Re: [PATCH] v1 pinctrl: mtk-eint: add eint new design for mt8196

[AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>]

Il 07/01/25 16:11, Linus Walleij ha scritto:
> On Thu, Dec 19, 2024 at 1:39 PM AngeloGioacchino Del Regno
> <angelogioacchino.delregno@collabora.com> wrote:
> 
>> Il 17/12/24 14:45, Linus Walleij ha scritto:
>>> Hi Chang,
>>>
>>> thanks for your patch!
>>>
>>> On Tue, Dec 3, 2024 at 2:13 PM chang hao <ot_chhao.chang@mediatek.com> wrote:
>>>
>>>> From: Chhao Chang <ot_chhao.chang@mediatek.com>
>>>>
>>>> Change 1: change EINT from 1 address to 5 addresses,
>>>> Eint number is stored on each base.
>>>> Change 2: Compatible with 1 address design
>>>>
>>>> Signed-off-by: Chhao Chang <ot_chhao.chang@mediatek.com>
>>>
>>> This patch looks good to me, as preparation for mt8196,
>>> but can one of the Mediatek experts please
>>> review it? If nothing happens I will just apply it I guess...
>>>
>>
>> Linus, that's something like the fourth time that he pushes variations of this
>> patch which do break all MediaTek SoCs in a way or another, leaving only MT8196
>> hopefully-functional.
> 
> That's unfortunate, and I see there is some annoyance building
> up.
> 
> The maintainers are here to protect the code and the change
> would have anyway just been reverted soon if it breaks stuff.
> 
> How can we get this contribution on a better path?
> 
> Chhao: maybe you can split your idea into smaller changes
> that can be reviewed and tested one by one?
> 

Splitting would be nice, but I'm not sure it's really doable... besides,
there's only one way to get this right: testing on older platforms!

Even just one done manually, like MT8173, would be fine - the others can be
tested in KernelCI as we do have many MTK SoCs in the lab.
Then it's just about following the review process and adding the suggested
changes - nothing in particular.

Besides, I'm sorry for the very (maybe too much!) "strong" reply :-)

Cheers,
Angelo

> Yours,
> Linus Walleij