Bug 31869 - Something has broken tint2, probably recent glib2.0 update
Summary: Something has broken tint2, probably recent glib2.0 update
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: All Linux
Priority: Normal normal
Target Milestone: ---
Assignee: Stig-Ørjan Smelror
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2023-05-05 13:21 CEST by Stig-Ørjan Smelror
Modified: 2023-07-07 16:33 CEST (History)
3 users (show)

See Also:
Source RPM: tint2-17.1.3-1.mga.src.rpm, glib2.0-2.76.2-2.mga9.src.rpm
CVE:
Status comment:


Attachments
tint2 gdb backtrace with debug packages installed (12.96 KB, text/plain)
2023-05-05 13:22 CEST, Stig-Ørjan Smelror
Details

Description Stig-Ørjan Smelror 2023-05-05 13:21:41 CEST
Hello.

After the recent updates, tint2 is broken. I can't figure out what's wrong.
Have tested on my daily driver and a fresh install in a virtual machine using the nonfree-netinstall iso.
Everything worked fine until the GNOME 44 updates started trickling in and I don't even use GNOME.

From my daily driver:
% tint2
tint2: Using glib slice allocator (default). Run tint2 with environment variable G_SLICE=always-malloc in case of strange behavior or crashes
tint2: xShm: Disabled
tint2: xRandr: Found crtc's: 4
tint2: xRandr: Linking output DVI-D-0 with crtc 0, resolution 2560x1440, DPI 108
tint2: xRandr: Linking output HDMI-0 with crtc 1, resolution 2560x1600, DPI 101
tint2: xRandr: crtc 2 seems disabled
tint2: xRandr: crtc 3 seems disabled
tint2: Loading config file: /home/stig/.config/tint2/tint2rc
tint2: tint2 has been compiled without battery support
tint2: real transparency off.... depth: 24
tint2: panel items: PTSBC
tint2: Systray composited rendering on
tint2: nb monitors 2, nb monitors used 2, nb desktops 4
tint2: panel 1 uses scale 1 
tint2: Loading icon theme cache...
Icon path not found in cache: theme = hicolor, icon = /usr/share/icons/mini/mageia.png, size = 0
tint2: Loading icon theme 'hicolor': 'hicolor'

Adding icon path to cache: theme = hicolor, icon = /usr/share/icons/mini/mageia.png, size = 0, path = /usr/share/icons/mini/mageia.png
tint2: panel 2 uses scale 1 
tint2: Kernel uevent interface initialized...
Icon path not found in cache: theme = hicolor, icon = /usr/share/icons/mini/mageia.png, size = 24
Adding icon path to cache: theme = hicolor, icon = /usr/share/icons/mini/mageia.png, size = 24, path = /usr/share/icons/mini/mageia.png
tint2: systray window 17825799
tint2: systray started
tint2: got 5 X errors
tint2: BadAlloc (insufficient resources for operation)
tint2: BadDrawable (invalid Pixmap or Window parameter)
tint2: BadAlloc (insufficient resources for operation)
tint2: BadDrawable (invalid Pixmap or Window parameter)
tint2: BadDrawable (invalid Pixmap or Window parameter)
add_icon: 6292376 (nxrunner.bin), pid 18789, visual 0x876550, colormap 6292375, depth 32, width 1, height 1
add_icon: 22020177 (), pid 18947, visual 0x876550, colormap 22020101, depth 32, width 22, height 22
tint2: got 1 X errors
tint2: BadDrawable (invalid Pixmap or Window parameter)
tint2: got 2 X errors
tint2: BadPixmap (invalid Pixmap parameter)
tint2: BadPixmap (invalid Pixmap parameter)
add_icon: 19922951 (pnmixer), pid 65523, visual 0x876550, colormap 19922950, depth 32, width 1, height 1
tint2: systray: error code 9
tint2: remove_icon: 19922951 (pnmixer)
[1]    1769426 segmentation fault (core dumped)  tint2
Comment 1 Stig-Ørjan Smelror 2023-05-05 13:22:24 CEST
Created attachment 13804 [details]
tint2 gdb backtrace with debug packages installed
Comment 2 Stig-Ørjan Smelror 2023-05-05 13:37:53 CEST
Addition to the attachment.

(gdb) where
#0  handle_x_event (e=0x7fffffffd240) at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:539
#1  0x000000000041e5a7 in handle_x_event (e=0x7fffffffd240) at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:447
#2  handle_x_events () at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:603
#3  0x000000000041eefa in run_tint2_event_loop () at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:781
#4  0x000000000041ef75 in tint2 (argc=<optimized out>, restart=0x7fffffffdbfc, argv=0x7fffffffdd28)
    at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:824
#5  tint2 (argc=<optimized out>, argv=0x7fffffffdd28, restart=0x7fffffffdbfc) at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/s
rc/main.c:812
#6  0x0000000000414957 in main (argc=1, argv=0x7fffffffdd28) at /usr/src/debug/tint2-17.1.3-1.mga9.x86_64/src/main.c:871
Comment 3 Stig-Ørjan Smelror 2023-05-05 13:42:30 CEST
The crash may be related to an updated glib2.
https://bbs.archlinux.org/viewtopic.php?id=284293
Comment 4 Lewis Smith 2023-05-05 20:35:40 CEST
Well discovered.

I think you are right. Glib2 was updated (with a lot of other stuff) on my system on 3 May. Tint2 is not the only application which may suffer.

That Arch reference is lengthy (avoid patches mentioned), and the conclusion:
"You need a patched version of tint2 to stop the tint2 from crashing"
is preceded by:
"glib2 won't fix, so unless they revert to the previous the g_slice behavior out of, i don't know, a minimal sense of responsibility to not go around and break stuff because tehy feel like and wrt"
which derides glib2 for a change that is not likely to be undone; and
"I don't want to keep ignoring glib2 or stand in the way of progress"
which is a resignation to live with the consequences - even if that means abandoning Tint2 panel/taskbar.

The solution of downgrading glib2 is short-term only.

It seems you are the tint2 packager, so giving you this bug. I suspect you can do nothing unless & until tint2 is updated to cope with the changed glib2. You can close it 'wontfix'; earmarked for Errata9.

CC'ing tmb & neoclust for info re glib2.0

Keywords: (none) => FOR_ERRATA9
CC: (none) => mageia, tmb
Assignee: bugsquad => smelror
Summary: Something has broken tint2, recent updates? => Something has broken tint2, probably recent glib2.0 update
Source RPM: tint2-17.1.3-1.mga.src.rpm => tint2-17.1.3-1.mga.src.rpm, glib2.0-2.76.2-2.mga9.src.rpm

Comment 5 Stig-Ørjan Smelror 2023-05-06 07:33:46 CEST
Reported it upstream.
https://gitlab.com/nick87720z/tint2/-/issues/4
Comment 6 Morgan Leijström 2023-07-07 15:42:56 CEST
I note in the upstream bug
"your patch is now applied to the Mageia package with a reference to your comment."

So, issue solved?

CC: (none) => fri

Comment 7 Stig-Ørjan Smelror 2023-07-07 16:32:49 CEST
Yes, issue solved. Thanks for the reminder.

Status: NEW => RESOLVED
Resolution: (none) => FIXED

Stig-Ørjan Smelror 2023-07-07 16:33:04 CEST

Keywords: FOR_ERRATA9 => (none)


Note You need to log in before you can comment on or make changes to this bug.