! Please note that this is a snapshot of our old Bugzilla server, which is read only since May 29, 2020. Please go to gitlab.xfce.org for our new server !
thunar-vfs-CRITICAL Segmentation fault
Status:
CLOSED: FIXED
Severity:
critical

Comments

Description xface 2006-08-27 12:30:43 CEST
User-Agent:       Opera/9.01 (X11; Linux i686; U; en)
Build Identifier: 

opening several windows (3) and closing with X button of frame causes seg fault.

Reproducible: Always

Steps to Reproduce:
1.thunar
2.file | open new window
3.close new window with X button on frame
4.file | open new window
5.close new window with X button on frame
6.bring up file menu ....


Actual Results:  
bash-3.1#thunar

file menu appears , thunar freezes for a couple of seconds the segfaults:


(thunar:3859): thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_name: assertion `*name != '\0'' failed
Segmentation fault



build 22899
Comment 1 xface 2006-08-27 12:36:52 CEST
slight correction. The segfault happens when I select open new window from the menu. Until then the menu is fully working. I can display help about etc. but as soon as I select open I get a partial display of new window then segs out.

Comment 2 Benedikt Meurer editbugs 2006-08-27 13:15:38 CEST
Post a backtrace.
Comment 3 xface 2006-08-27 14:54:35 CEST
(thunar:24716): thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_name: assertion `*name != '\0'' failed
[New LWP 24909]

Program received signal SIGSEGV, Segmentation fault.
[Switching to LWP 24909]
0xb77fcc08 in g_atomic_int_exchange_and_add () from /usr/lib/libglib-2.0.so.0
(gdb) bt
#0  0xb77fcc08 in g_atomic_int_exchange_and_add ()
   from /usr/lib/libglib-2.0.so.0
#1  0xb7ed45a5 in thunar_vfs_mime_info_unref ()
   from /usr/lib/libthunar-vfs-1.so.2
#2  0xb7ed4588 in thunar_vfs_mime_info_unref ()
   from /usr/lib/libthunar-vfs-1.so.2
#3  0xb7ec57b8 in thunar_vfs_volume_status_get_type ()
   from /usr/lib/libthunar-vfs-1.so.2
#4  0xb7ee4b54 in ?? () from /usr/lib/libthunar-vfs-1.so.2
#5  0x00000000 in ?? ()
(gdb) 
Comment 4 Benedikt Meurer editbugs 2006-08-27 15:20:23 CEST
That's not the thread that caused the crash.
Comment 5 Benedikt Meurer editbugs 2006-08-27 15:21:23 CEST
And btw. build with --enable-debug=full.
Comment 6 Benedikt Meurer editbugs 2006-09-03 16:42:02 CEST
Moving to 0.5.0rc2.
Comment 7 xface 2006-09-03 20:08:57 CEST
(gdb) run
Starting program: /usr/bin/thunar 
Failed to read a valid object file image from memory.
(no debugging symbols found)
thunar-apr-Message: Initializing ThunarApr extension
thunar-sbr-Message: Initializing ThunarSbr extension
thunar-uca-Message: Initializing ThunarUca extension
thunar-sbr-Message: Shutting down ThunarSbr extension
thunar-apr-Message: Shutting down ThunarApr extension

thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_name: assertion `*name != '\0'' failed
aborting...
[New LWP 21765]

Program received signal SIGABRT, Aborted.
[Switching to LWP 21765]
0xb7f6e410 in ?? ()
(gdb) bt
#0  0xb7f6e410 in ?? ()
#1  0xb416eb48 in ?? ()
#2  0x00000006 in ?? ()
#3  0x00005505 in ?? ()
#4  0xb7740161 in raise () from /lib/libc.so.6
#5  0xb7741d16 in abort () from /lib/libc.so.6
#6  0xb786c405 in g_logv () from /usr/lib/libglib-2.0.so.0
#7  0x00000000 in ?? ()
(gdb) q

Thunar 0.4.1svn-r23067 built with --enable-debug=full
Comment 8 Benedikt Meurer editbugs 2006-09-03 20:11:04 CEST
Uhm, try running Thunar separately and let it create a core dump. Then open the core dump in gdb and print a backtrace.
Comment 9 xface 2006-09-03 21:37:03 CEST
From dmesg after running thunar form console.

