OK, I need help with a Wacom Bamboo (Solved for one person)

[Meemaw]

[root@localhost meemaw]# wacdump /dev/input/event5
wacdump: wacusb.c:637: USBIdentifyModel: Assertion `nCnt == sizeof(keybits)' failed.
Aborted

Same for event6...... but event4 gives this;


MODEL=Unknown                           ROM=1.0-0
CLS=USB  VNDR=Unknown  DEV=Unknown  SUB=UNKNOWN

    LEFT=             MIDDLE=              RIGHT=              EXTRA=
    SIDE=              TOUCH=             STYLUS=            STYLUS2=
     BT0=                BT1=                BT2=                BT3=
     BT4=                BT5=                BT6=                BT7=
     BT8=                BT9=               BT10=               BT11=
    BT12=               BT13=               BT14=               BT15=
    BT16=               BT17=               BT18=               BT19=
    BT20=               BT21=               BT22=               BT23=

So, what's up with that?  There are no numbers at all - is it really event 4 or does the system save that from a previous try?  Is there a way to delete the ones that don't work?

Thanks for doing this during the holiday!!!! I really want to get it figured out for my daughter for Christmas rather than for her birthday in February!!!
 

[Toolfox]

Curiouser and curiouser...Let me verify that I've got this right so far:

You compiled and installed the 0.8.5-8 source code from the LinuxWacom Project? You do NOT have the LinuxWacom package from the PCLinuxOS Repository installed?

What version number does wacdump report? ( run either "wacdump -V " or "wacdump --version")

I just pulled down the 0.8.5-8 source code and took a quick look through it. When you compiled, did you include the option "--enable-wacom" when you ran ./configure to include building the kernel driver? The kernel module is NOT built by default; you might need an updated module to support the Bamboo Series. The kernel module is required to access UBS tablets, and wacdump depends on the kernel module to supply information to/from the tablet.

I think that accessing an input event node for a non-tablet device might give the wacdump results you posted; it did on my system, but I'm running the 0.8.2 version rom the PCLOS repository, which reports a wacdump version of 0.7.4. That version, according to the comments in the wacdump.c code file, dates back to May 2006. The code included in the most recent tarball reports version 0.8.2 is the most recent from December 2008. I just want to make sure that versions aren't being mixed together by accident.




Toolfox
 

[Meemaw]

Curiouser and curiouser...Let me verify that I've got this right so far:

You compiled and installed the 0.8.5-8 source code from the LinuxWacom Project? You do NOT have the LinuxWacom package from the PCLinuxOS Repository installed?

Correct - I removed ours and installed the 0.8.5-8

What version number does wacdump report? ( run either "wacdump -V " or "wacdump --version")

0.8.2

I just pulled down the 0.8.5-8 source code and took a quick look through it. When you compiled, did you include the option "--enable-wacom" when you ran ./configure to include building the kernel driver? The kernel module is NOT built by default; you might need an updated module to support the Bamboo Series. The kernel module is required to access UBS tablets, and wacdump depends on the kernel module to supply information to/from the tablet.

Yes, I used the command      ./configure --enable-wacom

I think that accessing an input event node for a non-tablet device might give the wacdump results you posted; it did on my system, but I'm running the 0.8.2 version rom the PCLOS repository, which reports a wacdump version of 0.7.4. That version, according to the comments in the wacdump.c code file, dates back to May 2006. The code included in the most recent tarball reports version 0.8.2 is the most recent from December 2008. I just want to make sure that versions aren't being mixed together by accident.

I don't think they are getting mixed up.

At this point I'm wondering if two modules are fighting to control the tablet and that's why I'm getting all the extra events - I just don't know how to fix it..... (of course, I'm sure I'm also still a beginner with huge gaps in what I know.......)
 

[Toolfox]

Three other thoughts:

1) Your mention of multiple drivers fighting over the tablet triggered an ancient memory. Check to see if there's a usbmouse kernel module running. If so, try removing it with either "rmmod usbmouse" or "modprobe -r usbmouse." I now remember having to do this in the past so my tablet is recognized as a tablet and not as a mouse. The permanent solution was to include "blacklist usbmouse" in the blacklist file under /etc/modprobe.d (but that's completely from memory without checking the actual locations -- I've got a Christmas dinner to get to.)

2) Recheck the report from /proc/bus/usb/devices. See if the line that contains "Driver=" reports "wacom," showing that the wacom kernel module has taken control of the tablet. There should also be information reporting the Vendor ID is 056a, which represents a device manufactured by Wacom.

