Arch Linux

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Touchpad freezes intermittently

Hey all,

I'm having an issue where my touchpad will be working perfectly, and then it will just stop moving the cursor for seconds at a time. clicking works, but anything tap-related is inop.
I had initially ruled this out as being hw or kernel-related, but now I wonder.
The wiki and several other sources indicate that xf86-input-libinput is the package to use for input-related devices for window compositors. The wiki page for libinput indicates that this is for both wayland and x.org, but the files in the package are definitely more oriented toward xorg.  This isn't really the point I'm trying to drive at, except to mention that this problem exists when using both x.org and Wayland.

$ pacman -Ql xf86-input-libinput
xf86-input-libinput /usr/
xf86-input-libinput /usr/include/
xf86-input-libinput /usr/include/xorg/
xf86-input-libinput /usr/include/xorg/libinput-properties.h
xf86-input-libinput /usr/lib/
xf86-input-libinput /usr/lib/pkgconfig/
xf86-input-libinput /usr/lib/pkgconfig/xorg-libinput.pc
xf86-input-libinput /usr/lib/xorg/
xf86-input-libinput /usr/lib/xorg/modules/
xf86-input-libinput /usr/lib/xorg/modules/input/
xf86-input-libinput /usr/lib/xorg/modules/input/libinput_drv.so
xf86-input-libinput /usr/share/
xf86-input-libinput /usr/share/X11/
xf86-input-libinput /usr/share/X11/xorg.conf.d/
xf86-input-libinput /usr/share/X11/xorg.conf.d/40-libinput.conf
xf86-input-libinput /usr/share/licenses/
xf86-input-libinput /usr/share/licenses/xf86-input-libinput/
xf86-input-libinput /usr/share/licenses/xf86-input-libinput/COPYING
xf86-input-libinput /usr/share/man/
xf86-input-libinput /usr/share/man/man4/
xf86-input-libinput /usr/share/man/man4/libinput.4.gz

I expect xf86-input-libinput just translates hw input into the context of the graphical environment.  So if I'm having problems with a touchpad, I guess I'm inclined to believe it's not libinput that is the culprit since it's intended to be generic/all-encompassing

So what if the touchpad is intermittently not responding? Being an i2c device, the kernel would probably pick-up on communications abnormalities.  Besides, the click input works even when the touch/drag input does not, so the i2c bus communication is fine?
Nothing is dmesg comes up when it's not responding, and nothing comes up in journalctl, either.

I'm posting relevant lines from dmesg from system boot here. Is it normal for a touchpad to be detected as both a mouse and a touchpad (as shown below).  There are two input devices related to the touchpad and I guess I'm considering whether both devices (input9 and input10) might be conflicting.

Any thoughts/help are appreciated.

asus_wmi: ASUS WMI generic driver loaded
asus_wmi: Initialization: 0x1
asus_wmi: BIOS WMI version: 9.0
asus_wmi: SFUN value: 0x21
asus-nb-wmi asus-nb-wmi: Detected ATK, not ASUSWMI, use DSTS
asus-nb-wmi asus-nb-wmi: Detected ATK, enable event queue
input: Asus WMI hotkeys as /devices/platform/asus-nb-wmi/input/input8
battery: new extension: ASUS Battery Extension
input: ELAN1300:00 04F3:3104 Mouse as /devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input9
EDAC amd64: F17h_M60h detected (node 0).
EDAC amd64: Node 0: DRAM ECC disabled.
input: ELAN1300:00 04F3:3104 Touchpad as /devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input10
hid-generic 0018:04F3:3104.0001: input,hidraw0: I2C HID v1.00 Mouse [ELAN1300:00 04F3:3104] on i2c-ELAN1300:00
EDAC amd64: F17h_M60h detected (node 0).
EDAC amd64: Node 0: DRAM ECC disabled.
input: ELAN1300:00 04F3:3104 Mouse as /devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input11
input: ELAN1300:00 04F3:3104 Touchpad as /devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input12
hid-multitouch 0018:04F3:3104.0001: input,hidraw0: I2C HID v1.00 Mouse [ELAN1300:00 04F3:3104] on i2c-ELAN1300:00

V1del

Forum Moderator

Registered: 2012-10-16

Posts: 23,209

Re: Touchpad freezes intermittently

Asus WMI is somewhat infamous for having an issue here:  https://bbs.archlinux.org/viewtopic.php … 8#p1926138

If it's not that, do you type and try to use the touchpad at the sameish time? might have to enable the flag to have it enabled while typing

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

Thanks for the reply.

I don't entirely understand why I would want to set the LIBINPUT_IGNORE_DEVICE environmental variable equal to 1, but based on the output of libinput and udevadm, I put together a similar rule for my device.  Because of the random/intermittent nature of the error, I will have to wait and see if the change has any impact, though.

/etc/udev/rules.d/51-asus_wmi.rules

ACTION=="add|change", KERNEL=="event9" ATTRS{name}=="ELAN1300:00 04F3:3104 Mouse", ENV{LIBINPUT_IGNORE_DEVICE}="1"

This my first time writing a udev rule, and I actually had to spend quite a bit of time reading through udev rules, their purpose, use cases, and their syntax so that I could understand what was written in the examples you linked to. 

