error "client bug: event processing lagging behind by ..ms, your system is too slow"

but I do not use a compositor (I don't use GNOME either).

From libinput's POV the client/caller/compositor is whatever process uses libinput. In your case, that's Xorg via the xf86-input-libinput driver. libinput isn't aware of what the caller is doing otherwise.

This has been a long-standing source of bug reports, but I don't know how to fix it better than we have now. What libinput does is:

every 10 calls to libinput_dispatch() we store the current time (or 0 in the other 9 cases)
while processing events and if our dispatch timestamp is not zero:
- check the kernel event timestamp
- calculate delta between kernel event and dispatch time
- if the delta is more than 20ms in the past, complain

IOW we only print that message if more than 20ms passed between the kernel generating the event and libinput processing this event. Also, this only happens on the first (==oldest) event we read for each libinput_dispatch().

We do have some features that use original timestamps after a timeout (e.g the middle button emulation does this) but they all bypass this code, this code is really only ever triggered when reading data on the fd from the device.

So... , either there's a bug in libinput that makes processing really slow and causes that delay or there is a 20ms+ gap between the kernel event and libinput processing it. Which can cause other issues since we rely on the timestamps to be accurate.

FWIW, another possibility is that the kernel generates incorrect event timestamps. @doraskayo mentioned having found a bug like that some time ago, what became of that?

Thanks you both, Peter and Michel, for details and conjecture on what's possibly happening here.

I've been following gnome-shell 3506 for some time now without any real feedback. Recently built a new i9-12900k system and was surprised to see these errors continuing to show up regularly, several each day. This is with Xorg/nVidia.

OS: Arch Linux x86_64
Kernel: 5.15.19-1-lts
Shell: bash 5.1.16
DE: GNOME 41.3
WM: Mutter
CPU: 12th Gen Intel i9-12900K (24) @ 6.500GHz
GPU: Intel AlderLake-S GT1
GPU: NVIDIA GeForce GTX 1050 Ti
Memory: 4884MiB / 64155MiB

Do the timestamps generated by the kernel and those used by libinput come from the same clock?

FYI, a couple of days ago, I found an issue in the atd daemon partly coming from the fact that two different clocks (with a small time shift between them) were used.

libinput uses CLOCK_MONOTONIC and sets that through libevdev on the event node, so I'm hoping this shouldn't be the issue.

OK. While there is an issue between CLOCK_REALTIME and the time() function, there doesn't seem to be a different clock similar to CLOCK_MONOTONIC (with the same epoch). So the problem must be somewhere else.

Well, the clock_gettime(2) man page says that CLOCK_MONOTONIC "is affected by the incremental adjustments performed by adjtime(3) and NTP". How much could that be in practice? However, I think that this should be rather uncommon and I have the impression that error messages from libinput occur too often for that to be the cause.

yeah, we used to have these messages ratelimited to 5 and hour or so (iirc) and that was still too common. Without the ratelimiting, you'll see lots of logspam so I doubt that adjtime() is the cause here. Plus - wouldn't in that case the kernel events see the same adjustment? adjtime falling into the magic window between the device event and userspace reading it would be rather exceptionally unlucky.

Though the window between the event and the read is very small in terms of code, an event is an interrupt, so that various things may occur at that moment, IMHO. However, keyboard-related interrupts should be much less common than other kinds of interrupts (disk, network...), so yes, I think that one would be very unlucky.

Hi, glad this problem is getting looked at. I see these errors almost every day, multiple times per day:

