linux-kernel - Re: [PATCH 00/17] Report power supply from hid-logitech-dj and others

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 23 Jan 2017 15:35:15 +0100
From:   Bastien Nocera <hadess@...ess.net>
To:     Benjamin Tissoires <benjamin.tissoires@...hat.com>,
        Jiri Kosina <jikos@...nel.org>,
        Peter Hutterer <peter.hutterer@...-t.net>,
        Nestor Lopez Casado <nlopezcasad@...itech.com>,
        Olivier Gay <ogay@...itech.com>,
        Simon Wood <simon@...gewell.org>
Cc:     linux-input@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 00/17] Report power supply from hid-logitech-dj and
 others

On Tue, 2017-01-17 at 15:35 +0100, Benjamin Tissoires wrote:
> Hey guys,
> 
> I tried to revive the in-kernel battery support for HID++ devices.
> I was thinking of doing just a few patches, but in the end I had to
> do
> cleanups and some more tweaks...
> 
> So, the final result is that now hid-logitech-hidpp should allow to
> handle any HID++ device, no matter which connection it uses.
> I was able to test it on some unifying devices, some USB and
> Bluetooth,
> but I'd like to get the confirmation from Simon that I did not break
> the G920.
> 
> Other than that, I implemented most features asked by Bastien during
> the
> last round:
> - have a sysfs file to indicate we are capable of power_supply
> - use ONLINE capability (not sure if I mess something up or if Gnome
> handles
>   it correctly)
> - report product, serial and manufacturer
> - report K750 battery info (not Lux, sorry)
> - report HID++ 1.0 battery info
> 
> 
<snip>

I've tested your patches with the kernel build that you kindly
provided. The output of "upower -d", here[1], shows both a K750
keyboard (the one with the solar charging) and a T650 touchpad (which
was plugged in to a separate power supply when testing).

Here's a jumble of notes:
- UPower expects the serial number to be available when the device is
created. This wasn't the case for the keyboard here, and we end up with
no serial number, even though the serial_number sysfs file is now
populated
- the K750's battery state doesn't seem to match that found by the
UPower code, eg. it's stuck in "Unknown" when upower could detect that
it is charging (it's sunny here). That might also be why the icon is
stuck at "battery-missing-symbolic".
- the model names of the batteries seem to have manufacturer
information prepended, eg. vendor: Logitech model_name: Logitech K750
I'd have expected to only have "K750" there.
- the touchpad is detected as a random "battery", but that's likely due
to the slightly dodgy code in UPower (look for "try to detect using the
device type" and cringe)
- the serial number is in a different format than in UPower:
  kernel: 4101-6f-63-fd-39
  UPower: 6F63FD39

I'll look at updating the UPower code, thanks.

UPower's "power_supply_ class code:
https://cgit.freedesktop.org/upower/tree/src/linux/up-device-supply.c
and its HID++ support:
https://cgit.freedesktop.org/upower/tree/src/linux/up-device-unifying.c

[1]: https://paste.fedoraproject.org/535093/51785481