# libinput list-devices
Device:           ELAN1300:00 04F3:3104 Touchpad
Kernel:           /dev/input/event10
Group:            7
Seat:             seat0, default
Size:             103x69mm
Capabilities:     pointer gesture
Tap-to-click:     disabled
Tap-and-drag:     enabled
Tap drag lock:    disabled
Left-handed:      disabled
Nat.scrolling:    disabled
Middle emulation: disabled
Calibration:      n/a
Scroll methods:   *two-finger edge 
Click methods:    *button-areas clickfinger 
Disable-w-typing: enabled
Accel profiles:   flat *adaptive
Rotation:         n/a

Device:           Asus WMI hotkeys
Kernel:           /dev/input/event8
Group:            8
Seat:             seat0, default
Capabilities:     keyboard 
Tap-to-click:     n/a
Tap-and-drag:     n/a
Tap drag lock:    n/a
Left-handed:      n/a
Nat.scrolling:    n/a
Middle emulation: n/a
Calibration:      n/a
Scroll methods:   none
Click methods:    none
Disable-w-typing: n/a
Accel profiles:   n/a
Rotation:         n/a

and now knowing about event9 and event10 (I've included event8 for kicks, since that's the Asus WMI hotkeys problem you mentioned), I ran:

# udevadm info -a -n /dev/input/event9

Udevadm info starts with the device specified by the devpath and then
walks up the chain of parent devices. It prints for every device
found, all possible attributes in the udev rules key format.
A rule to match, can be composed by the attributes of the device
and the attributes from one single parent device.

  looking at device '/devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input11/event9':
    KERNEL=="event9"
    SUBSYSTEM=="input"
    DRIVER==""
    ATTR{power/async}=="disabled"
    ATTR{power/control}=="auto"
    ATTR{power/runtime_active_kids}=="0"
    ATTR{power/runtime_active_time}=="0"
    ATTR{power/runtime_enabled}=="disabled"
    ATTR{power/runtime_status}=="unsupported"
    ATTR{power/runtime_suspended_time}=="0"
    ATTR{power/runtime_usage}=="0"

  looking at parent device '/devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001/input/input11':
    KERNELS=="input11"
    SUBSYSTEMS=="input"
    DRIVERS==""
    ATTRS{capabilities/abs}=="0"
    ATTRS{capabilities/ev}=="17"
    ATTRS{capabilities/ff}=="0"
    ATTRS{capabilities/key}=="30000 0 0 0 0"
    ATTRS{capabilities/led}=="0"
    ATTRS{capabilities/msc}=="10"
    ATTRS{capabilities/rel}=="1943"
    ATTRS{capabilities/snd}=="0"
    ATTRS{capabilities/sw}=="0"
    ATTRS{id/bustype}=="0018"
    ATTRS{id/product}=="3104"
    ATTRS{id/vendor}=="04f3"
    ATTRS{id/version}=="0100"
    ATTRS{name}=="ELAN1300:00 04F3:3104 Mouse"
    ATTRS{phys}=="i2c-ELAN1300:00"
    ATTRS{power/async}=="disabled"
    ATTRS{power/control}=="auto"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_active_time}=="0"
    ATTRS{power/runtime_enabled}=="disabled"
    ATTRS{power/runtime_status}=="unsupported"
    ATTRS{power/runtime_suspended_time}=="0"
    ATTRS{power/runtime_usage}=="0"
    ATTRS{properties}=="0"
    ATTRS{uniq}==""

  looking at parent device '/devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00/0018:04F3:3104.0001':
    KERNELS=="0018:04F3:3104.0001"
    SUBSYSTEMS=="hid"
    DRIVERS=="hid-multitouch"
    ATTRS{country}=="00"
    ATTRS{power/async}=="enabled"
    ATTRS{power/control}=="auto"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_active_time}=="0"
    ATTRS{power/runtime_enabled}=="disabled"
    ATTRS{power/runtime_status}=="unsupported"
    ATTRS{power/runtime_suspended_time}=="0"
    ATTRS{power/runtime_usage}=="0"
    ATTRS{quirks}=="334864"

  looking at parent device '/devices/platform/AMDI0010:03/i2c-0/i2c-ELAN1300:00':
    KERNELS=="i2c-ELAN1300:00"
    SUBSYSTEMS=="i2c"
    DRIVERS=="i2c_hid"
    ATTRS{name}=="ELAN1300:00"
    ATTRS{power/async}=="enabled"
    ATTRS{power/control}=="auto"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_active_time}=="0"
    ATTRS{power/runtime_enabled}=="disabled"
    ATTRS{power/runtime_status}=="unsupported"
    ATTRS{power/runtime_suspended_time}=="0"
    ATTRS{power/runtime_usage}=="0"

  looking at parent device '/devices/platform/AMDI0010:03/i2c-0':
    KERNELS=="i2c-0"
    SUBSYSTEMS=="i2c"
    DRIVERS==""
    ATTRS{name}=="Synopsys DesignWare I2C adapter"
    ATTRS{power/async}=="disabled"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_enabled}=="enabled"
    ATTRS{power/runtime_status}=="suspended"
    ATTRS{power/runtime_usage}=="0"

  looking at parent device '/devices/platform/AMDI0010:03':
    KERNELS=="AMDI0010:03"
    SUBSYSTEMS=="platform"
    DRIVERS=="i2c_designware"
    ATTRS{driver_override}=="(null)"
    ATTRS{power/async}=="disabled"
    ATTRS{power/autosuspend_delay_ms}=="1000"
    ATTRS{power/control}=="auto"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_active_time}=="131941"
    ATTRS{power/runtime_enabled}=="enabled"
    ATTRS{power/runtime_status}=="suspended"
    ATTRS{power/runtime_suspended_time}=="456577"
    ATTRS{power/runtime_usage}=="0"

  looking at parent device '/devices/platform':
    KERNELS=="platform"
    SUBSYSTEMS==""
    DRIVERS==""
    ATTRS{power/async}=="disabled"
    ATTRS{power/control}=="auto"
    ATTRS{power/runtime_active_kids}=="0"
    ATTRS{power/runtime_active_time}=="0"
    ATTRS{power/runtime_enabled}=="disabled"
    ATTRS{power/runtime_status}=="unsupported"
    ATTRS{power/runtime_suspended_time}=="0"
    ATTRS{power/runtime_usage}=="0"

I have two devices (event9 and event10 (mouse and touchpad, respectively)) and possibly 3 depending on if the Asus WMI hotkeys is to be considered in all of this, but I only created a rule for event9 so far.  It's definitely not the solution, since the problem has some up again while writing this response.

