[11:55] <ubotu> New bug: #150572 in libx11 (main) "gnome-panel crashed with SIGSEGV in _XimServerDestroy()" [Medium,New] https://launchpad.net/bugs/150572
[13:14] <ubotu> New bug: #137735 in xserver-xorg-video-via (main) "[gutsy] wine causes hard lockup - every time" [Undecided,Confirmed] https://launchpad.net/bugs/137735
[13:26] <ubotu> New bug: #132000 in kdebase (main) "Clicking on OpenGL in KInfoCenter freezes up the system in Gutsy (dup-of: 137735)" [Undecided,Confirmed] https://launchpad.net/bugs/132000
[13:26] <ubotu> New bug: #147269 in kipi-plugins (main) "[gutsy] When clicking on tools > Image Viewer; the PC freezes (dup-of: 137735)" [Undecided,Invalid] https://launchpad.net/bugs/147269
[14:46] <ubotu> New bug: #176555 in libx11 (main) "libx11-6 / libx11-data conflict during upgrade from Dapper" [Medium,Confirmed] https://launchpad.net/bugs/176555
[15:36] <ubotu> New bug: #176562 in linux-restricted-modules-2.6.22 (restricted) "Config fails in low memory machine (64MB) " [Undecided,New] https://launchpad.net/bugs/176562
[16:45] <ubotu> New bug: #149507 in xorg (main) "Amarok's "global hotkeys" become stuck often" [Undecided,New] https://launchpad.net/bugs/149507
[19:11] <pochu> hey there. Every video I play is played in black&white colours, but they are full-colour. slomo told me this was X fault, since I tried running them with a different engine from mplayer and it was the same. Also, while trying to reproduce them, I've had some X crashes, one of which I reported as bug 173265 (the backtrace is completely useless). Is there something I could do to help debugging this?
[19:11] <ubotu> Bug 173265 on http://launchpad.net/bugs/173265 is private
[19:11] <pochu> This is with an intel gma 915 and the -intel driver
[19:12] <pochu> (hardy up-to-date)
[19:13] <jcristau> which xv adaptor are you using (textured or video)?
[19:21] <pochu> I don't really know, and I can't find on X/Debugging how to find it.
[19:21] <pochu> (II) Loading extension XVideo
[19:21] <pochu> Not sure if that means video - that's in Xorg.0.log
[19:21] <jcristau> mplayer allows you to choose which one
[19:22] <pochu> ah
[19:23] <jcristau> mplayer -vo xv:port=<number>
[19:24] <jcristau> get the number from xvinfo
[19:24] <pochu> mplayer -vo xv /path/to/file
[19:24] <pochu> ok
[19:25] <jcristau> (each adaptor has a "port base" line)
[19:27] <pochu> mplayer -vo xv:port=73 Desktop/andando10.mp4 - black&white
[19:27] <pochu> I only have one adaptor so the number should be the good one
[19:28] <jcristau> hmm, i thought only the 965 didn't have a video adaptor
[19:29] <pochu> 915 mobile, that is
[19:29] <pochu> 00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Processor to DRAM Controller (rev 04)
[19:32] <jcristau> aiui, the textured adaptor is using the 3d engine to display video, so maybe there's a bug in that part of the driver on your hardware
[19:34] <pochu> I've seen more bugs on lp with similar descriptions (playing a video in totem and then crash!) - bug 175303, bug 174571
[19:34] <ubotu> Bug 175303 on http://launchpad.net/bugs/175303 is private
[19:34] <ubotu> Launchpad bug 174571 in xorg-server "Xorg crashed with SIGSEGV" [Medium,Incomplete] https://launchpad.net/bugs/174571
[19:34] <pochu> dunno if they have the same hardware or not
[19:35] <pochu> Is it crazy to try to get the crash when running X under gdb? I've never done that before
[19:35] <jcristau> you need to run gdb over ssh
[19:35] <jcristau> but, it usually works
[19:36] <pochu> I could do that, yes
[19:36] <pochu> do you think the backtrace will be helpful?
[19:36] <jcristau> make sure you have debugging symbols for the server and driver
[19:36] <pochu> -dbg or dbgsym?
[19:36] <pochu> or both? :)
[19:37] <jcristau> either, i think
[19:37] <jcristau> i've never used dbgsym
[20:01] <pochu_> let's get started :-)
[20:08] <pochu_> I'd have expected https://help.ubuntu.com/community/DebuggingSystemCrash to explain how to run X under gdb :-/
[20:09] <pochu_> https://wiki.ubuntu.com/DebuggingXorg explains how to attach to it though :-)
[20:14] <pochu_> ouch, ctrl+alt+f1 corrupted the screen
[20:14] <pochu_> hmm, I could have restarted X through ssh instead of a tty...
[20:24] <pochu_> Program received signal SIGSEGV, Segmentation fault.
[20:24] <pochu_> [Switching to Thread 0xb7c846b0 (LWP 5643)]
[20:24] <pochu_> 0xb7d9df8b in strlen () from /lib/tls/i686/cmov/libc.so.6
[20:24] <pochu_> \o/
[20:30] <pochu_> jcristau: I've pasted it in bug 173265, in case you want to have a look
[20:30] <ubotu> Bug 173265 on http://launchpad.net/bugs/173265 is private
[20:30] <pochu_> bug 173265
[20:30] <ubotu> Launchpad bug 173265 in xorg-server "Xorg crashed with SIGSEGV" [Medium,New] https://launchpad.net/bugs/173265
[20:32] <pochu_> I still have the connection opened, in case there is something useful I can do
[20:33] <pochu_> Should I attach /etc/X11/xorg.conf, /var/log/Xorg.0.log, /var/log/syslog, /var/log/dmesg ... ?
