Tixati causes segmentation fault in BSPWM

strammer · 2024-12-22 16:43:53

Tixati recently started causing segmentation faults upon opening the application without any apparent change that caused this. The GUI will open and crash as soon as I can start clicking. Once in a blue moon, Tixati does not crash.
I suspect an update may have caused this change but I don't know how verify this. Only the latest version is available on the Tixati website.

The journal reads:

дек 22 16:14:44 pc systemd-coredump[32993]: Process 32351 (tixati) of user 1000 terminated abnormally with signal 11/SEGV, processing...
дек 22 16:14:44 pc systemd[1]: Created slice Slice /system/systemd-coredump.
дек 22 16:14:44 pc systemd[1]: Started Process Core Dump (PID 32993/UID 0).
дек 22 16:14:46 pc systemd-coredump[32994]: [?] Process 32351 (tixati) of user 1000 dumped core.

                                               Stack trace of thread 32351:
                                               #0  0x000079ff14810e96 n/a (libgtk-3.so.0 + 0x410e96)
                                               #1  0x000079ff14dd9290 n/a (libgdk-3.so.0 + 0x7b290)
                                               #2  0x000079ff14de1041 n/a (libgdk-3.so.0 + 0x83041)
                                               #3  0x000079ff14d8268b gdk_display_get_event (libgdk-3.so.0 + 0x2468b)
                                               #4  0x000079ff14de1810 n/a (libgdk-3.so.0 + 0x83810)
                                               #5  0x000079ff150de559 n/a (libglib-2.0.so.0 + 0x5d559)
                                               #6  0x000079ff15141257 n/a (libglib-2.0.so.0 + 0xc0257)
                                               #7  0x000079ff150df287 g_main_loop_run (libglib-2.0.so.0 + 0x5e287)
                                               #8  0x000079ff145e4ebf gtk_main (libgtk-3.so.0 + 0x1e4ebf)
                                               #9  0x000000000081d978 n/a (n/a + 0x0)
                                               ELF object binary architecture: AMD x86-64
дек 22 16:14:46 pc systemd[1]: systemd-coredump@0-32993-0.service: Deactivated successfully.
дек 22 16:14:46 pc systemd[1]: systemd-coredump@0-32993-0.service: Consumed 1.356s CPU time, 679M memory peak.

```

The last comment on the AUR for Tixati is that it has moved to GTK3 while the dependencies for are it are in the GTK2 family.

What else can I do to debug this?

Last edited by strammer (2025-03-29 08:07:10)

WorMzy · 2024-12-22 18:04:12

Mod note: moving to AUR Issues

strammer · 2025-03-28 16:09:48

Update: It crashes in BSPWM but not in Cinnamon.

strammer · 2025-03-29 08:06:17

Support thread in the Tixati forum

seth · 2025-03-29 08:58:54

https://wiki.archlinux.org/title/Debuginfod
Does it help to "export GDK_CORE_DEVICE_EVENTS=1"?

strammer · 2025-04-02 06:09:48

No, it crashes all the same. Thanks I'll go through https://wiki.archlinux.org/title/GTK#Troubleshooting and report back

strammer · 2025-04-02 06:28:47

By the way, I don't know why you are linking to Debuginfod? While I haven ran gdb, I am sure that tixati has no debug symbols. It produces it's own crash report, like this:

hdr: 333 tixati_3.33.1_l64
bld: Linux 6.13.0 on 64-bit x86
sig: 11 11 0 1 0
time: 0000000000000000 0000000000000000
lbas: 0
