GNOME Bugzilla – Bug 774715
Segfault when doing layer arbitrary rotation with wacom tablet
Last modified: 2018-01-14 15:43:41 UTC
GIMP git master, Linux. When doing arbitrary rotation on a layer, using a wacom tablet, GIMP segfaults at the moment of clicking the 'rotate' button. It does not happen when using a mouse. The backtrace, not sure if meaningful:
+ Trace 236871
Hi, Thanks for reporting this crash. I just tested with a Wacom Bamboo (an old version) on a Linux 64 bit and could not reproduce the crash. Can you still reproduce with updated babl/GEGL/GIMP repositories? Does it happen every time? If so, could you give us accurate reproduction steps please? Thanks.
I upgraded everything babl/gegl/gimp from git and the same thing happens. It does happen every time. Here are my steps to reproduce: -Have tablet plugged in before starting gimp, configured etc. -Start gimp -Open any image or start a new one, and from here on, use only tablet, not mouse -Use the tablet to go Layer > Transform > Arbitrary rotation -Rotate the layer -Press 'Rotate' Backtrace: Thread 1 "gimp-2.9" received signal SIGSEGV, Segmentation fault. 0x00007ffff72c794f in ?? () from /usr/lib/x86_64-linux-gnu/libgdk-x11-2.0.so.0 (gdb) bt
+ Trace 236880
And this is on Debian testing.
Ok. I still can't reproduce. The trace is quite similar to the one in bug 733159. They both crash on the same line (gdk_device_get_state() call inside gimp_device_info_get_device_coords()). So that may be a Gdk bug. Which GTK+ version are you on?
I think this is the problem where we mis-assign GimpDeviceInfo to GdkDevice, and finally a user who can help debugging :) Can you please attach: 1. your devicerc from .config/GIMP/2.9 2. the output of "xinput list" with your tablet plugged
Created attachment 340720 [details] devicerc
gtk+ version is 2.24.31-1 on debian. devicerc is attached above. And here is the xinput output: ⎡ Virtual core pointer id=2 [master pointer (3)] ⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)] ⎜ ↳ Logitech M570 id=8 [slave pointer (2)] ⎜ ↳ Logitech K270 id=9 [slave pointer (2)] ⎜ ↳ Wacom Intuos S Pen stylus id=11 [slave pointer (2)] ⎜ ↳ Wacom Intuos S Pen eraser id=12 [slave pointer (2)] ⎜ ↳ Wacom Intuos S Pad pad id=13 [slave pointer (2)] ⎣ Virtual core keyboard id=3 [master keyboard (2)] ↳ Virtual core XTEST keyboard id=5 [slave keyboard (3)] ↳ Power Button id=6 [slave keyboard (3)] ↳ Sleep Button id=7 [slave keyboard (3)] ↳ Logitech K270 id=10 [slave keyboard (3)]
The newer bug 784480 suggests this depends on rulers being visible or not. Will, can you check this please?
(In reply to Michael Natterer from comment #8) > The newer bug 784480 suggests this depends on rulers being visible > or not. Will, can you check this please? This does appear to be the case, after a few tests. The crash happens when rulers are not visible.
Ok thanks, let's resolve as duplicate of the newer bug which figured the ruler dependency. *** This bug has been marked as a duplicate of bug 784480 ***