[i855] netbook launcher mouse focus is off

Thanks for reporting this bug. Seems like your on Intel graphics, if so, could you please see this bug https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/349314, and test if the kernel provided there works for you?

i have this too. HP ze4949, about 5 years old. tried the new kernel posted above, it fixes nothing. attached is my lspci.

Tried the above listed kernel as well on my IBM r51. Doesn't fix this issue at all. I plugged in a usb mouse and it has the same issues as the mousepad.

I think I have this bug as well on my Benq Joybook 6000.
The proposed new kernel seems to fix an issue with drawing speed.
I do not have any speed issues at all.
What it seems to be is that netbook-launcher seems to think my screen has a different wdht than it has in reality. When I move my mouse, icons are selected that are on a completely different X-position.
But it is not only the mouse!
I have five rows of icons on my main desktop, but netbook-laucher seems to think there are four!
So, down-arrow will move from 1st icon in 1st row to 5th icon in 1st row, to 4th icon in 2nd row, to 3rd icon in 3rd row, etc.
Hope this helps in finding the cause.

I have the same issue with Netbook remix 9.04 running on a Travelmate C110 tablet (standard 1024x768 screen). Mouse mappings seem to work perfectly in all of the windows (firefox etc) apart from the launcher / desktop itself. I would agree that the issue seems to be to do with screen width. It seems as if the launcher is expecting my monitor to be widescreen and has mapped accordingly. The central section has space for four icons. When my mouse is over the 1st, the 3rd is highlighted, 2nd to 4th. When it reaches halfway however, the problem reverses and becomes 3rd to 1 etc. It appears as if the screen is being mapped in from the sides as a widescreen and then when it runs out of space on my old width screen, bounces back to the edges.
This isn't a problem on the normal desktop for jaunty with my computer.
Is there a configuration file for the netbook launcher that tells it what the screen size is?

Upgraded to karmic on my dell D400 with intel graphics chip this is still a problem in the new version of netbook-launcher

Just installed UNR in a small VAIO VGN-T70B with Intel graphics 82852/855GM and a display with 1280x768 pixels. I observe the behavior of strange mouse offset but also somewhat inconsistent, like described by Henk de Leeuw and Gethin.

The issue your seeing is definitely a bug with the graphics driver. It also may be that the launcher is running in with software acceleration. Can you please attach the output of `glxinfo` to this bug, and also the output generated by the `glxgears` command (leave it running for about 20 seconds).

Same problem on dell inspiron 700m

name of display: :0.0
get fences failed: -1
param: 6, val: 0
display: :0 screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.2
server glx extensions:
    GLX_ARB_multisample, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_OML_swap_method, GLX_SGI_make_current_read, GLX_SGI_swap_control,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_visual_select_group
client glx vendor string: SGI
client glx version string: 1.4
client glx extensions:
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_allocate_memory,
    GLX_MESA_copy_sub_buffer, GLX_MESA_swap_control,
    GLX_MESA_swap_frame_usage, GLX_OML_swap_method, GLX_OML_sync_control,
    GLX_SGI_make_current_read, GLX_SGI_swap_control, GLX_SGI_video_sync,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer,
    GLX_SGIX_visual_select_group, GLX_EXT_texture_from_pixmap
GLX version: 1.2
GLX extensions:
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_MESA_swap_control, GLX_MESA_swap_frame_usage, GLX_OML_swap_method,
    GLX_SGI_make_current_read, GLX_SGI_swap_control, GLX_SGI_video_sync,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_visual_select_group
OpenGL vendor string: Tungsten Graphics, Inc
OpenGL renderer string: Mesa DRI Intel(R) 852GM/855GM GEM 20090326 2009Q1 RC2 x86/MMX/SSE2
OpenGL version string: 1.3 Mesa 7.4
OpenGL extensions:
    GL_ARB_multisample, GL_ARB_multitexture, GL_ARB_pixel_buffer_object,
    GL_ARB_point_parameters, GL_ARB_texture_border_clamp,
    GL_ARB_texture_compression, GL_ARB_texture_cube_map,
    GL_ARB_texture_env_add, GL_ARB_texture_env_combine,
    GL_ARB_texture_env_crossbar, GL_ARB_texture_env_dot3,
    GL_ARB_texture_mirrored_repeat, GL_ARB_texture_rectangle,
    GL_ARB_transpose_matrix, GL_ARB_vertex_buffer_object,
    GL_ARB_vertex_program, GL_ARB_window_pos, GL_EXT_abgr, GL_EXT_bgra,
    GL_EXT_blend_color, GL_EXT_blend_equation_separate,
    GL_EXT_blend_func_separate, GL_EXT_blend_logic_op, GL_EXT_blend_minmax,
    GL_EXT_blend_subtract, GL_EXT_clip_volume_hint, GL_EXT_cull_vertex,
    GL_EXT_compiled_vertex_array, GL_EXT_copy_texture,
    GL_EXT_draw_range_elements, GL_EXT_framebuffer_object, GL_EXT_fog_coord,
    GL_EXT_multi_draw_arrays, GL_EXT_packed_depth_stencil,
    GL_EXT_packed_pixels, GL_EXT_pixel_buffer_object, GL_EXT_point_parameters,
    GL_EXT_polygon_offset, GL_EXT_rescale_normal, GL_EXT_secondary_color,
    GL_EXT_separate_specular_color, GL_EXT_stencil_wrap, GL_EXT_subtexture,
    GL_EXT_texture, GL_EXT_texture3D, GL_EXT_texture_edge_clamp,
    GL_EXT_texture_env_add, GL_EXT_texture_env_combine,
    GL_EXT_texture_env_dot3, GL_EXT_texture_filter_anisotropic,
    GL_EXT_texture_lod_bias, GL_EXT_texture_object, GL_EXT_texture_rectangle,
    GL_EXT_vertex_array, GL_3DFX_texture_compression_FXT1,
    GL_APPLE_client_storage, GL_APPLE_packed_pixels,
    GL_ATI_blend_equation_se...

