Synaptics touchpad not working

[Mike Spencer]

Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
displays clock and xterms as specified in ~/.xinitrc, xterm with focus
accepts keystrokes, executes commands. X displays cursor but touchpad
doesn't move it or work at all.

Synaptics touchpad is seen by lsusb.

X is finding a driver and finding the touchpad (snip from Xorg.log
infra). Modprobing the synaptics_usb kernel module doesn't help.
Running xev from the xterm with focus, xev doesn't show any events
from fiddling with the touchpad.

lshw doesn't mention the touchpad.

Many queries on the net about Synaptics touchpad not working quite as
desired. Don't find anything about my situation.

Any suggestions?

Slackware 14.1, 3.10.17-smb 32-bit kernel.

--- Begin snip from Xorg.log ---

[ 590.131] (II) LoadModule: "synaptics"
[ 590.131] (II) Loading /usr/lib/xorg/modules/input/synaptics_drv.so
[ 590.131] (II) Module synaptics: vendor="X.Org Foundation"
[ 590.131] compiled for 1.14.2, module version = 1.7.1
[ 590.131] Module class: X.Org XInput Driver
[ 590.131] ABI class: X.Org XInput driver, version 19.1
[ 590.131] (II) Using input driver 'synaptics' for ' '
[ 590.131] (**) : always reports core events
[ 590.131] (**) Option "Device" "/dev/input/event11"
[ 590.157] (--) synaptics: : x-axis range 0 - 1236 (res 12)
[ 590.157] (--) synaptics: : y-axis range 0 - 898 (res 12)
[ 590.158] (II) synaptics: : device does not report pressure,
will use touch data.
[ 590.158] (II) synaptics: : device does not report finger width.
[ 590.158] (--) synaptics: : buttons: left double triple
[ 590.158] (--) synaptics: : Vendor 0x6cb Product 0x2970
[ 590.158] (--) synaptics: : invalid pressure range. defaulting
to 0 - 255
[ 590.158] (--) synaptics: : invalid finger width range.
defaulting to 0 - 15
[ 590.158] (**) Option "TapButton1" "1"
[ 590.158] (**) Option "TapButton2" "2"
[ 590.158] (**) Option "TapButton3" "3"
[ 590.158] (--) synaptics: : touchpad found
[ 590.158] (**) : always reports core events
[ 590.182] (**) Option "config_info"
"udev:/sys/devices/pci0000:00/0000:00:12.0/usb1/1-1/1-1.3/1-1.3:1.0/input/input\
11/event11"
[ 590.182] (II) XINPUT: Adding extended input device " " (type:
TOUCHPAD, id 11)
[ 590.182] (**) synaptics: : (accel) MinSpeed is now constant
deceleration 2.5
[ 590.182] (**) synaptics: : (accel) MaxSpeed is now 1.75
[ 590.182] (**) synaptics: : (accel) AccelFactor is now 0.131
[ 590.183] (**) : (accel) keeping acceleration scheme 1
[ 590.183] (**) : (accel) acceleration profile 1
[ 590.183] (**) : (accel) acceleration factor: 2.000
[ 590.183] (**) : (accel) acceleration threshold: 4
[ 590.183] (--) synaptics: : touchpad found

mds -- And mouse0 is corectly ignored


[ 590.184] (II) config/udev: Adding input device (/dev/input/mouse0)
[ 590.184] (II) No input driver specified, ignoring this device.




--
Mike Spencer Nova Scotia, Canada

[Michael Black]

On Tue, 12 Aug 2014, Mike Spencer wrote:

>
> Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
> displays clock and xterms as specified in ~/.xinitrc, xterm with focus
> accepts keystrokes, executes commands. X displays cursor but touchpad
> doesn't move it or work at all.
>
> Synaptics touchpad is seen by lsusb.
>
> X is finding a driver and finding the touchpad (snip from Xorg.log
> infra). Modprobing the synaptics_usb kernel module doesn't help.
> Running xev from the xterm with focus, xev doesn't show any events
> from fiddling with the touchpad.
>
> lshw doesn't mention the touchpad.
>
> Many queries on the net about Synaptics touchpad not working quite as
> desired. Don't find anything about my situation.
>
> Any suggestions?
>

No.

Two years ago, I got a USB keyboard that was like a laptop keyboard for
the desk. It included a touchpad, and that little joystick thing in
between the keys.

I expected to have to configure it, but was really surprised when I
plugged it in. It all worked fine with Slackware 14, I didn't have to
configure anything, or even tell anything to use the USB keyboard rather
than the PS/2 keyboard. They just worked in tandem.

So that's not going to help you one bit, other than it would seem
Slackware is set up for it out of the box.

Michael

[Frank P. Westlake]

Here are just the lines containing "syn" in my "Xorg.*.log" files; maybe
they might indicate something:

$ grep -i "syn" /var/log/Xorg.0.log
[ 124.326] Initializing built-in extension SYNC
[ 126.535] (II) config/udev: Adding input device SynPS/2 Synaptics
TouchPad (/dev/input/event8)
[ 126.535] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "evdev
touchpad catchall"
[ 126.535] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "touchpad"
[ 126.535] (II) LoadModule: "synaptics"
[ 126.535] (II) Loading /usr/lib64/xorg/modules/input/synaptics_drv.so
[ 126.564] (II) Module synaptics: vendor="X.Org Foundation"
[ 126.564] (II) Using input driver 'synaptics' for 'SynPS/2 Synaptics
TouchPad'
[ 126.564] (**) SynPS/2 Synaptics TouchPad: always reports core events
[ 126.585] (II) synaptics: SynPS/2 Synaptics TouchPad: ignoring touch
events for semi-multitouch device
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: x-axis range
1472 - 5776 (res 55)
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: y-axis range
1408 - 5012 (res 103)
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: pressure range
0 - 255
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: finger width
range 0 - 15
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: buttons: left
right double triple
[ 126.585] (--) synaptics: SynPS/2 Synaptics TouchPad: Vendor 0x2
Product 0x7
[ 126.586] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 126.586] (**) SynPS/2 Synaptics TouchPad: always reports core events
[ 126.601] (II) XINPUT: Adding extended input device "SynPS/2
Synaptics TouchPad" (type: TOUCHPAD, id 11)
[ 126.601] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)

