Bug 357881

Summary: Alps Touchpad Wrongly Treated as ImPS/2 Mouse
Product: [openSUSE] openSUSE 10.3 Reporter: Foolish Ewe <foolishewe>
Component: KernelAssignee: Stefan Dirsch <sndirsch>
Status: RESOLVED FIXED QA Contact: E-mail List <qa-bugs>
Severity: Normal    
Priority: P5 - None CC: alexander.egger, chrubis, foolishewe, ms, snwint
Version: Final   
Target Milestone: ---   
Hardware: x86-64   
OS: openSUSE 10.3   
Whiteboard:
Found By: Customer Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: /etc/xorg.conf
/var/log/SaX.log
output of "/usr/sbin/hwinfo --mouse --log=mouse.log"
output of "dmesg"
dmesg after an upgrade from OpenSuse 10.3->11.0 beta 2
/var/log/messages from time of 10.3->11.0 upgrade attempt
/var/log/messages using 10.3 kernel with i8042.debug parameter
Input: alps update
/var/log/messages using the kernel from comment 39
hardware information for Comment 39
dmesg output from most recent reboot
/etc/X11/xorg/conf as per comment 47
/var/log/Xorg.0.log as requested in Comment 47
/var/log/SaX.log as requested in comment 47.

Description Foolish Ewe 2008-02-01 05:55:34 UTC
I have a lap top with an ALPS touchpad/glidepad or whatever the right term is and would like to configure the sensitivity of it, would
ksynaptics (I use KDE) or gsynaptics be the customary tools for this?
My guess is that kernel device detection may have mislabeled it.

I'm running the most recent patches for OpenSuse 10.3 on an Acer 5720-4171, with Vista!/Opensuse Dual boot:
uname -a
Linux linux-8whl 2.6.22.16-0.1-default #1 SMP 2008/01/23 14:28:52 UTC x86_64 x86_64 x86_64 GNU/Linux
/usr/sbin/hwinfo --mouse
46: PS/2 00.0: 10500 PS/2 Mouse
[Created at input.159]
UDI: /org/freedesktop/Hal/devices/platform_i8042_i8042_AUX_port_logicaldev_input
Unique ID: AH6Q.hI1e_vHaTSC
Hardware Class: mouse
Model: "ImPS/2 Generic Wheel Mouse"
Vendor: int 0x0210
Device: int 0x0013 "ImPS/2 Generic Wheel Mouse"
Device File: /dev/input/mice (/dev/input/mouse0)
Device Files: /dev/input/mice, /dev/input/mouse0, /dev/input/event2, /dev/input/by-path/platform-i8042-serio-1-event-mouse, /dev/input/by-path/platform-i8042-serio-1-mouse
Device Number: char 13:63 (char 13:32)
Driver Info #0:
Buttons: 3
Wheels: 1
XFree86 Protocol: explorerps/2
GPM Protocol: exps2
Config Status: cfg=no, avail=yes, need=no, active=unknown


The only laptop modification was to replace a (currently) unsupported Atheros 5007EG based WIFI minipci-express card, I now use an Intel 3945 (drivers work flawlessly, but the hardware seems a bit wimpy on send/receive signal strength).
All hardware except the crystal eye camera function currently in OpenSuse,
but the touch pad is sensitive (the cursor "teleports" during typing,
even if I don't contact the pad sometimes), and sporadically has
strange behavior near its center.

The Vista side doesn't seem to give me full information and dmesg also is wanting. From Vista! I get:Device Description - Alps Pointing-device
Hardware IDs - ACPI\PNP0F13
*PNP0F13
Service - i8042prt
Device Class - Mouse
Device Class Guid - {4d36e96f-e325-11ce-bfc1-08002be10318}

Driver Key - {4d36e96f-e325-11ce-bfc1-08002be10318}\0000

ConfigFlags - 00000000
Manufacturer - Alps Electric
Physical Device Object name -\Device\0000005d
Capabilities -00000020
CM_DEVCAP_SILENTINSTALL
Display name - Alps Pointing-device
Device Instance Path - ACPI\PNP0F13\4&19120B6&0

The device sensitivity is also poorly configured in Vista, but I haven't found any tool that lets me configure the device in either Vista or OpenSuse 10.3.

