Since last reboot, my trusty laptop has the following annoying behavior:
mouse clicks don't take effect until the next event (typically moving
the mouse).

I checked with `xev` and indeed after a mouse click I see the following
output:

[...]

MotionNotify event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21834559, (83,68), root:(369,570),
state 0x0, is_hint 0, same_screen YES

ButtonPress event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21836087, (83,68), root:(369,570),
state 0x0, button 1, same_screen YES

and it sits there without showing any "ButtonRelease" even though
I definitely released the mouse button. When I then press a key
I finally see the ButtonRelease followed by the key press/release:

ButtonRelease event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837777, (83,68), root:(369,570),
state 0x100, button 1, same_screen YES

KeyPress event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837777, (83,68), root:(369,570),
state 0x0, keycode 111 (keysym 0xff52, Up), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False

KeyRelease event, serial 38, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837880, (83,68), root:(369,570),
state 0x0, keycode 111 (keysym 0xff52, Up), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: False

Any idea where the problem might be coming from, how to track it down,
or (even better) fix it?

This is a Thinkpad T61 running Debian testing, with XFCE running
on Xorg.


=== Stefan

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 2026-04-24 at 11:00, Stefan Monnier wrote:

Since last reboot, my trusty laptop has the following annoying
behavior: mouse clicks don't take effect until the next event
(typically moving the mouse).
Any idea where the problem might be coming from, how to track it
down, or (even better) fix it?

This is a Thinkpad T61 running Debian testing, with XFCE running on
Xorg.

My first reflex would be to look at the xserver-xorg-input-* packages.
If those don't seem to be it, the next thing I think of would probably
be the kernel input-device drivers, which I doubt are packaged
separately from the kernel proper.
There may well be other places in the stack that could be relevant, some
of them potentially between those two, but none that are springing
immediately to my mind.
--
The Wanderer
The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man. -- George Bernard Shaw


--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 4/24/26 20:00, Stefan Monnier wrote:

Since last reboot, my trusty laptop has the following annoying behavior: mouse clicks don't take effect until the next event (typically moving
the mouse).

I checked with `xev` and indeed after a mouse click I see the following output:

[...]

MotionNotify event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21834559, (83,68), root:(369,570),
state 0x0, is_hint 0, same_screen YES

ButtonPress event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21836087, (83,68), root:(369,570),
state 0x0, button 1, same_screen YES

and it sits there without showing any "ButtonRelease" even though
I definitely released the mouse button. When I then press a key
I finally see the ButtonRelease followed by the key press/release:

ButtonRelease event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837777, (83,68), root:(369,570),
state 0x100, button 1, same_screen YES

KeyPress event, serial 35, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837777, (83,68), root:(369,570),
state 0x0, keycode 111 (keysym 0xff52, Up), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False

KeyRelease event, serial 38, synthetic NO, window 0x3800001,
root 0xf6, subw 0x0, time 21837880, (83,68), root:(369,570),
state 0x0, keycode 111 (keysym 0xff52, Up), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: False

Any idea where the problem might be coming from, how to track it down,
or (even better) fix it?

This is a Thinkpad T61 running Debian testing, with XFCE running
on Xorg.

=== Stefan

The funky behavior like this almost always leads to some buggy
power-saving feature.
Try to disable power-saving within a kernel module which your mouse use.
If it is on usb bus:
$ echo -1 | sudo tee /sys/module/usbcore/parameters/autosuspend

--
With kindest regards, Alexander.
Debian - The universal operating system
https://www.debian.org


--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

The funky behavior like this almost always leads to some buggy
power-saving feature.
Try to disable power-saving within a kernel module which your mouse use.
If it is on usb bus:
$ echo -1 | sudo tee /sys/module/usbcore/parameters/autosuspend

Good idea, but this is not connected via USB (`lsusb` doesn't show
anything related to mouses or keyboards). It's the mouse button of my Thinkpad's trackpoint, which AFAICT is connected "internally". My web
searches suggest this uses the old i8042 PS/2 controller.
But I can't see any power management in there:

# find /sys/module/i8042/
/sys/module/i8042/
/sys/module/i8042/uevent
/sys/module/i8042/parameters
/sys/module/i8042/parameters/unmask_kbd_data
/sys/module/i8042/parameters/debug
#

Any idea how I could control the power saving (or even if there is any
power saving to control) in there?