In answer to your question about typing while dragging, the answer is that whether or not I sometimes type while using the touchpad, this is certainly happening when I am not using the keyboard at all.

Last edited by jdoggsc (2021-03-13 05:08:09)

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I put together separate udev rules for the mouse device and the touchpad device.  I referenced the link sent by V1del above. When I create a rule for udev to disable the mouse, I can't tell a difference--touchpad still works. When i create the rule for the touchpad, then the touchpad is rendered inoperable.  For now, I have just created a rule to disable the mouse, since I can't tell that the mouse device is even usable.
The original issue still persists, though.  would asus_wmi even be the culprit here?  This isn't related to keyboard function keys...

here's /etc/udev/rules.d/52_asus_mouse.rules

ACTION=="add|change", KERNEL=="event9" ATTRS{name}=="ELAN1300:00 04F3:3104 Mouse", ENV{LIBINPUT_IGNORE_DEVICE}="1"

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I'm still trying to piece this all together (never had to troubleshoot much hardware on laptops, thankfully), and this is what I'm discovering:

- the kernel module is hid
- this module simply reads in the hardware frames over i2c or whatever, but that's not sufficient to know how to move the cursor on the screen.  The kernel assigns dev files to the hardware.  In this case, /dev/input/event*.

- xorg has conf files in /usr/share/X11/xorg.conf.d, which is what binds an xorg driver to one of these /dev/input/event* input events from the kernel.   40-libinput.conf is only on such configuration (albeit the most popular now, from what I can surmise). Any driver provider (synaptics/libinput/other) can be candidates at this point.
  -libinput is assigned /dev/input/event* devices by 40-libinput.conf, so it takes over control of my touchpad.
-the libinput utility (different from xf86-input-libinput) has several utilities that can be used to debug/record/replay libinput's actiity

# libinput debug-events

or

# libinput debug-gui

These are both very interesting to watch while giving input on the trackpad--particularly the gui tool.

WHAT I'VE NOTICED:
when my trackpad stops responding to finger taps and drags, both debug commands above show that no activity is happening.  My immediate impression of this is that libinput is not getting any updates from the /dev/input/event9 file, which means the kernel is not providing input updates at that interface.

Although I got nothing from the libinputs debug-events command except for a dearth of updated coordinates while my finger was moving), I DID notice an error on stdout after closing the libinput debug-gui command:

event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 20ms, your system is too slow

Aha! Some feedback about the hardware!  But what to do about it now?
Following that lead, I found this other post which dead-ends because the OP's problem seems to have mysteriously solved itself...
The thread did point me to Xorg.0.log where I found similar messages.  Not realizing how all these modules are stacked, I had not thought beforehand to check Xorg.0.log for this stuff.

[   156.266] (EE) event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 12ms, your system is too slow
[   161.288] (EE) event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 12ms, your system is too slow
[   266.758] (EE) event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 11ms, your system is too slow
[   650.460] (EE) event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 11ms, your system is too slow
[   715.751] (EE) event9  - ELAN1300:00 04F3:3104 Touchpad: client bug: event processing lagging behind by 14ms, your system is too slow

My CPU is an AMD Ryzen5 4500U and this happens even when my system is idle.  It seems inconceivable that it wouldn't be able to give input processing due attention.  I never had this problem on 10-year-old i3 laptop that I died before I just bought this one...
I'll have to check into this more later. I'm falling asleep as I type this out right now.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I spent some time looking through the libinput source code, but not enough time to really get an idea of how it's put together or what's going on.  I was satisfied to know that libinput is reading events from the kernel and responding appropriately.

As I mentioned in my previous post, the stream at the kernel event handle (/dev/input/event*) simply stops, and so libinput has no information to interpret.

I looked into the event handle.  If one simple runs

$ cat /dev/input/event9

then the output stream will react to whatever is happening on the touchpad.  When a finger is on the pad--even if not moving at all--the stream is constant as it is reporting the simple presence of the finger on the pad.  When all fingers are removed, the output stream is silent/stops.
I checked the output of

# libinput debug-events

with the output of

cat /dev/input/event9

and checked to see if both fell silent at the same time.  If they didn't, then libinput is to blame.  If they did, then the kernel is the culprit.  They both fell silent at the same time.

So my conclusion is that the kernel is simply not reporting the touchpad input randomly, for several seconds at a time.  I therefore have opened a kernel bug report here at bugzilla.kernel.org

linos

Member

Registered: 2021-01-03

Posts: 8

Re: Touchpad freezes intermittently

I have exactly the same issue. I went through the same debug steps as you, before finally finding out about this thread.

Though only difference: I have a similar but slightly different hardware.

I have an Lenovo Yoga Slim 7 with AMD Ryzen 4800U an the Touchpad is a

ELAN0634:00 04F3:3124

Thanks for sharing your procedure with us here.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I'm sorry that you're having this issue, too. It makes the computer quite frustrating to use! 

Thanks for chiming-in on bugzilla.  There was good response when I first opened the ticket, but it's been silent for several days. Maybe with multiple people experiencing the problem they will show interest in it again.

Last edited by jdoggsc (2021-01-05 03:27:50)

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I'm concerned that all the back-and-forth on bugzilla.kernel.org will eventually be so much that devs won't want to invest the time to sift through out conversation to understand where this topic is at.  For that reason, I'd like to post discussion related to it here if it's not known to be a productive bit of information that devs can use to help narrow down the cause as well as a solution.