MinSpeed is now constant deceleration 2.5

[ 126.601] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)
MaxSpeed is now 1.75
[ 126.601] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)
AccelFactor is now 0.036
[ 126.602] (**) SynPS/2 Synaptics TouchPad: (accel) keeping
acceleration scheme 1
[ 126.602] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration profile 1
[ 126.602] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration
factor: 2.000
[ 126.602] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration
threshold: 4
[ 126.602] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 126.603] (II) config/udev: Adding input device SynPS/2 Synaptics
TouchPad (/dev/input/mouse0)
[ 4348.562] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 6922.412] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 14621.758] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 22008.349] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 27740.053] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 32069.333] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 37658.675] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 40916.165] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 46757.105] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 52299.497] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 54579.861] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 61322.873] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 67158.004] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 73002.953] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 78465.992] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 82351.794] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 87090.688] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 92443.973] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 97025.344] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[103716.619] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found

-----------------------------------------------------------------------

$ grep -i "syn" /var/log/Xorg.1.log
[ 28727.184] Initializing built-in extension SYNC
[ 28727.958] (II) config/udev: Adding input device SynPS/2 Synaptics
TouchPad (/dev/input/event6)
[ 28727.958] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "evdev
touchpad catchall"
[ 28727.958] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "touchpad"
[ 28727.958] (II) LoadModule: "synaptics"
[ 28727.959] (II) Loading /usr/lib64/xorg/modules/input/synaptics_drv.so
[ 28727.959] (II) Module synaptics: vendor="X.Org Foundation"
[ 28727.959] (II) Using input driver 'synaptics' for 'SynPS/2 Synaptics
TouchPad'
[ 28727.959] (**) SynPS/2 Synaptics TouchPad: always reports core events
[ 28727.980] (II) synaptics: SynPS/2 Synaptics TouchPad: ignoring touch
events for semi-multitouch device
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: x-axis range
1472 - 5776 (res 55)
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: y-axis range
1408 - 5012 (res 103)
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: pressure range
0 - 255
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: finger width
range 0 - 15
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: buttons: left
right double triple
[ 28727.980] (--) synaptics: SynPS/2 Synaptics TouchPad: Vendor 0x2
Product 0x7
[ 28727.981] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 28727.981] (**) SynPS/2 Synaptics TouchPad: always reports core events
[ 28727.993] (II) XINPUT: Adding extended input device "SynPS/2
Synaptics TouchPad" (type: TOUCHPAD, id 11)
[ 28727.993] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)

MinSpeed is now constant deceleration 2.5

[ 28727.993] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)
MaxSpeed is now 1.75
[ 28727.993] (**) synaptics: SynPS/2 Synaptics TouchPad: (accel)
AccelFactor is now 0.036
[ 28727.994] (**) SynPS/2 Synaptics TouchPad: (accel) keeping
acceleration scheme 1
[ 28727.994] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration profile 1
[ 28727.994] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration
factor: 2.000
[ 28727.994] (**) SynPS/2 Synaptics TouchPad: (accel) acceleration
threshold: 4
[ 28727.995] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 28727.996] (II) config/udev: Adding input device SynPS/2 Synaptics
TouchPad (/dev/input/mouse0)
[ 29033.619] (II) UnloadModule: "synaptics"

Frank

[Marco Maggi]

"Mike Spencer" wrote:
> Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
> displays clock and xterms as specified in ~/.xinitrc, xterm with focus
> accepts keystrokes, executes commands. X displays cursor but touchpad
> doesn't move it or work at all.

I use an Acer Aspire V5 and it has this thing:

Synaptics PS/2 Port TouchPad. Synaptics ClickPad V8.1. This is
actually a ``clickpad'': the whole rectangle is a big button and
there are no left, middle or right small buttons.

I am unable to tune the fine configuration parameters like I desire, but
at least it works. I suggest you to read the manual pages synclient(1)
and synaptics(4).

On my system I create the "xorg.conf" file with the configuration tool
and then tune it by hand, if needed. I use the following sections:

Section "ServerLayout"
...
InputDevice "Mouse0" "CorePointer"
...
EndSection

Section "InputDevice"
Identifier "Mouse0"
Driver "synaptics"
Option "Device" "/dev/input/mice"
Option "Protocol" "auto-dev"
Option "ClickPad" "on"

# Unfortunately this does not work.
# Option "SHMConfig" "yes"

# See the manpage section "Area Handling".
# Option "LeftEdge" "1765"
# Option "RightEdge" "5369"
# Option "TopEdge" "1635"
# Option "BottomEdge" "4429"

# Select which button is reported when tapping on the
# bottom-right corner.
Option "RBCornerButton" "2"

# See the manpage section "Tapping". Decide when
# tapping happens.
Option "FingerLow" "40"
Option "FingerHigh" "50"
# Option "FingerLow" "25"
# Option "FingerHigh" "30"
Option "MaxTapTime" "180"
Option "MaxTapMove" "233"
EndSection

HTH
--
"Now feel the funk blast!"
Rage Against the Machine - "Calm like a bomb"

[Ars Ivci]

On Wed, 13 Aug 2014 08:25:18 +0200
Marco Maggi <ma...@maggi.invalid> wrote:

> "Mike Spencer" wrote:
> > Trying to get X working on an Acer Aspire E5-721 laptop. X
> > starts, displays clock and xterms as specified in ~/.xinitrc, xterm
> > with focus accepts keystrokes, executes commands. X displays
> > cursor but touchpad doesn't move it or work at all.
>
> I use an Acer Aspire V5 and it has this thing:
>
> Synaptics PS/2 Port TouchPad. Synaptics ClickPad V8.1. This

Also, do not forget to "enable clicks" in the touchpad section via
control center or whatever, hardware, mouse. Tapping did not work until
I did this.

