Dell XPS 13 9350 touchpad is busted in 4.12-rc

Dell XPS 13 9350 touchpad is busted in 4.12-rc

[Andy Lutomirski]

I just build from Linus' tree from Saturday (ef82f1ad), and touchpad
input is busted on my laptop.  The device is still there and evtest
still shows events, but something is wrong and libinput either thinks
I'm not moving at all or I'm trying to teleport to the edge of the
screen.

Is this a known problem or should I try to track it down?

Thanks,
Andy

Re: Dell XPS 13 9350 touchpad is busted in 4.12-rc

[Jiri Kosina]

On Wed, 24 May 2017, Andy Lutomirski wrote:

> I just build from Linus' tree from Saturday (ef82f1ad), and touchpad
> input is busted on my laptop.  The device is still there and evtest
> still shows events, but something is wrong and libinput either thinks
> I'm not moving at all or I'm trying to teleport to the edge of the
> screen.
> 
> Is this a known problem or should I try to track it down?

Did this work properly in 4.11?

Is this G11 based chipset? (IOW, is this device driven by wacom driver?). 
If so, 6f107fab8 and/or 60a2218698 might be potential candidates.

Adding Jason, Aaron and Ping to CC.

-- 
Jiri Kosina
SUSE Labs

Re: Dell XPS 13 9350 touchpad is busted in 4.12-rc

[Andy Lutomirski]

On Wed, May 24, 2017 at 3:26 PM, Jiri Kosina <jikos@kernel.org> wrote:
> On Wed, 24 May 2017, Andy Lutomirski wrote:
>
>> I just build from Linus' tree from Saturday (ef82f1ad), and touchpad
>> input is busted on my laptop.  The device is still there and evtest
>> still shows events, but something is wrong and libinput either thinks
>> I'm not moving at all or I'm trying to teleport to the edge of the
>> screen.
>>
>> Is this a known problem or should I try to track it down?
>
> Did this work properly in 4.11?

I confess to not having tested 4.11, for 4.11-rc7 works.

>
> Is this G11 based chipset? (IOW, is this device driven by wacom driver?).
> If so, 6f107fab8 and/or 60a2218698 might be potential candidates.

I don't think so.

On a good kernel, I see:

/dev/input/event10:    DLL0704:01 06CB:76AE Touchpad

That thing is here, and the driver is hid-multitouch:

/sys/devices/pci0000:00/0000:00:15.1/i2c_designware.1/i2c-8/i2c-DLL0704:01/0018:06CB:76AE.0001/input/input13

On 4.12-rc, I see a SynPC/2 Synaptics Touchpad instead, attached to
i8042/serio1.

The culprit appears to be:

i2c_designware.1: controller timed out
i2c_hid i2c-DLL0704:01: hid_descr_cmd failed

I'm wondering whether something like
9143059fafd4eebed2d43ffb5455178d4010e60a could be involved.

Re: Dell XPS 13 9350 touchpad is busted in 4.12-rc

[Mika Westerberg]

On Wed, May 24, 2017 at 09:25:00PM -0700, Andy Lutomirski wrote:
> On Wed, May 24, 2017 at 3:26 PM, Jiri Kosina <jikos@kernel.org> wrote:
> > On Wed, 24 May 2017, Andy Lutomirski wrote:
> >
> >> I just build from Linus' tree from Saturday (ef82f1ad), and touchpad
> >> input is busted on my laptop.  The device is still there and evtest
> >> still shows events, but something is wrong and libinput either thinks
> >> I'm not moving at all or I'm trying to teleport to the edge of the
> >> screen.
> >>
> >> Is this a known problem or should I try to track it down?
> >
> > Did this work properly in 4.11?
> 
> I confess to not having tested 4.11, for 4.11-rc7 works.
> 
> >
> > Is this G11 based chipset? (IOW, is this device driven by wacom driver?).
> > If so, 6f107fab8 and/or 60a2218698 might be potential candidates.
> 
> I don't think so.
> 
> On a good kernel, I see:
> 
> /dev/input/event10:    DLL0704:01 06CB:76AE Touchpad
> 
> That thing is here, and the driver is hid-multitouch:
> 
> /sys/devices/pci0000:00/0000:00:15.1/i2c_designware.1/i2c-8/i2c-DLL0704:01/0018:06CB:76AE.0001/input/input13
> 
> On 4.12-rc, I see a SynPC/2 Synaptics Touchpad instead, attached to
> i8042/serio1.
> 
> The culprit appears to be:
> 
> i2c_designware.1: controller timed out
> i2c_hid i2c-DLL0704:01: hid_descr_cmd failed
> 
> I'm wondering whether something like
> 9143059fafd4eebed2d43ffb5455178d4010e60a could be involved.

You need the following, I think:

https://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git/commit/?h=i2c/for-current&id=e2c824924cdb41528932c550647406ad81336b18

Re: Dell XPS 13 9350 touchpad is busted in 4.12-rc

[Andy Lutomirski]

On Thu, May 25, 2017 at 12:25 AM, Mika Westerberg
<mika.westerberg@linux.intel.com> wrote:
> On Wed, May 24, 2017 at 09:25:00PM -0700, Andy Lutomirski wrote:
>> On Wed, May 24, 2017 at 3:26 PM, Jiri Kosina <jikos@kernel.org> wrote:
>> > On Wed, 24 May 2017, Andy Lutomirski wrote:
>> >
>> >> I just build from Linus' tree from Saturday (ef82f1ad), and touchpad
>> >> input is busted on my laptop.  The device is still there and evtest
>> >> still shows events, but something is wrong and libinput either thinks
>> >> I'm not moving at all or I'm trying to teleport to the edge of the
>> >> screen.
>> >>
>> >> Is this a known problem or should I try to track it down?
>> >
>> > Did this work properly in 4.11?
>>
>> I confess to not having tested 4.11, for 4.11-rc7 works.
>>
>> >
>> > Is this G11 based chipset? (IOW, is this device driven by wacom driver?).
>> > If so, 6f107fab8 and/or 60a2218698 might be potential candidates.
>>
>> I don't think so.
>>
>> On a good kernel, I see:
>>
>> /dev/input/event10:    DLL0704:01 06CB:76AE Touchpad
>>
>> That thing is here, and the driver is hid-multitouch:
>>
>> /sys/devices/pci0000:00/0000:00:15.1/i2c_designware.1/i2c-8/i2c-DLL0704:01/0018:06CB:76AE.0001/input/input13
>>
>> On 4.12-rc, I see a SynPC/2 Synaptics Touchpad instead, attached to
>> i8042/serio1.
>>
>> The culprit appears to be:
>>
>> i2c_designware.1: controller timed out
>> i2c_hid i2c-DLL0704:01: hid_descr_cmd failed
>>
>> I'm wondering whether something like
>> 9143059fafd4eebed2d43ffb5455178d4010e60a could be involved.
>
> You need the following, I think:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git/commit/?h=i2c/for-current&id=e2c824924cdb41528932c550647406ad81336b18

Yup, that works.  Thanks :)