Feb 23 11:42:57 /usr/lib/gdm-x-session[1180]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 29ms, your system is too slow
Feb 23 14:17:13 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 45ms, your system is too slow
Feb 23 14:29:41 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 35ms, your system is too slow
Feb 23 14:34:07 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 39ms, your system is too slow
Feb 23 14:34:13 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 45ms, your system is too slow
Feb 23 14:34:55 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 23ms, your system is too slow
Feb 24 06:23:30 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 31ms, your system is too slow
Feb 25 07:41:45 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 40ms, your system is too slow
Feb 25 07:52:02 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 30ms, your system is too slow
Feb 25 07:56:21 /usr/lib/gdm-x-session[1176]: (EE) event11 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 25ms, your system is too slow
Feb 25 15:41:57 /usr/lib/gdm-x-session[1208]: (EE) event10 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 22ms, your system is too slow
Feb 26 13:34:56 /usr/lib/gdm-x-session[1177]: (EE) event10 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 35ms, your system is too slow
Feb 26 14:48:28 /usr/lib/gdm-x-session[1177]: (EE) event10 - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 27ms, your system is too slow
Feb 28 18:02:56 /usr/lib/gdm-x-session[1143]: (EE) event4  - Microsoft Microsoft Ergonomic Keyboard: client bug: event processing lagging behind by 31ms, your system is too slow

Not sure what is indicated when the event number changes, if anything. Possibly random for different boots?

I add myself to the list.

A pretty complete log: (/var/log/Xorg.log.old)

[   145.873] Current Operating System: Linux inhost 5.16.8-arch1-1 #1 SMP PREEMPT Tue, 08 Feb 2022 21:21:08 +0000 x86_64
(...)
[   147.017] (--) NVIDIA(GPU-0): DFP-7: 165.0 MHz maximum pixel clock
[   147.017] (--) NVIDIA(GPU-0): 
[   207.878] (EE) event15 - USB Keyboard: client bug: event processing lagging behind by 18ms, your system is too slow
[   210.254] (EE) event15 - USB Keyboard: client bug: event processing lagging behind by 14ms, your system is too slow
[   222.696] (EE) event15 - USB Keyboard: client bug: event processing lagging behind by 36ms, your system is too slow
[   962.486] (EE) event15 - USB Keyboard: client bug: event processing lagging behind by 19ms, your system is too slow
[  1029.574] (EE) event15 - USB Keyboard: client bug: event processing lagging behind by 24ms, your system is too slow
[  1029.574] (EE) event15 - USB Keyboard: WARNING: log rate limit exceeded (5 msgs per 60min). Discarding future messages.
[  1318.623] (EE) 
[  1318.623] (EE) Backtrace:
[  1318.623] (EE) 0: /usr/lib/Xorg (xorg_backtrace+0x89) [0x560cd051d049]
[  1318.623] (EE) 1: /usr/lib/Xorg (0x560cd03cd000+0x15ae69) [0x560cd0527e69]
[  1318.624] (EE) 2: /usr/lib/libpthread.so.0 (0x7f960a81a000+0x13870) [0x7f960a82d870]
[  1318.624] (EE) 3: /usr/lib/Xorg (0x560cd03cd000+0x1e5378) [0x560cd05b2378]
[  1318.624] (EE) 4: /usr/lib/Xorg (0x560cd03cd000+0x8d375) [0x560cd045a375]
[  1318.624] (EE) 5: /usr/lib/Xorg (0x560cd03cd000+0x8d9bc) [0x560cd045a9bc]
[  1318.624] (EE) 6: /usr/lib/Xorg (0x560cd03cd000+0x8ebb3) [0x560cd045bbb3]
[  1318.624] (EE) 7: /usr/lib/Xorg (0x560cd03cd000+0x10cd4b) [0x560cd04d9d4b]
[  1318.624] (EE) 8: /usr/lib/Xorg (XkbHandleActions+0x1d4) [0x560cd05137b4]
[  1318.624] (EE) 9: /usr/lib/Xorg (0x560cd03cd000+0x1dcf12) [0x560cd05a9f12]
[  1318.624] (EE) 10: /usr/lib/Xorg (0x560cd03cd000+0x134c7e) [0x560cd0501c7e]
[  1318.624] (EE) 11: /usr/lib/Xorg (0x560cd03cd000+0x14e404) [0x560cd051b404]
[  1318.624] (EE) 12: /usr/lib/Xorg (WaitForSomething+0x2e2) [0x560cd051b712]
[  1318.624] (EE) 13: /usr/lib/Xorg (0x560cd03cd000+0x3c83b) [0x560cd040983b]
[  1318.624] (EE) 14: /usr/lib/libc.so.6 (__libc_start_main+0xd5) [0x7f960a670b25]
[  1318.624] (EE) 15: /usr/lib/Xorg (_start+0x2e) [0x560cd040a75e]
[  1318.624] (EE) 
[  1318.624] (EE) Segmentation fault at address 0x7ffd7289f000
[  1318.624] (EE) 
Fatal server error:
[  1318.624] (EE) Caught signal 11 (Segmentation fault). Server aborting
[  1318.624] (EE) 
[  1318.624] (EE) 
Please consult the The X.Org Foundation support 
	 at http://wiki.x.org
 for help. 
[  1318.624] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[  1318.624] (EE) 
[  1323.650] (WW) NVIDIA: Wait for channel idle timed out.
[  1323.692] (EE) Server terminated with error (1). Closing log file.