CPU:    0
EIP:    0060:[<c019fc2e>]    Tainted: P      VLI
EFLAGS: 00210286   (2.6.18-rc4-no1 #4) 
EIP is at mountstats_open+0x6e/0x120
eax: 00000000   ebx: f4e62050   ecx: f4e62118   edx: 00000001
esi: 00000000   edi: ee02d61c   ebp: dbfe0c60   esp: d9949eb0
ds: 007b   es: 007b   ss: 0068
Process thunar (pid: 25943, ti=d9948000 task=ef176050 task.ti=d9948000)
Stack: dbfe0c60 c04eedc0 ee905260 dbfe0c60 00000000 ee02d61c c019fbc0 c0164740 
       ee02d61c dbfe0c60 00000001 dbfe0c60 d9949f20 0000000c d9948000 c0164987 
       ee02fdac dfff3f20 00008000 dbfe0c60 00000000 00008000 c0164890 d9949f20 
Call Trace:
 [<c019fbc0>] mountstats_open+0x0/0x120
 [<c0164740>] __dentry_open+0x190/0x290
 [<c0164987>] nameidata_to_filp+0x37/0x50
 [<c0164890>] do_filp_open+0x50/0x60
 [<c0164a9f>] get_unused_fd+0x7f/0xe0
 [<c0164c83>] do_sys_open+0x63/0x110
 [<c0164d57>] sys_open+0x27/0x30
 [<c0103121>] sysenter_past_esp+0x56/0x79
Code: 8b 45 70 8b 54 24 20 89 44 24 08 8b 42 f0 31 d2 e8 c8 ba f8 ff 85 c0 89 c3
 74 63 b8 01 00 00 00 e8 e8 3a f7 ff 8b 83 48 04 00 00 <8b> 78 0c 85 ff 74 02 ff
 07 b8 01 00 00 00 e8 4f 3b f7 ff 89 e0 
EIP: [<c019fc2e>] mountstats_open+0x6e/0x120 SS:ESP 0068:d9949eb0
 <6>note: thunar[25943] exited with preempt_count 1


how do I open the core dump. :?
Comment 10 Benedikt Meurer editbugs 2006-09-03 21:39:23 CEST
Hm, a crash in open(), that looks like a bug in the kernel (well, actually every crash in the kernel is a bug in the kernel). You may want to fix a bug report.

To open the core dump, use gdb <path-to-binary> <path-to-core-file>.
Comment 11 Benedikt Meurer editbugs 2006-09-09 22:00:54 CEST
Any news?
Comment 12 xface 2006-09-09 23:13:16 CEST
just rebuild from cvs and get basically similar behaviour. After 3 or 4 operations of open and closing a new thunar window it starts to slow each time on opening the new window.

I got to about 8 reps and it froze for a minute or so before crashing completely.

bash-3.1#thunar
thunar-apr-Message: Initializing ThunarApr extension
thunar-sbr-Message: Initializing ThunarSbr extension
thunar-uca-Message: Initializing ThunarUca extension
thunar-sbr-Message: Shutting down ThunarSbr extension
thunar-apr-Message: Shutting down ThunarApr extension


Gtk-CRITICAL **: _gtk_tree_view_find_path: assertion `node != NULL' failed
aborting...
Trace/breakpoint trap

in relation to your last msg, I dont seem to find a core dump file. Where should I expect to find this.
Comment 13 Benedikt Meurer editbugs 2006-09-09 23:15:39 CEST
That's a completely different error. Do you use an unstable GTK+ version? Or another related library that is not really considered stable?

If ulimit -c unlimited is set, you'll find a core file in the folder from which Thunar was started.
Comment 14 xface 2006-09-09 23:26:40 CEST
 dev-libs/glib-2.12.3  
 x11-libs/gtk+-2.10.3  
 x11-libs/pango-1.14.3  
 dev-libs/atk-1.12.2  

I updated these packages over the last couple of days and also rebuilt xfce off snv :

 exo libxfce4util libxfce4mcs libxfcegui4 mousepad pyxfce xfce-extra/terminal thunar

so it's perhaps not surprising that it's different although globally I am seeing similar but not identical failure.
Comment 15 Benedikt Meurer editbugs 2006-09-09 23:30:27 CEST
It fails in completely different modules: The last one is in GtkTreeView. The first one is on ThunarVfsMimeDatabase. These two modules have absolutely nothing to do with each other.