--
t.
Weiner's Law of Libraries:
There are no answers, only cross references.

[Mike Spencer]

Michael Black <et...@ncf.ca> writes:

> On Tue, 12 Aug 2014, Mike Spencer wrote:
>
>> Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
>> displays clock and xterms as specified in ~/.xinitrc, xterm with focus
>> accepts keystrokes, executes commands. X displays cursor but touchpad
>> doesn't move it or work at all.
>>
>> Synaptics touchpad is seen by lsusb.
>>
>> X is finding a driver and finding the touchpad (snip from Xorg.log
>> infra). Modprobing the synaptics_usb kernel module doesn't help.
>> Running xev from the xterm with focus, xev doesn't show any events
>> from fiddling with the touchpad.

>> ...
>> Any suggestions?

> No.

Dang!

> So that's not going to help you one bit, other than it would seem
> Slackware is set up for it out of the box.

Yeah, it *looks* like that. There's a file,

/usr/share/X11/xorg.conf.d/50-synaptics.conf

which X seems to be reading. There's a synaptics_usb.ko kernel module.
X grovels through configuring the USB touchpad X driver, [1]

/usr/lib/xorg/modules/input/synaptics_drv.so


I *did* find what appears to be a contradistion in /proc/bus/input/devices:

I: Bus=0003 Vendor=06cb Product=2970 Version=0111
N: Name=" "
P: Phys=usb-0000:00:12.0-1.3/input0
S: Sysfs=/devices/pci0000:00/0000:00:12.0/usb3/3-1/3-1.3/3-1.3:1.0/
input/input11 [line broken for readability]
U: Uniq=
H: Handlers=mouse0 event11
B: PROP=1
B: EV=b
B: KEY=e520 0 10000 0 0 0 0 0 0 0 0
B: ABS=2608000 3

This is above my pay grade, but is this right? The xorg.log tries to
find a driver for mouse0 and fails. Why does X set up synaptics-usb,
then go looking for mouse0? Because it's specified in
/proc/bus/input/devices? The ...devices file correctly identifies the
USB ID 06cb:2970, same as seen by lsusb (which, incidentally, isn't in
the big USB-ID file) but then wants to configure mouse0, too.

"Frank P. Westlake" <frank.w...@gmail.com> wrote:

Here are just the lines containing "syn" in my "Xorg.*.log" files; maybe
they might indicate something:

> $ grep -i "syn" /var/log/Xorg.0.log
> [ 124.326] Initializing built-in extension SYNC
> [ 126.535] (II) config/udev: Adding input device SynPS/2 Synaptics
> TouchPad (/dev/input/event8)
> [ 126.535] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "evdev
> touchpad catchall"
> [ 126.535] (**) SynPS/2 Synaptics TouchPad: Applying InputClass "touchpad"
> [ 126.535] (II) LoadModule: "synaptics"
> [ 126.535] (II) Loading /usr/lib64/xorg/modules/input/synaptics_drv.so

That's a PS/2 touchpad, loading synaptics_drv. Mine is USB but loads
the same driver. Is that a clue? I don't have the X source on my HD
to look at.

Marco Maggi <ma...@maggi.invalid> wrote:

> I use an Acer Aspire V5 and it has this thing:
>

> Synaptics PS/2 Port TouchPad. Synaptics ClickPad V8.1. This is
> actually a ``clickpad'': the whole rectangle is a big button and
> there are no left, middle or right small buttons.

Yes, that's the physical bunny I have but (according to lsusb) its
USB, not PS/2.

Still stumped after trying various combinations of kernel modules,
xorg.conf variations. No mouse.

- Mike



[1] Relevant xorg.log lines

[ 229.380] (II) config/udev: Adding input device (/dev/input/event11)
[ 229.380] (**) : Applying InputClass "evdev touchpad catchall"
[ 229.380] (**) : Applying InputClass "touchpad"
[ 229.380] (**) : Applying InputClass "evdev touchpad catchall"
[ 229.380] (**) : Applying InputClass "touchpad"
[ 229.380] (II) LoadModule: "synaptics"
[ 229.380] (II) Loading /usr/lib/xorg/modules/input/synaptics_drv.so
[ 229.381] (II) Module synaptics: vendor="X.Org Foundation"
[ 229.381] compiled for 1.14.2, module version = 1.7.1
[ 229.381] Module class: X.Org XInput Driver
[ 229.381] ABI class: X.Org XInput driver, version 19.1
[ 229.381] (II) Using input driver 'synaptics' for ' '
[ 229.381] (**) : always reports core events
[ 229.381] (**) Option "Device" "/dev/input/event11"
[ 229.409] (--) synaptics: : x-axis range 0 - 1236 (res 12)
[ 229.409] (--) synaptics: : y-axis range 0 - 898 (res 12)
[ 229.409] (II) synaptics: : device does not report pressure, will use touch data.
[ 229.409] (II) synaptics: : device does not report finger width.
[ 229.409] (--) synaptics: : buttons: left double triple
[ 229.409] (--) synaptics: : Vendor 0x6cb Product 0x2970
[ 229.409] (--) synaptics: : invalid pressure range. defaulting to 0 - 255
[ 229.409] (--) synaptics: : invalid finger width range. defaulting to 0- 15
[ 229.409] (**) Option "TapButton1" "1"
[ 229.409] (**) Option "TapButton2" "2"
[ 229.409] (**) Option "TapButton3" "3"
[ 229.409] (--) synaptics: : touchpad found
[ 229.409] (**) : always reports core events
[ 229.421] (**) Option "config_info" "udev:/sys/devices/pci0000:00/0000:00:12.0/usb3/3-1/3-1.3/3-1.3:1.0/input/input11/event11"
[ 229.421] (II) XINPUT: Adding extended input device " " (type: TOUCHPAD, id 11)
[ 229.421] (**) synaptics: : (accel) MinSpeed is now constant deceleration 2.5
[ 229.421] (**) synaptics: : (accel) MaxSpeed is now 1.75
[ 229.421] (**) synaptics: : (accel) AccelFactor is now 0.131
[ 229.421] (**) : (accel) keeping acceleration scheme 1
[ 229.421] (**) : (accel) acceleration profile 1
[ 229.421] (**) : (accel) acceleration factor: 2.000
[ 229.421] (**) : (accel) acceleration threshold: 4
[ 229.421] (--) synaptics: : touchpad found
[ 229.422] (II) config/udev: Adding input device (/dev/input/mouse0)
[ 229.422] (II) No input driver specified, ignoring this device.
[ 229.422] (II) This device may have been added with another device file.

