You are not logged in.
Output of lsusb:
...
Bus 005 Device 003: ID 046d:c068 Logitech, Inc. G500 Laser Mouse
...
So I have to add the PID it to the patch's list?
Offline
So I digged a little into it and removed the kernel patch completely from my system.
The missing symlinks in /dev/input/by-id are coming from systemd update not from the kernel update.
With everything up-to-date I ran
pacman -U /var/cache/pacman/pkg/systemd-242.84-2-x86_64.pkg.tar.xz /var/cache/pacman/pkg/systemd-libs-242.84-2-x86_64.pkg.tar.xz /var/cache/pacman/pkg/systemd-sysvcompat-242.84-2-x86_64.pkg.tar.xz
and thus downgrading systemd to the previous version. This resolved the issue for now (as a workaround) (with kernel 5.2.14 running)
*Edit:
Seems like this is related: https://github.com/systemd/systemd/issues/13557
*Edit2:
This seems like the original issue: https://github.com/systemd/systemd/issues/13518
Last edited by Helloagain (2019-09-14 14:47:59)
Offline
After seeing the issue on 5.2.10 (seems to works sometimes, sometimes not), I tried the downgrade for systemd too. It turns out, this is the real issue for me as well.
Now with running kernel 5.2.14, I do see the boot message about the failed connect to hidpp, but afterwards everything works out.
Offline
You can check if it is 13557 by retrieving the PKGBUILD source using git then adding the following to the backports array
then building and testing the result.
_backports=(067cc51f2e6eb94305556597f0254f115857ff49
48d26c90852c22ec94be961f5fbdcf462bb9a6e8
)
https://github.com/systemd/systemd/comm … 115857ff49
https://github.com/systemd/systemd/comm … 462bb9a6e8
Offline
You can check if it is 13557 by retrieving the PKGBUILD source using git then adding the following to the backports array
then building and testing the result._backports=(067cc51f2e6eb94305556597f0254f115857ff49 48d26c90852c22ec94be961f5fbdcf462bb9a6e8 )
https://github.com/systemd/systemd/comm … 115857ff49
https://github.com/systemd/systemd/comm … 462bb9a6e8
That doesn't seem to change anything for me.
It boots normally, but the /by-id/ entries are missing and button inputs are ignored, just as before.
Edit: Looks like that pr isn't fixing the issue for other people as well https://github.com/systemd/systemd/issues/13553
Last edited by F43nd1r (2019-09-15 01:58:48)
Offline
For me backporting the relevant commit into the PKGBUILD
_backports=(
'067cc51f2e6eb94305556597f0254f115857ff49' # Fixes missing input symlinks in /dev/input/by-id
)
works.
(The other commit is just a test case, so I did not care about it)
Offline
For some reason it started working again, several hours after upgrading to kernel 5.2.14, and even without rebooting.
Rebooted, now it stopped working again.
Offline
Ketchup901 wrote:For some reason it started working again, several hours after upgrading to kernel 5.2.14, and even without rebooting.
Rebooted, now it stopped working again.
What is your systemd version? Did you try downgrading systemd or applying the backport above?
systemd-version:
systemctl --version
Offline
I seem to have fixed this for myself by blacklisting the module hid_logitech_hidpp. Don't know what else I may lose without that, but my G502's extra buttons are back to working as they did before.
Offline
Systemd downgrade (242.84-2-arch) did not fix my problem with mouse, I am also having issue with my steam controller not emulating xbox inputs in games. Both devices work sometimes and not others just like my mouse, the random nature of this bug makes a lot of false positives in testing. You really have to reboot the system dozens of times to check if the fix you applied actually works or not, and even then you can't be sure.
I believe this is a deeper kernel issue but I'm not smart enough to work out what it could be.
Offline
Ketchup901 wrote:Ketchup901 wrote:For some reason it started working again, several hours after upgrading to kernel 5.2.14, and even without rebooting.
Rebooted, now it stopped working again.
What is your systemd version? Did you try downgrading systemd or applying the backport above?
systemd-version:
systemctl --version
It just fixed itself again now, without updating and without rebooting. Wasn't working this morning, left it on for the day, and it now works magically.
I haven't tried doing anything.
Systemd version 243.
Offline
It just fixed itself again now, without updating and without rebooting. Wasn't working this morning, left it on for the day, and it now works magically.
I haven't tried doing anything.Systemd version 243.
I did see this issue again as well, systemd 242, kernel 5.2.14. An arbitrary number of reboots usually fixes the issue. Still annoying as hell.
Offline
Issue seems fixed now with linux 5.3.1 and systemd 243.51
Offline
Yeah, sorry for this. This bug was introduced due to a mistake from my part. Linux 5.3 should include the fix.
And sorry from my late response, I only saw this now. Next time feel free to ping me by email, github or irc.
Offline
I'm still encountering an issue with this on linux 5.7.9-arch1-1 and systemd 245.6-8-arch, using a Logitech G500s mouse. Doing an ls /dev/input/by-id shows the keyboard input:
usb-Logitech_G500s_Laser_Gaming_Mouse_46701E2DB60009-event-if01
usb-Logitech_G500s_Laser_Gaming_Mouse_46701E2DB60009-event-mouse
usb-Logitech_G500s_Laser_Gaming_Mouse_46701E2DB60009-if01-event-kbd
usb-Logitech_G500s_Laser_Gaming_Mouse_46701E2DB60009-mouse
In windows I mapped button 5 (I think) to button 2, but pressing it in xev doesn't register as anything being pressed. I don't know if this is because the G500s has a different ID from the G500 and needs to be added to some list, but if so:
ID 046d:c24e Logitech, Inc. G500s Laser Gaming Mous
Offline
Hey all,
I beg you to excuse my raising a dead topic, but it's been over a year now and the problem is still present, albeit not in Arch, but in Debian 11, with kernel 5.10.0-9 and systemd 247.3-6. I am posting this here only for reference, because this is the most relevant thread I found on the topic and I hope it will help others.
My solution was to blacklist both hid_logitech_hidpp AND hid_logitech_dj in /etc/modprobe.d/logitech-g502-lightspeed.conf:
blacklist hid_logitech_hidpp
blacklist hid_logitech_dj
After a reboot my G502 Lightspeed now works full throttle!
Again, please accept my apologies and I do hope this information helps others with this otherwise spectacular mouse.
Offline
This really doesn't help all that much, all of these software versions are severely outdated in relation to an Arch system and the bug has apparently been fixed when this surfaced originally.
Closing this old topic.
Offline