3) Also check (if I'm getting the syntax correct) "more /proc/bus/input/devices" to see if there is again a "Vendor=056a" to indicate Wacom as a manufacturer, "Name=Wacom [additional ID info]" to indicate that the kernel module recognizes your particular model, and "Handlers= [/dev/input/ node] to say which event node is connected between the hardware and software.

I'd almost bet cash money you're right: something silly is getting in the way of the Wacom kernel module. I know it's going to be a monster list, but maybe post your entire list of running kernel modules with the tablet plugged in.





Toolfox

[Meemaw]

1) Your mention of multiple drivers fighting over the tablet triggered an ancient memory. Check to see if there's a usbmouse kernel module running. If so, try removing it with either "rmmod usbmouse" or "modprobe -r usbmouse." I now remember having to do this in the past so my tablet is recognized as a tablet and not as a mouse. The permanent solution was to include "blacklist usbmouse" in the blacklist file under /etc/modprobe.d (but that's completely from memory without checking the actual locations -- I've got a Christmas dinner to get to.)

[root@localhost meemaw]# rmmod usbmouse
ERROR: Module usbmouse does not exist in /proc/modules
(so I'll have to do a module list.....)

2) Recheck the report from /proc/bus/usb/devices. See if the line that contains "Driver=" reports "wacom," showing that the wacom kernel module has taken control of the tablet. There should also be information reporting the Vendor ID is 056a, which represents a device manufactured by Wacom.

T:  Bus=04 Lev=01 Prnt=01 Port=01 Cnt=01 Dev#=  3 Spd=12  MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=056a ProdID=00d3 Rev= 1.06
S:  Manufacturer=Wacom Co.,Ltd.
S:  Product=CTH-661
C:* #Ifs= 2 Cfg#= 1 Atr=80 MxPwr= 98mA
I:* If#= 0 Alt= 0 #EPs= 1 Cls=03(HID  ) Sub=01 Prot=02 Driver=wacom
E:  Ad=81(I) Atr=03(Int.) MxPS=   9 Ivl=4ms
I:* If#= 1 Alt= 0 #EPs= 1 Cls=03(HID  ) Sub=00 Prot=00 Driver=wacom
E:  Ad=82(I) Atr=03(Int.) MxPS=  64 Ivl=4ms

3) Also check (if I'm getting the syntax correct) "more /proc/bus/input/devices" to see if there is again a "Vendor=056a" to indicate Wacom as a manufacturer, "Name=Wacom [additional ID info]" to indicate that the kernel module recognizes your particular model, and "Handlers= [/dev/input/ node] to say which event node is connected between the hardware and software.

This Vendor=056a ProdID=00d3   and this   Product=CTH-661 are correct.....

I'd almost bet cash money you're right: something silly is getting in the way of the Wacom kernel module. I know it's going to be a monster list, but maybe post your entire list of running kernel modules with the tablet plugged in.

I'll save it to a txt file and pm it to you (to save room in the forum)

Right now, you go enjoy Christmas!!!!  This doesn't have to be done today.... I appreciate all your wonderful help!!!!
Merry Christmas and enjoy that dinner and your family.......
 

[Toolfox]

I looked through your kernel module list...and I don't see anything that strikes me as a possible problem.

Here's another thought I just came up with concerning who or what might be blocking access to the tablet's interface:

XWindows itself.

If XWindows sees the hardware as a USB mouse, it might be attaching itself to the hardware, preventing wacdump from reading the output data stream. This has been a known situation with the wacom drivers/utilities.

I just tried this on my system: Boot your computer and select the "Safe Mode" option instead of normal boot. In the middle of booting text-mode, the system will drop into "Single-User Mode" with a command prompt that looks like a single "#" symbol. Run wacdump from there. You'll have to know which event node to use, so run the command "more /proc/bus/input/devices" first to get that information. The "more" utility pauses after each screenful of output information and continues to the next screenful with each press of the spacebar (you can get help on the more command in the middle of using it by typing "h" instead of <space>).

For example, the final device on my system is my Graphire4, which looks like this:

I: Bus=0003 Vendor=056a Product=0016 Version=0403
N: Name="Wacom Graphire4 6x8"
P: Phys=
S: Sysfs=/class/input/input6
U: Uniq=
H: Handlers=mouse1 event6
B: EV=1f
B: KEY=1c63 0 70011 0 0 0 0 0 0 0 0
B: REL=100
B: ABS=100 3000003
B: MSC=1

My handlers are mouse1 and event6, so I could test my tablet with /dev/input/event6 or /dev/input/mouse1. You'll need to plug in whatever values are particular to your system.




