gedit crashed with SIGSEGV in g_type_check_instance_is_a()

[Travis Watkins]

Was trying to open a file, seems to crash every time I try.

ProblemType: Crash
DistroRelease: Ubuntu 11.10
Package: gedit 3.1.2-0ubuntu1
ProcVersionSignature: Ubuntu 3.0.0-4.5-generic 3.0.0-rc6
Uname: Linux 3.0.0-4-generic x86_64
Architecture: amd64
CrashCounter: 1
Date: Tue Jul 12 16:12:49 2011
ExecutablePath: /usr/bin/gedit
InstallationMedia: Ubuntu 11.04 "Natty Narwhal" - Alpha amd64 (20110202)
ProcCmdline: gedit
ProcEnviron:
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0x7f3714289ff0 <g_type_check_instance_is_a+112>: testb $0x4,0x16(%rdi)
 PC (0x7f3714289ff0) ok
 source "$0x4" ok
 destination "0x16(%rdi)" (0x7f3600100016) not located in a known VMA region (needed writable region)!
SegvReason: writing unknown VMA
Signal: 11
SourcePackage: gedit
StacktraceTop:
 g_type_check_instance_is_a () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
 g_file_equal () from /usr/lib/x86_64-linux-gnu/libgio-2.0.so.0
 ?? () from /usr/lib/libgtk-3.so.0
 ?? () from /usr/lib/libgtk-3.so.0
 ?? () from /usr/lib/libgtk-3.so.0
Title: gedit crashed with SIGSEGV in g_type_check_instance_is_a()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[Apport retracing service (apport)]

StacktraceTop:
 g_type_check_instance_is_a (type_instance=<value optimized out>, iface_type=<value optimized out>) at /build/buildd/glib2.0-2.29.10/./gobject/gtype.c:3952
 g_file_equal (file1=0x2dd0e60, file2=0x2de2e80) at /build/buildd/glib2.0-2.29.10/./gio/gfile.c:585
 gtk_path_bar_finalize (object=0x28d9410) at /build/buildd/gtk+3.0-3.1.8/./gtk/gtkpathbar.c:282
 ?? ()
 ?? ()

[Sebastien Bacher (seb128)]

Thank you for your bug report. Is that still an issue? Could you attach the file to the bug?

[Matthew Wardrop (mister.wardrop)]

I get this crash once in a while too; I was going to auto-report it with apport; but this seems to be a duplicate bug :). It certainly does not happen regularly; but it does happen often. It happened in both Oneiric and now in Precise (which I run on this machine).

[Romanos Dodopoulos (rwmanos)]

Gedit crash to me too, with SIGSEGV in g_type_check_instance_is_a().

[Marcel Stimberg (marcelstimberg)]

The crash is reproducible on Precise (see my duplicate bug report #927150) -- I followed the instructions provided by fabrg in bug #926858:
Steps to reproduce:

1. Open gedit and display File Browser (side panel)
2. Open a structure folder with some folders and files (i suggest you /home/), and expand some folder till you see some files, like
-- /afolder/
--------afile
--------afile
-- /afolder/
--------afile
--------afile
--------afile
3. click "refresh view" button
4. the structure should collapse automaticaly. if not, try to refresh again untill it does. (probally gedit will crash)
5. when the folder collapse, expand it again.
6. repeat steps 3 to 5 till gedit crashes.

I had to repeat steps 3-5 quite a few times but in the end gedit crashed when I clicked on the expand arrow for a directory.

[Sebastien Bacher (seb128)]

Thanks Marcel but that bug you describe leads to a different stacktrace, somebody should report a new apport bug for it

[Marcel Stimberg (marcelstimberg)]

Hi Sebastien, the issue seems to be the same, a crash at g_type_check_instance_is_a, called by g_file_equal, the segvanalysis is bascially identical (see my bug #927150). But you are right, the stacktrace is not exactly identical (g_type_check_instance_is_a <- g_file_equal <- gtk_path_bar_finalize in this bug, g_type_check_instance_is_a <- g_file_equal <- g_hash_table_lookup_node in my case) -- so you think it is worth keeping the two bug reports separate? I'll unmark bug #927150 as a duplicate, then.

[Sebastien Bacher (seb128)]

well, that bug seems to be in the gtk fileselector where the one you described has comes from one of the gedit .so, they seems different enough to open a new bug, not to mention that the stacktrace here is incomplete so getting an updated one with current version will be useful in any case

[Marcel Stimberg (marcelstimberg)]

Got your point, I unmarked bug #927150 as a duplicate.

[dino99 (9d9)]

This version has expired