Re: [PATCH] HID: magicmouse: fix battery reporting for Bluetooth Magic Trackpad USB-C

["Jose Villaseñor Montfort" <pepemontfort@gmail.com>]

Hi Andrei,

Thanks for tackling this -- I hit the same bug on the Magic Trackpad 2
(USB-C, 004c:0324) over Bluetooth and posted a fix a few days earlier:

  [PATCH] HID: input: read battery capacity from its actual report offset
  https://lore.kernel.org/linux-input/20260702192139.114809-1-pepemontfort@gmail.com/

The two approaches fix the problem in different places:

  - Mine fixes the root cause in the generic query path.
    hidinput_query_battery_capacity() hard-codes buf[1] and ignores the
    battery field's offset within the report. I store the field offset at
    setup and read from it, so any device whose state-of-charge byte is
    not the first field -- like this trackpad, which prefixes it with a
    status byte -- is parsed correctly, with no per-device quirk.

  - Yours keeps the generic path as-is and routes this specific device
    around it via HID_BATTERY_QUIRK_AVOID_QUERY plus a probe-time fetch.

If the generic query is fixed, the AVOID_QUERY quirk shouldn't be needed
for this device. I'm happy to go with whichever the maintainers prefer;
if the generic fix is acceptable I'll respin it with a Fixes: tag and a
stable Cc, as you did. Your note about the connect-time report being
dropped under driver_input_lock is a good catch either way.

Thanks,
Jose