Toolfox

[Meemaw]

Well, I think my safe mode uses the older kernel and everything was installed to the newer kernel.... and so it won't do wacdump........ it doesn't see the wacom driver, but it sees two other instances of it.... both Wacom Bamboo P+T 6x8.... one says
Sysfs=class/input/input4       Handlers= mouse1 event 4

and the other says
Sysfs=class/input/input5       Handlers= mouse2 event 5

(fun, fun)

[Toolfox]

If you need a safemode entry for the newer kernel, try creating one using the "Boot --> Set Up Boot System" utility in the Control Center. Drill through the wizard until you get to the listing of boot options. Create a new option -- maybe call it "2.6.27 failsafe" or something that makes sense to you. Copy the settings from the normal boot option for the newer kernel except for the append options; take those from the old kernel failsafe.

In my system, the failsafe append options number 2:

failsafe acpi=on

Another solution would be to temporarily disable autobooting into X. That's set as a part of the autologin system, also located in the Control Center's Boot area. It's called "Set Up Autologin to automatically log in," and there's two radio buttons for "Autologin the following user" and "No, I don't want to autologin." The checkbox for "Launch the graphical environment when your system starts" can be cleared to prevent XWindows from autostarting. To start XWindows after you're done experimenting, type "startx" and you'll be back in a familiar desktop setting. Don't forget to reset your boot options.

Don't use both options simultaneously -- this is an either/or chioce.




Toolfox

[Meemaw]

Thanks, but at this point, it's on her laptop..... doing the same thing as on mine - the system sees it in PCC as two different mouse entries and one tablet entry, so we have the exact same problem. If we can run wacdump to figure out which event we're supposed to use with the wacom driver, how do we get rid of the others? (They say the modules they are using are mouse1 and mouse2......)

[Toolfox]

How about this, then:

Post the results of running the following command in a terminal:

cat /proc/bus/input/devices

This will tell us which event (or events) the Wacom module is using to connect the hardware to the system. It will also tell us ALL of the hardware that requires an event hook and where its going.



Toolfox

[coyoteawooo]

MeeMaw's breaking my computer again. 


I: Bus=0011 Vendor=0002 Product=0007 Version=25b1
N: Name="SynPS/2 Synaptics TouchPad"
P: Phys=isa0060/serio4/input0
S: Sysfs=/class/input/input0
U: Uniq=
H: Handlers=mouse0 event0
B: EV=b
B: KEY=6420 0 70000 0 0 0 0 0 0 0 0
B: ABS=11000003

I: Bus=0011 Vendor=0001 Product=0001 Version=ab41
N: Name="AT Translated Set 2 keyboard"
P: Phys=isa0060/serio0/input0
S: Sysfs=/class/input/input1
U: Uniq=
H: Handlers=kbd event1
B: EV=120013
B: KEY=4 2000000 3803078 f800d001 feffffdf ffefffff ffffffff fffffffe
B: MSC=10
B: LED=7

I: Bus=0019 Vendor=0000 Product=0002 Version=0000
N: Name="Power Button (FF)"
P: Phys=LNXPWRBN/button/input0
S: Sysfs=/class/input/input2
U: Uniq=
H: Handlers=kbd event2
B: EV=3
B: KEY=100000 0 0 0

I: Bus=0019 Vendor=0000 Product=0005 Version=0000
N: Name="Lid Switch"
P: Phys=PNP0C0D/button/input0
S: Sysfs=/class/input/input3
U: Uniq=
H: Handlers=event3
B: EV=21
B: SW=1

I: Bus=0019 Vendor=0000 Product=0001 Version=0000
N: Name="Power Button (CM)"
P: Phys=PNP0C0C/button/input0
S: Sysfs=/class/input/input4
U: Uniq=
H: Handlers=kbd event4
B: EV=3
B: KEY=100000 0 0 0

I: Bus=0019 Vendor=0000 Product=0006 Version=0000
N: Name="Video Bus"
P: Phys=/video/input0
S: Sysfs=/class/input/input5
U: Uniq=
H: Handlers=kbd event5
B: EV=3
B: KEY=3f000b 0 0 0 0 0 0 0

I: Bus=0019 Vendor=0000 Product=0006 Version=0000
N: Name="Video Bus"
P: Phys=/video/input0
S: Sysfs=/class/input/input6
U: Uniq=
H: Handlers=kbd event6
B: EV=3
B: KEY=3f000b 0 0 0 0 0 0 0