[Jerry Peters]

This doesn't seem to be important. I get the same messages and my TP
works:

[ 12.549] (--) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
[ 12.550] (II) config/udev: Adding input device SynPS/2 Synaptics
TouchPad (/dev/input/mouse0)
[ 12.550] (II) No input driver specified, ignoring this device.
[ 12.550] (II) This device may have been added with another device
file.

There wouldn't happen to be a key combination to enable/disable the TP
would there? On my AA1 Fn-F7 toggles the TP on/off.

[frank.w...@gmail.com]

From JerryPeters :

>There wouldn't happen to be a key combination to enable/disable the TP
>would there? On my AA1 Fn-F7 toggles the TP on/off.

I have not used it but mine has a slight recess at the
upper left corner which I think is a configurable on/off
button. I think a long-press on that recess might toggle
the pad.

Frank

[Mike Spencer]

Jerry Peters <je...@example.invalid> writes:

> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>
>> Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
>> displays clock and xterms as specified in ~/.xinitrc, xterm with focus
>> accepts keystrokes, executes commands. X displays cursor but touchpad
>> doesn't move it or work at all.
>

> There wouldn't happen to be a key combination to enable/disable the TP
> would there? On my AA1 Fn-F7 toggles the TP on/off.

Very astute. Yes, there is such a function specified in the (l)user
manual. It has no discernable effect, pressing Fn-F7 either before
startx or after X is running. Presumably Windoes, rather than
hardware, dependent.

Then frank.w...@gmail.com wrote:

> I have not used it but mine has a slight recess at the
> upper left corner which I think is a configurable on/off
> button. I think a long-press on that recess might toggle
> the pad.

No such recess here. Xev(1) desn't report any kind of fiddling with
the touchpad (clickpad) -- tap, press, click, long press, swipe are
not seen as X events.

If you have a Synaptics USB touchpad, what USB ID does lsusb report?
06CB:????. That would tell us if it's reasonable to compare yours to
mine.


I think I need to contact a real guru. This device may not be
supported by any extant drivers. Documentation/sourc for drivers that
mentions USB IDs at all mentions:

06CB:0001 Synaptics USB TouchPad
06CB:0002 Synaptics Integrated USB TouchPad
06CB:0003 Synaptics cPad
06CB:0006 Synaptics TouchScreen
06CB:0007 Synaptics USB Styk
06CB:0008 Synaptics USB WheelPad
06CB:0009 Synaptics Composite USB TouchPad
06CB:0010 Synaptics Wireless TouchPad
06CB:0013 Synaptics DisplayPad

This device is USB ID 06CB:2970 which isn't even listed in
/usr/share/hwdata/usb.ids.

Worse, there's a kernel driver that may or may not work and also an X
driver that may or may not work.

Still stuck, but thanks.

[Jerry Peters]

Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>
> Jerry Peters <je...@example.invalid> writes:
>
>> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>>
>>> Trying to get X working on an Acer Aspire E5-721 laptop. X starts,
>>> displays clock and xterms as specified in ~/.xinitrc, xterm with focus
>>> accepts keystrokes, executes commands. X displays cursor but touchpad
>>> doesn't move it or work at all.
>>
>> There wouldn't happen to be a key combination to enable/disable the TP
>> would there? On my AA1 Fn-F7 toggles the TP on/off.
>
> Very astute. Yes, there is such a function specified in the (l)user
> manual. It has no discernable effect, pressing Fn-F7 either before
> startx or after X is running. Presumably Windoes, rather than
> hardware, dependent.

OK, the AA1 is hardware, or at least it works with Linux.

Instead of xev, try cat /dev/whatever and see if you're getting
anything from the TP. That tells you whether it's X or the kernel
that's the problem.

[Mike Spencer]

Jerry Peters <je...@example.invalid> writes:

> Instead of xev, try cat /dev/whatever and see if you're getting
> anything from the TP. That tells you whether it's X or the kernel
> that's the problem.

Excellent clue. Several features of the newer systems are new to me
and I find myself lost in twisty little passages.

An external USB mouse works without further ado. And there's helpful
info in /dev/input/by-id. From that I can cat /dev/input/event12 or
/dev/input/mouse0 and get hentracks. No response from event13 or
mouse1 attached to the touchpad.

Inference is that it's the kernel driver that's failing.

