Open andyprice opened 5 months ago
It looks like the second p_object
is corrupted already, its vtable is _vptr$Object = 0x0
and other values are garbage too. Give the reproduction probability, could it be a thread issue where the pointer that has been freed at another thread?
I'm unable to reproduce it at will but it occurs maybe once or twice a day.
I actually haven't seen this crash in a while now. I suspect it was a specific large inherited scene (of a 3D model) in my project that was triggering it and it stopped crashing after I removed the scene to redesign part of the project, but I can't be certain. I'll report back if it starts crashing again.
Tested versions
System information
Godot v4.3.dev (e92d55bbf) - Fedora Linux 39 (Workstation Edition) - X11 - Vulkan (Forward+) - integrated Intel(R) Xe Graphics (TGL GT2) () - 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz (8 Threads)
Issue description
I have encountered this crash a few times recently while testing out the master branch. Stack traces from two of them with p_object contents printed:
In case it's relevant: my project features a 3D scene inheriting from an imported .blend file containing multiple meshes. Another 3D scene (the game level) has an instance of the inherited scene in its tree.
Steps to reproduce
I'm unable to reproduce it at will but it occurs maybe once or twice a day.
Minimal reproduction project (MRP)
(No MRP)