[SOLVED] Laggy, sticky mouse input

Salkay · 2020-08-30 15:28:56

A bit of a long shot, but anyone else notice a laggy mouse cursor in the last few days?

Everything is fine when using a mouse normally. However, if I don't touch the mouse for a few seconds, and then move it, it "sticks" in the stationary position for 0.5–1 second, then moves after that. The hand movement seems to have been "stored", so it might be a graphic artefact instead of an input problem.

I've checked both /var/log/Xorg.0.log and journalctl, but nothing appears when the lag occurs. I do see the following in /var/log/Xorg.0.log, but it's there from half an hour ago, and doesn't appear when I see this problem specifically.

[ 38377.568] (EE) event20 - MOSART Semi. 2.4G Wireless Mouse: client bug: event processing lagging behind by 18ms, your system is too slow
[ 38798.532] (EE) event20 - MOSART Semi. 2.4G Wireless Mouse: client bug: event processing lagging behind by 26ms, your system is too slow
[ 39191.241] (EE) event20 - MOSART Semi. 2.4G Wireless Mouse: client bug: event processing lagging behind by 28ms, your system is too slow
[ 39195.378] (EE) event20 - MOSART Semi. 2.4G Wireless Mouse: client bug: event processing lagging behind by 29ms, your system is too slow

FWIW I'm using KDE and Nvidia, all of which (along with the kernel) had updates in the last few days. I can see this issue in both linux and linux-lts kernels. Everything is up-to-date.

Last edited by Salkay (2020-12-11 12:19:32)

V1del · 2020-08-30 15:44:37

Sounds like USB autosuspend kicking in on the dongle, try disabling that: https://wiki.archlinux.org/index.php/Po … utosuspend

Salkay · 2020-08-30 15:48:02

Thanks V1del. I'll give that a try and report back in a few days. The problem appears to be intermittent and seems to be much better now (after a suspend-resume cycle), so I might not be able to replicate it immediately.

EDIT: actually that wiki page is a bit confusing to me. If there is no explicit udev rule, is autosuspend on or off?

Last edited by Salkay (2020-08-30 15:49:29)

V1del · 2020-08-30 15:57:21

It should be on by default if it isn't handled by anything else (... e.g. TLP or laptop-mode tools or so)

Salkay · 2020-09-08 01:20:26

Just a bit of an update on this.

I had a read of the kernel documentation linked by that wiki page. It looks like USB autosuspend was already disabled for the troublesome mouse (as per /sys/bus/usb/devices/*/power/control), but it was enabled for the USB hub (0424:2514 Microchip Technology, Inc. (formerly SMSC) USB 2.0 Hub).

I created /etc/udev/rules.d/50-usb_power_save.rules with the following content

ACTION=="add", SUBSYSTEM=="usb", TEST=="power/control", ATTR{power/control}="on"

restarted the system, and this disabled the USB hub. I haven't noticed the stickiness any more, so I'll tentatively mark this as solved. It's a bit hard to definitively know this, because the stickiness was intermittent.

Thanks again for the assistance V1del.

Salkay · 2020-09-08 09:26:04

And in a timely fashion, I'm still getting the jerkiness. It might be happening more after I resume from suspend, but it's hard to know. In this particular case, it only lasted about a minute before going away. FWIW it's a wireless mouse with a USB dongle.

Z2h-A6n · 2020-09-09 03:16:50

This may be a bit silly, but these sound similar to the symptoms a low battery in the wireless mice I've used. The fact that the stickiness is intermittent makes this seem a bit less likely to me, but I think it doesn't rule it out.

Salkay · 2020-09-09 03:24:39

@Z2h-A6n No, definitely a good question, but yes, I did check the batteries already.

seth · 2020-09-09 05:58:57

Looks like https://bbs.archlinux.org/viewtopic.php?id=186844 ("solved" by replacing the mouse)
The mouse (meaning the remote device) probably goes into power saving mode and re-pairs with its dongle on interaction and that pairing seems too slow.

If you click it, that will probably be delayed as well, but the moment the click takes action you can normally move the mouse?

Salkay · 2020-09-09 13:07:39

@seth yes I guess that's plausible, but I'm suppose there's no way to tell for sure if it's a software or hardware problem.

I've had the mouse for a few years, and it's only exhibited this behaviour now, so I don't think it goes into power-saving mode by design.

It's plausible however that this is a new problem with the mouse. It's intermittent, and rare, which doesn't necessarily mean it's not a hardware issue, but it's certainly unclear!