loc: 00000000000000f8 00000000000000f8 0

stk: 0000000000829bdf
stk: 000072f976816cd0
stk: 000072f977250fd6
stk: 000072f977576d90
stk: 000072f9775792c1
stk: 000072f97751768b
stk: 000072f977578fd0
stk: 000072f977871104
stk: 000072f9778d4d57
stk: 000072f977871de7
stk: 000072f977024dff
stk: 00000000008316f8
stk: 0000000000530f9f
stk: 000072f976800488
stk: 000072f97680054c
stk: 000000000056a0be

  #0  00829bdf  tixati@00400000+0x429bdf
  #1  76816cd0  /usr/lib/libc.so.6@767d9000+0x3dcd0
  #2  77250fd6  /usr/lib/libgtk-3.so.0@76c00000+0x650fd6
  #3  77576d90  /usr/lib/libgdk-3.so.0@774f3000+0x83d90
  #4  775792c1  /usr/lib/libgdk-3.so.0@774f3000+0x862c1
  #5  7751768b  /usr/lib/libgdk-3.so.0@774f3000+0x2468b  (gdk_display_get_event@775175f0+0x9b)
  #6  77578fd0  /usr/lib/libgdk-3.so.0@774f3000+0x85fd0
  #7  77871104  /usr/lib/libglib-2.0.so.0@77813000+0x5e104
  #8  778d4d57  /usr/lib/libglib-2.0.so.0@77813000+0xc1d57
  #9  77871de7  /usr/lib/libglib-2.0.so.0@77813000+0x5ede7  (g_main_loop_run@77871cb0+0x137)
  #10  77024dff  /usr/lib/libgtk-3.so.0@76c00000+0x424dff  (gtk_main@77024d70+0x8f)
  #11  008316f8  tixati@00400000+0x4316f8
  #12  00530f9f  tixati@00400000+0x130f9f
  #13  76800488  /usr/lib/libc.so.6@767d9000+0x27488
  #14  7680054c  /usr/lib/libc.so.6@767d9000+0x2754c  (__libc_start_main@768004c0+0x8c)
  #15  0056a0be  tixati@00400000+0x16a0be

exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 00000000000003e8 1000
exd: 0000000000000000 0
exd: 0000000001000000 16777216
exd: 00000000000028e4 10468
exd: 0000000000000000 0
exd: 0000000000000000 0
exd: 0000000000000000 0

esd:
esd:
esd: 1856849869
esd: []
esd: [s_kin,1,s_name,0]
esd: {69677=[193500566,1],type=[torrent]}
esd: {type=[peers,trackers,trackergroup]}
esd: [{type=peers},[3083228241,[s_name,0,s_progress,1,s_status,0]],{type=piecemanager},[3083228241,[s_name,0]],{type=[trackers,trackergroup]},[3083228241,[s_priority,0]],{type=outputfolders},[3083228241,[s_name,0]],{type=outputfolder},[3083228241,[s_name,0]],{type=torrent},[1486997330,[{type=outputfolders},{type=trackers},{type=peers},{type=piecemanager}]]]
esd: 0
esd: 0
esd: 1856849869
esd: 0
esd:
esd: 1856849869
esd:
esd:

You may notice it's basically the core dump from my original message and some stuff that probably only makes sense to the developers.

seth · 2025-04-02 07:49:41

By the way, I don't know why you are linking to Debuginfod? While I haven ran gdb, I am sure that tixati has no debug symbols.

The backtrace goes through glib, gdk & gtk, there're debug packages for those and it would certainly be helpful to know what those stack positions are.

strammer · 2025-04-07 15:42:42

gdb tixati

GNU gdb (GDB) 16.2
# ...
Enable debuginfod for this session? (y or [n]) y
Debuginfod has been enabled.
To make this setting permanent, add 'set debuginfod enabled on' to .gdbinit.
(No debugging symbols found in tixati)
(gdb) run
Starting program: /usr/bin/tixati
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/usr/lib/libthread_db.so.1".
# ...
(tixati:225969): Gtk-WARNING **: 18:34:29.125: Theme parsing error: gtk-contained-dark.css:2871:228: Missing closing bracket for :not()
# ...
Thread 1 "tixati" received signal SIGSEGV, Segmentation fault.
gtk_tray_icon_manager_filter (xevent=0x7fffffffa6a0,
    event=<optimized out>, user_data=0x51f3c00)
