Bug 1256314
| Summary: | Touchpad won't work n HP Pavilion x360 11-n001ng | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Matthias31 <fiedler_matthias> |
| Component: | kernel | Assignee: | Kernel Maintainer List <kernel-maint> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 23 | CC: | bnocera, btissoir, fiedler_matthias, gansalmon, itamar, jonathan, kernel-maint, madhu.chinakonda, mchehab, michael.b.peters13, muhammad.saad, peter.hutterer |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2016-04-07 01:46:37 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Attachments: | |||
|
Description
Matthias31
2015-08-24 11:04:13 UTC
has this worked in f22 or did you only try f23? can you attach the dmesg output please? when you run evemu-record with no arguments, does it list a touchpad device? if so, what's the description when you select it? it did only on f23 not work. under f22 it is working without any problems Is this still a problem with the released version of f23? If it is still a problem, I'll need the evemu-describe from that touchpad. Created attachment 1102311 [details]
dmesg output
Hi, I think I havethe same (or similar) bug. HP x360 paviion. Fedora 23 fresh install. I attached my dmesg output. evemu doesn`t display any available devices at all. Initially, my pad an keyboard worked, but a suspend broke them. Restarts did nothing. Then i re-installed, and they didn`t work in the installer either,though they did the first time. Once booted they worked again, but suspend broke them. This time a reboot fixed them ...until the next reboot broke them. did you run evemu-describe as root? I'vbe tested the fedora 23 Mate and the fedora 23 KDE Spin on both keybord and touchpad are working. Created attachment 1103089 [details]
evemu-describe
Sorry, didn't run as root. Here's evemu-describe:
Available devices:
/dev/input/event0: Power Button
/dev/input/event1: Lid Switch
/dev/input/event2: Power Button
/dev/input/event3: AT Translated Set 2 keyboard
/dev/input/event4: Video Bus
/dev/input/event5: SynPS/2 Synaptics TouchPad
/dev/input/event6: HP Wireless hotkeys
/dev/input/event7: ST LIS3LV02DL Accelerometer
/dev/input/event8: ELAN0732:00 04F3:0358
/dev/input/event9: HDA Intel PCH Mic
/dev/input/event10: HDA Intel PCH Headphone
/dev/input/event11: HDA Intel PCH HDMI/DP,pcm=3
/dev/input/event12: HP WMI hotkeys
/dev/input/event13: HP Truevision HD
I'll attach the output of evemu-describe /dev/input/event5
@Michael: in my case nothing did work. No keybord no tuchbpad, and in BETA from f23 the tochscreen din't woork to. I must Plug an USB mouse and USB keybord to work with my HP x360 pavilion. Touchpad looks normal and should come up normal too (it does here when I emulate it). Does it show up when you run sudo libinput-debug-events, and does it show events as you use it? Can you attach your xorg.log and your dmesg? Created attachment 1103459 [details]
dmesg after keyboard and touchpad not working on boot
It looks like the devices always don't work on the first boot, but then work after a restart (even though they didn't the first time I installed), and then any suspend breaks them. I've attached dmesg, and I'll grab my xorg log too. I'll also attach the output of libinput-debug-events. When I ran it while the devices weren't working, it didn't display any events from them. Then I rebooted and they worked, and I've attached the output from that too. Let me know if you need anything else. Created attachment 1103461 [details]
xorg log
Created attachment 1103462 [details]
libinput-debug-events from a cold boot with devices not working
Created attachment 1103463 [details]
libinput-debug-events after reboot with devices working
The events are:
- Draging finger over touchpad
- Moving window using the touch-screen
- Hitting keys on the keyboard
Created attachment 1103464 [details]
Xorg log after suspend broke keyboard and touchpad
I decided to attach a xorg log from after a suspend in case that helps.