With that in mind, I added "i8042.noloop i8042.nomux i8042.nopnp i8042.reset" to my kernel boot options and rebooted. I still have the problem.  I had tried .noloop and .reset about a month ago, but hadn't tried .nopnp because I guess I didn't know if the AUX port disables was actually correct.

Continuing down the rabbit hole I've been going down in adding debug messages to the i2c-hid module code,  I'm getting this info:

[ 1078.254196] i2c_hid: entered i2c_hid_init_irq
[ 1078.254199] i2c_hid: passing function call i2c_hid_irq to "request_threaded_irq" through function i2c_hid_init_irq
[ 1078.254200] ret = request_threaded_irq(client->irq, NULL, i2c_hid_irq,irqflags | IRQF_ONESHOT, client->name, ihid);
[ 1078.254201] client->irq: 76
[ 1078.254203] irqflags: 0
[ 1078.254204] client->name: ELAN1300:00

The point I was trying to find was client->irq: 76.

The number assigned is not guaranteed to be the same at each boot, but for this session, it was enough so that I could go to /sys/kernel/irq/76/ and cat each of the files in that list
(one could have just as soon found it by cd'ing to /sys/kernel/irq and running

grep -r "ELAN"

.
I got here because I'm trying to hunt down the source of the interrupt so that I can hopefully identify what might be preventing the interrupt.

$ cat /sys/kernel/irq/76/hwirq
16

The kernel interrupt (76 for this boot session) is mapped to hardware interrupt (hwirq) 16.  I've found it's consistently 16 on my computer. I'm still learning about this all, so I don't really know what its significance is

so this is the trail I've found.  Hw interrupt (hwirq) 16---> maps to kernel IRQ 76---> which is linked to client->name ($ cat /sys/kernel/irq/76/action)---> which will trigger a call to function i2c_hid_irq() in i2c-hid-core.c --> which will call i2c_hid_get_input() in i2c-hid-core.c, which will call i2c_master_recv(), which will return the 16-byte buffer of data that libinput can use.

some other interesting info can be obtained by cat'ing /sys/kernel/debug/hid/0018\:04F3\:3104.0003/events.  It looks like actually deciphers the stream from i2c so you can see things like number of fingers on the pad, button pushes, and some positional information.  This stream of data stops when the cursor stops responding, so this is still downstream of the actual problem.

I have no idea if this a productive direction to be going, but I'm learning how this aspect of the kernel is put together, and I've learned how to recompile kernel modules and insert them back into the kernel

I'm still hunting for the point in code where Jiaxun Yang's comment about VPCCMD_R_TOUCHPAD might be relevant for our situation.

Last edited by jdoggsc (2021-03-05 05:07:42)

lepantella

Member

Registered: 2021-01-08

Posts: 1

Re: Touchpad freezes intermittently

Hi, I don't have any tech expertise but just want to confirm that I have the same touchpad problem. ELAN1300:00 04F3:3104 Touchpad on ASUS VivoBook 15 M513. On Windows it works totally fine so it's not a hardware issue. Is there any progress to this or is the source of the error still unknown?

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

Hey thanks for chiming in 
You have the exact same laptop that I have, and the same assessment--works fine in Windows.  Unfortunately, the cause is still unknown. Linos and I have been checking into things daily for the last week while we wait for support from kernel devs.

Any info you can dig up, even if you don't think you have relevant technical experience, is still helpful.

linos

Member

Registered: 2021-01-03

Posts: 8

Re: Touchpad freezes intermittently

Side note from me:
I had to blacklist the module "elan_i2c" to let "i2c_hid" control the touchpad. Otherwise it would not be recognized at all.
Booting with the kernel parameters "i8042.noloop i8042.nomux i8042.nopnp i8042.reset" (I may have to check which one is really necessary) I don't have to blacklist "elan_i2c" anymore and i8042 discovers an AUX port (touhpad).

I just found out about debug-info parameters of the corresponding kernel modules and hid-recorder and will see if I can collect some useful information here.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

hmm.  your laptop situation is slightly different from mine, then.  I never had elan_i2c load automatically--frankly, I didn't even know about it until your last post.

I loaded it to see if it would change anything, but I don't think it will:

$ sudo modprobe elan_i2c
$ lsmod | grep elan_i2c
elan_i2c               53248  0<----- it's not being used

In the course of finding what was what, I removed hid modules one-by-one to see if there was any impact to usability.  Since I determined that the following were unneeded, I blacklisted them (for now) to narrow the scope of what I'm investigating (I don't want to have to consider any interactions between them)
mac_hid:            (this is for apple computers.  useless for me)
hid_generic:       (this module WILL allow trackpad use if loaded instead of i2c_hid, but trackpad usability is quite poor.  every tap is considered a click, and touchpad touch location is considered an absolute coordinate on your screen.
hid:                     (technically this is the parent (?) module for i2c_hid, but there seemed to be no difference when it was unloaded. Unloading all modules with "hid" in the name disables cursor movement completely.  If you add i2c_hid back in, it does not pull hid back in with it)

if you run the following commands, it will allow you to enable debug output for i2c_hid without having to reboot and edit the kernel-load line in GRUB

# rmmod i2c_hid
# modprobe i2c_hid debug=1

you'll see a lot more happening in dmesg

I assume you can do the same with i8042, but I didn't try it this way--I just added the line to GRUB and rebooted

Last edited by jdoggsc (2021-01-08 17:02:25)

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

Hi, I don't have any tech expertise but just want to confirm that I have the same touchpad problem. ELAN1300:00 04F3:3104 Touchpad on ASUS VivoBook 15 M513. On Windows it works totally fine so it's not a hardware issue. Is there any progress to this or is the source of the error still unknown?