But most of time it's just something like

[   715.652] (EE) event3  - Gaming KB Gaming KB: client bug: event processing lagging behind by 26ms, your system is too slow

My OS now:

OS: Arch Linux x86_64  
Kernel: 5.17.1-arch1-1
Terminal shell: zsh 5.8.1 (x86_64-pc-linux-gnu)
`/bin/sh`: GNU bash, version 5.1.16(1)-release (x86_64-pc-linux-gnu)
WM: dwm-6.2
CPU: AMD Ryzen 7 1700 (16) @ 2.999GHz
GPU: NVIDIA GeForce GTX 1060 3GB 
Memory: 7962MiB

My systems hangs randomly from about 4 months ago.

Can move mouse
Music (audio) continue barely playing
Plasma doesn't react to "Ctrl BackSpc , BackSpc"
"Caps lock" or any other key with LED associated, doesn't change when I hit that keys
Only way to reset is with the "Magic keys" (SysReq) or remotely via ssh. These are some possible interesting outputs (including the "client bug ...":

tail -f  /var/log/Xorg.0.log | xargs -IL date +"%Y%m%d_%H%M%S:L"

(--) NVIDIA(GPU-0): Lenovo Group Limited LEN L193pC (DFP-0): connected
(--) NVIDIA(GPU-0): Lenovo Group Limited LEN L193pC (DFP-0): Internal TMDS
(--) NVIDIA(GPU-0): Lenovo Group Limited LEN L193pC (DFP-0): 330.0 MHz maximum pixel clock
(EE) event5  - ELECOM ELECOM TrackBall Mouse: client bug: event processing lagging behind by 26ms, your system is too slow
(EE) event6  - Dell Dell Smart Card Reader Keyboard: client bug: event processing lagging behind by 39ms, your system is too slow
(EE) client bug: timer event5 debounce short: scheduled expiry is in the past (-27ms), your system is too slow

While in remote terminal connected to my "hanged" PC via ssh:

Message from syslogd@localhost at Mar 30 13:36:29 ...
kernel:[15352.323550][    C3] watchdog: BUG: soft lockup - CPU#3 stuck for 26s! [gdbus:2048]
kernel:[79568.626222][    C7] watchdog: BUG: soft lockup - CPU#7 stuck for 27s! [kmail:2115]
kernel:[340461.130524][   C1] watchdog: BUG: soft lockup - CPU#1 stuck for 26s! [QSGRenderThread:2858]