=== Stefan

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Don't know if my idea is relevant; A long time ago I had some weirdness
with interrupts, where the mouse became extremely sluggish when that
acted up. I think that one complained in dmesg. Maybe it is something to
look at?
-- nwe


--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Don't know if my idea is relevant; A long time ago I had some weirdness with interrupts, where the mouse became extremely sluggish when that acted
up. I think that one complained in dmesg. Maybe it is something to look at?

Good idea, thanks.
This said, I can't see any complaint in the dmesg log that would seem related. OTOH, I noticed

[ 6.428328] input: TPPS/2 IBM TrackPoint as /devices/platform/i8042/serio1/serio2/input/input5

so I went to see that /sys node and tried

echo on > /sys/devices/platform/i8042/serio1/serio2/input/input5/power/control

But that didn't seem to make any difference ?


=== Stefan

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Hmm...

To track things down further, I decided to install an older kernel to
see if that fixes the problem. After rebooting into that old kernel,
things work great again. So I started to "bisect" except that I haven't
been able to reproduce the problem with a newer kernel either.

The upside is that the problem seems to be fixed. Not sure if "for
good" or if it's still out there lurking, but I'll enjoy it while
it lasts. Thanks for the help, in any case,


=== Stefan


Stefan Monnier [2026-04-28 16:42:41] wrote:

Don't know if my idea is relevant; A long time ago I had some weirdness with >> interrupts, where the mouse became extremely sluggish when that acted
up. I think that one complained in dmesg. Maybe it is something to look at?

Good idea, thanks. This said, I can't see any complaint in the dmesg
log that would seem related. OTOH, I noticed

[ 6.428328] input: TPPS/2 IBM TrackPoint as /devices/platform/i8042/serio1/serio2/input/input5

so I went to see that /sys node and tried

echo on > /sys/devices/platform/i8042/serio1/serio2/input/input5/power/control

But that didn't seem to make any difference ?

=== Stefan

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 28/04/26 at 23:18, Stefan Monnier wrote:

Hmm...

To track things down further, I decided to install an older kernel to
see if that fixes the problem. After rebooting into that old kernel,
things work great again. So I started to "bisect" except that I haven't
been able to reproduce the problem with a newer kernel either.

The upside is that the problem seems to be fixed. Not sure if "for
good" or if it's still out there lurking, but I'll enjoy it while
it lasts. Thanks for the help, in any case,

=== Stefan

Perhaps a good thing to do now is to test the latest back-ported kernel
and if it also has the issue submit a bug report against the "linux-image-6.12.74+deb13+1-amd64" package (I guess you use the 64bit kernel).

Although your it's not an usb mouse issue maybe it could be related to
the #1134813 bug report, I found that by searching for "mouse" in
"reportbug".

Just my 2› hint, good luck.
--
Franco Martelli

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

To track things down further, I decided to install an older kernel to
see if that fixes the problem. After rebooting into that old kernel,
things work great again. So I started to "bisect" except that I haven't
been able to reproduce the problem with a newer kernel either.

That was too good to be true: the problem is back, and I confirmed that
it doesn't appear right after boot but only "after a while".
So far it has appeared after a suspend+resume but not just after any suspend/resume.

I looked at the dmesg output (see below) of the suspend+resume after
which the problem (re)appeared but couldn't spot anything that seemed
relevant: in that output, there are two rounds of suspend+resume
(delimited by diff3 markers, so I could use `smerge-mode` in Emacs to
look at the word-level diffs ?). The mouse buttons still worked
correctly after the first suspend+resume but not after the second, but
if you compare the two parts of the log, there are virtually no
differences. ?

Franco Martelli [2026-04-29 21:13:22] wrote:

Perhaps a good thing to do now is to test the latest back-ported kernel and if it also has the issue submit a bug report against the "linux-image-6.12.74+deb13+1-amd64" package (I guess you use the 64bit kernel).

I'm on "testing" so I'm currently on 6.19.13. Now that I know it shows
up only after a while, I may (re)try bisecting, but the bisection will
be slooow.

Although your it's not an usb mouse issue maybe it could be related to the #1134813 bug report, I found that by searching for "mouse" in "reportbug".

The reporter closed it saying that it was a problem on his side, so it
seems unlikely to be related, but thanks for the suggestion.