Sounds like you've been dealing with this possibly longer than I have.  You come from Ubuntu and register on the Manjaro and arch forums just so you can subscribe to these threads and keep abreast of any progress that's being made.  Clearly this is driving you crazy, too. I was hoping Coiby on the Manjaro thread was on to something but that lead went cold 2 months ago.  I only found that thread today.  His comment about the AMD GPIO interrupt controller driver is interesting.  To me, this is clearly an issue with how the kernel is processing the interrupts, and all of us apparently have AMD processors (4th-gen Ryzen, IIRC), so that was interesting to hear him echo a similar sentiment.  He clearly knows more about it, though.  The problem could be with the AMD drivers and have nothing to do with the elantech touchpads...

linos

Member

Registered: 2021-01-03

Posts: 8

Re: Touchpad freezes intermittently

II just tried out linux 5.11 rc2, both on manjaro and arch. The changelog of Linux 5.11 seems like 2/3 of the commits are concerning the new AMD Platforms.

Quick summary: My touchpad is recognized but doesn't work at all. Could you guys try as well, an lets share observations? I will write details later in the evening.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

I downloaded the source from kernel.org and compiled with a pkgbuild as described in the the wiki, but I had problems booting.  I had problems with character sets, so I couldn't mount /boot, and there were other problems related to amdgpu.  I couldn't get as far as booting to a GUI.  Could you share your PKGBUILD?

Last edited by jdoggsc (2021-01-09 22:25:44)

loqs

Member

Registered: 2014-03-06

Posts: 18,047

Re: Touchpad freezes intermittently

jdoggsc please post the command you used to mount / boot and its full output.  Did you disable any modules in the config or use make modconfig to reduce the number of modules built?  Those would be my initial thoughts as to cause.

Last edited by loqs (2021-01-09 22:34:20)

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

jdoggsc please post the command you used to mount / boot and its full output.  Did you disable any modules in the config or use make modconfig to reduce the number of modules built?  Those would be my initial thoughts as to cause.

Sorry, but I'm not around my laptop right now. I can answer some of those questions, though. I pulled the config from /proc/config.gz used for the arch repo kernel 5.10.4. I gunzipped it in the PKGBUILD folder to use in the script.  I didn't change the config at all.

I manually modprobed bday so I could mount /boot using

 mount /boot

since it's in the fstab (which I had access to in the recovery console).  It complained about
"IO charset iso88559-1 not found", so it couldn't process the line for /boot in fstab

I don't recall what exactly the error was that was related to amdgpu

loqs

Member

Registered: 2014-03-06

Posts: 18,047

Re: Touchpad freezes intermittently

When you can check the config the relevant option for the charset is CONFIG_NLS_ISO8859_1 /  `modinfo nls_iso8859-1`

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

$ cat .config | grep CONFIG_NLS_ISO8859_1
CONFIG_NLS_ISO8859_1=m
CONFIG_NLS_ISO8859_13=m
CONFIG_NLS_ISO8859_14=m
CONFIG_NLS_ISO8859_15=m

not sure what m is supposed to represent, but "M" in

 make menuconfig

is the same as selecting the option.  so it's not like this option was deselected.  I think it's a symptom of something else that's wrong.
Here's the first part of my PKGBUILD

pkgbase=linux-custom
pkgver=5.11.rc2
pkgrel=1
pkgdesc='Linux'
_srctag=v${pkgver%.*}-${pkgver##*.}
url="https://git.archlinux.org/linux.git/log/?h=$_srctag"
arch=(x86_64)
license=(GPL2)
makedepends=(
  bc kmod libelf pahole cpio perl tar xz
  xmlto python-sphinx python-sphinx_rtd_theme graphviz imagemagick
  git
)
options=('!strip')
_srcname=linux-5.11-rc2
source=(
  "https://git.kernel.org/torvalds/t/linux-5.11-rc2.tar.gz"
)
validpgpkeys=(
  'ABAF11C65A2970B130ABE3C479BE3E4300411886'  # Linus Torvalds
  '647F28654894E3BD457199BE38DBBDC86092693E'  # Greg Kroah-Hartman
  'A2FF3A36AAA56654109064AB19802F8B0D70FC30'  # Jan Alexander Steffens (heftig)
)
sha256sums=('SKIP')

export KBUILD_BUILD_HOST=archlinux
export KBUILD_BUILD_USER=$pkgbase
export KBUILD_BUILD_TIMESTAMP="$(date -Ru${SOURCE_DATE_EPOCH:+d @$SOURCE_DATE_EPOCH})"

prepare() {
  cd $_srcname

I didn't touch anything below that.
Really, the only things I changed at all were pkgbase, pkgver ,source, shs256sums, and _srcname.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

5.11 rc2 is promising!

After getting past some issues with booting, and rebooting to recover from  the divide-by-zero error that caused my system to fall into a strangely semi-responsive state (below), I've been using the computer for about an hour now and haven't noticed anything amiss with the touchpad.

In the 2.5 months that I've had this laptop, there has been exactly one other time where I have gone this long without experiencing any touchpad problems.  Obviously, since this feels like a delicious breath of fresh air to be able to freely direct the cursor around the screen, I'm going to keep using 5.11 rc2 for a while.  I hope to be able to determine if this is the light at the end of the tunnel, or if this moment of good behavior has been purely circumstantial.

the code below is more a commentary about the state of 5.11 rc2 than anything else. 