>> Very astute. Yes, there is such a function specified in the (l)user
>> manual. It has no discernable effect, pressing Fn-F7 either before
>> startx or after X is running. Presumably Windoes, rather than
>> hardware, dependent.
>
> OK, the AA1 is hardware, or at least it works with Linux.

The Aspire E17 manual also has a Fn-F7 combo that is supposed to
toggle the touchpad on/off. Doesn't affect the /dev/input cat tests.

Well, lsusb says the device ID is 06cb:2970. I'll have to try to run
down someone who's currently coding USB drivers. I had email from a
guy who worked on synaptics_usb.c but its is, he says, 10 year old
code. And the USB IDs referenced in the source are all 06cb:xxx, xxx <
0013.

Progress but still stuck. Thanks.

[Jim Diamond]

Mike,

sorry you're having a hard time with this. I have no really good
suggestions (except maybe more e-mailing with kernel driver writers),
but a maybe not bad idea would be to use some USB bus sniffing program
to see if there is any traffic on the USB bus when you use the
touchpad. If there isn't, then (Fn-F7 notwithstanding) you may need
to do something else to initialize / wake up the touchpad. That
probably won't be easy, but there is no point looking at the
synaptics driver if it isn't getting any data.

(I don't recall the previous message here, and can't easily see them
while typing this, so sorry if these next two thoughts have been
discussed.)

Did you look in /var/log/dmesg to see what your kernel says about the
touchpad (if anything) on booting? There might be some useful info
there. (In particular, if it says something like "I see a Synaptics
touchpad, but I don't know the model number" that could be a help.

It is possible that if you blacklist the synaptics-usb kernel module,
the generic mouse one might take over the touchpad. You wouldn't get
specific touchpad functions, but it might be better than nothing.


Good luck.

Jim

[Jim Diamond]

On 2014-08-16 at 09:34 ADT, Jim Diamond <Jim.D...@deletethis.AcadiaU.ca> wrote:
> On 2014-08-16 at 04:29 ADT, Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>>
> sorry you're having a hard time with this. I have no really good
> suggestions (except maybe more e-mailing with kernel driver writers),
> but a maybe not bad idea would be to use some USB bus sniffing program
> to see if there is any traffic on the USB bus when you use the
> touchpad. If there isn't, then (Fn-F7 notwithstanding) you may need
> to do something else to initialize / wake up the touchpad. That
> probably won't be easy, but there is no point looking at the
> synaptics driver if it isn't getting any data.

Check out synaptics-test-1.1.tar.bz2 at
http://jan-steinhoff.de/linux/synaptics-test.html
(No idea if it works, but...)

[Mike Spencer]

Jim Diamond <Jim.D...@deletethis.AcadiaU.ca> writes:

> sorry you're having a hard time with this.

Thank you for the sympathy. :-)

> ...maybe not bad idea would be to use some USB bus sniffing program

> to see if there is any traffic on the USB bus when you use the
> touchpad.

New concept for me. I'll have a look for such things.

> If there isn't, then (Fn-F7 notwithstanding) you may need to do
> something else to initialize / wake up the touchpad. That probably

> won't be easy...

The kernel finds it (see infra) but is it "awake". Another new concept
for me to investigate. A way to address the Fn-F7 function from
software? Getting a little hairy for me here. :-)

lsusb does say (inter alia, of course),

bDeviceClass 0 (Defined at Interface level)
...
Configuration Descriptor:
...
bmAttributes 0xa0
(Bus Powered)
Remote Wakeup <<<< You mentioned wakeup? Huh.

Whatever that means. Deep water for me.

> ...but there is no point looking at the synaptics driver if it isn't
> getting any data.

Right.

> Did you look in /var/log/dmesg to see what your kernel says about the
> touchpad (if anything) on booting? There might be some useful info
> there. (In particular, if it says something like "I see a Synaptics
> touchpad, but I don't know the model number" that could be a help.

dmesg says:

usb 3-1.3: New USB device found, idVendor=06cb,idProduct=2970
usb 3-1.3: New USB device strings: Mfr=1, Product=2,SerialNumber=0
usb 3-1.3: Product:
usb 3-1.3: Manufacturer:

The empty Product: & Manufacturer: fields are because this device
isn't in /usr/share/hwdata/usb.ids. FWIW, I submitted its ID to
http://www.linux-usb.org/usb-ids.html. And further down,

[External mouse]
input: Logitech USB-PS/2 Optical Mouse as /devices/pci0000:00/00\
00:00:10.0/usb1/1-2/1-2:1.0/input/input11
hid-generic 0003:046D:C03D.0001: input,hidraw0: USB HID v1.10 Mo\
use [Logitech USB-PS/2 Optical Mouse] on usb-0000:00:10.0-2/input0

[touchpad]
input: as /devices/pci0000:00/0000:00:12.0/usb3/3-1/3-1.3/3-\
1.3:1.0/input/input12
hid-multitouch 0003:06CB:2970.0002:input,hiddev0,hidraw1: USB H\
ID v1.11 Mouse [ ] on usb-0000:00:12.0-1.3/input0

No other comments in dmesg. Wild guess here: "ordinary" USB mice do
PS/2-over-USB somehow while 06CB:2970 does something different.

> It is possible that if you blacklist the synaptics-usb kernel module,
> the generic mouse one might take over the touchpad. You wouldn't get
> specific touchpad functions, but it might be better than nothing.

Huh. I never thought of that, never had to do that before. I'll look
at it.

> Good luck.

Tnanks.

