Comment 20 for bug 447209

Revision history for this message
Steve Dodier-Lazaro (sidi) wrote :

I feel like a comment is needed from me, since I'm the person who was meant to fix this bug a long time ago. Here is why it has not and will probably never be fixed:

The bug is triggered on a limited set of hardware configurations, of which I own none. On the computer on which xvd was developed, it never crashed. On my alternative laptop, it crashed 4 times in six months, and I suspect this was due to another problem related to my not-so-stable Pulseaudio + Wine + crappy BIOS/Kernel setup.

The exact problem, as far as I can diagnose it, is a crash on track volume querying. I do not know any more. I can not use a debugger to find where it crashes since I can't reproduce the bug. I almost know nothing of GStreamer, so I am not able to quote reasons that would lead to a crash when a client application queries volume for an apparently valid track. I believe I assigned one of the duplicate reports to the GStreamer team, but got no answer.

#8, Frank, all these reports are automated reports of the same bug happening on different setups. Considering the number of reports, it affects a lot of people, yet a much bigger lot of people uses xvd without any problem. Do not think the problem is being disregarded, I just can not help it (see above as for why). It is rather pretentious of you to write "stop saying you'll fix it, and fix it". I spent hours looking for solutions to this problem, and if I could not find one, it's certainly not my fault.

#9, Teej, I should have assigned myself to the upstream project instead of the Ubuntu package, I'll grant you that. This is now done. Anyone able to diagnose the bug better than me and to step in for fixing it is free to assign oneself instead of me.

#12, Lionel, that would have caused a SIGFPE and not a SIGSEGV. The problem is below, unfortunately. However, you're right about the lack of sturdiness of the code (eventhough I know no configuration where the volume range could be 0). Opening a new report about that. Thanks.

#16, Frank, thanks for the hardware configuration. Would you be kind enough to check (with the lshw command) what model of sound card do you exactly own? Also, kernel versions (uname -a) and gstreamer versions (apt-cache policy libgstreamer-plugins-base0.10-0) might help any passer-by developer who knows about GStreamer.

#19, Frank, the other problem seems completely unrelated. Please open a new report against Xubuntu with the *exact* error messages you get.

Final comment: I need exact hardware/kernel/gstreamer configurations. There is no way for me to fix the bug, but anybody with the same configuration and who knows about debugging/development may be able to have a look. If you can reproduce the bug, a GDB trace of where it crashes (with the GStreamer debugging symbols installed, a nice function stack, and possibly some prints of the function parameters / local variables in where it crashes and the 1 or 2 functions above, plus the last function located in xvd) is more than welcome.