I've tried SaX2 -r and it gives the same results. In SaX2, under mouse types I've tried the PS/2 Mouse Glidepoint-PS/2
but it seems to be some kind of generic device, my xorg.conf reads afterwards:
Section "InputDevice"
Driver "mouse"
Identifier "Mouse[1]"
Option "Device" "/dev/input/mice"
Option "Name" "GlidePointPS/2"
Option "Protocol" "GlidePointPS/2"
Option "Vendor" "PS/2-Mouse"
Option "ZAxisMapping" "4 5"
EndSection

And I have also tried the Sax2 ALPS Touchpad mouse setup but that
gives me a failure message saying that I should look at /var/log/Xorg.99.log (it is long, so I'll only post a snippet)
On the surface it appears to try to use the Synaptics Driver, the tail end of the log reads:
(II) Synaptics touchpad driver version 0.14.6 (1406)
Mouse[1] waiting 100 msec to become devices ready
Mouse[1] waiting time total: 100
Mouse[1] waiting 100 msec to become devices ready
Many of These Messages Deleted
Mouse[1] waiting time total: 2100
Mouse[1] no synaptics event device found (checked 17 nodes, waited 2100 msec)
(**) Option "Device" "/dev/input/mice"
(**) Option "SHMConfig" "on"
(**) Option "LeftEdge" "120"
(**) Option "RightEdge" "830"
(**) Option "TopEdge" "120"
(**) Option "BottomEdge" "650"
(**) Option "FingerLow" "14"
(**) Option "FingerHigh" "15"
(**) Option "MaxTapTime" "180"
(**) Option "MaxTapMove" "110"
(**) Option "EmulateMidButtonTime" "75"
(**) Option "VertScrollDelta" "20"
(**) Option "HorizScrollDelta" "20"
(**) Option "EdgeMotionMinSpeed" "15"
(**) Option "EdgeMotionMaxSpeed" "15"
(**) Option "UpDownScrolling" "1"
(**) Option "CircularScrolling" "1"
(**) Option "CircScrollTrigger" "2"
Query no Synaptics: 6003C8
(EE) Mouse[1] no synaptics touchpad detected and no repeater device
(EE) Mouse[1] Unable to query/initialize Synaptics hardware.
(EE) PreInit failed for input device "Mouse[1]"
(II) UnloadModule: "synaptics"
(WW) <default pointer>: No Device specified, looking for one...
(II) <default pointer>: Setting Device option to "/dev/input/mice"
(--) <default pointer>: Device: "/dev/input/mice"
(==) <default pointer>: Protocol: "Auto"
(**) Option "AlwaysCore"
(**) <default pointer>: always reports core events
(==) <default pointer>: Emulate3Buttons, Emulate3Timeout: 50
(**) <default pointer>: ZAxisMapping: buttons 4 and 5
(**) <default pointer>: Buttons: 9
(**) <default pointer>: Sensitivity: 1
(WW) No core pointer registered
(II) XINPUT: Adding extended input device "<default pointer>" (type: MOUSE)
(II) XINPUT: Adding extended input device "Keyboard[0]" (type: KEYBOARD)
(--) <default pointer>: PnP-detected protocol: "ExplorerPS/2"
(II) <default pointer>: ps2EnableDataReporting: succeeded
No core pointer

Fatal server error:
failed to initialize core devices
(II) intel(0): xf86UnbindGARTMemory: unbind key 2
(II) intel(0): xf86UnbindGARTMemory: unbind key 3

Synaptics touchpad also failed as an option. Neither Ksynaptics nor gsynaptics will start up.