Jan 09 22:56:22 kernel: divide error: 0000 [#1] PREEMPT SMP NOPTI
Jan 09 22:56:22 kernel: CPU: 4 PID: 816 Comm: Xorg Tainted: G            E     5.11.0-rc2-1-custom #1
Jan 09 22:56:22 kernel: Hardware name: ASUSTeK COMPUTER INC. VivoBook_ASUSLaptop X513IA_M513IA/X513IA, BIOS X513IA.303 07/08/2020
Jan 09 22:56:22 kernel: RIP: 0010:CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel: Code: 00 00 00 00 48 8b 84 24 c0 00 00 00 c7 00 00 00 00 00 84 d2 0f 84 0f 03 00 00 41 89 f0 31 c9 41 be 00 01 00 00 44 89 f0 31 d2 <f7> f3 31 d2 41 f7 f0 48 8b 94 24 b8 00 00 00 89 02 40 84 ed 0f 84
Jan 09 22:56:22 kernel: RSP: 0018:ffffa0b600ee35b0 EFLAGS: 00010246
Jan 09 22:56:22 kernel: RAX: 0000000000010000 RBX: 0000000000000000 RCX: 0000000000000000
Jan 09 22:56:22 kernel: RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffff8c5605382090
Jan 09 22:56:22 kernel: RBP: 0000000000000001 R08: 0000000000000080 R09: 0000000000000000
Jan 09 22:56:22 kernel: R10: 0000000000000000 R11: ffff8c560538a058 R12: 0000000000000000
Jan 09 22:56:22 kernel: R13: ffff8c5605388e58 R14: 0000000000010000 R15: 0000000000000019
Jan 09 22:56:22 kernel: FS:  00007f18357f82c0(0000) GS:ffff8c5707700000(0000) knlGS:0000000000000000
Jan 09 22:56:22 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 09 22:56:22 kernel: CR2: 00007f3f19344f60 CR3: 000000019b562000 CR4: 0000000000350ee0
Jan 09 22:56:22 kernel: Call Trace:
Jan 09 22:56:22 kernel:  dml21_ModeSupportAndSystemConfigurationFull+0x3857/0x5d10 [amdgpu]
Jan 09 22:56:22 kernel:  ? kfree+0x422/0x460
Jan 09 22:56:22 kernel:  ? __update_load_avg_cfs_rq+0x28c/0x310
Jan 09 22:56:22 kernel:  ? dcn20_populate_dml_pipes_from_context+0xa84/0xc00 [amdgpu]
Jan 09 22:56:22 kernel:  dml_get_voltage_level+0x138/0x1d0 [amdgpu]
Jan 09 22:56:22 kernel:  dcn21_validate_bandwidth+0x16e/0x720 [amdgpu]
Jan 09 22:56:22 kernel:  dc_validate_global_state+0x2f2/0x390 [amdgpu]
Jan 09 22:56:22 kernel:  amdgpu_dm_atomic_check+0xbb3/0xcc0 [amdgpu]
Jan 09 22:56:22 kernel:  drm_atomic_check_only+0x5db/0x8d0 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_object_put.part.0+0x1f/0x50 [drm]
Jan 09 22:56:22 kernel:  ? drm_atomic_set_property+0x8b/0xb30 [drm]
Jan 09 22:56:22 kernel:  drm_atomic_commit+0x13/0x50 [drm]
Jan 09 22:56:22 kernel:  drm_mode_obj_set_property_ioctl+0x156/0x3d0 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_obj_find_prop_id+0x40/0x40 [drm]
Jan 09 22:56:22 kernel:  drm_ioctl_kernel+0xb2/0x100 [drm]
Jan 09 22:56:22 kernel:  drm_ioctl+0x215/0x390 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_obj_find_prop_id+0x40/0x40 [drm]
Jan 09 22:56:22 kernel:  amdgpu_drm_ioctl+0x49/0x80 [amdgpu]
Jan 09 22:56:22 kernel:  __x64_sys_ioctl+0x83/0xb0
Jan 09 22:56:22 kernel:  do_syscall_64+0x33/0x40
Jan 09 22:56:22 kernel:  entry_SYSCALL_64_after_hwframe+0x44/0xa9
Jan 09 22:56:22 kernel: RIP: 0033:0x7f18361c4f6b
Jan 09 22:56:22 kernel: Code: 89 d8 49 8d 3c 1c 48 f7 d8 49 39 c4 72 b5 e8 1c ff ff ff 85 c0 78 ba 4c 89 e0 5b 5d 41 5c c3 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d d5 ae 0c 00 f7 d8 64 89 01 48
Jan 09 22:56:22 kernel: RSP: 002b:00007ffc46d8da28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
Jan 09 22:56:22 kernel: RAX: ffffffffffffffda RBX: 00007ffc46d8da60 RCX: 00007f18361c4f6b
Jan 09 22:56:22 kernel: RDX: 00007ffc46d8da60 RSI: 00000000c01864ba RDI: 000000000000000d
Jan 09 22:56:22 kernel: RBP: 00000000c01864ba R08: 0000000000000066 R09: 00000000cccccccc
Jan 09 22:56:22 kernel: R10: 0000000000000fff R11: 0000000000000246 R12: 00005619da62cf20
Jan 09 22:56:22 kernel: R13: 000000000000000d R14: 0000000000000000 R15: 0000000000000003
Jan 09 22:56:22 kernel: Modules linked in: rfcomm algif_hash algif_skcipher af_alg bnep md4 cmac ccm intel_rapl_msr intel_rapl_common btusb edac_mce_amd uvcvideo kvm_amd btrtl btbcm videobuf2_vmalloc btintel videobuf2_memops>
Jan 09 22:56:22 kernel:  acpi_cpufreq pinctrl_amd rng_core pkcs8_key_parser snd_aloop snd_pcm snd_timer snd soundcore videodev drm mc sg fuse crypto_user agpgart ip_tables x_tables ext4 crc32c_generic crc16 mbcache jbd2 xhci>
Jan 09 22:56:22 kernel: ---[ end trace 0c771852397a2e85 ]---