    at ../gtk/gtk/deprecated/gtktrayicon-x11.c:725
725	  else if (xev->xany.window == icon->priv->manager_window)
(gdb) c
Continuing.

Thread 1 "tixati" received signal SIGSEGV, Segmentation fault.
gtk_tray_icon_manager_filter (xevent=0x7fffffffa6a0,
    event=<optimized out>, user_data=0x51f3c00)
    at ../gtk/gtk/deprecated/gtktrayicon-x11.c:725
725	  else if (xev->xany.window == icon->priv->manager_window)
(gdb) c
Continuing.
Couldn't get registers: No such process.
# ...
Program terminated with signal SIGSEGV, Segmentation fault.
The program no longer exists.

The program is not being run.
(gdb) q

Maybe tixati is calling a deprecated method `gtk_tray_icon_manager_filter` that is now also removed in GTK.

seth · 2025-04-08 07:08:01

It most certainly does - and the symbol isn't even in older gtk3 libraries anymore.
https://support.tixati.com/settings%20- … 20behavior so apparently you can disable the systray, but unless there's a commandline switch or documentation of the settings, that's gonna be a hen-egg problem.

Maybe it helps to run it in a context w/o a systray (startx xterm)?

strammer · 2025-04-11 08:14:13

This is very cool. I disabled the tray icon (in a Cinnamon session). Still the same issue though.

Thread 1 "tixati" received signal SIGSEGV, Segmentation fault.
gtk_tray_icon_manager_filter (xevent=0x7fffffff8d10,
    event=<optimized out>, user_data=0x51f3a90)
    at ../gtk/gtk/deprecated/gtktrayicon-x11.c:725
725       else if (xev->xany.window == icon->priv->manager_window)
(gdb) c
Continuing.
[Thread 0x7fffaffff6c0 (LWP 2968482) exited]

Thread 1 "tixati" received signal SIGSEGV, Segmentation fault.
gtk_tray_icon_manager_filter (xevent=0x7fffffff8d10,
    event=<optimized out>, user_data=0x51f3a90)
    at ../gtk/gtk/deprecated/gtktrayicon-x11.c:725
725       else if (xev->xany.window == icon->priv->manager_window)

Last edited by strammer (2025-04-11 09:04:42)

seth · 2025-04-11 08:25:19

Unfortunately it still tries to use the tray.
If you don't know where its config is located,

strace tixati |& grep $USER | grep -vE '(font|icon)'

If you're lucky, it's pain-text and has a "usetray=yes" key or so in it…

strammer · 2025-04-11 08:51:27

This is obviously a tixati issue, not an Arch issue, GNU issue or Linux issue. I don't mean for it to be resolved here. Thanks for debugging this for my benefit.

Arch Linux

#1 2024-12-22 16:43:53

Tixati causes segmentation fault in BSPWM

#2 2024-12-22 18:04:12

Re: Tixati causes segmentation fault in BSPWM

#3 2025-03-28 16:09:48

Re: Tixati causes segmentation fault in BSPWM

#4 2025-03-29 08:06:17

Re: Tixati causes segmentation fault in BSPWM

#5 2025-03-29 08:58:54

Re: Tixati causes segmentation fault in BSPWM

#6 2025-04-02 06:09:48

Re: Tixati causes segmentation fault in BSPWM

#7 2025-04-02 06:28:47

Re: Tixati causes segmentation fault in BSPWM

#8 2025-04-02 07:49:41

Re: Tixati causes segmentation fault in BSPWM

#9 2025-04-07 15:42:42

Re: Tixati causes segmentation fault in BSPWM

#10 2025-04-08 07:08:01

Re: Tixati causes segmentation fault in BSPWM

#11 2025-04-11 08:14:13

Re: Tixati causes segmentation fault in BSPWM

#12 2025-04-11 08:25:19

Re: Tixati causes segmentation fault in BSPWM

#13 2025-04-11 08:51:27

Re: Tixati causes segmentation fault in BSPWM

Board footer