, Sergiu.Moga at microchip.com wrote:
>>> On 22.12.2022 15:36, Marek Vasut wrote:
>>>> On 12/22/22 11:53, Sergiu Moga wrote:
>>>>> This series of patches is meant to add support for USB Mass Storage
>>>>> on SAM9X60, SAMA7G5 and SAMA5D2 boards and register ohci-at91 driver
>>>>> into
>>>>> Driver Model. In order for this to be achieved, the respective
>>>>> DT nodes have been added, the USB clock has been registered into CCF
>>>>> and the required defconfigs have been added to the boards' defconfig.
>>>>> What is more, in order for the VBUS to stay enabled, a
>>>>> `child_pre_probe`
>>>>> method has been added to overcome the DM core disabling it in
>>>>> `usb_scan_device`: when the generic `device_probe` method is called,
>>>>> the pinctrl is processed once again, undoing whatever changes have
>>>>> been made in our driver's probe method.
>>>>> In order to enable USB on SAMA7G5 the addition of RSTC and USB 2.0 PHY
>>>>> drivers were required.
>>>>
>>>> Please split the series into more manageable parts -- architecture and
>>>> DT bits, board, clock, usb .
>>>
>>> Hi, I am sorry but I believe that they are ordered correctly and they
>>> all should be part of this series as it ensures usb across all boards
>>> and most of them depend on each other.
>>> I think the order in which they have been sent makes the most sense:
>>> adding what is needed by sam9x60, followed by sama7g5, sama5d2 and
>>> finally enabling them all in the defconfigs.
>>>
>>> I am not sure that sending multiple series that are dependent on each
>>> other would make it easier.
>>
>> The USB patches go through USB tree, they seem independent from the rest
>> of the config changes and DTs. Please send them separately.
> 
> Hi Marek,
> 
> All previous revisions were assigned to me in patchwork. The split can
> be done at patchwork level if you feel they should go through different
> trees
> I can continue picking them up through at91 if you are fine, as this was
> the patchwork assignment. Otherwise feel free to review and take the USB
> patches through the usb tree (and assign those to you in patchwork ),
> and I will handle the DT/configs , at91 part.

I don't understand the opposition to splitting drivers and architecture 
core patches from one massive series.