Jan 09 22:56:22 kernel: ------------[ cut here ]------------
Jan 09 22:56:22 kernel: WARNING: CPU: 4 PID: 816 at kernel/rcu/tree_plugin.h:297 rcu_note_context_switch+0x42/0x480
Jan 09 22:56:22 kernel: Modules linked in: rfcomm algif_hash algif_skcipher af_alg bnep md4 cmac ccm intel_rapl_msr intel_rapl_common btusb edac_mce_amd uvcvideo kvm_amd btrtl btbcm videobuf2_vmalloc btintel videobuf2_memops>
Jan 09 22:56:22 kernel:  acpi_cpufreq pinctrl_amd rng_core pkcs8_key_parser snd_aloop snd_pcm snd_timer snd soundcore videodev drm mc sg fuse crypto_user agpgart ip_tables x_tables ext4 crc32c_generic crc16 mbcache jbd2 xhci>
Jan 09 22:56:22 kernel: CPU: 4 PID: 816 Comm: Xorg Tainted: G      D     E     5.11.0-rc2-1-custom #1
Jan 09 22:56:22 kernel: Hardware name: ASUSTeK COMPUTER INC. VivoBook_ASUSLaptop X513IA_M513IA/X513IA, BIOS X513IA.303 07/08/2020
Jan 09 22:56:22 kernel: RIP: 0010:rcu_note_context_switch+0x42/0x480
Jan 09 22:56:22 kernel: Code: c3 c0 d1 02 00 65 48 03 1d 53 bd ef 44 0f 1f 44 00 00 40 84 ed 75 15 65 48 8b 04 25 c0 7b 01 00 8b 90 fc 03 00 00 85 d2 7e 02 <0f> 0b 65 48 8b 04 25 c0 7b 01 00 8b 80 fc 03 00 00 85 c0 7e 0a 41
Jan 09 22:56:22 kernel: RSP: 0018:ffffa0b600ee3000 EFLAGS: 00010002
Jan 09 22:56:22 kernel: RAX: ffff8c5603a31f40 RBX: ffff8c570772d1c0 RCX: 0000000000000000
Jan 09 22:56:22 kernel: RDX: 0000000000000001 RSI: ffffffffbba9316b RDI: 0000000000000000
Jan 09 22:56:22 kernel: RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000400
Jan 09 22:56:22 kernel: R10: abcc77118461cefd R11: ffffa0b600ee32ef R12: ffff8c570772c3c0
Jan 09 22:56:22 kernel: R13: ffff8c5603a31f40 R14: ffff8c5603a31f40 R15: ffff8c5603a1d000
Jan 09 22:56:22 kernel: FS:  00007f18357f82c0(0000) GS:ffff8c5707700000(0000) knlGS:0000000000000000
Jan 09 22:56:22 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 09 22:56:22 kernel: CR2: 00007f3f19344f60 CR3: 000000019b562000 CR4: 0000000000350ee0
Jan 09 22:56:22 kernel: Call Trace:
Jan 09 22:56:22 kernel:  __schedule+0xac/0x8b0
Jan 09 22:56:22 kernel:  ? sched_clock_cpu+0x58/0xb0
Jan 09 22:56:22 kernel:  schedule+0x5b/0xc0
Jan 09 22:56:22 kernel:  schedule_timeout+0x11c/0x160
Jan 09 22:56:22 kernel:  wait_for_completion+0x9e/0x100
Jan 09 22:56:22 kernel:  virt_efi_query_variable_info+0x141/0x150
Jan 09 22:56:22 kernel:  efi_query_variable_store+0x7b/0x1e0
Jan 09 22:56:22 kernel:  ? vsnprintf+0x6e/0x4f0
Jan 09 22:56:22 kernel:  efivar_entry_set_safe+0xbd/0x210
Jan 09 22:56:22 kernel:  efi_pstore_write+0x12d/0x1a0
Jan 09 22:56:22 kernel:  pstore_dump+0x1b3/0x2f0
Jan 09 22:56:22 kernel:  kmsg_dump+0xa8/0xd0
Jan 09 22:56:22 kernel:  oops_end+0x61/0xd0
Jan 09 22:56:22 kernel:  do_trap+0xcb/0x120
Jan 09 22:56:22 kernel:  ? CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel:  do_error_trap+0x64/0xa0
Jan 09 22:56:22 kernel:  ? CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel:  exc_divide_error+0x36/0x50
Jan 09 22:56:22 kernel:  ? CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel:  asm_exc_divide_error+0x12/0x20
Jan 09 22:56:22 kernel: RIP: 0010:CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel: Code: 00 00 00 00 48 8b 84 24 c0 00 00 00 c7 00 00 00 00 00 84 d2 0f 84 0f 03 00 00 41 89 f0 31 c9 41 be 00 01 00 00 44 89 f0 31 d2 <f7> f3 31 d2 41 f7 f0 48 8b 94 24 b8 00 00 00 89 02 40 84 ed 0f 84
Jan 09 22:56:22 kernel: RSP: 0018:ffffa0b600ee35b0 EFLAGS: 00010246
Jan 09 22:56:22 kernel: RAX: 0000000000010000 RBX: 0000000000000000 RCX: 0000000000000000
Jan 09 22:56:22 kernel: RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffff8c5605382090
Jan 09 22:56:22 kernel: RBP: 0000000000000001 R08: 0000000000000080 R09: 0000000000000000
Jan 09 22:56:22 kernel: R10: 0000000000000000 R11: ffff8c560538a058 R12: 0000000000000000
Jan 09 22:56:22 kernel: R13: ffff8c5605388e58 R14: 0000000000010000 R15: 0000000000000019
Jan 09 22:56:22 kernel:  dml21_ModeSupportAndSystemConfigurationFull+0x3857/0x5d10 [amdgpu]
Jan 09 22:56:22 kernel:  ? kfree+0x422/0x460
Jan 09 22:56:22 kernel:  ? __update_load_avg_cfs_rq+0x28c/0x310
Jan 09 22:56:22 kernel:  ? dcn20_populate_dml_pipes_from_context+0xa84/0xc00 [amdgpu]