=== Stefan



<<<<<<< So-far-so-good
[ 619.518449] perf: interrupt took too long (6210 > 6200), lowering kernel.perf_event_max_sample_rate to 32000
[ 626.935733] warning: `atop' uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211
[ 683.771047] wls3: deauthenticating from c8:91:f9:3b:10:cb by local choice (Reason: 3=DEAUTH_LEAVING)
[ 683.782726] audit: type=1400 audit(1777561643.638:1348): apparmor="ALLOWED" operation="open" class="file" profile="Xorg" name="/" pid=1255 comm="Xorg" requested_mask="r" denied_mask="r" fsuid=0 ouid=0
[ 684.231370] PM: suspend entry (deep)
[ 684.294295] Filesystems sync: 0.062 seconds
[ 684.356976] Freezing user space processes
[ 684.358641] Freezing user space processes completed (elapsed 0.001 seconds) [ 684.358652] OOM killer disabled.
[ 684.358655] Freezing remaining freezable tasks
[ 684.359856] Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
[ 684.359968] printk: Suspending console(s) (use no_console_suspend to debug) [ 684.365012] e1000e: EEE TX LPI TIMER: 00000000
[ 684.532806] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[ 684.539170] ata1.00: Entering standby power mode
[ 684.674906] ACPI: EC: interrupt blocked
[ 684.691147] ACPI: PM: Preparing to enter system sleep state S3
[ 684.701618] ACPI: EC: event blocked
[ 684.701620] ACPI: EC: EC stopped
[ 684.701621] ACPI: PM: Saving platform NVS memory
[ 684.701881] Disabling non-boot CPUs ...
[ 684.703477] smpboot: CPU 1 is now offline
[ 684.708688] ACPI: PM: Low-level resume complete
[ 684.708705] ACPI: EC: EC started
[ 684.708705] ACPI: PM: Restoring platform NVS memory
[ 684.708705] Enabling non-boot CPUs ...
[ 684.708705] smpboot: Booting Node 0 Processor 1 APIC 0x1
[ 684.710838] CPU1 is up
[ 684.717086] ACPI: PM: Waking up from system sleep state S3
[ 684.741581] ACPI: EC: interrupt unblocked
[ 684.748011] ACPI: EC: event unblocked
[ 684.748648] usb usb3: root hub lost power or was reset
[ 684.748800] usb usb4: root hub lost power or was reset
[ 684.749643] usb usb6: root hub lost power or was reset
[ 684.760878] usb usb5: root hub lost power or was reset
[ 684.761415] usb usb7: root hub lost power or was reset
[ 684.768288] tpm tpm0: TPM is disabled/deactivated (0x6)
[ 684.772422] thinkpad_acpi: ACPI backlight control delay disabled
[ 685.016853] usb 3-1: reset full-speed USB device number 2 using uhci_hcd
[ 685.016866] usb 6-2: reset full-speed USB device number 2 using uhci_hcd
[ 685.072834] ata3: SATA link down (SStatus 0 SControl 300)
[ 685.072900] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[ 685.075508] ata1.00: ACPI cmd f5/00:00:00:00:00:a0(SECURITY FREEZE LOCK) filtered out
[ 685.075517] ata1.00: ACPI cmd ef/10:03:00:00:00:a0(SET FEATURES) filtered out
[ 685.085564] sd 0:0:0:0: [sda] Starting disk
[ 685.095114] ata1.00: ACPI cmd f5/00:00:00:00:00:a0(SECURITY FREEZE LOCK) filtered out
[ 685.095129] ata1.00: ACPI cmd ef/10:03:00:00:00:a0(SET FEATURES) filtered out
[ 685.105547] ata1.00: configured for UDMA/133
[ 685.147786] ata4.00: ACPI cmd ef/03:42:00:00:00:a0(SET FEATURES) filtered out
[ 685.147800] ata4.00: ACPI cmd ef/03:0c:00:00:00:a0(SET FEATURES) filtered out
[ 685.288929] usb 3-2: reset full-speed USB device number 3 using uhci_hcd
[ 686.384948] pcieport 0000:00:1c.4: pciehp: Timeout on hotplug command 0x0018 (issued 1644 msec ago)