dell 700m glxgears 283 frames in 5.0 seconds = 54.557 fps

same problem here with ubuntu 9.10 alpha 6 on ibm x40

same problem here with 9.04 on thinkpad x40

Same problem here with ubuntu 9.10 alpha on Acer TravelMate 370

Same problem here with Ubuntu 9.10 release candidate on old Fujitsu-Siemens Lifebook P7010.

It seems like the bug is confirmed on different configurations with Intel cards. Is there a possible woakround at UNR's level?

No, the issue is with gl_read_pixels inside of mesa. The only way around it is to change the entire input scheme in clutter (which they have honestly considered). I don't think there is anything we can do about this one, the drivers have to implement gl_read_pixels properly.

On side note, this is *most* common on systems with L shaped memory (e.g. a 1GB and a 2GB memory module). Users may experience better results with rectangular memory. I believe the behavior of gl_read_pixels on the intel drivers with L shaped memory is still "undefined" aka "broken" but my information is about a month old. Still I think thats relevant to 9.10's drivers.

The L shaped memory thing probably goes a long way to explain why my (quick) glance at the models of the notebooks of those effected are not netbooks.

The Lifebook P7010 that I have this bug on is configured with a single 512MB module.

So if UNR is not going to fix this, is it mesa or a driver that should be fixed? And if so, can we have this bug moved to that?

Same problem have been identified on a Toshiba NB200-10Z (Same at NB205 - just with Scandinavian Keybord).

Neil,

It might be easiest to rule out driver issues (or rule in) by providing some patched clutter packages that provide some additional debugging in this area. It would be cool if we could convince it to save the offscreen hit detection render with a marker for where it thinks the cursor is. It should more or less confirm the gl_read_pixels diagnosis.

Hi,
I have this problem on a Fujitsu P7010 with twin matched 512Mb sticks of RAM. I haven't tried using just the keyboard, but I have the mouse issue.
This was a problem for me in Jaunty and is still a problem in Karmic. It worked with Intrepid.
What can I (IE how do I) help you rule in/out driver issues.

Regards
Otto

I gotta work with neil to figure out how to proceed. He knows clutter internals pretty well and this is something that is rather difficult to hunt down.

I will also be happy to help resolve this issue. Just let me know what to do.