[Jerry Peters]

Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>
> Jerry Peters <je...@example.invalid> writes:
>
>> Instead of xev, try cat /dev/whatever and see if you're getting
>> anything from the TP. That tells you whether it's X or the kernel
>> that's the problem.
>
> Excellent clue. Several features of the newer systems are new to me
> and I find myself lost in twisty little passages.
>
> An external USB mouse works without further ado. And there's helpful
> info in /dev/input/by-id. From that I can cat /dev/input/event12 or
> /dev/input/mouse0 and get hentracks. No response from event13 or
> mouse1 attached to the touchpad.
>
> Inference is that it's the kernel driver that's failing.
>

That sounds like there's some initialization specific to that TP that
needs to be done by the driver.

What kernel version are you on BTW. IIRC the defaut for 14.1 was
3.10.17, while the current kernel is 3.16.1 which will work for
14.1.

[buck]

Jerry Peters <je...@example.invalid> wrote in news:lsofgk$t4p$1@dont-
email.me:

Mike,
I think that the advice to try a "new" LTS kernel is excellent.

May I suggest you install the "Pi" (3.14.16) huge kernel and
corresponding modules from -current? Because the computer is new,
there is a good possibility that a new kernel will handle the
Synaptics TP.
--
buck

[Mike Spencer]

Jerry Peters <je...@example.invalid> writes:

> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>

>> An external USB mouse works without further ado. And there's helpful
>> info in /dev/input/by-id. From that I can cat /dev/input/event12 or
>> /dev/input/mouse0 and get hentracks. No response from event13 or
>> mouse1 attached to the touchpad.
>>
>> Inference is that it's the kernel driver that's failing.
>
> That sounds like there's some initialization specific to that TP that
> needs to be done by the driver.

Right. [1]

> What kernel version are you on BTW. IIRC the defaut for 14.1 was
> 3.10.17, while the current kernel is 3.16.1 which will work for
> 14.1.

Yes, 3.10.17.

The current project is to get this laptop working well enough that I
can venture out from my dial-up badger hole to wifi-enabled watering
holes to do such things as download a new kernel along with all the
attendant bits such as source, includes, system-map etc.

I have a feeler out with a friend who's a Debian maintainer, hoping to
get a clue on whether any support for USB ID 06cb:2970 is out there at
all. Don't want to do a tedious admin chore if it's known beforehand
not to work.

Never thought of updating the kernel from what, for me, is the shiny
new version. (Slack 11 and 12.1 doing just fine on other boxen.)

Thanks.


[1] I'm seeing a similar problem with sound. System initializes on the
HDMI card. Not an urgent matter for this box.

[Mike Spencer]

buck <bu...@private.mil> writes:

> Mike,
>
> I think that the advice to try a "new" LTS kernel is excellent.
>
> May I suggest you install the "Pi" (3.14.16) huge kernel and
> corresponding modules from -current? Because the computer is new,
> there is a good possibility that a new kernel will handle the
> Synaptics TP.

LTS kernel? Tikky-tikky-click.....Oh, right. LTS is definitely for
me. :-)

I'll give that a try.

I've confirmed that I can do wi-fi with this machine so I should be
able to do that.

I'm wondering if the Fn-F7 "turn touch pad ON/OFF" facility may have
some significance here. Has no obsrveable effect on this box. I
should try to research that, too. I've never seen any discussion of
the Fn-F? key combos anywhere. They mostly work as expected on an old
Panasonic laptop but not on the Acer Aspire.

Tnx,

[Jerry Peters]

Sure it's Fn-F7? I looking at some Acers at Walmart last week & it
looked like it might be Fn-F4, though interpreting the icon is tricky.

On this AA1 the icon is a rectangle with a hand pointiing at it from
the upper right corner. Of course it's obvious, once someone tells you
what it's for.

[Ars Ivci]

On Tue, 19 Aug 2014 20:43:18 +0000 (UTC)
Jerry Peters <je...@example.invalid> wrote:

> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
> >
> > buck <bu...@private.mil> writes:

[...]

> >
> > I'm wondering if the Fn-F7 "turn touch pad ON/OFF" facility may have
> > some significance here. Has no obsrveable effect on this box. I
> > should try to research that, too. I've never seen any discussion of
> > the Fn-F? key combos anywhere. They mostly work as expected on an
> > old Panasonic laptop but not on the Acer Aspire.
> >
> > Tnx,
>
> Sure it's Fn-F7? I looking at some Acers at Walmart last week & it
> looked like it might be Fn-F4, though interpreting the icon is tricky.
>
> On this AA1 the icon is a rectangle with a hand pointiing at it from
> the upper right corner. Of course it's obvious, once someone tells you
> what it's for.
>

It's Fn-F5 here, Toshiba. I had a Thinkpad and it was Fn-F8. No
standards...

[Mike Spencer]

Jerry Peters <je...@example.invalid> writes:

> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>

>> I'm wondering if the Fn-F7 "turn touch pad ON/OFF" facility may have
>> some significance here. Has no obsrveable effect on this box. I
>> should try to research that, too. I've never seen any discussion of
>> the Fn-F? key combos anywhere. They mostly work as expected on an old
>> Panasonic laptop but not on the Acer Aspire.
>
>

> Sure it's Fn-F7? I looking at some Acers at Walmart last week & it
> looked like it might be Fn-F4, though interpreting the icon is tricky.

Well, as certain as I can be without original Windoes. I have the
manual which shows the icon as you describe associated with F7. The
F7 key bears that icon.

> On this AA1 the icon is a rectangle with a hand pointiing at it from
> the upper right corner. Of course it's obvious, once someone tells you
> what it's for.

You know the old joke about the prof's blackboard full of math, formal
logic etc...."And so it's obvious that...um...ummm...". Leaves
classroom, gone 10 minutes, back, "Yes, it *is* obvious that..." Only
if they tell you. :-)

And as another poster observes, there seems not to be any standard for
this.

BTW, googling for "Asus Aspire E17" turns up mostly hits in .HU and
.SK. Odd, that. Do I have a rara avis? Works great except for the
{touch,click}pad.

