Re: PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

Re: PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

[Tommy Will]

Hi Vasily,

Thanks for your mail & information. We would start the investigation
and give you feedback later.
Please wait.

> The issues occurs on every kernel after
> 1302bac33d9e88cd43e482191a806998f3ed43cc commit. Before the commit, this touchpad device was detected as PS/2 mouse and "cat /dev/input/mice" shows data when the touchpad is used.
> After the commit, there are no events from the touchpad in /dev/input/mice or /dev/input/mouseX files anymore.
>
> According to drivers/input/mouse/alps.c source file, all devices with signatures "ec[0] == 0x88 && ec[1] == 0x08" are treated as "Rushmore touchpads". This touchpad has E7 signature of "73-03-0A" and EC signature of "88-08-22", and it looks like that V5 protocol (not
> "Rushmore") should be used here. If V5 protocol is forced to use for this device then everything works fine.
>
> The patch is trivial, but I'm not sure how to properly separate "Rushmore" and "V5" devices. Are there any datasheets or experts to help me with it?
>

I checked the E7 & EC information and think it should be a Rushmore device.
Could you please let me know after you use V5 protocol, if touchpad
still be recognized as "AlpsPS/2 ALPS DualPoint TouchPad" or returns
to "PS/2 Generic Mouse" ?

Thanks
-- 
Best Regards,
Tommy

Re: PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

[Vasily Titskiy]

Hi Tommy,

This is the output of `dmesg | grep -i 'input\|alps'` command after V5 enforced:
[    1.592880] input: Lid Switch as
/devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input0
[    1.593928] input: Power Button as
/devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input1
[    1.593964] input: Sleep Button as
/devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
[    1.593999] input: Power Button as
/devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
[    1.788587] input: AT Translated Set 2 keyboard as
/devices/platform/i8042/serio0/input/input4
[    2.661783] input: PS/2 Mouse as /devices/platform/i8042/serio1/input/input5
[    2.674857] input: AlpsPS/2 ALPS GlidePoint as
/devices/platform/i8042/serio1/input/input6

...and before my patch (current kernel behavior):
[    1.604029] input: Lid Switch as
/devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input0
[    1.606813] input: Power Button as
/devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input1
[    1.608991] input: Sleep Button as
/devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
[    1.610997] input: Power Button as
/devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
[    3.636959] input: AT Translated Set 2 keyboard as
/devices/platform/i8042/serio0/input/input4
[    8.377859] input: DualPoint Stick as
/devices/platform/i8042/serio1/input/input5
[    8.542194] input: AlpsPS/2 ALPS DualPoint TouchPad as
/devices/platform/i8042/serio1/input/input6

  Vasily Titskiy


On Fri, Feb 7, 2014 at 3:18 AM, Tommy Will <tommywill2011@gmail.com> wrote:
> Hi Vasily,
>
> Thanks for your mail & information. We would start the investigation
> and give you feedback later.
> Please wait.
>
>> The issues occurs on every kernel after
>> 1302bac33d9e88cd43e482191a806998f3ed43cc commit. Before the commit, this touchpad device was detected as PS/2 mouse and "cat /dev/input/mice" shows data when the touchpad is used.
>> After the commit, there are no events from the touchpad in /dev/input/mice or /dev/input/mouseX files anymore.
>>
>> According to drivers/input/mouse/alps.c source file, all devices with signatures "ec[0] == 0x88 && ec[1] == 0x08" are treated as "Rushmore touchpads". This touchpad has E7 signature of "73-03-0A" and EC signature of "88-08-22", and it looks like that V5 protocol (not
>> "Rushmore") should be used here. If V5 protocol is forced to use for this device then everything works fine.
>>
>> The patch is trivial, but I'm not sure how to properly separate "Rushmore" and "V5" devices. Are there any datasheets or experts to help me with it?
>>
>
> I checked the E7 & EC information and think it should be a Rushmore device.
> Could you please let me know after you use V5 protocol, if touchpad
> still be recognized as "AlpsPS/2 ALPS DualPoint TouchPad" or returns
> to "PS/2 Generic Mouse" ?
>
> Thanks
> --
> Best Regards,
> Tommy

Re: PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

[Kevin Cernekee]

