Kernels post 5.0.21 breaks Dell XPS 13" 9360 (2016/2017) trackpad

Tap to click does not work and scrolling direction does nothing

Rather the scrolling direction option does not change the scrolling direction

Have a look here
https://wiki.archlinux.org/index.php/Dell_XPS_13_(9360)#Touchpad

I am not getting any errors when i run dmesg | grep -i psmouse and the trackpad works fine with 5.0.2. This bug occurs with 5.1 and I'm able to reproduce it

Checking in again here. The XPS13 is one of the premier go-to linux laptops, so I'm a bit surprised by this regression. The kernel that operates the trackpad properly is no longer supported. Can anyone confirm tap to click and scroll direction is is working on the XPS 9360 post 5.0.2?

so 4.19 LTS is no good for the touchpad either? try it with the XPS tweaks bogdan links to above.

Here is the dmesg from 5.0.21

[ 2.852484] psmouse serio1: synaptics: queried min coordinates: x [1276..], y [1118..]
[ 2.852494] psmouse serio1: synaptics: Your touchpad (PNP: DLL082a PNP0f13) says it can support a different bus. If i2c-hid and hid-rmi are not used, you might want to try setting psmouse.synaptics_intertouch to 1 and report this to linux-input@vger.kernel.org.
[ 2.852786] iTCO_wdt: Intel TCO WatchDog Timer Driver v1.11
[ 2.852940] iTCO_wdt: Found a Intel PCH TCO device (Version=4, TCOBASE=0x0400)
[ 2.853099] iTCO_wdt: initialized. heartbeat=30 sec (nowayout=0)
[ 2.857478] uvcvideo 1-5:1.0: Entity type for entity Extension 4 was not initialized!
[ 2.857481] uvcvideo 1-5:1.0: Entity type for entity Extension 3 was not initialized!
[ 2.857483] uvcvideo 1-5:1.0: Entity type for entity Processing 2 was not initialized!
[ 2.857485] uvcvideo 1-5:1.0: Entity type for entity Camera 1 was not initialized!
[ 2.857574] input: Integrated_Webcam_HD: Integrate as /devices/pci0000:00/0000:00:14.0/usb1/1-5/1-5:1.0/input/input14
[ 2.857650] usbcore: registered new interface driver uvcvideo
[ 2.857651] USB Video Class driver (1.1.1)
[ 2.876111] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:1f.3/sound/card0/input13
[ 2.877914] input: HDA Intel PCH Headphone Mic as /devices/pci0000:00/0000:00:1f.3/sound/card0/input15
[ 2.877999] input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input16
[ 2.878102] input: HDA Intel PCH HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input17
[ 2.878192] input: HDA Intel PCH HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input18
[ 2.878292] input: HDA Intel PCH HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input19
[ 2.878381] input: HDA Intel PCH HDMI/DP,pcm=10 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input20
[ 2.909247] psmouse serio1: synaptics: Touchpad model: 1, fw: 8.2, id: 0x1e2a1, caps: 0xf00323/0x840300/0x12e800/0x0, board id: 3038, fw id: 2375007

Here is the dmesg from 5.1.16

[ 2.865082] psmouse serio1: synaptics: queried max coordinates: x [..5666], y [..4734]
[ 2.879094] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:1f.3/sound/card0/input13
[ 2.879190] input: HDA Intel PCH Headphone Mic as /devices/pci0000:00/0000:00:1f.3/sound/card0/input15
[ 2.879270] input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input16
[ 2.879344] input: HDA Intel PCH HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input17
[ 2.879418] input: HDA Intel PCH HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input18
[ 2.879490] input: HDA Intel PCH HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input19
[ 2.879558] input: HDA Intel PCH HDMI/DP,pcm=10 as /devices/pci0000:00/0000:00:1f.3/sound/card0/input20
[ 2.898658] psmouse serio1: synaptics: queried min coordinates: x [1276..], y [1118..]
[ 2.898666] psmouse serio1: synaptics: Your touchpad (PNP: DLL082a PNP0f13) says it can support a different bus. If i2c-hid and hid-rmi are not used, you might want to try setting psmouse.synaptics_intertouch to 1 and report this to linux-input@vger.kernel.org.
[ 2.955245] psmouse serio1: synaptics: Touchpad model: 1, fw: 8.2, id: 0x1e2a1, caps: 0xf00323/0x840300/0x12e800/0x0, board id: 3038, fw id: 2375007

Thanks for responding, but I am not certain I understand how 4.19 helps. 5.0.21 works. Kernels after 5.0.21 do not work properly. This is a repeatable bug that I'm trying to illuminate on a very popular machine.

Most changes came in 5.1-rc1. We have to see if there are public reports similar to your problem. Is the regression still present in 5.2 kernel series?

I will check now

Still broken in 5.2.0-1

Well the reason I asked about 4.19 is because it is LTS and still supported for around 18 months. 5.0 series is end of life so you are vulnerable to unpatched security risks. 5.1 and 5.2 are mainline and not LTS kernels. At least if you do have to switch to something else you have support until December 2020 with 4.19. That should be more than enough time to fix this issue in the meantime. I suggest keeping and using 4.19 but have 5.2 alongside as well it for testing purposes with the touchpad until a resolution can be found.

Additionally, it is not necessary to run the latest kernel release just because it's there. Okay you may lose some power optimisations but if hardware works best with older LTS kernels then it's worth the trade off. 4.19 has backported hardware support so there should be no issues with any other component in your laptop.

Finally, Have you tried switching to the libinput driver from synaptics?

Seems you still use the synaptics driver. We recommend to unistall it. Normally libinput should be enough. See also here.

Wow. I am stunned and dumbfounded that the expectation is still so much on the enduser to research and solve Linux UI basics, especially against a known Linux laptop.

I took time to test and report this issue, and I can repeat my test results. The expectation is that I should lift a finger to get tap to click or scrolling to work on an XPS?

I appreciate the work everyone does, please don't get me wrong. Can we try a different approach?

Can someone verify if tap to click on kernels after 5.0.21 work or can future kernels detect and handle appropriately? Certainly this should not be an enduser responsibility, especially when 5.0.21 worked and continues to work fine.

It is not a UI issue, it's hardware compatibility.

Your laptop being available for shipment with Ubuntu LTS factory installed and drivers available for RHEL does not guarantee full functionality with all other or just newer distributions and kernel modules.

While you may not like spending any more time on this issue yourself, actually it is the only way it will get resolved because without your ongoing cooperation nothing further will happen here.

Well, please do a git-bisect on the 5.1 kernel series then and tell me the exact commit which might had introduced the regression. You can take a look here to see what was needed to fix an issue we had in the past for i686 architecture ...

1 Like

Yeah the best way to find the culprit. No wonder it is some merge commit! :face_with_hand_over_mouth:

Forum kindly sponsored by Bytemark