Having said that, if the initial mouse movement is "saved" and occurs a split second later, doesn't that suggest it's a software issue? If it's (hardware) auto-sleeping, then presumably that original movement would be lost. I would have thought if the mouse were power-saving and incapable of transmitting the signal, it would also be incapable of capturing the movement. Also, when I experience the problem, I trigger it quickly, i.e. every few seconds, so presumably power-saving wouldn't kick in that rapidly.

I haven't tried clicking to test yet.

Last edited by Salkay (2020-09-09 13:10:55)

seth · 2020-09-09 13:51:20

The theory is that the nouse wakes up (because it registers the motion - it cannot no track motion during PS, just not wifi around) and starts pairing with it's dongle. The motion is cached until the pairing is complete and then unrolled.

Afaiu it's the same model and shows the same symptoms across 6 years - so if there's a bug, it must have survived all that time.
If the behavior is new to you, you should cross-test w/ the LTS kernel to see whether it's indeed a regression.

Ropid · 2020-09-09 13:52:57

Maybe this problem is caused by swapping? I remember seeing something like what you describe, and I think that was caused by stuff having been swapped out. What's the memory usage when the problem happens? Do you see swap being in use in the output of "free -h"?

Last edited by Ropid (2020-09-09 13:54:09)

Salkay · 2020-09-09 23:02:41

seth wrote:

The theory is that the nouse wakes up (because it registers the motion - it cannot no track motion during PS, just not wifi around) and starts pairing with it's dongle. The motion is cached until the pairing is complete and then unrolled.

Ah okay, that makes sense.

seth wrote:

Afaiu it's the same model and shows the same symptoms across 6 years - so if there's a bug, it must have survived all that time.
If the behavior is new to you, you should cross-test w/ the LTS kernel to see whether it's indeed a regression.

Yes, I actually did try with the LTS kernel (as per my original post). Unfortunately it made no difference. There were a few recent (at the time) updates to KDE and Nvidia, so it might also be a regression there, but I didn't test those. (I find it very fiddly to revert those packages.)

Ropid wrote:

Maybe this problem is caused by swapping? I remember seeing something like what you describe, and I think that was caused by stuff having been swapped out. What's the memory usage when the problem happens? Do you see swap being in use in the output of "free -h"?

Thanks for the suggestion, but I don't use swap on this system. I have 32 GB RAM, and it's never close to being full, apart from the occasional memory leak from Firefox, KDE Plasma, Handbrake, etc.!

seth · 2020-09-10 06:23:51

xorg-server and xorg-input-* packages might be inolved, but I don't see how KDE or the nvidia blob can interere here.
You could install and try https://wiki.archlinux.org/index.php/Ge … pose_mouse to see whether it also happens on the console (and if so, rule out the GUI packages as culprit)

Salkay · 2020-09-10 06:52:18

Thanks @seth. It's a bit hard to troubleshoot in that fashion, because it's so intermittent. I also feel like it's becoming less frequent since my original post, occurring once every few days. Hence, even if the issue were still present on the console, it's likely that I'd have to wait a few days before I'd notice it. I'll keep it in mind if it becomes extremely frequent though.

Flapper · 2020-09-16 09:11:41

I'm getting the same sort of messages in Xorg.0.log, but I'm using USB keyboard and mouse.
I don't notice any mouse lag, but when the screenlocker has been active, the keyboard doesn't respond, until I click the mouse in another window then back.

[127259.117] (EE) event2  - Logitech USB Optical Mouse: client bug: event processing lagging behind by 17ms, your system is too slow
[135572.419] (EE) event3  - Logitech USB Keyboard: client bug: event processing lagging behind by 23ms, your system is too slow
[135766.312] (EE) event3  - Logitech USB Keyboard: client bug: event processing lagging behind by 23ms, your system is too slow
[135810.023] (EE) event3  - Logitech USB Keyboard: client bug: event processing lagging behind by 34ms, your system is too slow

Salkay · 2020-09-16 09:22:45

Flapper wrote:

I'm getting the same sort of messages in Xorg.0.log, but I'm using USB keyboard and mouse.

I suspect that these messages are unrelated, because I get them even when the mouse is behaving properly.

Flapper wrote:

I don't notice any mouse lag, but when the screenlocker has been active, the keyboard doesn't respond, until I click the mouse in another window then back.

Hmmm, that sounds pretty weird, but quite different to what I'm getting.

Subject-17 · 2020-10-07 01:49:30

This has absolutely been happening to me too, glad I'm not alone. It started a bit over a month ago.