dmesg -wH

[  +0.000195]  ? _nv037748rm+0xb3/0x150 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000193]  ? _nv037747rm+0x297/0x4e0 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000192]  ? _nv036106rm+0xbe/0x140 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000129]  ? _nv036107rm+0x42/0x70 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000128]  ? _nv000566rm+0x41/0x50 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000130]  ? _nv000723rm+0x73a/0xa80 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000171]  ? _nv000723rm+0x38/0xa80 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000172]  ? rm_ioctl+0x54/0xb0 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000170]  ? nvidia_ioctl+0x68a/0x890 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000102]  ? nvidia_frontend_unlocked_ioctl+0x34/0x50 [nvidia 34a6a4e24d96610e3f8f22e8426a6bbaff870838]
[  +0.000104]  ? __x64_sys_ioctl+0x7f/0xb0
[  +0.000002]  ? do_syscall_64+0x59/0x80
[  +0.000003]  ? syscall_exit_to_user_mode+0x18/0x40
[  +0.000001]  ? do_syscall_64+0x69/0x80
[  +0.000002]  ? syscall_exit_to_user_mode+0x18/0x40
[  +0.000000]  ? do_syscall_64+0x69/0x80
[  +0.000002]  ? do_syscall_64+0x69/0x80
[  +0.000001]  ? syscall_exit_to_user_mode+0x18/0x40
[  +0.000001]  ? do_syscall_64+0x69/0x80
[  +0.000002]  ? syscall_exit_to_user_mode+0x18/0x40
[  +0.000001]  ? do_syscall_64+0x69/0x80
[  +0.000001]  ? do_syscall_64+0x69/0x80
[  +0.000001]  ? entry_SYSCALL_64_after_hwframe+0x44/0xae
[  +0.000003]  </TASK>

dmesg -wH

[  +0,948649] [drm:drm_new_set_master [drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Failed to grab modeset ownership 
[  +0,000554] [drm:drm_new_set_master [drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Failed to grab modeset ownership 
[  +0,000167] [drm:drm_new_set_master [drm]] *ERROR* [nvidia-drm] [GPU ID 0x00000100] Failed to grab modeset ownership

I tried with another keyboard/mouse and even with each and all of this kernel parameters: pci=roteirq pci=noacpi acpi=off irqpoll noapic pci=biosirq

But always ends hanged randomly. I suspect is audio related, cause seems to be more frequent with audio, but not sure about that.

Operating System: openSUSE Tumbleweed 20220502 KDE Plasma Version: 5.24.4 KDE Frameworks Version: 5.93.0 Qt Version: 5.15.2 Host: HP Z1 Entry Tower G5 Kernel: 5.16.4-1-default Shell: bash 5.1.16 CPU: Intel i7-9700K (8) @ 4.900GHz GPU: Intel CoffeeLake-S GT2 [UHD Graphics 630] GPU: NVIDIA GeForce RTX 2060 Rev. A Memory: 5513MiB / 15766MiB

If I can help with more info, please, ask to me, cause I'm desperate.

watchdog: BUG: soft lockup - CPU#3 stuck for 26s!

your CPU is getting stuck. And unless I'm totally wrong, nvidia_ioctl implies the NVIDIA binary driver, at which point all bets are off.

Concerning the last two reports (from @charlubermensch and @rafael.linux.user), there is no reason to think that the crashes are related to the error messages: they do not occur at the same time, and in most cases (including my machines), there are no issues except these error messages.

I must add that those messages Message from syslogd@localhost at Mar 30 13:36:29 ... kernel:[15352.323550][ C3] watchdog: BUG: soft lockup - CPU#3 stuck for 26s! [gdbus:2048] kernel:[79568.626222][ C7] watchdog: BUG: soft lockup - CPU#7 stuck for 27s! [kmail:2115] kernel:[340461.130524][ C1] watchdog: BUG: soft lockup - CPU#1 stuck for 26s! [QSGRenderThread:2858]

appears after desktop is freezed (only mouse movemente) in another PC while logged thru a ssh terminal to the "hanged" PC. If fact, Xorg.bin it's showed consuming > 100% by nmon. If I kill Xorg.bin, then Plasmashell is over 100% CPU usage, but CAN'T KILL in any way Plasmashell (I tried ALL possible ways).

A last curiosity is that I use 2 monitors and when desktop "hangs" the mouse is visible and movable in MAIN monitor, but doesn't appear in secondary monitor.