Thanks for all your help, sorry to be so long winded.

Bill M.
Comment 1 Foolish Ewe 2008-02-01 05:57:26 UTC
I guess I can limp on if it doesn't get fixed, but it would be great if we
could resolve this one, it would make this new laptop much more fun!

Thanks:

Bill M.
Comment 2 Marcus Schaefer 2008-02-01 09:06:02 UTC
Hmm, could you provide the following information:

  ===> save your current xorg.conf somewhere

  init 3
  sax2

  ===> click on change configuration
  ===> click on ok -> save

Provide the files:

/var/log/SaX.log
/etc/X11/xorg.conf

Thanks
Comment 3 Foolish Ewe 2008-02-01 16:34:04 UTC
Created attachment 192808 [details]
/etc/xorg.conf

Just to be sure, I removed the original and created a version with the only change being 3d acceleration being enabled.  I saved it and then reran Sax2 using the new file (so your Sax2.log won't have the missing xorg.conf messages in it).
Comment 4 Foolish Ewe 2008-02-01 16:37:13 UTC
Created attachment 192809 [details]
/var/log/SaX.log

this is the requested /var/log/SaX.log
Comment 5 Foolish Ewe 2008-02-01 16:41:43 UTC
Just adding this comment to be sure that the NEEDINFO status gets cleared.

Comments #3 and #4 should have the attachments, please let me know if you have any trouble reading the files.

Thanks:

Bill M.
Comment 6 Marcus Schaefer 2008-02-04 14:31:54 UTC
01-Feb 11:12:47 <I> [ Sysp: Mouse detection data
	Mouse0    =>  Protocol   : explorerps/2
	Mouse0    =>  Device     : /dev/input/mice
	Mouse0    =>  Buttons    : 5
	Mouse0    =>  Wheel      : 1
	Mouse0    =>  Emulate    : 0
	Mouse0    =>  Name       : ImPS/2 Generic Wheel Mouse
	Mouse0    =>  VendorID   : 0x0210
	Mouse0    =>  DeviceID   : 0x0013
	Mouse0    =>  Profile    : <undefined>
	Mouse0    =>  RealDevice : <undefined>
	Mouse0    =>  NutShell   : 0
]

the ID's doesn't point to an alps device ?

Steffen what do you think ?
Comment 7 Foolish Ewe 2008-02-04 16:28:06 UTC
I've noticed the status went back to NeedsInfo, is there anything I can do to help, should I provide any additional data?

Regards:

Bill M.
Comment 8 Marcus Schaefer 2008-02-05 08:22:28 UTC
I just set needinfo to Steffen Winterfeldt... we should wait for his input :)
Comment 9 Steffen Winterfeldt 2008-02-06 13:34:36 UTC
Please run 'hwinfo --mouse --log=foo' and attach the log.
Comment 10 Foolish Ewe 2008-02-07 07:07:32 UTC
Created attachment 193540 [details]
output of "/usr/sbin/hwinfo --mouse --log=mouse.log"

Hello Steffen:

Thank you for looking into this.  Due to recently released security updates, my kernel has been upgraded before running the hwinfo command to generate the mouse.log attachment.  The current version information reads:
uname -a
Linux linux-8whl 2.6.22.16-0.2-default #1 SMP 2008/02/01 19:36:55 UTC x86_64 x86_64 x86_64 GNU/Linux

If you require additional information or would like me to regenerate the previously attachments (xorg.conf and SaX.log), please let me know.

Regards:

Bill M.
Comment 11 Foolish Ewe 2008-02-07 07:10:08 UTC
I'm not sure if the NEEDINFO was set for me at this point or is still set for Steffen.   For now I'll leave it asserted, just in case.