I'm running a threadripper 1950X with a 2080ti, "raw" i3-wm (via startx and a barren .xinitrc, no compositing, no power saving or screen locking etc save for i3status), 1440x3440@144hz screen, and a wired Logitech Logitech G903 Wired @ 1ms (1000hz) poll rate.

I tried changing mouse poll to lower settings, but 1. it didn't help and 2. I don't want to use a lower setting.

I've just added `drm_kms_helper.poll=0` to my boot entry, and while I haven't gotten the absurd-several-seconds-long lag since rebooting, moving the mouse in circles can spike firefox from <10% on a single CPU to 70%.

Right after rebooting, running nothing other than termite and i3 status, moving the mouse in quick circles can spike Xorg from ~2% to 50% of 1 CPU.

No Errors in Xorg.log other than the "mouse slow"/keyboard slow information. Nothing plugged in other than a yubi key, mouse, and keyboard. Using `nvidia` drivers

xinput:
```

⎡ Virtual core pointer id=2 [master pointer (3)]
⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)]
⎜ ↳ Razer Razer BlackWidow Tournament Edition Chroma Consumer Control id=10 [slave pointer (2)]
⎜ ↳ Razer Razer BlackWidow Tournament Edition Chroma id=12 [slave pointer (2)]
⎜ ↳ Logitech Logitech G903 Wired/Wireless Gaming Mouse id=13 [slave pointer (2)]
⎜ ↳ Logitech G903 Wired/Wireless Gaming Mouse id=14 [slave pointer (2)]
⎣ Virtual core keyboard    id=3 [master keyboard (2)]
↳ Virtual core XTEST keyboard    id=5 [slave keyboard (3)]
↳ Power Button id=6 [slave keyboard (3)]
↳ Power Button id=7 [slave keyboard (3)]
↳ Razer Razer BlackWidow Tournament Edition Chroma id=8 [slave keyboard (3)]
↳ Razer Razer BlackWidow Tournament Edition Chroma Keyboard id=9 [slave keyboard (3)]
↳ Razer Razer BlackWidow Tournament Edition Chroma System Control id=11 [slave keyboard (3)]
↳ Yubico Yubikey 4 OTP+U2F+CCID    id=15 [slave keyboard (3)]
↳ Razer Razer BlackWidow Tournament Edition Chroma Consumer Control id=16 [slave keyboard (3)]
↳ Logitech G903 Wired/Wireless Gaming Mouse id=17 [slave keyboard (3)]
```

Last edited by Subject-17 (2020-10-07 01:51:16)

Salkay · 2020-10-07 03:52:34

FWIW I think my issues have gone. I haven't noticed them for the last 2–3 weeks. I'm not sure if it was some package update or possibly even radio interference?

oberstal · 2020-10-11 01:58:52

Same thing here, it's one of the updates in the last few weeks.

At first I thought it was battery, but mouse is charged to full. I dual-boot into Windows 10 on this machine, and the issues do not happen there.

Kernel 5.8.14-arch1-1
CPU: Ryzen 9 3900x
GPU: RTX 2080 Super
Mem: 64GB, only 4GB used as I type this.

Mouse: Logitech MX Master 3 using the USB dongle instead of bluetooth.

seth · 2020-10-11 06:33:27

Subject-17 & oberstal use the same mouse

@oberstal, did you try adding "drm_kms_helper.poll=0" to the kernel?

@Subject-17, did the cpu load only appear w/ "drm_kms_helper.poll=0"?
https://wiki.archlinux.org/index.php/ke … _and_dmesg

Moving the mouse above or among clients will cause responses in the clients or the WM (in doubt repaints, banging the Xorg server) which can easily rise the CPU load.
But: this has nothing to do w/ the mouse behavior at all. t's like saying "Swinging my bat mysteriously causes people to die" - it's that the bat hits them in the head, and the resulting brain trauma kills them. Just swinging a bat itself is not lethal at all.

Zod · 2020-10-11 12:44:16

As an aside, probably has nothing to do with what you guys are discussing BUT!

Logitech M325

I notice the stickiness only with Firefox.

With Chromium or Opera it is not there.

If you search "firefox mouse wheel lag" it seems there is some history with this issue.