Comment #14 suggests that there are no events coming out of the kernel when the devices aren't working, so I'm punting this to the kernel for now. I should mention my laptop actually has a different model number to the original one in this bug report. It's a HP Pavilion x360 11-n116TU rather than a HP Pavilion x360 11-n001ng. Also, I just installed Fedora 23 MATE-Compiz theme and there was no sign of this problem, it worked fine (even suspend). Then I tried installing the group "Fedora Workstation" to install gnome, and the exact same bug re-appeared. Sorry, by "MATE-Compiz theme" I meant "MATE-Compiz spin". As in, I did a clean install from a disk. @peter i think it isn't the kernel, The MATE Spin , KDE Spin etc has the same kernel and the bug didn't appear weird. libinput-debug-events sits below the X server and receives the same events that the server would receive (and both can run at the same time). The only thing below it is the kernel itself, if libinput-debug-events doesn't see events then none are coming out of the kernel. Short of some really whacky udev mess I don't see any other trigger, but since the device shows up in both outputs (comments #14 and #15), it's not a udev setup issue. Let's try something else: do you see events with sudo evemu-record when the touchpad is dead? Running sudo evemu-record I don't see any events when they're dead. The events show up when they're working though. The other explanation is that when gnome does the suspend, it triggers a special (ACPI?) command that make the interrupt controller basically dead. Thus, you wouldn't get any input from keyboard/touchscreen/touchpad. It's however weird that MATE doesn't express the same behavior. I think it would worth testing a rawhide kernel (4.4-rc*) to see if there has not been any changes that would solve this. Okay... So. I was wondering what package specifically was triggering the bug when I installed gnome, and I think I managed to narrow it down to iio-sensor-proxy, which I guess makes sense. Installing the latest rawhide kernel 4.4.0-0.rc4* seems to fix the problem. If I have to switch back to the stable kernel for some reason, is there a way to disable the accelerometer driver or something? Cheers! If it really is a problem with iio-sensor-proxy, please look at getting the information in the README for it: https://github.com/hadess/iio-sensor-proxy This should tell us whether you have a detected accelerometer, and what it might be doing. If it's fixed in a newer kernel, this rather points to a kernel bug... Yep, there's definitely an accelerometer detected: [michael@192-168-1-18 ~]$ monitor-sensor ** Message: Accelerometer orientation changed: left-up ** Message: Accelerometer orientation changed: normal I'll attach the output of those other commands. Created attachment 1104397 [details]
udevadm info
Output of udevadm info --export-db
Created attachment 1104399 [details]
gdbus introspect
Output of:
gdbus introspect --system --dest net.hadess.SensorProxy --object-path /net/hadess/SensorProxy
Right, I don't see anything here that would point to a bug in iio-sensor-proxy per se, probably a bug in hid subsystem which handles both the accelerometer via the IIO subsystem, and the touchpads/keyboards. So I think that iio-sensor-proxy is rather a way to reproduce the bug rather than where it is. You can get touchpad and keybord running when you Download the rawhide kernel repo´ s from https://alt.fedoraproject.org/pub/alt/rawhide-kernel-nodebug/fedora-rawhide-kernel-nodebug.repo and copy this to /etc/yum.repos.d and run: dnf clean all dnf update wait until finished and reboot. Do not remove the stable kernel in case you need them later. Same problem on HP Notebook 1000 Touchpad does not work at cold boot. One thing I noticed in addition to the reboot thing: You can also make it work after a suspend/resume. Created attachment 1108323 [details]
dmesg output at cold boot
Output of dmesg right after a cold boot when the touchpad wasn't working.
Created attachment 1108324 [details]
dmesg output after suspend/resume
Output of dmesg after a suspend/resume. The touchpad started working after the resume.
Created attachment 1108325 [details]
evemu-record after a cold boot
Output of evemu-record after a cold boot. Device not working.
List of available devices shows touchpad as "ImPS/2 Synaptics TouchPad".
Available devices:
/dev/input/event0: Lid Switch
/dev/input/event1: Power Button
/dev/input/event2: Power Button
/dev/input/event3: AT Translated Set 2 keyboard
/dev/input/event4: ImPS/2 Synaptics TouchPad
/dev/input/event5: Video Bus
/dev/input/event6: HP Wireless hotkeys
/dev/input/event7: HDA Intel PCH Mic
/dev/input/event8: HDA Intel PCH Headphone
/dev/input/event9: HDA Intel PCH HDMI/DP,pcm=3
/dev/input/event10: HP WMI hotkeys
/dev/input/event11: HP TrueVision HD
Created attachment 1108326 [details]
evemu-record after suspend/resume
Output of evemu-record after a suspend/resume. Device working. Listed as "SynPS/2 Synaptics TouchPad".