I: Bus=0003 Vendor=056a Product=00d3 Version=0106
N: Name="Wacom Bamboo P&T 6x8"
P: Phys=
S: Sysfs=/class/input/input7
U: Uniq=
H: Handlers=mouse1 event7
B: EV=1f
B: KEY=1c63 0 70033 0 0 0 0 0 0 0 0
B: REL=100
B: ABS=100 3000103
B: MSC=1

I: Bus=0003 Vendor=056a Product=00d3 Version=0106
N: Name="Wacom Bamboo P&T 6x8"
P: Phys=
S: Sysfs=/class/input/input8
U: Uniq=
H: Handlers=mouse2 event8
B: EV=1f
B: KEY=1c63 0 70033 0 0 0 0 0 0 0 0
B: REL=100
B: ABS=100 3000103
B: MSC=1

[Meemaw]

MeeMaw's breaking my computer again. 

Well, I wouldn't say that, but I seem to have bought something that won't work without LOTS of configuration!!!!

Thanks a million, Toolfox!!!
 

[Toolfox]

Now that I've got some more info to go on, I think I see what's going on:

This tablet is two devices in one -- a traditional stylus/puck tablet and a finger-driven touchpad (hence the "P&T" in the name...no relation to Penn & Teller, though... ). The two devices use two different event inputs -- one for the tablet with pressure and tilt, one for the touchpad.

The two inputs on Coyoteawooo's system are going to be /dev/input/event7 and /dev/input/event8.

Try running wacdump with each of the above input nodes. Does the tablet's pen respond to either input node? Does the tablet's touchpad respond to either input node?

Try running xxd with each of the above input nodes. Does the tablet's pen respond to either input node? Does the tablet's touchpad respond to either input node?

Try running the system without the X server running. You can do that in failsafe mode...as I detailed before. If you don't have a failsafe mode for the kernel you have to use, you can temporarily create a boot entry to do that...as I detailed before. If you don't want to create a new boot entry, you can set the system to not start XWindows automatically...again, as I detailed above.

We could also just plunge ahead and setup xorg.conf for one of the inputs. If it doesn't work, we can edit the input references to point to the other one.

Let's see if we can get any response out of the tablet directly, first.




Toolfox

[coyoteawooo]

For the wacdump on event7 or event8, it says  "wacdump: wacusb.c:637: USBIdentifyModel: Assertion 'nCnt == sizeof(keybits)' failed.  Aborted"  And there is no response from the pen on either.  When we run the xxd, it says nothing at all, and the stylus shows no response.

When we run wacdump in safe mode, it says "ERROR: Failed to open /dev/input/event7 No such file or directory WacomOpenTablet No such file or directory."  And the same for event8 or mouse1. But if we run event4 we get the above "Aborted" message and event5 gives the Unknown Device with the chart that's supposed to have all the numbers on it.

[Toolfox]

I thought the target machine is using an updated kernel (2.6.27.31)...which means that the failsafe boot option uses the older kernel (2.6.26.. The wacom modules were NOT compiled for the older kernel, which would explain why you don't have the right event nodes.

As I've said before, try disabling the autostart of XWindows in the Control Center --> Boot -->  Set up autologin --> Clear the "Launch the graphical environment" checkbox -->  Click the "OK" button. Now restart your system normally (NOT safemode) with the tablet plugged in. You should get a console login where you can type your username + enter key and password + enter key to log in.

Verify that the wacom module loaded:

lsmod | grep wacom

(that character between lsmod and grep is the pipeline symbol: a vertical line. On my keyboard, it's a Capital+Backslash)

Run that "more /proc/bus/input/devices" command to look over what devices are using what event inputs (they should be somewhat in numerical order). Use the spacebar to display the next screenful of output (the "more" command lets you look at text output one screen at a time. You can hit the "h" key in the middle of this to bring up More's help information). Compare these results to the one you reported the other day:

Event0 = Laptop Touchpad
Event1 = keyboard
Event2 = Power Button
Event3 = Lid Switch
Event4 = Power Button
Event5 = Video
Event6 = Video
Event7 = Wacom Tablet
Event8 = Wacom Tablet

Let's see if the order and number of devices stay the same; they should.

If all is well enough, try running wacdump or xxd. Eliminating XWindows might be what we need to verify functionality of the hardware and drivers. My current thought is that XWindows might be seeing the tablet as a mouse and is attaching to it exclusively. This is why wacdump or xxd can't get any response; XWindows would be diverting all info through itself.




Toolfox


P.S. -- To get back into XWindows after these experiments are completed, type "startx"