Processed: Re: Bug#1029584: Info received (Bug#1029584: Info received (Bug#1029584: Acknowledgement (linux-image-6.1.0-1-amd64: Mouswheel reports way too many events with Debian kernel config)))

2023-01-25 Thread Debian Bug Tracking System
Processing control commands:

> tag -1 upstream
Bug #1029584 [src:linux] linux-image-6.1.0-1-amd64: Mouswheel reports way too 
many events with Debian kernel config
Added tag(s) upstream.
> forwarded -1 https://gitlab.freedesktop.org/libinput/libinput/-/issues/852
Bug #1029584 [src:linux] linux-image-6.1.0-1-amd64: Mouswheel reports way too 
many events with Debian kernel config
Set Bug forwarded-to-address to 
'https://gitlab.freedesktop.org/libinput/libinput/-/issues/852'.

-- 
1029584: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1029584
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems



Bug#1029584: Info received (Bug#1029584: Info received (Bug#1029584: Acknowledgement (linux-image-6.1.0-1-amd64: Mouswheel reports way too many events with Debian kernel config)))

2023-01-25 Thread Diederik de Haas
Control: tag -1 upstream
Control: forwarded -1 
https://gitlab.freedesktop.org/libinput/libinput/-/issues/852

On 25 January 2023 17:52:10 CET, Tobias Klausmann  
wrote:
>Also reported to libinput upstream:
>
>https://gitlab.freedesktop.org/libinput/libinput/-/issues/852

Thanks! Updated bug metadata accoedingly



Bug#1029584: Info received (Bug#1029584: Info received (Bug#1029584: Acknowledgement (linux-image-6.1.0-1-amd64: Mouswheel reports way too many events with Debian kernel config)))

2023-01-25 Thread Tobias Klausmann
Hi! 

Also reported to libinput upstream:

https://gitlab.freedesktop.org/libinput/libinput/-/issues/852



Bug#1029584: Info received (Bug#1029584: Acknowledgement (linux-image-6.1.0-1-amd64: Mouswheel reports way too many events with Debian kernel config))

2023-01-25 Thread Tobias Klausmann
Hi! 

After some digging, I have found the kernel option/mmodule that causes
this: HID_LOGITECH_HIDPP. I presume my mouse (Logitech 903) is handled
by this module (hid_logitech_hidpp.ko), though it's not strictly needed.

Things that work to make the original problem go away:

- Build a kernel with CONFIG_HID_LOGITECH_HIDPP=n
- Blacklisting the module (add `modprobe.blacklist=hid_logitech_hidpp` to
  GRUB_CMDLINE_LINUX in  /etc/default/grub

Things that _do not_ work:

- `rmmod hid_logitech_hidpp` (no error reporting it's in use, I presume
  libinput just makes a decision on whether it's there when starting and
  never reverts, even if the module goes away).

My best guess is that not all of userspace is ready yet for this hi-res
stuff --- not that I found anything that worked really well with it
(tried: Vivaldi, a Chromium-based browser, assorted Qt and GTK apps).

Since there is an acceptable way out for me, this bug could be closed.
OTOH, this is very hard to debug breakage, so just relying on people
finding this bug is maybe a bit harsh. Your call.

Best,
Tobias