On Mon, Aug 21, 2023 at 12:32 PM Andy Shevchenko
[off-list ref] wrote:

On Mon, Aug 21, 2023 at 12:19:39PM +0200, Benjamin Tissoires wrote:

quoted

On Mon, Aug 21, 2023 at 11:35 AM Andy Shevchenko
[off-list ref] wrote:

quoted

On Mon, Aug 21, 2023 at 12:34:30PM +0300, Andy Shevchenko wrote:

quoted

On Mon, Aug 21, 2023 at 10:51:04AM +0200, Benjamin Tissoires wrote:

quoted

On Aug 21 2023, Andy Shevchenko wrote:

...

quoted

quoted

quoted

quoted

Long story short, I'm not able to test it right now (and I got quite
some backlog as you can imagine). IIRC the code was fine, so I think we
can just take the series as is, and work on the quirks (if any) later.

Thank you!

The thing that might be broken is interrupts handling. If that works,
I'm pretty confident with the rest.

I.o.w. first 5 patches to test is already 98% of guarantee that everything
is fine.

Actually I applied you series locally, and applied Danny's patches on
top, and I could run your series in qemu with the cp2112 as USB
passthrough.

Everything is working fine, so I can take this one just now.

Thank you! I assume you have some IRQ (like GPIO button) to test with that.

Yeah, binding a test i2c-hid touchpad on top of hid forces you to use
GPIOs. Otherwise you are polling, and it's not allowed in i2c-hid
anymore IIRC :)

If no, it's easily to describe (in ACPI, see [1]) and use a wire to emulate
the button presses. In that case the /proc/interrupts should show the
different numbers.

Thanks, but again, the GPIO is tested just by checking if the touchpad
can send events when touched.

Now I need to update my CI to rely on danny's patches and a DSDT overwrite :)

Cheers,
Benjamin

[1]: https://github.com/westeri/meta-acpi/blob/master/recipes-bsp/acpi-tables/samples/edison/buttons.asli

--
With Best Regards,
Andy Shevchenko