[buck]

Mike Spencer <m...@bogus.nodomain.nowhere> wrote in
news:87lhqk5...@bogus.nodomain.nowhere:

>
> buck <bu...@private.mil> writes:
>
>> Mike,
>>
>> I think that the advice to try a "new" LTS kernel is excellent.
>>
>> May I suggest you install the "Pi" (3.14.16) huge kernel and
>> corresponding modules from -current? Because the computer is new,
>> there is a good possibility that a new kernel will handle the
>> Synaptics TP.
>
> LTS kernel? Tikky-tikky-click.....Oh, right. LTS is definitely for
> me. :-)

LTS=Long Term Support ;)

Which is why I suggested the -current kernel...
--
buck

[Michael Black]

Is this the netbook? You've been having so much trouble, I've forgotten.

I plugged in my USB keyboard that has the touchpad, it worked immediately,
I didn't even have to restart X. But it comes in at 06CB:009, which I
gather from your accumualted postings is well defined. I can't believe
there is a lot of variation, I found some broken laptops last year and
wondered about using one of the touchpads on the desk, and I looked up
part numbers and found only a few variations, such as one for PS2 and
another for USB. I would have thought once they got it right, they'd not
need to keep chaning things.

That said, if this is the Acer Aspire One netbook, then I'll make the
effort to fire up mine, which is the one with only 8gig "hard drive" and
that weird Linux distribution of Linux installed. Maybe looking at dmesg
there will provide some information. That's now an old distribution, so
if it worked then, I'm not sure why a new kernel would be needed.

Michael

[Jerry Peters]

It's not, AA1 connects the TP via PS/2.

[Jerry Peters]

Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>
> Jerry Peters <je...@example.invalid> writes:
>
>> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>>
>>> I'm wondering if the Fn-F7 "turn touch pad ON/OFF" facility may have
>>> some significance here. Has no obsrveable effect on this box. I
>>> should try to research that, too. I've never seen any discussion of
>>> the Fn-F? key combos anywhere. They mostly work as expected on an old
>>> Panasonic laptop but not on the Acer Aspire.
>>
>>
>> Sure it's Fn-F7? I looking at some Acers at Walmart last week & it
>> looked like it might be Fn-F4, though interpreting the icon is tricky.
>
> Well, as certain as I can be without original Windoes. I have the
> manual which shows the icon as you describe associated with F7. The
> F7 key bears that icon.
>
>> On this AA1 the icon is a rectangle with a hand pointiing at it from
>> the upper right corner. Of course it's obvious, once someone tells you
>> what it's for.
>
> You know the old joke about the prof's blackboard full of math, formal
> logic etc...."And so it's obvious that...um...ummm...". Leaves
> classroom, gone 10 minutes, back, "Yes, it *is* obvious that..." Only
> if they tell you. :-)

Yes, there were regular postings in the AA1 forums where someone would
accidently turn off the TP.

>
> And as another poster observes, there seems not to be any standard for
> this.
>
> BTW, googling for "Asus Aspire E17" turns up mostly hits in .HU and
> .SK. Odd, that. Do I have a rara avis? Works great except for the
> {touch,click}pad.
>

Yeah, I tried googling on the USB id but didn't find anything that
really stands out.

Just a thought, but have you tried blacklisting the synaptics-usb
module and letting the regular HID stack takeover?

[Mike Spencer]

Michael Black <et...@ncf.ca> writes:

> On Tue, 19 Aug 2014, Mike Spencer wrote:
>>
>> BTW, googling for "Asus Aspire E17" turns up mostly hits in .HU and
>> .SK. Odd, that. Do I have a rara avis? Works great except for the
>> {touch,click}pad.
>>
> Is this the netbook? You've been having so much trouble, I've forgotten.

I think this must be a very new model, one which was produced thinking
that people were like me: want (a) big screen, (b) DVD drive, which
were my key purchase decision items. And then the market turned to (a)
big is clunky, want small: netbook or tablet or smart phone plz and
(b) DVDs are sooo 20th c., I stream all my media. Found some pages in
Czech and Finnish, too. :-o

Anyhow, Asus Aspire E17 (E5-721-492D), AMD quad-core CPU, 500G HD, 4G
RAM, AMD Radeon graphics, DVD drive, 4 USB ports, ethernet port, some
others (HDMI?). Nominal 17" screen (dunno the aspect ratio but way wider than
4:3), about 17" x 13" x 1" overall. Everything works great except
touchpad (not at all) and sound (not as I'd like but it's okay.)

> I plugged in my USB keyboard that has the touchpad, it worked immediately,
> I didn't even have to restart X.

Now there's a thought. Non-optimal but as it is, I have to plug in an
external mouse anyhow.

> But it comes in at 06CB:009, which I gather from your accumulated
> postings is well defined.

Yes. Listed in usb.ids and referenced explicitly in module src
synaptics_usb.c. 06CB:2970 is not in either place nor was it on the
website that maintains usb.ids (until I put it there.)

> I can't believe there is a lot of variation, I found some broken
> laptops last year and wondered about using one of the touchpads on
> the desk, and I looked up part numbers and found only a few
> variations, such as one for PS2 and another for USB. I would have

> thought once they got it right, they'd not need to keep changing
> things.

That seems reasonable. So why am I stuck? I'm about to grovel through
the src on the 3.14.17 and 3.16.1 kernels to see if there's any
explicit mention of anything that would be a clue. (I couldn't easily
do this were the new machine not working great at a friendly WAP.)

I'm guessing that either (a) its new and as yet unsupported or (b)
it's just like the older models and there some arcane little gotcha.

> That said, if this is the Acer Aspire One netbook...

Not. Several years newer, much bigger. AIUI, the Aspire One had ca. 8
hrs battery life. This has about 4 hrs with steady but light use.