[ 686.388937] pcieport 0000:00:1c.2: pciehp: Timeout on hotplug command 0x0018 (issued 1648 msec ago)
[ 686.388943] pcieport 0000:00:1c.1: pciehp: Timeout on hotplug command 0x0018 (issued 1648 msec ago)
[ 686.390035] ath: phy0: ASPM enabled: 0x43
[ 686.396719] pcieport 0000:00:1c.3: pciehp: Timeout on hotplug command 0x0018 (issued 1656 msec ago)
[ 686.408939] pcieport 0000:00:1c.0: pciehp: Timeout on hotplug command 0x0018 (issued 1668 msec ago)
[ 686.410920] OOM killer enabled.
[ 686.410930] Restarting tasks: Starting
[ 686.413913] Restarting tasks: Done
[ 686.413946] random: crng reseeded on system resumption
[ 686.448589] video LNXVIDEO:00: Restoring backlight state
[ 686.449448] PM: suspend exit
[ 686.481086] Bluetooth: MGMT ver 1.23
[ 687.028979] e1000e 0000:00:19.0 enp0s25: NIC Link is Down
[ 689.650527] wls3: authenticate with c8:91:f9:3b:10:cb (local address=00:1c:b3:c4:42:82)
[ 689.650539] wls3: send auth to c8:91:f9:3b:10:cb (try 1/3)
[ 689.653995] wls3: authenticated
[ 689.656740] wls3: associate with c8:91:f9:3b:10:cb (try 1/3)
[ 689.659861] wls3: RX AssocResp from c8:91:f9:3b:10:cb (capab=0x11 status=0 aid=6)
[ 689.660080] wls3: associated
[ 689.660177] ath: EEPROM regdomain: 0x807c
[ 689.660180] ath: EEPROM indicates we should expect a country code
[ 689.660181] ath: doing EEPROM country->regdmn map search
[ 689.660183] ath: country maps to regdmn code: 0x3a
[ 689.660184] ath: Country alpha2 being used: CA
[ 689.660186] ath: Regpair used: 0x3a
[ 689.660187] ath: regdomain 0x807c dynamically updated by country element
[ 1173.323125] systemd-journald[322]: Time jumped backwards, rotating.
[ 1818.226562] perf: interrupt took too long (8285 > 7762), lowering kernel.perf_event_max_sample_rate to 24000
=======
[ 2683.140452] audit: type=1400 audit(1777563648.373:1349): apparmor="ALLOWED" operation="open" class="file" profile="Xorg" name="/" pid=1255 comm="Xorg" requested_mask="r" denied_mask="r" fsuid=0 ouid=0
[ 2683.170621] wls3: deauthenticating from c8:91:f9:3b:10:cb by local choice (Reason: 3=DEAUTH_LEAVING)
[ 2683.628192] PM: suspend entry (deep)
[ 2683.671085] Filesystems sync: 0.042 seconds
[ 2683.737413] Freezing user space processes
[ 2683.739642] Freezing user space processes completed (elapsed 0.002 seconds) [ 2683.739651] OOM killer disabled.
[ 2683.739654] Freezing remaining freezable tasks
[ 2683.740789] Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
[ 2683.740882] printk: Suspending console(s) (use no_console_suspend to debug) [ 2683.746411] e1000e: EEE TX LPI TIMER: 00000000
[ 2683.912854] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[ 2683.916919] ata1.00: Entering standby power mode
[ 2684.032583] ACPI: EC: interrupt blocked
[ 2684.045174] ACPI: PM: Preparing to enter system sleep state S3
[ 2684.056423] ACPI: EC: event blocked
[ 2684.056425] ACPI: EC: EC stopped
[ 2684.056426] ACPI: PM: Saving platform NVS memory
[ 2684.056732] Disabling non-boot CPUs ...
[ 2684.058279] smpboot: CPU 1 is now offline
[ 2684.061225] ACPI: PM: Low-level resume complete
[ 2684.061225] ACPI: EC: EC started
[ 2684.061225] ACPI: PM: Restoring platform NVS memory
[ 2684.061225] Enabling non-boot CPUs ...
[ 2684.061225] smpboot: Booting Node 0 Processor 1 APIC 0x1
[ 2684.066483] CPU1 is up