Anyway, I'm using chromium now.

j1simon · 2020-10-17 08:36:21

Operating System: Arch Linux
KDE Plasma Version: 5.20.0
KDE Frameworks Version: 5.75.0
Qt Version: 5.15.1
X.Org version: 1.20.9 
Kernel Version: 5.8.14-zen1-1-zen
OS Type: 64-bit
Processors: 64 × AMD Ryzen Threadripper 3970X 32-Core Processor
Memory: 62.7 GiB of RAM
Graphics Processor: Radeon RX 580 Series
Mouse: Logitech MX Master 3
Keyboard: Logitech MX Keys

Same problem:

...
[   353.753] (EE) event2  - Logitech MX Keys: client bug: event processing lagging behind by 27ms, your system is too slow
[   415.627] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 28ms, your system is too slow
[   417.053] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 26ms, your system is too slow
[   417.318] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 20ms, your system is too slow
[   417.875] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 32ms, your system is too slow
[   454.034] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-5ms), your system is too slow
[   454.034] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-18ms), your system is too slow
[   458.357] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[   458.357] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[   462.618] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-12ms), your system is too slow
[   504.494] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-11ms), your system is too slow
[  1873.290] (EE) event2  - Logitech MX Keys: client bug: event processing lagging behind by 28ms, your system is too slow
[  2469.880] (EE) event2  - Logitech MX Keys: client bug: event processing lagging behind by 26ms, your system is too slow
[  2590.519] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-0ms), your system is too slow
[  2590.519] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-13ms), your system is too slow
[  3086.530] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 15ms, your system is too slow
[  3098.946] (II) config/udev: Adding input device Logitech USB Headset Logitech USB Headset (/dev/input/event14)
[  3098.946] (**) Logitech USB Headset Logitech USB Headset: Applying InputClass "evdev keyboard catchall"
[  3098.946] (**) Logitech USB Headset Logitech USB Headset: Applying InputClass "libinput keyboard catchall"
[  3098.946] (**) Logitech USB Headset Logitech USB Headset: Applying InputClass "system-keyboard"
[  3098.946] (II) Using input driver 'libinput' for 'Logitech USB Headset Logitech USB Headset'
[  3098.978] (II) systemd-logind: got fd for /dev/input/event14 13:78 fd 81 paused 0
[  3098.978] (**) Logitech USB Headset Logitech USB Headset: always reports core events
[  3098.978] (**) Option "Device" "/dev/input/event14"
[  3098.978] (**) Option "_source" "server/udev"
[  3098.980] (II) event14 - Logitech USB Headset Logitech USB Headset: is tagged by udev as: Keyboard
[  3098.980] (II) event14 - Logitech USB Headset Logitech USB Headset: device is a keyboard
[  3098.980] (II) event14 - Logitech USB Headset Logitech USB Headset: device removed
[  3098.980] (**) Option "config_info" "udev:/sys/devices/pci0000:40/0000:40:01.1/0000:41:00.0/0000:42:08.0/0000:45:00.3/usb9/9-6/9-6:1.3/0003:046D:0A8F.000B/input/input26/event14"
[  3098.980] (II) XINPUT: Adding extended input device "Logitech USB Headset Logitech USB Headset" (type: KEYBOARD, id 16)
[  3098.980] (**) Option "xkb_layout" "es"
[  3098.982] (II) event14 - Logitech USB Headset Logitech USB Headset: is tagged by udev as: Keyboard
[  3098.982] (II) event14 - Logitech USB Headset Logitech USB Headset: device is a keyboard
[  3237.835] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-4ms), your system is too slow
[  3237.835] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-17ms), your system is too slow
[  3515.825] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[  3516.381] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 24ms, your system is too slow
[  3516.381] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-12ms), your system is too slow
[  3646.245] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-8ms), your system is too slow
[  3646.245] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-21ms), your system is too slow
[  3744.154] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 20ms, your system is too slow
[  3744.405] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-7ms), your system is too slow
[  3744.405] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-20ms), your system is too slow
[  4123.110] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[  4123.296] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-3ms), your system is too slow
[  4123.296] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-16ms), your system is too slow
[  4207.756] (II) config/udev: removing device Logitech USB Headset Logitech USB Headset
[  4207.756] (**) Option "fd" "81"
[  4207.757] (II) event14 - Logitech USB Headset Logitech USB Headset: device removed
[  4207.757] (II) UnloadModule: "libinput"
[  4207.757] (II) systemd-logind: releasing fd for 13:78
[ 19530.913] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 17ms, your system is too slow
[ 19531.100] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 33ms, your system is too slow
[ 19531.768] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 19970.777] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 13ms, your system is too slow
[ 19970.881] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 15ms, your system is too slow
[ 19978.477] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-11ms), your system is too slow
[ 20024.468] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 20024.944] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 20025.138] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 11ms, your system is too slow
[ 20025.138] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 20029.682] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-9ms), your system is too slow
[ 20031.809] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 17ms, your system is too slow
[ 20032.245] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 11ms, your system is too slow
[ 20032.996] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 13ms, your system is too slow
[ 20033.616] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 20053.807] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 18ms, your system is too slow
[ 20053.807] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 20058.263] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-1ms), your system is too slow
[ 20060.789] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-11ms), your system is too slow
[ 20204.672] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-11ms), your system is too slow
[ 20641.553] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 13ms, your system is too slow
[ 20642.577] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 17ms, your system is too slow
[ 20642.715] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 19ms, your system is too slow
[ 20643.899] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-1ms), your system is too slow
[ 21260.510] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 16ms, your system is too slow
[ 21467.622] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-8ms), your system is too slow
[ 21482.215] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[ 21482.736] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 23ms, your system is too slow
[ 21482.736] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-11ms), your system is too slow
[ 21615.104] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[ 21615.854] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 11ms, your system is too slow
[ 21615.926] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 15ms, your system is too slow
[ 21618.658] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 27ms, your system is too slow
[ 21618.884] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 15ms, your system is too slow
[ 21618.884] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 21647.478] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-6ms), your system is too slow
[ 21722.506] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[ 21724.070] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[ 21726.146] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 14ms, your system is too slow
[ 21726.790] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 12ms, your system is too slow
[ 21728.370] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 28ms, your system is too slow
[ 21728.370] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 21781.558] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-2ms), your system is too slow
[ 21843.332] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-3ms), your system is too slow
[ 21843.332] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-16ms), your system is too slow
[ 21845.808] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-0ms), your system is too slow
[ 21848.878] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-7ms), your system is too slow
[ 21848.878] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-20ms), your system is too slow
[ 21850.692] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-1ms), your system is too slow
[ 21850.692] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-14ms), your system is too slow
[ 21853.922] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-1ms), your system is too slow
[ 21853.922] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-14ms), your system is too slow
[ 21970.590] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 24ms, your system is too slow
[ 21970.782] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-0ms), your system is too slow
[ 21976.702] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-2ms), your system is too slow
[ 22274.998] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 11ms, your system is too slow
[ 22275.450] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 22276.038] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 31ms, your system is too slow
[ 22276.264] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 19ms, your system is too slow
[ 22276.810] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 21ms, your system is too slow
[ 22276.810] (EE) event3  - Logitech MX Master 3: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 22277.094] (EE) client bug: timer event3 debounce: scheduled expiry is in the past (-8ms), your system is too slow
[ 22277.094] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-22ms), your system is too slow
[ 22344.048] (EE) event3  - Logitech MX Master 3: client bug: event processing lagging behind by 13ms, your system is too slow
[ 22344.830] (EE) client bug: timer event3 debounce short: scheduled expiry is in the past (-1ms), your system is too slow
...