Regards:

Bill M.
Comment 12 Marcus Schaefer 2008-02-07 12:57:42 UTC
the logs really doesn't point to an alps touchpad device... strange
Comment 13 Foolish Ewe 2008-02-07 15:09:55 UTC
Created attachment 193636 [details]
output of "dmesg"

Thanks for your comments Marcus and Steffen.  I'm puzzled as well.  I have a few questions:
1) Is there some way I can actually know who makes my touchpad.  As far as I know the main vendors are Synaptics and Alps, but the only real clue as to vendor comes from my Windows configuration, there is no eternal mark or other indication of the manufacturer and/or part number.
2) Regarding PS/2 mouse interface devices, how does Linux identify the device?  Could the ALPs device be mislabeled or sending identifying information outside the normal channel?
3) Besides the identification issue, is it possible to override the identification?
4) Just in case, I've attached the output of dmesg in desg.log.
5) Would additional information from the Vista driver help?

Regards:

Bill M.
Comment 14 Foolish Ewe 2008-02-12 15:44:48 UTC
Hello All:

Some quick remarks:

1) I meant "external mark" not "eternal mark" in point (1) of comment 13.
   I guess the unlucky number 13 put a hex on the 'X' key when typing :-).

2) I found a quick description of the PS/2 Mouse interface at:
   http://www.computer-engineering.org/ps2mouse/.  Is Linux following
   the protocol described there for device identification?
3) If (2) is correct, are you suggesting that my Laptop's touchpad
   is faking out the protocol by returning to Linux an ato answer of 0x03
   for device ID (indicating a Microsoft scrolling wheel mouse) for the
   mouse ID after setting the sample rates to first 200, then 100 and then 80?
4) How do other ps/2 devices (i.e. touchpads) negotiate the identification
   protocol?  In particular what is the Alps protocol?
5) Would I be right in thinking that Cirque actually designs and sells
   the touchpads sold by Alps?  Have you tried contacting Cirque to get
   developer support (i.e. go to http://www.cirque.com/pages/?section=12&page=54)
6) I've contacted my lap-top vendor (Acer) and have requested vendor and
   model data of my touchpad. 


Regards:

Bill M.
Comment 15 Foolish Ewe 2008-02-15 06:05:16 UTC
Markus, Steffen:

I realize you folks are busy, but do you have any thoughts or progress on this?

I was able to find the configuration tool under Vista, but only in administrator
mode.  It appears to be similar to ksynaptics in control from what I remember
of it, with the exception that I didn't notice a method for adjusting pressure
sensitivity.

Regards:

Bill M.
Comment 16 Foolish Ewe 2008-02-15 13:16:54 UTC
I took the recent security upgrades to the Linux kernel, which did not impact this behavior.

uname -a
Linux linux-8whl 2.6.22.17-0.1-default #1 SMP 2008/02/10 20:01:04 UTC x86_64 x86_64 x86_64 GNU/Linux


Regards:

Bill M.
Comment 17 Foolish Ewe 2008-02-23 03:36:56 UTC
Hello:

The OpenSuSE folks have been kind of quiet about this, I hope I haven't
upset them.  I have not gotten satisfaction from ACER on an exact identification
of my touchpad, but for now I'm goign forward on the assumption that it is
manufactured for Alps and is a Cirque design.

Then there are several touch pad kits at Cirque, however mine has 3 mechanical
buttons, which would eliminate all OEM options except the
TSM9957 Touchpad or a TSM3957 Touchpad.

Is it possible that the alps.c mouse driver might
not have support for this touchpad in the alphs_model_info table?
This table is accessed in the function alps_init for component detection.
I'd really appreciate some activity from the OpenSuSE folks, they had a nice
start, and know more about this sort of stuff than I would (which is why I'm guessing at an approach).

Regards:

Bill M.
Comment 18 Steffen Winterfeldt 2008-02-25 10:45:23 UTC
Sorry for the delay, I'll look into this ASAP, but really didn't have time
to during the last days.
Comment 19 Foolish Ewe 2008-02-26 03:48:29 UTC
Thanks for the update Steffen, I realize it can be hectic.