I have noticed this behaviour for a couple months now running an up-to-date gentoo on two separate sets of hardware. It seems to always happen when I am pressing the volume up or down media keys on my keyboard (two different models). One system has an nvidia card, the other has intel graphics. The log I captured today from the intel machine was as follows:

[1044063.488] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 22ms, your system is too slow
[1045937.520] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 33ms, your system is too slow
[1047240.904] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 22ms, your system is too slow
[1048605.039] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 22ms, your system is too slow
[1051392.484] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 30ms, your system is too slow
[1051397.004] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 22ms, your system is too slow
[1051446.557] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 30ms, your system is too slow
[1122207.970] (EE) event2  - Microsoft Wired Keyboard 600: client bug: event pro
cessing lagging behind by 27ms, your system is too slow
[1122270.259] (EE) 
[1122270.259] (EE) Backtrace:
[1122270.259] (EE) 0: /usr/bin/X (xorg_backtrace+0x5b) [0x5635cf3a03bb]
[1122270.260] (EE) 1: /usr/bin/X (0x5635cf267000+0x13d025) [0x5635cf3a4025]
[1122270.260] (EE) 2: /lib64/libc.so.6 (0x7f5f2c1b2000+0x3d730) [0x7f5f2c1ef730]
[1122270.260] (EE) 3: /lib64/libc.so.6 (0x7f5f2c1b2000+0x1744db) [0x7f5f2c3264db]
[1122270.260] (EE) 4: /usr/bin/X (0x5635cf267000+0x7d557) [0x5635cf2e4557]
[1122270.260] (EE) 5: /usr/bin/X (0x5635cf267000+0x7ef71) [0x5635cf2e5f71]
[1122270.260] (EE) 6: /usr/bin/X (0x5635cf267000+0x7f46d) [0x5635cf2e646d]
[1122270.260] (EE) 7: /usr/bin/X (0x5635cf267000+0x80492) [0x5635cf2e7492]
[1122270.260] (EE) 8: /usr/bin/X (0x5635cf267000+0xf7fcd) [0x5635cf35efcd]
[1122270.260] (EE) 9: /usr/bin/X (XkbHandleActions+0x2a8) [0x5635cf392438]
[1122270.261] (EE) 10: /usr/bin/X (0x5635cf267000+0x1246e1) [0x5635cf38b6e1]
[1122270.261] (EE) 11: /usr/bin/X (0x5635cf267000+0x1248ca) [0x5635cf38b8ca]
[1122270.261] (EE) 12: /usr/bin/X (0x5635cf267000+0x1369e0) [0x5635cf39d9e0]
[1122270.261] (EE) 13: /usr/bin/X (WaitForSomething+0x258) [0x5635cf39dc68]
[1122270.261] (EE) 14: /usr/bin/X (0x5635cf267000+0x72c33) [0x5635cf2d9c33]
[1122270.261] (EE) 15: /usr/bin/X (0x5635cf267000+0x76b8b) [0x5635cf2ddb8b]
[1122270.261] (EE) 16: /lib64/libc.so.6 (0x7f5f2c1b2000+0x2931a) [0x7f5f2c1db31a]
[1122270.262] (EE) 17: /lib64/libc.so.6 (__libc_start_main+0x7c) [0x7f5f2c1db3cc]
[1122270.262] (EE) 18: /usr/bin/X (_start+0x21) [0x5635cf2a3071]
[1122270.262] (EE) 
[1122270.262] (EE) Segmentation fault at address 0x7ffc228d82d0
[1122270.262] (EE) 
Fatal server error:
[1122270.262] (EE) Caught signal 11 (Segmentation fault). Server aborting
[1122270.262] (EE) 
[1122270.262] (EE) 
Please consult the The X.Org Foundation support 
         at http://wiki.x.org
 for help. 
[1122270.262] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[1122270.262] (EE) 
[1122270.262] (II) AIGLX: Suspending AIGLX clients for VT switch
[1122270.323] (EE) Server terminated with error (1). Closing log file.

This machine is currently running xorg-server-21.1.3, and libinput-1.20.1 with xf86-input-libinput-1.2.0 on 5.15.41 kernel.