I have disabled autosuspend with uedv rule:

$ cat /etc/udev/rules.d/50-usb_power_save.rules
# blacklist autosuspend for keyboard/mouse
ACTION=="add", SUBSYSTEM=="usb", ATTR{idVendor}=="046d", TEST=="power/autosuspend", ATTR{power/autosuspend}="-1"

And the udev rule works well:

$ cat /sys/bus/usb/devices/9-2.3/manufacturer
Logitech
$ cat /sys/bus/usb/devices/9-2.3/power/autosuspend
-1
...

But it still fails. Generally the mouse/keyboard works fine but there are times that I have to press a key 2 times or press the left mouse button 2 times for it to respond.
Where is the error? In "libinput" or "xf86-input-libinput"? Should I report my error here: https://gitlab.gnome.org/GNOME/gnome-sh … ssues/2041 ?

Subject-17 · 2020-10-17 19:30:31

@Seth: No, CPU load was there before, but not sure if it was significantly more/less.

I enabled it via

# cat /boot/loader/entries/arch.conf , matches with /proc/cmdline
title ArchLinux
linux /vmlinuz-linux
initrd /initramfs-linux.img
options cryptdevice=PARTUUID=REDACTED_UUID:cryptdisk:allow-discards root=/dev/mapper/vgroup-lvroot rw splash rootwait amd_iommu=on iommu=pt nvidia-drm.modeset=1 drm_kms_helper.poll=0

However, I don't see it listed in sysctl (is that expected?)

sudo sysctl -a | grep poll
fs.epoll.max_user_watches = 6696960
net.core.busy_poll = 0

@ diagnostics:
I've taken to leaving `tail -f ~/.local/share/xorg/Xorg.0.log` up. Just Syu'd & rebooted last night. The below logs displayed cleanly after/during some mouse lag.