Regards:


Bill M.
Comment 20 Steffen Winterfeldt 2008-03-04 14:28:18 UTC
The kernel input device driver reprots it as 'ImPS/2 Generic Wheel Mouse'.
Before we start to apply strange workarounds, the kernel people should
have a look if that can't be improved.
Comment 21 Foolish Ewe 2008-03-05 00:55:06 UTC
Hi Steffen:

Thanks for the update.

My previous thoughts that Cirque was providing touchpads for laptops was wrong, they are not in any laptops according to Cirque (I'm not sure why I thought that, perhaps it was the fact that ALPs talks about glidepoint on their web pages which when I looked it up was mentioned as a Cirque product).

At this point, my best guess is that it is an ALPS, but I really have no more data than the original post.  However, my best guess is still that the device is identified in the alps.c source code module, and that may be a good place to look.
My current guess is that unless the device is initialized using the ALPS protocol, it will default to ImPS/2 emulation, so my previous guess that there may be a missing table entry might be worth considering.

Regards:

Bill M.
Comment 22 Foolish Ewe 2008-03-29 16:47:01 UTC
Hello Steffen and Markus:

It has been almost 1 month since the last update, so I thought I'd ask how things are going.  Have the kernel folks had a chance to look yet?

Currently, neither Acer nor ALPS has given me very specific information in response to my queries.

Regards:

Bill M.
Comment 23 Foolish Ewe 2008-05-02 02:27:04 UTC
Hello All:

I appreciate all of your efforts, but I haven't heard anything in the last 2 months or so and was hoping to see if there was any news.  Let me know if there
is any additional data or ways that I can cooperate to help move this along.

Thanks:

Bill M.
Comment 24 Stefan Dirsch 2008-05-08 08:56:30 UTC
(In reply to comment #20 from Steffen Winterfeldt)
> The kernel input device driver reports it as 'ImPS/2 Generic Wheel Mouse'.
> Before we start to apply strange workarounds, the kernel people should
> have a look if that can't be improved.

We could reassign to kernel component now, but the issue might have been fixed meanwhile. Therefore I suggest to give openSUSE 11.0 >= Beta2 a try first.


Comment 25 Foolish Ewe 2008-05-09 00:29:27 UTC
Thank you Stefan:

I'm currently using my laptop to do some deadline oriented work, so I'd like to consider my options.  Would it be enough to install a recent version of the kernel and boot it under 10.3?  Would it be possible to upgrade my installation,
or must I do a new install?

Regards:

Bill
Comment 26 Stefan Dirsch 2008-05-09 07:00:27 UTC
Not sure if you can simply install the kernel of 11.0 on 10.3. I didn't try this yet. Therefore I cannot recommend it. Upgrading your installationn would
be enough, but at the moment this would be a beta version. Maybe it would be
the best to wait for the upgrade until 11.0 has been released. But if you have
a partition left on your harddisk you could test with a new installation right now.
Comment 27 Alexander Egger 2008-05-09 20:23:33 UTC
I had the same problem. The reason is that the kernel does not detect the Alps touchpad correctly. There is a trivial fix see: 

http://lkml.org/lkml/2007/8/4/41

Works great on my Dell Vostro 1700
Comment 28 Stefan Dirsch 2008-05-09 22:51:37 UTC
> http://lkml.org/lkml/2007/8/4/41
The patch is included in Kernel of openSUSE 11.0. 
Comment 29 Foolish Ewe 2008-05-10 13:40:43 UTC
Created attachment 214185 [details]
dmesg after an upgrade from OpenSuse 10.3->11.0 beta 2