Architecture: i386
DistroRelease: Ubuntu 9.10
GConfNonDefault:
 /apps/netbook-launcher/disable_single_instance=false
 /apps/netbook-launcher/force_low_graphics=false
LiveMediaBuild: Ubuntu-Netbook-Remix 9.10 "Karmic Koala" - Release i386 (20091028.4)
Lsusb:
 Bus 001 Device 002: ID 13fe:3123 Kingston Technology Company Inc.
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: Dell Computer Corporation Inspiron 500m
Package: netbook-launcher 2.1.12-0ubuntu3
PackageArchitecture: i386
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
ProcCmdLine: initrd=/ubninit file=/cdrom/preseed/netbook-remix.seed boot=casper quiet splash -- BOOT_IMAGE=/ubnkern
ProcEnviron:
 SHELL=/bin/bash
 LANG=en_US.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
RelatedPackageVersions:
 xserver-xorg 1:7.4+3ubuntu7
 libgl1-mesa-glx 7.6.0-1ubuntu4
 libdrm2 2.4.14-1ubuntu1
 xserver-xorg-video-intel 2:2.9.0-1ubuntu2
 xserver-xorg-video-ati 1:6.12.99+git20090929.7968e1fb-0ubuntu1
Tags: ubuntu-unr ubuntu-unr
Uname: Linux 2.6.31-14-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
XorgConf: Error: [Errno 2] No such file or directory: '/etc/X11/xorg.conf'
XorgLogOld: Error: [Errno 2] No such file or directory: '/var/log/Xorg.0.log.old'
dmi.bios.date: 07/14/2004
dmi.bios.vendor: Dell Computer Corporation
dmi.bios.version: A08
dmi.board.vendor: Dell Computer Corporation
dmi.chassis.type: 8
dmi.chassis.vendor: Dell Computer Corporation
dmi.modalias: dmi:bvnDellComputerCorporation:bvrA08:bd07/14/2004:svnDellComputerCorporation:pnInspiron500m:pvr:rvnDellComputerCorporation:rn:rvr:cvnDellComputerCorporation:ct8:cvr:
dmi.product.name: Inspiron 500m
dmi.sys.vendor: Dell Computer Corporation
fglrx: Not loaded
system:
 distro: Ubuntu
 architecture: i686kernel: 2.6.31-14-generic

I hope the above attachments sent by apport-collect are appropriate. I tried to follow David Barth's request in https://bugs.launchpad.net/netbook-remix-launcher/+bug/456229/comments/1 and the instructions at https://help.ubuntu.com/community/ReportingBugs.

On my Dell Inspiron 500m, the problem existed already in 9.04 Jaunty, and got worse in 9.10 Karmic beta and release. I'm running 9.10 Karmic release from the USB stick created with unetbootin, and 9.10 Karmic beta installed on a 4 GiB CompactFlash card replacing the hard disk with an adapter.

Here are my results with different memory configurations and UNR versions.
DIMM A / DIMM B / UNR version: symptoms (see descriptions below)
-----------------------------------------------------------------
1 GiB 256 MiB 9.04 release: A (bad)
1 GiB 256 MiB 9.10 release: B (worse)
1 GiB none 9.10 release: B (worse)
none 1 GiB 9.10 release: B (worse)
256 MiB none 9.10 release: B/C [very slow of course, unsure]
1 GiB 256 MiB 9.10 release: B (worse) or C (worst)
256 MiB 1 GiB 9.10 release: B (worse)

Symptom A (bad): In Jaunty (1GiB+256MiB), application icons did react when the mouse cursor was somewhere way off to the left from the actual icon. IIRC, the category tabs worked quite ok. I think this is the same as the original report and most comments above (e.g. #7).

Symptom B (worse): In Karmic beta and release, the category tabs only highlight with the mouse cursor on or to the left of the category icon. No highlighting happens if the cursor is on the right side of the icon, the label or elsewhere in the tab. Also, it seems that the "active area" is horizontally "striped", i.e. the tab reacts at only certain vertical coordinates. Clicking on application icons doesn't work at all. Icons are never highlighted when hovering on them, and they don't respond to clicks. Changing "Visual effects" from "None" to "Normal" doesn't change the behavior. Desktop effects work ok in other application windows.

Symptom C (worst): On two occasions in Karmic release, the category tabs reacted extremely sporadically, and the application icons never reacted at all. Changing "Visual effects" from "None" to "Normal", the launcher was "faded" (partially transparent) and totally unresponsive to the mouse and the keyboard for about 1-2 minutes. After it recovered, the behavior was the same as with no visual effects. Desktop effects worked ok in other application windows.