Basicly that means either there's some kind of memory corruption in Thunar or one of the libraries in use, or your system failing (i.e. bad RAM, overheated CPU, etc.). The latter is pretty unlikely. Most probably memory corruption or some currently unknown reason. 
Comment 16 xface 2006-09-10 00:00:32 CEST
I'll run an extended memtest86+ tonight tho' I'm pretty sure it's solid. I've done some extended compiling the last few days and that normally stresses the system a lot more than opening a few thunar windows.

Also , can you advise on building thunar from snv.

Earlier I used the Gentoo "cvs" ebuilds for the packages I indicated (with debugging=full added).

It may be more repeatable if I compiled directly in my svn archive but I get 

make[3]: *** No rule to make target `thunarx-aliasdef.c', needed by `libthunarx_1_la-thunarx-aliasdef.lo'.  Stop.

Is there an easy way to rebuild all of xfce in one hit or do I need to build each package manaully?
Comment 17 Benedikt Meurer editbugs 2006-09-10 00:02:16 CEST
When building from SVN directly, you need to pass --enable-maintainer-mode to autogen.sh.

You may ask on the xfce mailinglist or on the xfce forum. Several people have posted xfce build scripts in the past.
Comment 18 xface 2006-09-10 09:24:53 CEST
5h30 memtest86+ no errors.
installed xfce4-4.4.0_rc1 et al

back to original error after 4 reps.


(thunar:15227): thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_na
me: assertion `*name != '\0'' failed
Segmentation fault
Comment 19 Benedikt Meurer editbugs 2006-09-10 09:48:43 CEST
Ok, that should have created a core file (or core.<pid>). Post the backtrace here.
Comment 20 xface 2006-09-10 10:05:03 CEST
still cant find the core but here the bt on a re-run 
(gdb) r
Starting program: /usr/bin/thunar 
Failed to read a valid object file image from memory.
(no debugging symbols found)

(thunar:12812): thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_na
me: assertion `*name != '\0'' failed
[New LWP 12965]

Program received signal SIGSEGV, Segmentation fault.
[Switching to LWP 12965]
0xb77e7a49 in g_atomic_int_exchange_and_add () from /usr/lib/libglib-2.0.so.0
(gdb) bt
#0  0xb77e7a49 in g_atomic_int_exchange_and_add ()
   from /usr/lib/libglib-2.0.so.0
#1  0xb7edaae7 in IA__thunar_vfs_mime_info_unref (info=0x0)
    at thunar-vfs-mime-info.c:165
#2  0xb7ecc6ac in _thunar_vfs_io_local_get_info (path=0x0, 
    absolute_path=0xb3a202a0 "/home/prof/Desktop/%2f", error=0x0)
    at thunar-vfs-io-local.c:423
#3  0xb7ecc8da in _thunar_vfs_io_local_listdir (path=0x8741810, 
    error=0x8741810) at thunar-vfs-io-local.c:550
#4  0xb7ecb8c7 in _thunar_vfs_io_jobs_listdir (job=0x844e0e0, 
    param_values=0x0, n_param_values=1, error=0xb3a2139c)
    at thunar-vfs-io-jobs.c:613
#5  0xb7ee09fc in thunar_vfs_simple_job_execute (job=0x844e0e0)
    at thunar-vfs-simple-job.c:127
#6  0xb7ed3b58 in thunar_vfs_job_execute (data=0x844e0e0, user_data=0x0)
    at thunar-vfs-job.c:367
#7  0xb7827f84 in g_thread_pool_free () from /usr/lib/libglib-2.0.so.0
#8  0xb7825d69 in g_static_private_free () from /usr/lib/libglib-2.0.so.0
#9  0xb78e1604 in __pthread_initialize_minimal () from /lib/libpthread.so.0
#10 0xb7775e6e in clone () from /lib/libc.so.6
(gdb) 
Comment 21 Benedikt Meurer editbugs 2006-09-10 11:28:22 CEST
Uhm, do you really have a file named %2f on your desktop?
Comment 22 xface 2006-09-10 12:01:05 CEST
apparently so! 