This is the dmesg after an upgrade from OpenSuse 10.3->11.0 beta 2
Comment 30 Foolish Ewe 2008-05-10 14:24:34 UTC
Based on the comments here, I bit the bullet and tried the upgrade from 10.3->11.0 Beta 2.  I may need to roll back but I thought I'd report a
few experiences first and see if I could get help.
1) My wireless no longer works.  I've attached the dmesg after the upgrade. This problem is a showstopper, any help would be appreciated.  If this problem
is not due to some brain damage on my part, I would recommend fixing this prior to shipping OpenSuse 11.0.
2) My machine locked hard when I went into Yast2->Hardware->Mouse Model and changed the model to Alps there is one choice, touchpad.  The first time my machine locked hard (required a hard reset with the power button/switch).  I'll try again after I send this and scan /var/log/messages for any clues.
3) ksynaptics and Sax2 have long had a troubling interaction that I don't know how to solve.  ksynaptics when I go to configure the touch pad (currently set up as a im/PS2 mouse) gives a popup complaining "Shared Memory is not accessible Please add the option 'SHMConfig "on"' into the touchpad section of /etc/X11/xorg.conf.
Historically I've noticed that my xorg.conf file would get overwritten every time X started, so I never could get the change to really take effect.  Is there a known solution for this.  If not, I'd like to request that this be addressed in the OpenSuse 11.0 release.
4) I had numerous packages added (mostly from packman) that were removed during the upgrade, I have not tried to adjust my repositories to fix this (yet), but this might just be part of the joy of going beta to fix a problem like this.

So in summary, I can't see any goodness in the new kernel, but that might be due to application side/beta testing issues.

Regards:

Bill M.
Comment 31 Foolish Ewe 2008-05-10 14:56:31 UTC
Hello:

As promised here is the follow up.
1) I tried testing the Alps Touchpad again.  It locked X (the timer did not advance and the mouse was non-responsive.  I was able to hotkey to a console
window and opened a root session, did an init 3 followed by an init 5 which
froze my system.
2) I noticed my time was advanced about 4 hours, I don't know why (I'm on U.S. east coast NY time).  I didn't explicitly set my clock and it apparently was set back 4 hours and advanced 4 hours in the logs.  I'll attach the logs after posting this.

I've attached the relevant part of /var/log/messages, beginning with the shut down before the upgrade was done, perhaps that will help.  If I can get a quick enough respnose, I could try a few things, but I may need to roll this machine back to 10.3 relatively soon given the state that it is in.

Regards:

Bill M. 
Comment 32 Foolish Ewe 2008-05-10 15:01:24 UTC
Created attachment 214193 [details]
/var/log/messages from time of 10.3->11.0 upgrade attempt

Logs from the time of shutdown till current time for the system after the upgrade.  The clock was not intentionally reset from 10:25 to 6:39 or to 14:30 or so.  I did intentionally reset the clock back to 10:34 or so.  Batteries are dying, got to go.
Comment 33 Stefan Dirsch 2008-05-10 21:25:17 UTC
So no improvements with kernel of 11.0. The other issue you've encountered when testing 11.0 needs to be reported serperately. Reassigning to kernel component.
Comment 34 Stefan Dirsch 2008-05-10 21:27:24 UTC
Kerel team, please see cocmment #20.
Comment 35 Foolish Ewe 2008-05-11 17:30:58 UTC
In comment 33, Stefan Dirsch recommends some issuing a separate report,
there are 2 issues not related to the mouse that I mention in comment 30:
1) Loss of wireless networking support using my intel 3945 card when going from 10.3->11.0 Beta 2.  Connectivity issues with Intel wireless chipsets appear in many other bugs, in my case I was an IPW user.  Searching, I've discovered numerous reports of others having similar problems, e.g. 38364, 372679, 366506, 377650, 360569, et. most likely mine is due to some related reason.  I've reverted to 10.3 since I felt it was unlikely to be fixed within the time frame I'd need wireless.
2) Issue number 3 in comment 30 about Sax2 and the SHMConfig On setting is really a 10.3 thing, I didn't get far enough to try it on 11.0, but would be willing to upgrade to 11.0 beta if the wireless networking issue was fixed and explore this further.  I noticed in Bug Report 32789 that Markus Schaefer indicated that this may be a symptom of the touchpad not being properly recognized.