[ 49948.050] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 36ms, your system is too slow
[ 49948.050] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 49977.379] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 49983.867] (EE) client bug: timer event8 debounce: scheduled expiry is in the past (-7ms), your system is too slow
[ 49983.867] (EE) client bug: timer event8 debounce short: scheduled expiry is in the past (-20ms), your system is too slow
[ 49997.532] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 13ms, your system is too slow
[ 49998.227] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 41ms, your system is too slow
[ 49998.227] (EE) client bug: timer event8 debounce: scheduled expiry is in the past (-16ms), your system is too slow
[ 50001.421] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 19ms, your system is too slow
[ 50001.664] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 63ms, your system is too slow
[ 50003.806] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: client bug: event processing lagging behind by 28ms, your system is too slow
[ 50003.806] (EE) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: WARNING: log rate limit exceeded (5 msgs per 60s). Discarding future messages.
[ 50007.387] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 50011.303] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 50011.434] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 50011.530] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 50011.678] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: SYN_DROPPED event - some input events have been lost.
[ 50011.678] (II) event8  - Logitech Logitech G903 Wired/Wireless Gaming Mouse: WARNING: log rate limit exceeded (5 msgs per 30s). Discarding future messages.

The

SYN_DROPPED event - some input events have been lost.

part is (seemingly) new.

I'm using my mouse in wired mode. I'll try switching mice and see if that does anything.

Last edited by Subject-17 (2020-10-17 19:33:31)

seth · 2020-10-17 21:00:49

drm_kms_helper is a module, so you want to check

systool -vm drm_kms_helper

but the parameter isn't necessarily exported.
Remove it from the bootloader config, reboot and check whether "moving the mouse in circles" still "can spike firefox from <10% on a single CPU to 70%" (I can do the same to chromium w/o any special parameters and different hardware - the browser just takes some cpu to handle the input events)

Arch Linux

#1 2020-08-30 15:28:56

[SOLVED] Laggy, sticky mouse input

#2 2020-08-30 15:44:37

Re: [SOLVED] Laggy, sticky mouse input

#3 2020-08-30 15:48:02

Re: [SOLVED] Laggy, sticky mouse input

#4 2020-08-30 15:57:21

Re: [SOLVED] Laggy, sticky mouse input

#5 2020-09-08 01:20:26

Re: [SOLVED] Laggy, sticky mouse input

#6 2020-09-08 09:26:04

Re: [SOLVED] Laggy, sticky mouse input

#7 2020-09-09 03:16:50

Re: [SOLVED] Laggy, sticky mouse input

#8 2020-09-09 03:24:39

Re: [SOLVED] Laggy, sticky mouse input

#9 2020-09-09 05:58:57

Re: [SOLVED] Laggy, sticky mouse input

#10 2020-09-09 13:07:39

Re: [SOLVED] Laggy, sticky mouse input

#11 2020-09-09 13:51:20

Re: [SOLVED] Laggy, sticky mouse input

#12 2020-09-09 13:52:57

Re: [SOLVED] Laggy, sticky mouse input

#13 2020-09-09 23:02:41

Re: [SOLVED] Laggy, sticky mouse input

#14 2020-09-10 06:23:51

Re: [SOLVED] Laggy, sticky mouse input

#15 2020-09-10 06:52:18

Re: [SOLVED] Laggy, sticky mouse input

#16 2020-09-16 09:11:41

Re: [SOLVED] Laggy, sticky mouse input

#17 2020-09-16 09:22:45

Re: [SOLVED] Laggy, sticky mouse input

#18 2020-10-07 01:49:30

Re: [SOLVED] Laggy, sticky mouse input

#19 2020-10-07 03:52:34

Re: [SOLVED] Laggy, sticky mouse input

#20 2020-10-11 01:58:52

Re: [SOLVED] Laggy, sticky mouse input

#21 2020-10-11 06:33:27

Re: [SOLVED] Laggy, sticky mouse input

#22 2020-10-11 12:44:16

Re: [SOLVED] Laggy, sticky mouse input

#23 2020-10-17 08:36:21

Re: [SOLVED] Laggy, sticky mouse input

#24 2020-10-17 19:30:31

Re: [SOLVED] Laggy, sticky mouse input

#25 2020-10-17 21:00:49

Re: [SOLVED] Laggy, sticky mouse input

Board footer