On Wed, Aug 11, 2021 at 3:02 AM Chunfeng Yun (云春峰)
[off-list ref] wrote:

On Fri, 2021-08-06 at 14:43 -0600, Rob Herring wrote:

quoted

On Fri, Jul 30, 2021 at 04:49:54PM +0800, Chunfeng Yun wrote:

quoted

There are 4 USB controllers on MT8195, the controllers (IP1~IP3,
exclude IP0) have a wrong default SOF/ITP interval which is
calculated from the frame counter clock 24Mhz by default, but
in fact, the frame counter clock is 48Mhz, so we should set
the accurate interval according to 48Mhz. Here add a new compatible
for MT8195, it's also supported in driver. But the first controller
(IP0) has no such issue, we prefer to use generic compatible,
e.g. mt8192's compatible.

That only works until you find some 8195 bug common to all
instances.

It's also OK for IP0 to use mt8195's compatible, these setting value is
the same as IP0's default value, use mt8192's may avoid these dummy
setting.

I still don't understand. By use mt8192's compatible, that means you
have for IP0:

compatible = "mediatek,mt8192-xhci", "mediatek,mtk-xhci";

And for the rest:
compatible = "mediatek,mt8195-xhci", "mediatek,mtk-xhci";

If there's a 8195 quirk you need to work around, then you can't on
IP0. You need to be able to address quirks in the future without
changing the DTB. That is why we require SoC specific compatibles even
when IP blocks are 'the same'.

Rob