I spotted that as well but decided not to change anything since this oddity may simply be exposing a bug that needs fixing anyway.

also this last run was done as root , so I dont know the reason for it scanning prof's h.d. You will know if that is expected but I point it out.

I dont use the desktop for anything more than a bg image to work against. For me the desktop is what my monitor sits on - and that's full of clutter ;)

I can only assume that erroneous filename it due to one of the crashes of thunar or some other software. It may be old , I never ls ~/.desktop and it could have gone unseen.

Comment 23 Benedikt Meurer editbugs 2006-09-10 12:03:42 CEST
It may be possible that %2f causes the file manager to crash, but I doubt it. Anyway try removing the file from ~/Desktop and see if its still crashing, but keep a backup of the file (-> content)!
Comment 24 xface 2006-09-10 12:12:04 CEST
stranger still, I moved the file to ~/tmp but get the same crash and the file is still in the bt 


(gdb) run
Starting program: /usr/bin/thunar 
Failed to read a valid object file image from memory.
(no debugging symbols found)

(thunar:7315): thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_nam
e: assertion `*name != '\0'' failed
[New LWP 7449]

Program received signal SIGSEGV, Segmentation fault.
[Switching to LWP 7449]
0xb7865a49 in g_atomic_int_exchange_and_add () from /usr/lib/libglib-2.0.so.0
(gdb) bt
#0  0xb7865a49 in g_atomic_int_exchange_and_add ()
   from /usr/lib/libglib-2.0.so.0
#1  0xb7f58ae7 in IA__thunar_vfs_mime_info_unref (info=0x0)
    at thunar-vfs-mime-info.c:165
#2  0xb7f4a6ac in _thunar_vfs_io_local_get_info (path=0x0, 
    absolute_path=0xb2a9c2a0 "/home/prof/tmp/%2f", error=0x0)
    at thunar-vfs-io-local.c:423
#3  0xb7f4a8da in _thunar_vfs_io_local_listdir (path=0x8842b50, 
    error=0x8842b50) at thunar-vfs-io-local.c:550
#4  0xb7f498c7 in _thunar_vfs_io_jobs_listdir (job=0x8469c80, 
    param_values=0x0, n_param_values=1, error=0xb2a9d39c)
    at thunar-vfs-io-jobs.c:613
#5  0xb7f5e9fc in thunar_vfs_simple_job_execute (job=0x8469c80)
    at thunar-vfs-simple-job.c:127
#6  0xb7f51b58 in thunar_vfs_job_execute (data=0x8469c80, user_data=0x0)
    at thunar-vfs-job.c:367
#7  0xb78a5f84 in g_thread_pool_free () from /usr/lib/libglib-2.0.so.0
#8  0xb78a3d69 in g_static_private_free () from /usr/lib/libglib-2.0.so.0
#9  0xb795f604 in __pthread_initialize_minimal () from /lib/libpthread.so.0
#10 0xb77f3e6e in clone () from /lib/libc.so.6
(gdb) q
The program is running.  Exit anyway? (y or n) y
Comment 25 Benedikt Meurer editbugs 2006-09-10 12:14:14 CEST
Because ~/tmp was listed (probably because you are using the tree pane). Anyway, try to rename the file to a different name.
Comment 26 xface 2006-09-10 12:23:30 CEST
mv file to another partn and umounted it.

similar crash but I had to cntl-c in gdb, it hung but did not seg as before.