[20:41] <ubotu> New bug: #173265 in xorg-server (main) "Xorg crashed with SIGSEGV in strlen" [Medium,New] https://launchpad.net/bugs/173265
[20:45] <pochu> back
[20:46] <pochu> I lost the connection - I shouldn't probably do this kind of things via wifi :P
[20:51] <jcristau> heh
[20:54] <jcristau> can you go to the i830_unbind_memory frame, and print the value of *mem?
[20:55] <pochu> I'm afraid I can't - I've restarted my laptop since I lost the connection :(
[20:56] <pochu> I could try to get the crash again, now wired
[21:06]  * pochu vs X - round 2
[21:06] <pochu> Let's give this another try
[21:11] <bryce> hi guys
[21:11] <pochu_> hey bryce 
[21:12] <pochu> heya bryce 
[21:12] <pochu> :-)
[21:13] <pochu_> woah, this was faster!
[21:13] <pochu_> [Switching to Thread 0xb7ca36b0 (LWP 5634)]
[21:13] <pochu_> 0xb7b7810e in i830_free_memory (pScrn=0x821be58, mem=0x86d87d8) at ../../src/i830_memory.c:289
[21:13] <pochu_> 289     ../../src/i830_memory.c: No such file or directory.
[21:16] <pochu_> jcristau: http://paste.ubuntu.com/2779/plain/
[21:17] <tormod> pochu_: type "up" until you are in i830_unbind_memory, then: print *mem
[21:20] <pochu_> I had to go down :) but here it is
[21:20] <pochu_> (gdb) print *mem
[21:20] <pochu_> $1 = {offset = 35, end = 0, size = 32, allocated_size = 0, bus_addr = 593808606253023233, key = 0, bound = 0, agp_offset = 4294967295, tiling = TILE_NONE,  fence_nr = 0, pitch = 0, name = 0x0, next = 0x0, prev = 0x40}
[21:22] <bryce> jcristau: btw, I was trying to build current xserver git-head with the current debian rules file, but it's failing in some of the glx code
[21:22] <bryce> jcristau: xserver compiles fine using a default ./configure, though
[21:22] <bryce> jcristau: I'm wondering if it's a mesa version mismatch or something?
[21:26]  * tormod updates DebuggingSystemCrash with links
[21:41] <pochu_> any idea? :)
[21:47] <tormod> I don't know what bus_addr is, but that's a seriously big number :)
[21:47] <pochu_> heh
[21:47] <pochu_> I have 1GB of ram, fwiw
[21:49] <tormod> pochu_: did you follow DebuggingXorg? Was it clear? Just would like feedback since I wrote it. Any suggestions are welcome.
[21:49] <pochu_> yeah, I did, and it looks perfect :)
[21:49] <tormod> well that's what I wanted to hear :)
[21:50] <pochu_> although you might want to say in the part of restarting X after modifying xorg.conf that it's a good idea to do it over ssh, since you have to connect to ssh anyway
[21:50] <pochu_> since ctrl+alt+f1 corrupted my screen and I had to reboot to restart it ;)
[21:53] <Q-FUNK> bryce: there's been progress on the -amd driver. it appears that changes in VBE since X core 1.3 might make -amd and several other marginal drivers (such as the siliconmotion on my laptop, also) fail DDC parsing.
[21:54] <tormod> pochu_: if ctrl-alt-f1 corrupts your screen that's another problem, but it's a good idea to suggest doing it remotely, thanks.
[21:59]  * pochu_ looks at fdo to see if this has been reported there
[22:05] <pochu_> doesn't look like
[22:18] <jcristau> bryce: xserver master needs mesa master
[22:19] <jcristau> pochu_: it's crashing because mem->name is NULL...
[22:20] <jcristau> but, the mprotect() call just before that shouldn't have failed, i think
[22:20] <jcristau> anyway, it'd be good to report the crash upstream
[22:22] <pochu_> any X expert wants to do that? I'm afraid a developer will ask something I won't know how to answer ;)
[22:24] <jcristau> don't worry about that, there will be people to help :)
[22:24] <pochu_> :-)
[22:25] <jcristau> weird, your two backtraces on #173265 are different
[22:27] <pochu_> the first one looks a bit similar since frame 8 though
[22:27] <pochu_> although mem isn't null in that one
[22:27] <jcristau> not the same line number
[22:27] <pochu_> oh, right
[22:28] <pochu_> Do you want me to get a third one? ;-)
[22:29] <jcristau> heh
[22:30] <pochu_> not kidding :P
[22:30] <pochu_> are they two different bugs, or could they be the same one?
[22:31] <jcristau> i don't know
[22:41] <ubotu> New bug: #132171 in xserver-xorg-video-via (main) "AntSpotlight and other screensaver freezes the system" [Medium,New] https://launchpad.net/bugs/132171
[23:01] <ubotu> New bug: #70731 in xscreensaver (main) "Crash (dup-of: 122999)" [Undecided,New] https://launchpad.net/bugs/70731
[23:16] <ubotu> New bug: #111645 in xserver-xorg-video-via (main) "crash when screensaver is activated" [Undecided,New] https://launchpad.net/bugs/111645
[23:34] <jcristau> pochu: actually your crashes look like they're the same as https://bugs.freedesktop.org/show_bug.cgi?id=13108
[23:34] <ubotu> Freedesktop bug 13108 in Driver/intel "[overlay] XV window hidden for some seconds causes SEGFAULT when taken into foreground again" [Major,New]