Re: [PATCH v3 0/4] Add basic support for Mediatek MT8183 SoC

[Matthias Brugger]

On 21/07/18 09:28, Erin Lo wrote:
> On Fri, 2018-07-20 at 12:44 +0200, Matthias Brugger wrote:
>>
>> On 20/07/18 10:19, Erin Lo wrote:
>>> On Mon, 2018-07-16 at 11:28 +0200, Matthias Brugger wrote:
>>>> Hi Erin,
>>>>
>>>> On 17/05/18 08:22, Erin Lo wrote:
>>>>> MT8183 is a SoC based on 64bit ARMv8 architecture.
>>>>> It contains 4 CA53 and 4 CA73 cores.
>>>>> MT8183 share many HW IP with MT65xx series.
>>>>> This patchset was tested on MT8183 evaluation board, and boot to shell ok.
>>>>>
>>>>> This series contains document bindings, device tree including interrupt, uart.
>>>>>
>>>>> Change in v3:
>>>>> 1. Fill out GICC, GICH, GICV regions
>>>>> 2. Update Copyright to 2018
>>>>>
>>>>> Change in v2:
>>>>> 1. Split dt-bindings into different patches
>>>>> 2. Correct bindings for supported SoCs (mtk-uart.txt)
>>>>>
>>>>> Ben Ho (1):
>>>>>   arm64: dts: Add Mediatek SoC MT8183 and evaluation board dts and
>>>>>     Makefile
>>>>>
>>>>> Erin Lo (3):
>>>>>   dt-bindings: arm: Add bindings for Mediatek MT8183 SoC Platform
>>>>>   dt-bindings: mtk-sysirq: Add compatible for Mediatek MT8183
>>>>>   dt-bindings: serial: Add compatible for Mediatek MT8183
>>>>>
>>>>
>>>> I'm a bit reluctant to take this series, as it will only enable the EVB board to
>>>> boot into a serial console. Are you planning to add support for other devices of
>>>> this SoC?
>>>>
>>>> Apart please take into account that there is an issue with the dts file, as you
>>>> were told by the kbuild test robot.
>>>>
>>>> Regards,
>>>> Matthias
>>>>
>>>
>>> Hi, Matthias
>>> Sorry for missing this letter...since mail proxy server.
>>> We plan to add support all the devices of MT8183 in serious.
>>> We have implemented the clock and pinctrl driver for upstream and they
>>> are in internal review right now.
>>>
>>
>> Nice to hear that :)
>>
>>> About the dts issue... do you suggest me to send new patch right now or
>>> wait for clock and pinctrl driver ready then send them together?
>>>
>>
>> I would prefer that you send at least the clock controller together, so that we
>> don't have any dummy clocks in the basic device tree.
>>
>> Regards,
>> Matthias
>>
> 
> Got it! Next patch, we will send them with clock controller without
> dummy clocks in the basic device tree.
> By the way, the clock controller driver of MT8183 is a little bit
> different from former ICs, so we need more time to prepare them.
> We will send them to public as soon as possible.
> 

Thanks for the info.

Best regards,
Matthias