On Fri, Feb 7, 2014 at 7:18 AM, Vasily Titskiy <qehgt0@gmail.com> wrote:
> Hi Tommy,
>
> This is the output of `dmesg | grep -i 'input\|alps'` command after V5 enforced:
> [    1.592880] input: Lid Switch as
> /devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input0
> [    1.593928] input: Power Button as
> /devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input1
> [    1.593964] input: Sleep Button as
> /devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
> [    1.593999] input: Power Button as
> /devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
> [    1.788587] input: AT Translated Set 2 keyboard as
> /devices/platform/i8042/serio0/input/input4
> [    2.661783] input: PS/2 Mouse as /devices/platform/i8042/serio1/input/input5
> [    2.674857] input: AlpsPS/2 ALPS GlidePoint as
> /devices/platform/i8042/serio1/input/input6

Could you confirm that you're seeing actual ALPS V5 absolute reports
with the patched driver, not just emulated PS/2 mouse reports?

Do edge scrolling and multitouch work correctly on the patched driver?

> ...and before my patch (current kernel behavior):
> [    1.604029] input: Lid Switch as
> /devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input0
> [    1.606813] input: Power Button as
> /devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input1
> [    1.608991] input: Sleep Button as
> /devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
> [    1.610997] input: Power Button as
> /devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
> [    3.636959] input: AT Translated Set 2 keyboard as
> /devices/platform/i8042/serio0/input/input4
> [    8.377859] input: DualPoint Stick as
> /devices/platform/i8042/serio1/input/input5
> [    8.542194] input: AlpsPS/2 ALPS DualPoint TouchPad as
> /devices/platform/i8042/serio1/input/input6

This laptop does have a trackstick, correct?

(I see it on the Dell spec sheet but I'd like to confirm your unit has it.)

Thanks.

Re: PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

[Vasily Titskiy]

Hi all,

On Fri, Feb 07, 2014 at 08:46:08AM -0800, Kevin Cernekee wrote:
> On Fri, Feb 7, 2014 at 7:18 AM, Vasily Titskiy <qehgt0@gmail.com> wrote:
> > Hi Tommy,
> >
> > [    2.674857] input: AlpsPS/2 ALPS GlidePoint as
> > /devices/platform/i8042/serio1/input/input6
> 
> Could you confirm that you're seeing actual ALPS V5 absolute reports
> with the patched driver, not just emulated PS/2 mouse reports?
> 
> Do edge scrolling and multitouch work correctly on the patched driver?
What is the easiest way to do it?

> 
> > ...and before my patch (current kernel behavior):
> > [    1.604029] input: Lid Switch as
> > /devices/LNXSYSTM:00/device:00/PNP0C0D:00/input/input0
> > [    1.606813] input: Power Button as
> > /devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input1
> > [    1.608991] input: Sleep Button as
> > /devices/LNXSYSTM:00/device:00/PNP0C0E:00/input/input2
> > [    1.610997] input: Power Button as
> > /devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
> > [    3.636959] input: AT Translated Set 2 keyboard as
> > /devices/platform/i8042/serio0/input/input4
> > [    8.377859] input: DualPoint Stick as
> > /devices/platform/i8042/serio1/input/input5
> > [    8.542194] input: AlpsPS/2 ALPS DualPoint TouchPad as
> > /devices/platform/i8042/serio1/input/input6
> 
> This laptop does have a trackstick, correct?
> 
> (I see it on the Dell spec sheet but I'd like to confirm your unit has it.)
Yes, it does. There is a trackstick beetween G,H,B keys.

--
  Vasily

PROBLEM: [Dell LATITUDE E7440] ALPS touchpad events are not routed into /dev/input/mice and /dev/input/mouseX

[Vasily Titskiy]

The issues occurs on every kernel after
1302bac33d9e88cd43e482191a806998f3ed43cc commit. Before the commit,
this touchpad device was detected as PS/2 mouse and "cat
/dev/input/mice" shows data when the touchpad is used.
After the commit, there are no events from the touchpad in
/dev/input/mice or /dev/input/mouseX files anymore.

According to drivers/input/mouse/alps.c source file, all devices with
signatures "ec[0] == 0x88 && ec[1] == 0x08" are treated as "Rushmore
touchpads". This touchpad has E7 signature of "73-03-0A" and EC
signature of "88-08-22", and it looks like that V5 protocol (not
"Rushmore") should be used here. If V5 protocol is forced to use for
this device then everything works fine.

The patch is trivial, but I'm not sure how to properly separate
"Rushmore" and "V5" devices. Are there any datasheets or experts to
help me with it?

Thanks.
--
  Vasily Titskiy