The client bug: event processing lagging behind by messages and the crash are probably unrelated: they do not occur at the same time, and other users get these messages but no crashes.

Here are some logs from a Debian 11 system running KDE Plasma Desktop 5.20 with Xorg 1.20.11.

[ 25385.383] (EE) event25 - Logitech Wireless Mouse MX Master 3: client bug: event processing lagging behind by 27ms, your system is too slow
[ 25385.383] (EE) event25 - Logitech Wireless Mouse MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60min). Discarding future messages.
[ 25394.637] (EE) client bug: timer event25 debounce short: scheduled expiry is in the past (-8ms), your system is too slow
[ 25421.103] (EE) client bug: timer event25 debounce: scheduled expiry is in the past (-13ms), your system is too slow
[ 25421.103] (EE) client bug: timer event25 debounce short: scheduled expiry is in the past (-26ms), your system is too slow
[ 25491.763] (EE) event20 - Dell Dell Universal Receiver: client bug: event processing lagging behind by 39ms, your system is too slow
[ 25616.756] (EE) event20 - Dell Dell Universal Receiver: client bug: event processing lagging behind by 36ms, your system is too slow
[ 25636.645] (EE) event20 - Dell Dell Universal Receiver: client bug: event processing lagging behind by 24ms, your system is too slow
[ 25636.645] (EE) event20 - Dell Dell Universal Receiver: WARNING: log rate limit exceeded (5 msgs per 60min). Discarding future messages.
[ 25726.431] (EE) client bug: timer event25 debounce short: scheduled expiry is in the past (-2ms), your system is too slow
[ 25789.590] (EE) client bug: timer event25 debounce: scheduled expiry is in the past (-12ms), your system is too slow
[ 25789.590] (EE) client bug: timer event25 debounce short: scheduled expiry is in the past (-25ms), your system is too slow

The messages come in waves and it doesn't seem to be load related - the system is pretty much idle all the time.

I noticed the debounce errors appear most frequently when holding a mouse button pressed for over a second and then releasing it.

The system otherwise runs perfectly. No missed mouse/keyboard events, no crashes, etc.

In my case, system not responding is gone, after uninstall Nvidia drivers!!! However, all messages about Mouse: client bug: event processing lagging behind by ...

continues appearing.

Is there an existing test program that could measure the libinput latency for some time, in isolation? That would make it a lot easier to isolate this (and then debug, with e.g. a perf trace running). This problem continually happens for me, so it shouldn't take too long to reproduce - if it's a bug in libinput or kernel.

best bet is probably starting libinput debug-events since that's about the most cut-down program that handles libinput events that you can get. You rule out virtually all other influences since all it does is print (to /dev/null if you really want to cut down on things).

Just make sure you supply the right options on startup to match whatever you have configured in your DE.

I think this maybe a problem with pipewire. I fixed this issue on Endeavour OS (Arch based) with

sudo pacman -S pulseaudio

which removes the conflicting pipewire-pulse package.

I started having this problem in in June in Ubuntu 22.04. Two different wireless keyboards and a USB wired keyboard, all had the same problem.

journalctl -b | grep lagging

Jun 10 18:40:13 desktop /usr/libexec/gdm-x-session[1752]: (EE) event20 - Microsoft Microsoft® 2.4GHz Transceiver v8.0: client bug: event processing lagging behind by 21ms, your system is too slow

Jun 10 19:05:36 desktop /usr/libexec/gdm-x-session[1752]: (EE) event2 - HP HP Wireless KB and MS: client bug: event processing lagging behind by 38ms, your system is too slow

Jun 10 19:20:32 desktop /usr/libexec/gdm-x-session[1752]: (EE) event18 - Microsoft Microsoft® SiderWinderTM X4 Keyboard: client bug: event processing lagging behind by 22ms, your system is too slow

Some things I tried stopped the error appearing but nothing I tried fixed the issue:

Switching to Wayland
Switching to KDE
Swithing to evdev for mouse and keyboard with .conf files
- /etc/X11/xorg.conf.d/
Editing grub GRUB_CMDLINE_LINUX_DEFAULT="iommu=soft"
Disabling USB auto suspend
- /sys/module/usbcore/parameters/autosuspend
Toggled every USB option in bios individually
Clean install of Fedora
Clean install of Endeavour OS

I've seen some people mention nvidia. I am running the Nvidia proprietary driver and using audio through HDMI plugged into the nvidia graphics card. Maybe a conflict with nvidia and pipewire?

