GNOME Bugzilla – Bug 101678
[0.6.1 candidate] "gnome-sound-recorder -r" segfaults without a filename.
Last modified: 2004-12-22 21:47:04 UTC
Package: gnome-media Severity: normal Version: 2.1.0 Synopsis: "gnome-sound-recorder -r" segfaults without a filename. Bugzilla-Product: gnome-media Bugzilla-Component: Gnome-Sound-Recorder Description: "gnome-sound-recorder -r" segfaults without a filename. Same with -f Debugging Information: (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)...[New Thread 16384 (LWP 3066)] (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... (no debugging symbols found)...(no debugging symbols found)... 0x408be269 in wait4 () from /lib/libc.so.6
+ Trace 31908
Thread 1 (Thread 16384 (LWP 3066))
------- Bug moved to this database by unknown@bugzilla.gnome.org 2002-12-20 07:47 ------- The original reporter (9915104t@yku.kobe-u.ac.jp) of this bug does not have an account here. Reassigning to the exporter, unknown@bugzilla.gnome.org. Reassigning to the default owner of the component, bugs@prettypeople.org.
I am attaching one more stack trace for the same problem, the source taken from the head
+ Trace 31909
Thanks
Adding keyword and changing priority
There is no -r argument. I think this is GStreamer aborting because there is no -r argument, which I think it should just quit rather than abort. Assigning to GStreamer
Created attachment 15338 [details] [review] Here's a patch for 0.6 I cooked up without access to a compiler. Could somebody please test this?
Created attachment 15473 [details] [review] more important fix - invalid use of g_error in gst_init
We should never use exit() inside a library. Just return or use g_error() or so. Imo, using g_error here is fine.
1) gst_init parses command line parameters. 2) gst_init only returns if it could initialize safely (currently that includes parsing all command line parameters). That's according to the docs. 3) g_error is "A convenience function/macro to log an error message. Error messages are always fatal, resulting in a call to abort() to terminate the application. This function will result in a core dump; don't use it for errors you expect. Using this function indicates a bug in your program, i.e. an assertion failure." And no, I don't want a SIGSEGV when somebody passed a wrong argument to a gst program.
But then you're saying that we should use g_error() (it is exactly what we want here if something goes fatally wrong in gst_init() when it shouldn't return) but that gst_init() shouldn't use g_error() on a bad commandline parameter, and that's not something inside gst_init(), but in the popt callbacks, right?
OK, second one committed to HEAD, marking as 0.6.1 candidate here.
Applied to 0.6.x CVS.
*** Bug 111652 has been marked as a duplicate of this bug. ***
*** Bug 112018 has been marked as a duplicate of this bug. ***
*** Bug 107404 has been marked as a duplicate of this bug. ***
*** Bug 111061 has been marked as a duplicate of this bug. ***
*** Bug 111174 has been marked as a duplicate of this bug. ***
*** Bug 111301 has been marked as a duplicate of this bug. ***
*** Bug 107158 has been marked as a duplicate of this bug. ***
*** Bug 113634 has been marked as a duplicate of this bug. ***
*** Bug 114714 has been marked as a duplicate of this bug. ***
*** Bug 114981 has been marked as a duplicate of this bug. ***
*** Bug 115850 has been marked as a duplicate of this bug. ***
*** Bug 115707 has been marked as a duplicate of this bug. ***
*** Bug 116747 has been marked as a duplicate of this bug. ***
*** Bug 118703 has been marked as a duplicate of this bug. ***
*** Bug 119703 has been marked as a duplicate of this bug. ***
*** Bug 119741 has been marked as a duplicate of this bug. ***
*** Bug 120555 has been marked as a duplicate of this bug. ***
*** Bug 120748 has been marked as a duplicate of this bug. ***
*** Bug 120795 has been marked as a duplicate of this bug. ***
*** Bug 120978 has been marked as a duplicate of this bug. ***
*** Bug 126275 has been marked as a duplicate of this bug. ***
*** Bug 122273 has been marked as a duplicate of this bug. ***
*** Bug 128860 has been marked as a duplicate of this bug. ***
*** Bug 128627 has been marked as a duplicate of this bug. ***
*** Bug 123112 has been marked as a duplicate of this bug. ***
*** Bug 126105 has been marked as a duplicate of this bug. ***
*** Bug 140177 has been marked as a duplicate of this bug. ***
*** Bug 145735 has been marked as a duplicate of this bug. ***