[ 2684.072773] ACPI: PM: Waking up from system sleep state S3
[ 2684.097624] ACPI: EC: interrupt unblocked
[ 2684.102742] usb usb6: root hub lost power or was reset
[ 2684.102905] usb usb7: root hub lost power or was reset
[ 2684.103597] ACPI: EC: event unblocked
[ 2684.124773] usb usb5: root hub lost power or was reset
[ 2684.125419] usb usb3: root hub lost power or was reset
[ 2684.126000] usb usb4: root hub lost power or was reset
[ 2684.131474] tpm tpm0: TPM is disabled/deactivated (0x6)
[ 2684.135909] thinkpad_acpi: ACPI backlight control delay disabled
[ 2684.381098] usb 6-2: reset full-speed USB device number 2 using uhci_hcd
[ 2684.385093] usb 3-2: reset full-speed USB device number 3 using uhci_hcd
[ 2684.408767] ata3: SATA link down (SStatus 0 SControl 300)
[ 2684.417660] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[ 2684.419257] ata1.00: ACPI cmd f5/00:00:00:00:00:a0(SECURITY FREEZE LOCK) filtered out
[ 2684.419271] ata1.00: ACPI cmd ef/10:03:00:00:00:a0(SET FEATURES) filtered out
[ 2684.429822] sd 0:0:0:0: [sda] Starting disk
[ 2684.439272] ata1.00: ACPI cmd f5/00:00:00:00:00:a0(SECURITY FREEZE LOCK) filtered out
[ 2684.439287] ata1.00: ACPI cmd ef/10:03:00:00:00:a0(SET FEATURES) filtered out
[ 2684.449676] ata1.00: configured for UDMA/133
[ 2684.507838] ata4.00: ACPI cmd ef/03:42:00:00:00:a0(SET FEATURES) filtered out
[ 2684.507853] ata4.00: ACPI cmd ef/03:0c:00:00:00:a0(SET FEATURES) filtered out
[ 2684.652822] usb 3-1: reset full-speed USB device number 2 using uhci_hcd
[ 2685.828937] pcieport 0000:00:1c.1: pciehp: Timeout on hotplug command 0x0018 (issued 1732 msec ago)
[ 2685.830058] ath: phy0: ASPM enabled: 0x43
[ 2685.833269] pcieport 0000:00:1c.4: pciehp: Timeout on hotplug command 0x0018 (issued 1736 msec ago)
[ 2685.837024] pcieport 0000:00:1c.3: pciehp: Timeout on hotplug command 0x0018 (issued 1740 msec ago)
[ 2685.873009] pcieport 0000:00:1c.2: pciehp: Timeout on hotplug command 0x0018 (issued 1776 msec ago)
[ 2685.877004] pcieport 0000:00:1c.0: pciehp: Timeout on hotplug command 0x0018 (issued 1780 msec ago)
[ 2685.878139] OOM killer enabled.
[ 2685.878145] Restarting tasks: Starting
[ 2685.907249] Restarting tasks: Done
[ 2685.907291] random: crng reseeded on system resumption
[ 2685.953965] video LNXVIDEO:00: Restoring backlight state
[ 2685.954374] Bluetooth: MGMT ver 1.23
[ 2685.954850] PM: suspend exit
[ 2686.764986] e1000e 0000:00:19.0 enp0s25: NIC Link is Down
[ 2689.355196] wls3: authenticate with c8:91:f9:3b:10:cb (local address=00:1c:b3:c4:42:82)
[ 2689.355207] wls3: send auth to c8:91:f9:3b:10:cb (try 1/3)
[ 2689.358194] wls3: authenticated
[ 2689.360814] wls3: associate with c8:91:f9:3b:10:cb (try 1/3)
[ 2689.363318] wls3: RX AssocResp from c8:91:f9:3b:10:cb (capab=0x11 status=0 aid=6)
[ 2689.363531] wls3: associated
[ 2689.363632] ath: EEPROM regdomain: 0x807c
[ 2689.363635] ath: EEPROM indicates we should expect a country code
[ 2689.363637] ath: doing EEPROM country->regdmn map search
[ 2689.363638] ath: country maps to regdmn code: 0x3a
[ 2689.363639] ath: Country alpha2 being used: CA
[ 2689.363641] ath: Regpair used: 0x3a
[ 2689.363644] ath: regdomain 0x807c dynamically updated by country element >>>>>>> Resume-after-which-the-problem-appeared

--- PyGate Linux v1.5.14
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)