/srv/irclogs.ubuntu.com/2008/03/17/#ubuntu-x.txt

ubotuNew bug: #199455 in mythtv (restricted) "mythfrontend.real crashed with SIGSEGV in memset()" [Undecided,Invalid] https://launchpad.net/bugs/19945500:20
seb128bryce: I fixed the gnome-settings-daemon crasher, that was due to conflicting variable names14:21
ubotuNew bug: #197759 in gnome-settings-daemon (main) "Unable to start the settings manager 'gnome-settings-daemon'. (dup-of: 198951)" [Undecided,Incomplete] https://launchpad.net/bugs/19775914:48
ubotuNew bug: #203181 in xorg (main) "Xorg crashed with SIGSEGV in free()" [Undecided,New] https://launchpad.net/bugs/20318115:30
bryceseb128: excellent16:03
seb128bryce: that's a 3 line changes ;-)16:04
seb128bryce: I wrote some explanations on the bug if you are interested16:04
seb128there is still the gnome-display-properties crasher though16:05
seb128I think you can reproduce it easily by using Xnest16:05
seb128I might look at this one too later but I've spent already quite some time on the g-s-d issue and I've other things to do this week16:06
brycelikely it needs the same fix16:06
bryceyes I know the feeling16:06
seb128not likely16:07
seb128the gsd crasher was due to a variable name conflict16:07
seb128the other one seems to be an another issue16:07
brycedo you have the name of the XError in this case?  16:09
seb128yes16:09
seb128be back in a minute16:09
seb128will give you that then16:09
seb128bryce: bug #199960 is an interesting one16:31
ubotuLaunchpad bug 199960 in gnome-settings-daemon "error starting GNOME Settings Daemon" [High,Confirmed] https://launchpad.net/bugs/19996016:31
seb128bryce: XRRGetScreenSizeRange() seems to be the crasher for this user16:31
bryceseb128: what is the proper way for people to get debug symbols for gsd and gnome-desktop?  comment #28 said he installed debug symbols, but his traces only have symbols for Xorg calls17:29
seb128bryce: add the dbgsym source as indicated on the wiki page about crashes and install libgtk2.0-0-dbgsym libgnome-desktop-2-dbgsym and gnome-settings-daemon-dbgsym17:30
=== asac_ is now known as asac
ubotuNew bug: #197161 in xorg (main) "[Hardy] Macbook trackpad not detected as trackpad" [Undecided,New] https://launchpad.net/bugs/19716119:23
ubotuNew bug: #135023 in xorg-server (main) "wrong resolution autodetected" [Undecided,Incomplete] https://launchpad.net/bugs/13502319:26
ubotuNew bug: #159634 in xorg-server (main) "Display detected as 1680x1050 but is 1680x1040" [Undecided,Incomplete] https://launchpad.net/bugs/15963419:27
ubotuNew bug: #197740 in xorg-server (main) "[Hardy] gnome-display-properties does crazy things" [Undecided,Incomplete] https://launchpad.net/bugs/19774019:43
ubotuNew bug: #203296 in xserver-xorg-input-synaptics "[hardy] Drag fails with keyboard stick and touchpad buttons" [Undecided,New] https://launchpad.net/bugs/20329620:26
bryceseb128: on 199960, the original reporter's issue has been solved.  But the commenter that provided the backtrace appears to be using a non-xrandr 1.2 driver, which seems to need that gtk.patch from ssp that I forwarded to you earlier.  I also attached it to the bug for you.21:09
seb128bryce: that is not clear to me, what difference would it make?21:15
seb128bryce: I'm not comfortable using those gtk changes in hardy and fedora doesn't do that either21:15
brycesee Soren's comment at the end of http://bugzilla.gnome.org/show_bug.cgi?id=52137121:15
ubotuGnome bug 521371 in gdk "XRRGetScreenResources error (libXrandr.so.2)" [Normal,Unconfirmed] 21:15
seb128bryce: right, what soren wrote that is that the gtk call to XRRGetScreenResources is conditional to the tests before21:17
seb128bryce: the bug you pointed is a similar gtk crash, not the g-s-d one21:18
seb128bryce: which means libgnomedesktop should have tests similar to the ones he copied in the comment21:18
seb128libgnome-desktop doesn't use the gtk xrandr functions, otherwise it would not build without patching gtk21:18
seb128we need to check whether the libgnome-desktop checks are correct21:19
seb128and if they are and there is some buggy drivers we need to trap the gdk errors as I told you the other days (which is what sorens mentions on the bug too)21:19
seb128bryce: does it make sense to you?21:19
seb128bryce: ok, libgnome-desktop has no such tests21:24
bryceI am getting so frustrated by this21:25
seb128why?21:26
seb128the main crasher is fixed now21:26
seb128the other issue is that the code dpes xrandr 1.2 only21:27
seb128does21:27
brycewell you keep threatening to drop the code unless this is done and that is done, and keep throwing aside patches I send21:27
seb128hum21:27
seb128that's not really true21:27
seb128I said we need to get the crasher at startup fixed, which we mainly did21:28
seb128you don't argue for shipping hardy crashing out of the box for radeonhd, i855, etc users, do you?21:28
seb128and I don't reject patches for the fun of it21:29
bryceno, which is why I put in several days last week putting in a patch to only run on ati and intel which we know work (which was difficult, being sick), but you dropped it21:29
seb128but I try to minimize the api changes and the distro specific code we have to support21:29
bryceI'm not saying you're wrong, just that I'm feeling frustrated21:29
seb128I'm sorry about that21:30
brycein any case, I put in a 1.2 check in apply_settings() in gsd-xrandr-manager.  Maybe that just needs to be done throughout.21:30
seb128but parsing the xorg log to determine if the card used is in a list is just not scalable and ugly21:30
brycebut it looks to me like this gtk.patch takes care of it globally without having to patch checks into a bunch of different functions21:31
seb128we would need to make gnome-desktop use those functions then21:31
seb128and gtk is one of our base libs21:32
seb128I'm not a big fan of changing behaviour over upstream and carry this patch21:32
seb128and if the redhat guys are not shipping it but are shipping the capplet they might have a reason21:32
seb128I would like to try to figure that before running into issues with gtk21:32
ubotuNew bug: #202121 in linux-restricted-modules-2.6.24 (restricted) "Latest linux-restricted-modules-2.6.24-12-generic breaks FGLRX?" [Undecided,New] https://launchpad.net/bugs/20212122:41
ubotuNew bug: #203327 in xorg (main) "Can`t change refresh rate" [Low,Invalid] https://launchpad.net/bugs/20332723:16
ubotuNew bug: #192144 in xorg-server (main) "Xorg crashed with SIGSEGV in _dl_relocate_object()" [Undecided,New] https://launchpad.net/bugs/19214423:46

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!