#11756 NORM 12.1.0: XO-1.75 - Add ELAN touchpad driver support in kernels (was: Add ELAN touchpad driver support in kernels)

Zarro Boogs per Child bugtracker at laptop.org
Wed Jun 6 12:54:14 EDT 2012

#11756: XO-1.75 - Add ELAN touchpad driver support in kernels
           Reporter:  martin.langhoff  |       Owner:  martin.langhoff
               Type:  task             |      Status:  new            
           Priority:  normal           |   Milestone:  12.1.0         
          Component:  kernel           |     Version:  not specified  
         Resolution:                   |    Keywords:                 
        Next_action:  diagnose         |    Verified:  0              
Deployment_affected:                   |   Blockedby:                 
           Blocking:                   |  

Comment(by martin.langhoff):

 Attached zImage-ELAN-patch, which is the latest XO-1.75 kernel from the
 12.1.0 OS plus the provided patch for ELAN. As dsd points out, it is a
 backport of a few patches already upstream.

 To test

  * Install latest 12.1.0 image (tested on "os12")
  * On a terminal

  rm /bootpart/boot/vmlinuz
  cp /run/media/olpc/MYUSB/zImage-ELAN-patch /bootpart/boot/vmlinuz

  * After reboot, the touchpad will be very sensitive, but tap-to-click
 will be disabled. In a terminal...

  # this command should show kernel version 3.0.19-01537-g3d86262
  uname -a
  # tweak the mousepad acceleration with xset. I get reasonable
 acceleration with
  xset m 1/6 1
  # check dmesg for correct detection
  dmesg | grep '\(lantech\|psmouse\)'
  # check the sysfs node for things like protocol
  cat /sys/bus/serio/drivers/psmouse/serio1/protocol



  * Very jittery. A "standing finger" shows a jittery pointer, even on
 battery. Looks like it will need tweaking.
  * S/R works but the "wakeup" stroke is completely lost. Maybe there is an
 issue with buffering?
  * Can we control the acceleration with commands to the sysfs node? This
 would be better than xset as we can do it from a udev rule.
  * Initial tests indicate we can we quiesce the device, as we do with AVC
  * We need to keep an eye on the "lost sync" issue dsd spotted. Running
 "dmesg | grep lost\ sync" did not show any hits in my testing.

Ticket URL: <http://dev.laptop.org/ticket/11756#comment:12>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system

More information about the Bugs mailing list