(gdb) r
Starting program: /usr/bin/thunar 
Failed to read a valid object file image from memory.
(no debugging symbols found)

(thunar:14085): Gtk-CRITICAL **: _gtk_tree_view_find_path: assertion `node != NULL' failed

(thunar:14085): Gtk-CRITICAL **: _gtk_tree_view_find_path: assertion `node != NULL' failed

(thunar:14085): Gtk-CRITICAL **: gtk_tree_model_get_iter: assertion `path->depth > 0' failed

(thunar:14085): Thunar-CRITICAL **: thunar_list_model_get_file: assertion `iter->stamp == store->stamp' failed

(thunar:14085): GLib-GObject-CRITICAL **: g_object_unref: assertion `G_IS_OBJECT (object)' failed

Program received signal SIGINT, Interrupt.
0xb78af818 in g_source_new () from /usr/lib/libglib-2.0.so.0
(gdb) bt
#0  0xb78af818 in g_source_new () from /usr/lib/libglib-2.0.so.0
#1  0xb78b40c7 in g_source_attach () from /usr/lib/libglib-2.0.so.0
#2  0xb7f7f115 in IA__thunar_vfs_job_launch (job=0xa093288) at thunar-vfs-job.c:540
#3  0xb7f8bba4 in thunar_vfs_simple_job_launch (func=0xffffff9c, n_param_values=1)
    at thunar-vfs-simple-job.c:197
#4  0xb7f92a78 in IA__thunar_vfs_listdir (path=0x9e7c680, error=0x0) at thunar-vfs.c:144
#5  0x08074e32 in ?? ()
#6  0x08d8c270 in ?? ()
#7  0x00000000 in ?? ()
(gdb) 
Comment 27 xface 2006-09-10 12:24:46 CEST
ps as you suspected , using short-cuts panel this does not happen.
Comment 28 Benedikt Meurer editbugs 2006-09-10 12:26:35 CEST
Hm, the tree pane related warnings look like an old GTK+-2.10 related bug. Are you sure that you are running the latest version of Thunar? I.e. is /usr/bin/thunar RC1 or later, or is an old version from your distribution and you installed the new snapshots into /usr/local?

BTW: Please attach this %2f file here.
Comment 29 xface 2006-09-10 12:30:55 CEST
help | about shows Thunar 0.4.0rc1

which thunar
/usr/bin/thunar
bash-3.1#
Comment 30 Benedikt Meurer editbugs 2006-09-10 12:33:46 CEST
Ok, then rebuild Thunar with --enable-debug=full (the full is important) so it'll crash on first Gtk-CRITICAL.

And attach the %2f file here.
Comment 31 xface 2006-09-10 13:01:45 CEST
Created attachment 784 
%2f

nothing mysterious it seems.
Comment 32 xface 2006-09-10 13:06:48 CEST
dont know if this build how you want. :?

two runs , one with %2f present , then umount it out of sight.

second run thunar still freezes but I need to cont-c it. Dont seem to get any db info.


thunar-vfs-CRITICAL **: thunar_vfs_mime_database_get_info_for_name: assertion `*name != '\0'' failed
aborting...
[New LWP 2165]

Program received signal SIGSEGV, Segmentation fault.
[Switching to LWP 2165]
0xb788da49 in g_atomic_int_exchange_and_add () from /usr/lib/libglib-2.0.so.0
(gdb) bt
#0  0xb788da49 in g_atomic_int_exchange_and_add () from /usr/lib/libglib-2.0.so.0
#1  0xb7f81467 in IA__thunar_vfs_mime_info_unref (info=0x0) at thunar-vfs-mime-info.c:165
#2  0xb7f72aac in _thunar_vfs_io_local_get_info (path=0x0, absolute_path=0xb42ea2a0 "/zic/%2f", error=0x0)
    at thunar-vfs-io-local.c:423
#3  0xb7f72cda in _thunar_vfs_io_local_listdir (path=0x8352c40, error=0x8352c40) at thunar-vfs-io-local.c:550
#4  0xb7f71c87 in _thunar_vfs_io_jobs_listdir (job=0x8320900, param_values=0x0, n_param_values=1, error=0xb42eb39c)
    at thunar-vfs-io-jobs.c:613
#5  0xb7f879c9 in thunar_vfs_simple_job_execute (job=0x8320900) at thunar-vfs-simple-job.c:127
#6  0xb7f7a0a8 in thunar_vfs_job_execute (data=0xffffffff, user_data=0x0) at thunar-vfs-job.c:367
#7  0xb78cdf84 in g_thread_pool_free () from /usr/lib/libglib-2.0.so.0
#8  0xb78cbd69 in g_static_private_free () from /usr/lib/libglib-2.0.so.0
#9  0xb7987604 in __pthread_initialize_minimal () from /lib/libpthread.so.0
#10 0xb781be6e in clone () from /lib/libc.so.6
(gdb) q
The program is running.  Exit anyway? (y or n) y
bash-3.1#umount /zic
bash-3.1#gdb thunar
GNU gdb 6.5
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i686-pc-linux-gnu"...(no debugging symbols found)
Using host libthread_db library "/lib/libthread_db.so.1".