Then Jerry Peters <je...@example.invalid> wrote:

jp> Just a thought, but have you tried blacklisting the synaptics-usb
jp> module and letting the regular HID stack takeover?

That module isn't loaded automatically by the kernel. AFAICT, no
module/driver takes possession of the device and it makes no
difference whether I modprobe that module or I don't.

Getting out of my depth here: might the older Synaptics USB pad have
been doing like PS/2-over-USB while this one is doing a new, non-PS/2
protocol? Maybe that's silly.

Tnx,

[Mike Spencer]

Okay, I've learned new stuff so I'll risk following up to myself.

I finally got *something* from the touchpad.

In [kernel-src]/Documentation/usbmon.txt there are instructions to do
something similar to tcpdump(1) on the USB bus. If I follow those
directions (summarized here very briefly):

% mount -t debugfs none_debugs /sys/kernel/debug
% modprobe usbmon

Verify that bus sockets are present.

% ls /sys/kernel/debug/usb/usbmon

Find which bus connects to the desired device, then

% cat /sys/kernel/debug/usb/usbmon/3u

where the 3 is because the touchpad is on USB bus #003,

then I get streams of data when stroking or clicking the touchpad. So
it is in fact talking to the USB bus.

Now:

Over in [kernel-src]/drivers/platform/x86/ there's the source for
acer-wmi.ko (acer-wmi.c) which is auto-loaded at boot. It has an array
of struct key_entry (defined in [kernel-src]
/include/linux/input/sparse_keymap.h). This appears to be an attempt
to deal with the Fn-F? key combos. A snippet:

static const struct key_entry acer_wmi_keymap[] = {
{KE_KEY, 0x01, {KEY_WLAN} }, /* WiFi */
{KE_KEY, 0x03, {KEY_WLAN} }, /* WiFi */
{KE_KEY, 0x04, {KEY_WLAN} }, /* WiFi */
{KE_KEY, 0x12, {KEY_BLUETOOTH} }, /* BT */
[snip]
{KE_KEY, 0x29, {KEY_PROG3} }, /* P_Key for TM8372 */
{KE_IGNORE, 0x41, {KEY_MUTE} },
{KE_IGNORE, 0x42, {KEY_PREVIOUSSONG} },
[snip]
>> {KE_KEY, 0x82, {KEY_TOUCHPAD_TOGGLE} }, /* Touch Pad Toggle */
>> {KE_KEY, KEY_TOUCHPAD_ON, {KEY_TOUCHPAD_ON} },
>> {KE_KEY, KEY_TOUCHPAD_OFF, {KEY_TOUCHPAD_OFF} },
>> {KE_IGNORE, 0x83, {KEY_TOUCHPAD_TOGGLE} },
>> {KE_KEY, 0x85, {KEY_TOUCHPAD_TOGGLE} },
{KE_END, 0}
};

and further down,

case KEY_TOUCHPAD_TOGGLE:
scancode = (device_state & ACER_WMID3_GDS_TOUCHPAD) ?
KEY_TOUCHPAD_ON : KEY_TOUCHPAD_OFF;

So it appears that the author expects Fn-F7 (or some key) to send 0x82
or 0x85 (while ignoring 0x83) and to toggle the touchpad.

Only that doesn't happen.

The only access I have to keycodes (AFAIK at the moment) is xev(1).
xev says that the Windoes key is sending keycode 133 (0x85) but that
doesn't toggle the touchpad. Fn-F7 isn't reported by xev at all.
And although I can read C code, unpacking kernel source is getting
over my head.

So I'm a bit ahead but still stuck.

- Mike

[Jerry Peters]

Could be ACPI is interpreting the Fn-F7 and it's never even getting to
the input subsystem. You might have to look at ACPI events. Unknown
ACPI events should get logged to syslog in Slackware. Then again, it
could happen entirely in the firmware. Although in the AA1 the Fn-F7
keypresses show up in xev.

[Mike Spencer]

Jerry Peters <je...@example.invalid> writes:

> Mike Spencer <m...@bogus.nodomain.nowhere> wrote:
>>
>> Okay, I've learned new stuff so I'll risk following up to myself.
>>
>> I finally got *something* from the touchpad.
>>
>> In [kernel-src]/Documentation/usbmon.txt there are instructions to
>> do something similar to tcpdump(1) on the USB bus. If I follow

>> those directions....then I get streams of data when stroking or

>> clicking the touchpad. So it is in fact talking to the USB bus.
>>
>> Now:
>>
>> Over in [kernel-src]/drivers/platform/x86/ there's the source for
>> acer-wmi.ko (acer-wmi.c) which is auto-loaded at boot.
>>

>> [SNIP: doesn't work]

>
> Could be ACPI is interpreting the Fn-F7 and it's never even getting to
> the input subsystem.

I finally found some info in /sys/devices/virtual/input/input9:
"cat /dev/input/event9" responds with raw binary data when I hit
Fn-F7. So I infer that it's getting to the input subsystem.

> You might have to look at ACPI events. Unknown ACPI events should
> get logged to syslog in Slackware.

Grovel, grovel. Nothing that appears relevant. Also ran acpi_listen.
I don't understand acpi. The acpid(8) manpage isn't very helpful.

> Then again, it could happen entirely in the firmware.

Yeah. But since I can see response from the Fn-F7 key and the touchpad
with devious tricks (see above), it seems like *something* should be
trapping that data. Maaybe I just need to know where (and how) to
poke the firmware with a pointy stick. :-)

> Although in the AA1 the Fn-F7 keypresses show up in xev.
>
>> So I'm a bit ahead but still stuck.

I'm trying to email the guy who, AFAICT, last worked on relevant code
but he may have gone ont other things. Reading up on acpi.

Tnx,