On Fri, 2021-01-08 at 12:42 +0100, Bastien Nocera wrote: > On Fri, 2021-01-08 at 12:23 +0100, Bastien Nocera wrote: > > Hey, > > > > I have recently bought a Logitech Marathon Mouse/Performance Plus > > M705, > > and I ended up with 2 power_supply class batteries for it when it > > was > > turned off already when using kernel 5.9.13. > > > > Using a newer kernel (5.11-rc2), the power_supply class device > > doesn't > > disappear either: > > <snip> > > I'll look into the upower problem soon. > > Filed as: > https://gitlab.freedesktop.org/upower/upower/-/issues/134 I actually don't think this is a bug. The docs for "online" mention: " Indicates if VBUS is present for the supply. When the supply is online, and the supply allows it, then it's possible to switch between online states (e.g. Fixed -> Programmable for a PD_PPS USB supply so voltage and current can be controlled). " So this isn't really something we can use to signal that a device is plugged in or not. And the docs for "present" say: " Reports whether a battery is present or not in the system. " Well, that would mean that the device is there, but no battery is present. eg. an XBox joypad without batteries but plugged in to the system via USB. Looks like a bug in the hidpp kernel code. This is what "solaar show" shows for the device: " 3: Marathon Mouse M705 (M-R0073) Device path : /dev/hidraw4 WPID : 406D Codename : M705 (M-R0073) Kind : mouse Protocol : HID++ 4.5 Polling rate : 8 ms (125Hz) Serial number: B14D65DA The power switch is located on the base. Battery: unknown (device is offline). " Cheers