Available devices:
/dev/input/event0: Lid Switch
/dev/input/event1: Power Button
/dev/input/event2: Power Button
/dev/input/event3: AT Translated Set 2 keyboard
/dev/input/event4: SynPS/2 Synaptics TouchPad
/dev/input/event5: Video Bus
/dev/input/event6: HP Wireless hotkeys
/dev/input/event7: HDA Intel PCH Mic
/dev/input/event8: HDA Intel PCH Headphone
/dev/input/event9: HDA Intel PCH HDMI/DP,pcm=3
/dev/input/event10: HP WMI hotkeys
/dev/input/event11: HP TrueVision HD
Created attachment 1108327 [details]
libinput-debug-events after a cold boot
Output of libinput-debug-events after a cold boot. Device not working. Listed as "ImPS/2 Synaptics TouchPad".
[root@hal9000 touchpad]libinput-debug-events
event2 DEVICE_ADDED Power Button seat0 default group1 cap:k
event5 DEVICE_ADDED Video Bus seat0 default group2 cap:k
event1 DEVICE_ADDED Power Button seat0 default group3 cap:k
event11 DEVICE_ADDED HP TrueVision HD seat0 default group4 cap:k
event3 DEVICE_ADDED AT Translated Set 2 keyboard seat0 default group5 cap:k
event4 DEVICE_ADDED ImPS/2 Synaptics TouchPad seat0 default group6 cap:p left scroll-nat scroll-button
event10 DEVICE_ADDED HP WMI hotkeys seat0 default group7 cap:k
event6 DEVICE_ADDED HP Wireless hotkeys seat0 default group8 cap:k
event3 KEYBOARD_KEY +19.97s KEY_LEFTCTRL (29) pressed
event3 KEYBOARD_KEY +20.46s KEY_LEFTCTRL (29) released
event3 KEYBOARD_KEY +22.65s KEY_LEFTCTRL (29) pressed
event3 KEYBOARD_KEY +22.75s KEY_C (46) pressed
Created attachment 1108329 [details]
libinput-debug-events after suspend/resume
Output of libinput-debug-events after suspend/resume. Device working. Listed as "SynPS/2 Synaptics TouchPad".
[root@hal9000 touchpad]# libinput-debug-events
event2 DEVICE_ADDED Power Button seat0 default group1 cap:k
event5 DEVICE_ADDED Video Bus seat0 default group2 cap:k
event1 DEVICE_ADDED Power Button seat0 default group3 cap:k
event11 DEVICE_ADDED HP TrueVision HD seat0 default group4 cap:k
event3 DEVICE_ADDED AT Translated Set 2 keyboard seat0 default group5 cap:k
event4 DEVICE_ADDED SynPS/2 Synaptics TouchPad seat0 default group6 cap:p size 95.66/42.89mm tap(dl off) left scroll-nat scroll-2fg-edge dwt-on
event10 DEVICE_ADDED HP WMI hotkeys seat0 default group7 cap:k
event6 DEVICE_ADDED HP Wireless hotkeys seat0 default group8 cap:k
event4 POINTER_MOTION +0.84s -1.33/ 2.84
event4 POINTER_MOTION +0.85s -8.11/ 8.48
event4 POINTER_MOTION +0.86s -22.58/ 13.72
event4 POINTER_MOTION +0.87s -35.94/ 14.30
event4 POINTER_MOTION +0.89s -34.68/ 13.57
event4 POINTER_MOTION +0.90s -21.12/ 8.99
event4 POINTER_MOTION +0.91s -14.17/ 5.26
@saad have you tryed to install the kernel-4.4 rc 5? @Matthias31 Not yet. Let me try. @saad tell me when you lucky or not, thank Merry Christmas @Matthias31 Looks like I am. It's working well so far. Thanks for the suggestion! :) maybe i am the men for redhat *lol* so i am happy when i can help you pls do not delete the original kernel in case you need him later Yes, I know. Usual precautions. :) So far I have turned the computer off and on 3 times and the bug did not appear again. Hopefully, it has been resolved. I wonder what exactly was causing it. @saad: yes it seems so as when them problem is gone, thats nice @Peter Hutterer: the kernel-4.4.0.0 shuld be proted for fedora 23 as soon as released, in the moment i see only the rc6 release for f24. You cann see that the Bug not appears when kernel-4-.4.0 is installed Unfortunately the kernel has its own timeline because it covers such a vast array of use-cases and devices. We can't push 4.4 to F23 because of a touchpad issue, but if you can find the patches that fixed the bug we can look at backporting them. @peter: the main thing is that is first found a workaround for the problem Fedora backported a series of patches for bug 1275718, so that may have fixed this issue. I'm currently on 4.4.6-300.fc23.x86_64, so given the comments 38 to 44 I'm closing this as fixed. |