All I do know is after 3 months I've finally fixed it by simply removing pipewire.

Yes, it could be due to pipewire. I removed pipewire from my machines in May 2022, and I can notice that the error no longer occurs (but I don't know since when).

I tried what you think: I changed to PipeWire, but nothing changed. However, each time I reinstall Nvidia propietary drivers, again randomly same issue. It's a headache.

Anyway, PipeWire is the right choice for audio, you can install it and leave it installed.

@rafael.linux.user I have the same problem, started around April 2021 with Manjaro, stopped for a while, came back with the latest January big updates. The system became frozen, couldn't move mouse or keyboard, especially when using Audacious.

Nothing I did helped, until mesa and vulkan drivers update yesterday. So I guess it's probably the driver problem.

There is a difference: System let me move mouse (but nothing more) in my primary monitor (it didn't reach my secondary monitor) and system never was absolutely freezed, cause I was able to reach it thru SSH and reboot from the console.

Currently, I have the nvidia driver installed again (remember I'm using Tumbleweed) and the crashes have been reduced a lot, but they haven't disappeared. In any case, I think it is indeed related to the audio and the Nvidia driver, probably something to do with system interruptions.

@rafael.linux.user This is the opposite. To solve the issue, you need to make sure that pipewire is not installed. But note again, that the crashes have nothing to do with this issue.

But I had issue with Pulseaudio, so I removed it and installed Pipewire trying to solve issue. Issue was not solved after that change, but Pipewire works much better than Pulseaudio, so I will not install Pulseaudio again.

This is still an issue with version 1.4.0. I started encountering it after upgrading from KDE5 to KDE6 and moving from X.org to wayland. I was using pipewire before and am using it still.

13:33:58.987600 kwin_wayland[1205]: kwin_libinput: Libinput: event20 - MX Anywhere 3 Mouse: client bug: event processing lagging behind by 33ms, your system is too slow
13:34:17.738486 kwin_wayland[1205]: kwin_libinput: Libinput: event20 - MX Anywhere 3 Mouse: client bug: event processing lagging behind by 34ms, your system is too slow
13:34:17.738516 kwin_wayland[1205]: kwin_libinput: Libinput: event20 - MX Anywhere 3 Mouse: WARNING: log rate limit exceeded (5 msgs per 60min). Discarding future messages.

I understand that this issue is hard to diagnose. I still hope this data point is of some use.

In any case, what should be easy to do is to downgrade the log level. This is clearly not an error and likely doesn't even merit being shown at the warning level. The info level seems reasonable.

Error still exists (Arch Linux, Plasma 6.2.3, kernel 6.11.8/6.12-rc7, xf86-input-libinput 1.5.0)

❯ journalctl -b -p2
lis 15 17:17:09 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 22ms, your system is too slow
lis 15 17:22:57 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 21ms, your system is too slow
lis 15 17:52:11 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 22ms, your system is too slow
lis 15 18:23:52 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 23ms, your system is too slow
lis 15 18:28:01 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 24ms, your system is too slow
lis 15 18:28:26 cachyos kwin_wayland[1008]: kwin_libinput: Libinput: event4  -  USB OPTICAL MOUSE: client bug: event processing lagging behind by 48ms, your system is too slow

error "client bug: event processing lagging behind by ..ms, your system is too slow"

Child items ...

Activity

journalctl -b | grep lagging

Admin message

Admin message

error "client bug: event processing lagging behind by ..ms, your system is too slow"

Activity

journalctl -b | grep lagging