Jan 09 22:56:22 kernel:  dml_get_voltage_level+0x138/0x1d0 [amdgpu]
Jan 09 22:56:22 kernel:  dcn21_validate_bandwidth+0x16e/0x720 [amdgpu]
Jan 09 22:56:22 kernel:  dc_validate_global_state+0x2f2/0x390 [amdgpu]
Jan 09 22:56:22 kernel:  amdgpu_dm_atomic_check+0xbb3/0xcc0 [amdgpu]
Jan 09 22:56:22 kernel:  drm_atomic_check_only+0x5db/0x8d0 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_object_put.part.0+0x1f/0x50 [drm]
Jan 09 22:56:22 kernel:  ? drm_atomic_set_property+0x8b/0xb30 [drm]
Jan 09 22:56:22 kernel:  drm_atomic_commit+0x13/0x50 [drm]
Jan 09 22:56:22 kernel:  drm_mode_obj_set_property_ioctl+0x156/0x3d0 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_obj_find_prop_id+0x40/0x40 [drm]
Jan 09 22:56:22 kernel:  drm_ioctl_kernel+0xb2/0x100 [drm]
Jan 09 22:56:22 kernel:  drm_ioctl+0x215/0x390 [drm]
Jan 09 22:56:22 kernel:  ? drm_mode_obj_find_prop_id+0x40/0x40 [drm]
Jan 09 22:56:22 kernel:  amdgpu_drm_ioctl+0x49/0x80 [amdgpu]
Jan 09 22:56:22 kernel:  __x64_sys_ioctl+0x83/0xb0
Jan 09 22:56:22 kernel:  do_syscall_64+0x33/0x40
Jan 09 22:56:22 kernel:  entry_SYSCALL_64_after_hwframe+0x44/0xa9
Jan 09 22:56:22 kernel: RIP: 0033:0x7f18361c4f6b
Jan 09 22:56:22 kernel: Code: 89 d8 49 8d 3c 1c 48 f7 d8 49 39 c4 72 b5 e8 1c ff ff ff 85 c0 78 ba 4c 89 e0 5b 5d 41 5c c3 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d d5 ae 0c 00 f7 d8 64 89 01 48
Jan 09 22:56:22 kernel: RSP: 002b:00007ffc46d8da28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
Jan 09 22:56:22 kernel: RAX: ffffffffffffffda RBX: 00007ffc46d8da60 RCX: 00007f18361c4f6b
Jan 09 22:56:22 kernel: RDX: 00007ffc46d8da60 RSI: 00000000c01864ba RDI: 000000000000000d
Jan 09 22:56:22 kernel: RBP: 00000000c01864ba R08: 0000000000000066 R09: 00000000cccccccc
Jan 09 22:56:22 kernel: R10: 0000000000000fff R11: 0000000000000246 R12: 00005619da62cf20
Jan 09 22:56:22 kernel: R13: 000000000000000d R14: 0000000000000000 R15: 0000000000000003
Jan 09 22:56:22 kernel: ---[ end trace 0c771852397a2e86 ]---
Jan 09 22:56:22 kernel: RIP: 0010:CalculateVMAndRowBytes.constprop.0+0x449/0xb20 [amdgpu]
Jan 09 22:56:22 kernel: Code: 00 00 00 00 48 8b 84 24 c0 00 00 00 c7 00 00 00 00 00 84 d2 0f 84 0f 03 00 00 41 89 f0 31 c9 41 be 00 01 00 00 44 89 f0 31 d2 <f7> f3 31 d2 41 f7 f0 48 8b 94 24 b8 00 00 00 89 02 40 84 ed 0f 84
Jan 09 22:56:22 kernel: RSP: 0018:ffffa0b600ee35b0 EFLAGS: 00010246
Jan 09 22:56:22 kernel: RAX: 0000000000010000 RBX: 0000000000000000 RCX: 0000000000000000
Jan 09 22:56:22 kernel: RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffff8c5605382090
Jan 09 22:56:22 kernel: RBP: 0000000000000001 R08: 0000000000000080 R09: 0000000000000000
Jan 09 22:56:22 kernel: R10: 0000000000000000 R11: ffff8c560538a058 R12: 0000000000000000
Jan 09 22:56:22 kernel: R13: ffff8c5605388e58 R14: 0000000000010000 R15: 0000000000000019
Jan 09 22:56:22 kernel: FS:  00007f18357f82c0(0000) GS:ffff8c5707700000(0000) knlGS:0000000000000000
Jan 09 22:56:22 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 09 22:56:22 kernel: CR2: 00007f3f19344f60 CR3: 000000019b562000 CR4: 0000000000350ee0

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

of course right after I push the "submit" button it acts up again

linos

Member

Registered: 2021-01-03

Posts: 8

Re: Touchpad freezes intermittently

Happend to me as well, but right before I clicked submit. It's so frustrating, I don't know how to keep working on this.

jdoggsc

Member

Registered: 2011-08-15

Posts: 98

Re: Touchpad freezes intermittently

Same here. Did you just try 5.11-rc4?  I was going to try that today.

When you said 2/3 of the 5.11 commits were related to AMD, where did you find that info?  You really got my hopes up when you said that, so I want to look through them and see if any of the are related to the AMD interrupt handlers like Coiby mentioned was a known problem.

linos

Member

Registered: 2021-01-03

Posts: 8

Re: Touchpad freezes intermittently

I don't remember the topic, but read about the changes of 5.11 somewhere on kernel.org.

One additional observation: while the bug happens, and the cursor stops moving, until I had lifted all finger, hid-recorder still shows "normal" input.