Regards:

Bill M.
Comment 36 Stefan Dirsch 2008-05-11 17:48:55 UTC
Thanks a lot for the summary, Bill!
Comment 37 Jiri Kosina 2008-05-20 21:17:54 UTC
It is possible that you have model of alps that does need a special handling in the alps driver.

Could you please attach the output of kernel booted with i8042.debug commandline parameter?

Apparently, according to the hwinfo you have posted, the alps driver doesn't claim the device, so we'd need to add support for your device.

Thanks, and sorry for the delays you have experienced in this bug.
Comment 38 Foolish Ewe 2008-05-22 01:00:26 UTC
Created attachment 217436 [details]
/var/log/messages using 10.3 kernel with i8042.debug parameter

Thanks Jiri:

Please find attached the requested log of the latest 10.3 kernel with patches applied and i8042.debug command line boot parameter, 
> uname -a
Linux linux-5dp6 2.6.22.17-0.1-default #1 SMP 2008/02/10 20:01:04 UTC x86_64 x86_64 x86_64 GNU/Linux

The /var/log/messages were trimmed to only include the relevant records, and
I've tried to anonymize the log.

Regards:

Bill M.
Comment 39 Jiri Kosina 2008-05-23 12:34:50 UTC
Created attachment 217782 [details]
Input: alps update

Could you please test whether the kernel from 

     http://labs.suse.cz/jikos/download/bug-357881/

that has the attached patch applied makes your touchpad be driven by alps driver?

Thanks.
Comment 40 Foolish Ewe 2008-05-24 15:25:48 UTC
Created attachment 217969 [details]
/var/log/messages using the kernel from comment 39

Hello Jiri:

As per your suggestion in Comment 39, it appears that my touchpad is recognized After booting, with the i8042.debug option enabled, I used Yast2->Hardware->Mouse Model change selected ALPS and then only model is touchpad and it took.  In my next posting, I will attach the Yast2->Hardware->Hardware Information save to file output.

> uname -a
Linux linux-5dp6 2.6.22.18-0.1-default #1 SMP 2008-05-20 13:14:32 +0200 x86_64 x86_64 x86_64 GNU/Linux


My touchpad is working, but ksynaptics still gives the "Shared Memory is not accesssible. Please add the option 'SHMConfig "on"' intothe touchpad section of /etc/X11/xorg.conf".

Is ksynaptics the way to adjust the settings even on an ALPS touchpad?

Thanks:

Bill M.
Please find attached the relevant snippet of the log file.
Comment 41 Foolish Ewe 2008-05-24 15:38:10 UTC
Created attachment 217971 [details]
hardware information for Comment 39

Hello Jiri:

This is in response to Comment 39. Please find attached the hardware information mentioned in the previous comment.

Regards:

Bill M.
Comment 42 Jiri Kosina 2008-05-26 13:24:19 UTC
So is the touchpad now driven properly by alps driver instead of psmouse, when using the new kernel? (it can't be seen from the /var/log/messages you provided, as it unfortunately doesn't seem to contain the complete boot messages). Sorry, I got slightly confused by your last two comments, so I just want to make sure that everything is working as expected on kernel side with the new kernel.

Thanks.
Comment 43 Foolish Ewe 2008-05-26 18:40:38 UTC
Created attachment 218188 [details]
dmesg output from most recent reboot

Hi Jiri:

Let me try to be more clear and to make sure I understand what you want.

First let's address your request, would a dmesg log work?
In my current dmesg, I see lines like:
mice: PS/2 mouse device common for all mice
input: AT Translated Set 2 keyboard as /class/input/input0
input: PC Speaker as /class/input/input1
input: PS/2 Mouse as /class/input/input2
input: AlpsPS/2 ALPS GlidePoint as /class/input/input3

I've attached the output of dmesg from my most recent reboot.