In the panel and all other applications the mouse works perfectly.

I always get the same symptoms with the native 1400x1050 and also a lower 1024x768 resolution.

Drivers from the xorg-edgers PPA (https://launchpad.net/~xorg-edgers/+archive/ppa) seem to work (at least worked for me)

opened a task against intel xorg driver since it looks increasingly likely to be the cause

Julian, thanks for your post. Updating after adding the PPA has fixed it for me. Looking forward to this fix reaching main stream.

The xorg-edgers driver doesn't fix the issue on my system:

$ sudo add-apt-repository ppa:xorg-edgers/ppa
$ sudo apt-get update
$ sudo apt-get install xserver-xorg-video-intel
$ dpkg -s xserver-xorg-video-intel|grep Version:
Version: 2:2.9.0+git20091026.10946118-0ubuntu0tormod
$ sudo service gdm restart

I still have symptom B (see comment #45). I also tried logging out and back in again, didn't help.

Is there something else I have to do to in a live CD session to make sure the upgraded driver is running?

I got it working on my Fujitsu P7010D with the following (similar to Otto):

$ sudo add-apt-repository ppa:ubuntu-x-swat/x-updates
$ sudo apt-get update
$ sudo apt-get upgrade
$ sudo service gdm restart

Don't know about any possible unstablity or other side effects with this yet... Hope for none!

I first tried the ubuntu-x-swat/x-updates ppa, as suggested by Konsuli, and it didn't help.

I then tried the xorg-edgers/ppa, as suggested by Antti, and it did the trick.
I did a full upgrade, but I assume it was the xserver-xorg-video-intel upgrade that did it.

As this bug is really critical for those affected, I hope the necessary changes from the xorg-edgers/ppa is pushed to karmic-updates soon.

Like itchyd wrote.

Dell inspiron 700m with Ubuntu Netbook 9.10 still with the problem.

The problem persist in the launcher. Once an application is opened, the mouse pointer responds ok.

Hi,
with a VIA Technologies, Inc. VT8623 [Apollo CLE266] integrated CastleRock graphics (rev 03)
I've the same problem..

I can confirm that upgrading to the xorg-edgers PPA fixed this problem for me
on a Dell Latitude D400.

  sudo add-apt-repository ppa:xorg-edgers/ppa
  sudo apt-get update
  sudo apt-get upgrade

and then restart gdm or the whole machine.

On a Gateway 400VTX laptop with an "Intel Corporation 82852/855GM Integrated Graphics Device rev 1"

I did this a few days ago:

sudo add-apt-repository ppa:xorg-edgers/ppa
sudo apt-get update
sudo apt-get install xserver-xorg-video-intel

and it did not fix the issue then, but today after the last update and a reboot, it works fine.

dpkg -s xserver-xorg-video-intel|grep Version:
Version: 2:2.9.0+git20091125.917f9bb2-0ubuntu0tormod~karmic

xorg-edgers/ppa fixed the issue on IBM x40.

Performing the following fixed this issue on my Dell 700m with Ubuntu 9.10 Netbook Remix (with all available updates installed):

sudo add-apt-repository ppa:xorg-edgers/ppa
sudo apt-get update
sudo apt-get upgrade

Now to fix a few other issues including Sound Card Gain... somehow mine is way too loud which is not the usual case :[

Hi,

I've seen your post on launchpad and regarding the second issue with
sound, did you try alsamixer from the command line?

It helped me resolve an issue with sound on my netbook. If you play with
the various levels, you may well find what you need.

regards,
Jacques

-------- Message initial --------
De: polskikrol <email address hidden>
Reply-to: Bug 368394 <email address hidden>
À: <email address hidden>
Sujet: [Bug 368394] Re: netbook launcher mouse focus is off
Date: Mon, 14 Dec 2009 02:00:00 -0000

Performing the following fixed this issue on my Dell 700m with Ubuntu
9.10 Netbook Remix (with all available updates installed):

sudo add-apt-repository ppa:xorg-edgers/ppa
sudo apt-get update
sudo apt-get upgrade

Now to fix a few other issues including Sound Card Gain... somehow mine
is way too loud which is not the usual case :[

Hi, I have Dell Latitude E4200 with an integrated intel graphics, latest Karmic; I had the same problem, mouse clicks in launcher worked only in in certain places, both in tabs and for application icons. Mouse highlight worked ok, though.