(gdb) r
Starting program: /usr/bin/thunar 
Failed to read a valid object file image from memory.
(no debugging symbols found)
thunar-apr-Message: Initializing ThunarApr extension
thunar-sbr-Message: Initializing ThunarSbr extension
thunar-uca-Message: Initializing ThunarUca extension
thunar-sbr-Message: Shutting down ThunarSbr extension
thunar-apr-Message: Shutting down ThunarApr extension

Program received signal SIGINT, Interrupt.
0xb782a818 in g_source_new () from /usr/lib/libglib-2.0.so.0
(gdb) q
The program is running.  Exit anyway? (y or n) y
Comment 33 xface 2006-09-10 13:09:10 CEST
#18 0xb7d4f42a in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#19 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#20 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#21 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#22 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#23 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#24 0xb7d4f46e in gtk_tree_model_rows_reordered () from /usr/lib/libgtk-x11-2.0.so.0
#25 0xb7d4f543 in gtk_tree_model_foreach () from /usr/lib/libgtk-x11-2.0.so.0
#26 0xb7d56111 in gtk_tree_model_filter_refilter () from /usr/lib/libgtk-x11-2.0.so.0
#27 0x080bed0c in ?? ()
#28 0x0858b540 in ?? ()
#29 0x081fa1a8 in ?? ()
#30 0x00000001 in ?? ()
#31 0xb7954650 in ?? () from /usr/lib/libgobject-2.0.so.0
#32 0xb7954650 in ?? () from /usr/lib/libgobject-2.0.so.0
#33 0x00000002 in ?? ()
#34 0xbfdedae8 in ?? ()
#35 0xb792548a in g_object_set_property () from /usr/lib/libgobject-2.0.so.0
#36 0xb792548a in g_object_set_property () from /usr/lib/libgobject-2.0.so.0
#37 0xb7eb01bf in exo_bind_properties_transfer (src_object=0x84efa70, src_pspec=0x81f9780, dst_object=0xffffff9c, 
    dst_pspec=0x81f98c0, transform=0xb7946a40 <g_value_transform>, user_data=0xffffff9c) at exo-binding.c:63
#38 0xb7eb0242 in exo_bind_properties_notify (src_object=0x84efa70, src_pspec=0x81f9780, data=0x8750ac8) at exo-binding.c:82
#39 0xb793a354 in g_cclosure_marshal_VOID__PARAM () from /usr/lib/libgobject-2.0.so.0
#40 0xb791faad in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0
#41 0xb7938020 in g_signal_has_handler_pending () from /usr/lib/libgobject-2.0.so.0
#42 0xb7938f34 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0
#43 0xb7939326 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0
#44 0xb792228f in g_object_interface_list_properties () from /usr/lib/libgobject-2.0.so.0
#45 0xb792125f in g_value_get_flags () from /usr/lib/libgobject-2.0.so.0
#46 0xb7925671 in g_object_set_property () from /usr/lib/libgobject-2.0.so.0
#47 0xb7eb01bf in exo_bind_properties_transfer (src_object=0x810e918, src_pspec=0x8126150, dst_object=0xffffff9c, 
    dst_pspec=0x81f9780, transform=0xb7946a40 <g_value_transform>, user_data=0xffffff9c) at exo-binding.c:63
---Type <return> to continue, or q <return> to quit---
#48 0xb7eb060a in IA__exo_binding_new_full (src_object=0x810e918, src_property=0x80df925 "show-hidden", dst_object=0x84efa70, 
    dst_property=0xffffff9c <Address 0xffffff9c out of bounds>, transform=0xb7946a40 <g_value_transform>, destroy_notify=0xffffff9c, 
    user_data=0x0) at exo-binding.c:278
#49 0xb7eb073e in IA__exo_binding_new (src_object=0xffffff9c, src_property=0xffffff9c <Address 0xffffff9c out of bounds>, 
    dst_object=0xffffff9c, dst_property=0xffffff9c <Address 0xffffff9c out of bounds>) at exo-binding.c:229
#50 0x080c4aa8 in ?? ()
#51 0x0810e918 in ?? ()
#52 0x080df925 in _IO_stdin_used ()
#53 0x084efa70 in ?? ()
#54 0x080df925 in _IO_stdin_used ()
#55 0x08a89868 in ?? ()
#56 0x00000020 in ?? ()
#57 0x00000000 in ?? ()
(gdb) 
Comment 34 Benedikt Meurer editbugs 2006-09-10 13:10:32 CEST
Created attachment 785 
Suggested fix