Regarding my previous messages, I booted the system using the new kernel and
then used yast to change the mouse settings to the Alps and reported the /var/log/messages.   I cannot remember if I rebooted between the yast configuration of the mouse.

Regards:

Bill M.
Comment 44 Jiri Kosina 2008-05-28 12:43:56 UTC
So, as far as I understand correctly, the kernel part is fixed with the kernel I provided (your touchpad is now driven by alps driver properly), but there is still some problem on the X side, right?
If so, I will commit the patch to our kernel repository and re-assign to X people again.

Thanks.
Comment 45 Foolish Ewe 2008-05-28 12:54:18 UTC
Thanks Jiri for all your help.  I believe the touchpad is being detected by the kernel based on what I'm seeing, which is confirmed in your remarks in Comment 44. I'm not sure if more needs to be done.  Interestingly on this touch pad the middle "button" actually could be pushed on the top edge or bottom edge during intellimouse emulation and acted like scroll up/down, which doesn't appear supported when testing under the ALPS Touchpad settings in SaX2.  My best guess is that the patched kernel provided is doing the right thing but either my understanding of how to configure X isn't so good or X might not properly handle the setup on my laptop.  Given this, I'm going to clear the needinfo flag.

By the way Jiri, you are fast! I had to cancel a previous message since we collided (this bug tracking system is smart that way :-)).

Bill M.
Comment 47 Stefan Dirsch 2008-05-28 13:32:54 UTC
The remaining issue seems to be configuration related. Maybe synaptics driver is not in use at all. This needs to be investigated. For this please attach /etc/X11/xorg.conf, /var/log/Xorg.0.log and /var/log/SaX.log.
Comment 48 Foolish Ewe 2008-05-28 23:54:27 UTC
Created attachment 218756 [details]
/etc/X11/xorg/conf as per comment 47

Stefan:

Thanks for taking the initiative on this, the requested files will be attached
during my next feew comments.

Please find attached the /etc/X11/xorg.conf requested by Stefan Dirsch in Comment #47.

Regards:

Bill M.
Comment 49 Foolish Ewe 2008-05-28 23:57:46 UTC
Created attachment 218757 [details]
/var/log/Xorg.0.log as requested in Comment 47

Hello Stefan:

Please find attached /var/log/Xorg.0.log as per Comment 47.

Regards:

Bill M.
Comment 50 Foolish Ewe 2008-05-29 00:00:00 UTC
Created attachment 218758 [details]
/var/log/SaX.log as requested in comment 47.

Hello Stefan:

I think this completes the requested information, please find attached the /var/log/SaX.log as requested in Comment 47.

Regards:

Bill M.
Comment 51 Stefan Dirsch 2008-05-29 01:29:54 UTC
synaptics driver is configured and in use.

> Interestingly on this touch pad the middle "button" actually could be pushed
> on the top edge or bottom edge during intellimouse emulation and acted like
> scroll up/down, which doesn't appear supported when testing under the ALPS
> Touchpad settings in SaX2. 

Please check the synaptics manual page ("man synaptics") about the features of the synaptics driver. With synclient you can change settings on the fly. There are also graphical frontends, which influence some settings: gsynaptics for Gnome and ksynaptics for KDE.

I think we can close this one now as FIXED.
Comment 52 Foolish Ewe 2008-05-30 00:19:45 UTC
Hello Stefan, Jiri and Steffen:

Funny, just after I got Stefan's e-mail, I wanted to check if it was resolved (I was still nervous about KSynaptics and thought it might not work properly), but lo and behold, the ksynaptics tool works now, with much rejoicing!  Thanks Jiri, Steffen, Stefan and any other contributors I might have overlooked.  I realize how hard it must be to track these sorts of things down (especially without hands on hardware access).

Thanks:

Bill M.