Upgrade of only the xorg-edgers PPA xserver-xorg-video-intel did not help, but apt-get upgrade did.
Thanks!

This seems to be fixed in Lucid with xserver-xorg-video-intel 2:2.9.1-1ubuntu1 and netbook-launcher 2.1.12-0ubuntu4.

The consensus seems to be that this is a video driver bug and that it seems to be fixed in Lucid as Antti and others pointed out. I'm therefore marking it as Invalid for netbook-launcher and Fix Released in xserver-xorg-video-intel. Please feel free to reopen the bug if this is still a problem in Lucid.

How does a fix in Lucid make this bug "invalid" on karmic? I cannot believe that an officially supported OS release is not supposed to be working. Is there any kind of official policy that warants this type of bug report handling?

It is clear that if it has been verified to be a bug in xserver-xorg-video-intel, and even fixed as such in lucid, it should be set to Invalid for netbook-launcher. But xserver-xorg-video-intel, should be set to confirmed or wontfix, as there currently is not a fix for karmic.

It is not invalid in karmic, it is invalid for netbook-launcher because it was a driver bug, not a netbook-launcher bug.

I don't think the exact fix has been identified, nor whether it would be easy to fix the bug in karmic, but given that the lucid release is in 2 months, I don't think it will be fixed in karmic.

Looks like the nUbuntu bug task was a typo.

No matter what, it how does the status "Fix Released" status for xserver-xorg-video-intel correspond to the status in karmic?

Also, is the support status of an officially supported release really reduced like you suggest when a new release is out, and even before it is released?

According to https://wiki.ubuntu.com/Releases
Karmic (being only 4 months old!) will reach it's end of life date on April 2011.
Is there a different date for when bugfixes will not be done anymore, like 2 months before next release?

I understand that if no easy bugfix is possible for karmic, it will not be fixed. But the mere fact that a new release is close hardly seems like a valid reason for rejecting a bug report.

We try to fix bugs in the development distro; fixing them in releases causes work duplication since we also have to fix them in the devel distro. We have to stop spending time on fixing bugs in releases at some point or we wouldn't ever make progress in the devel distro.

Most updates after a release are usually done in the month subsequent to release; of course we will continue updating the releases for e.g. security updates, or critical bug fixes such as timezone information updates.

The fact that a new release with the fix is approaching at least gives people with a bug a forward path which they will have to follow at some point anyway.

I understand your point.

It just really surprises me that the point where bugfixes other than security updates are stopped is reached even before the next release is out. I actually thought that Canonical promised to maintain Ubuntu Linux until end-of-life date was reached. I guess I was wrong.

I sincerely hope that LTS releases are not handled in the same way.

Certainly if a patch for this fix materialized it would probably be a good candidate for karmic-updates but it doesn't seem that a specific fix has been found. It does however appear that the bug got fixed somewhere in updating the driver to entirely new versions. Ubuntu's policy is not to push an entire version, which may contain more regressions, to fix a single bug. A specific patch addressing this bug would need to be isolated.

Right now the bug status seems to accurately reflect what's going on. This isn't a bug in netbook launcher so I closed all those tasks as Invalid. I marked the xserver-xorg-video-intel task as "Fix Released" because the default status tracks the development release (Lucid right now). We track stable release updates via release nominations. So that should be all good. The only remaining consideration is if we should open an additional task for Karmic. Loïc declined the nomination because there is not yet an actionable patch that we can push to karmic-updates. If such a patch is made then certainly a Karmic bug task can be opened.

I hope this makes more sense. You can read Ubuntu's update's policy here: https://wiki.ubuntu.com/StableReleaseUpdates . I would imagine that this bug could perhaps warrant an update, but only if a specific patch is isolated and has been tested.

As Sergi Sales already wrote on 2009-11-15, it seems that the same problems occur with Via graphic cards. When I am using openchrome drivers I am unable to select any icons from netbook-launcher. When I change the openchrome drivers to vesa I am able to select icons. Tested on both 9.10 and 10.04.

lspci:
01:00.0 VGA compatible controller: VIA Technologies, Inc. CN700/P4M800 Pro/P4M800 CE/VN800 [S3 UniChrome Pro] (rev 01)