Ok, one by one. This patch should fix the crash on %2f problem. (BTW: The problem with the file is the slash in the Name field, which made Thunar crash).
Comment 35 xface 2006-09-10 13:23:00 CEST
OK. it seems you are just filtering out one explicit special case. Doesn't this imply that gtk_utf8_validate should be doing more?

btw I actually thought a double // was valid in a path. Maybe this is just not getting parsed properly but also %xx ought to be valid also.

have not looked into what  gtk_utf8_validate claims to do but if you send it "/" as a filename it should spit back, no?
Comment 36 Benedikt Meurer editbugs 2006-09-10 13:27:18 CEST
No, "/" is perfectly valid UTF-8. But "/" is not an allowed filename in Unix/Linux. Filenames may contain any character, except for "/".

Did you try with the patch and the %2f file?
Comment 37 xface 2006-09-10 14:11:46 CEST
Seems to have taken out the filename bug.

hang remains the same.


#21 0xb78f3326 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0
#22 0xb78dc28f in g_object_interface_list_properties () from /usr/lib/libgobject-2.0.so.0
---Type <return> to continue, or q <return> to quit---
#23 0xb78db25f in g_value_get_flags () from /usr/lib/libgobject-2.0.so.0
#24 0xb78e0a67 in g_object_notify () from /usr/lib/libgobject-2.0.so.0
#25 0xb78f3976 in g_cclosure_marshal_VOID__VOID () from /usr/lib/libgobject-2.0.so.0
#26 0xb78d9aad in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0
#27 0xb78f2020 in g_signal_has_handler_pending () from /usr/lib/libgobject-2.0.so.0
#28 0xb78f2f34 in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0
#29 0xb78f3326 in g_signal_emit () from /usr/lib/libgobject-2.0.so.0
#30 0xb7eb42ec in thunar_vfs_job_source_dispatch (source=0x91b18c0, callback=0, user_data=0x0) at thunar-vfs-job.c:453
#31 0xb77e67d2 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#32 0xb77e8307 in g_main_context_acquire () from /usr/lib/libglib-2.0.so.0
#33 0xb77e863a in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#34 0xb7bf8673 in gtk_main () from /usr/lib/libgtk-x11-2.0.so.0
#35 0x0805f5a2 in ?? ()
#36 0x08327fe8 in ?? ()
#37 0x00000000 in ?? ()
Comment 38 Benedikt Meurer editbugs 2006-09-14 13:27:32 CEST
That's a different issue. File a new bug report with an up2date backtrace and logged messages of the problem.
Comment 39 Benedikt Meurer editbugs 2006-09-14 13:28:42 CEST
Committed with revision 23158.

2006-09-14	Benedikt Meurer <benny@xfce.org>

	* thunar-vfs/thunar-vfs-io-local.c(_thunar_vfs_io_local_get_info):
	  Properly validate the Name of .desktop files. Bug #2227.

Bug #2227

Reported by:
xface
Reported on: 2006-08-27
Last modified on: 2009-07-17

People

Assignee:
Jannis Pohlmann
CC List:
0 users

Version

Version:
0.3.3svn

Attachments

%2f (114 bytes, text/plain)
2006-09-10 13:01 CEST , xface
no flags
Suggested fix (755 bytes, patch)
2006-09-10 13:10 CEST , Benedikt Meurer
no flags

Additional information