2025-08-02 07:15 CEST

View Issue Details Jump to Notes ]
IDProjectCategoryView StatusLast Update
0002247Spring engineGfx driver issuespublic2010-12-09 11:14
Reportert0 
Assigned Toabma 
PrioritynormalSeveritycrashReproducibilityN/A
StatusresolvedResolutionwon't fix 
Product Version0.82.6.1 
Target VersionFixed in Version 
Summary0002247: crashed at exit in 3DModel.cpp:144
DescriptionI don't remember it i looked on statistic or not but when my springlobby crashed i looked for some info and found nice jusy core file from spring.
Additional InformationInfolog looks ok but I'm attached it anyway. This happend under mintlinux(almost like ubuntu 10.10) spring was builded by me, uset mt version. I dont know how to get nice pstack under linux so if needed i will need that info.
TagsNo tags attached.
Checked infolog.txt for Errors
Attached Files

-Relationships
+Relationships

-Notes

~0006027

hoijui (reporter)

are you sure the core file is from that run? cause the infolog suggests the game ended fine.
when you compiled spring yourself, you should have debug symbols packed together with the binaries already, which means it would successfully auto-translate the stack trace on a crash, and write it to infolog.

sooo.. as spring seems not to be crashed, but SL yes, maybe you should check for error info there, and report it on their site.

~0006030

t0 (reporter)

It crashed again. Core is that same (size) but infolog is diffrent(attached), probably i messed something up with previous. SL didnt crashed.

~0006031

hoijui (reporter)

Last edited: 2010-12-05 22:51

from 2nd infolog:

[ 25252] Segmentation fault (SIGSEGV) in spring 0.82.6.1 (MT-Sim)
[ 25252] Stacktrace:
[ 25252] bin/spring-multithreaded() [0xa8609f]
[ 25252] /lib/libc.so.6(+0x33c20) [0x7fab1e6d2c20]
[ 25252] bin/spring-multithreaded() [0x85ca66]
[ 25252] bin/spring-multithreaded() [0x85c90f]
[ 25252] Translated Stacktrace:
[ 25265] spring_0.82.6.1/rts/System/Platform/Linux/CrashHandler.cpp:250
[ 25265] ??:0
[ 25265] spring_0.82.6.1/rts/Rendering/Models/3DModel.cpp:144
[ 25265] /usr/include/c++/4.4/bits/stl_vector.h:533

the interesting line:
https://github.com/spring/spring/blob/0.82.6.1/rts/Rendering/Models/3DModel.cpp#L144

... then again, the stack trace looks quite screwed. maybe you should run in gdb.
i usually use the GUI front-end ddd. basically what you do is, start a game normally with the lobby (just to get a script.txt (that file defines the game settings, like number of players, sides, start points, ...)). then you stop the game, and run it in gdb or ddd again:
gdb /home/t0/progi/bin/spring-multithreaded
> run /full/path/to/script.txt
... play until crash
> bt

edit:
.. maybe also try if non-MT spring works, though i guess it is not that.

~0006032

abma (administrator)

or attach gdb to the running process:
$ gdb $(which spring) -p $(pgrep -x spring)

~0006044

t0 (reporter)

ok i found possible solution - new ati drivers. I played today watched many games and didn't happed anything bad. Tomorrow I will check how it is working my own mt version not that provided by ppa repo. Do you need anything more to close that bug?

~0006048

abma (administrator)

shame on ati for these buggy opengl drivers!
+Notes

-Issue History
Date Modified Username Field Change
2010-12-05 01:16 t0 New Issue
2010-12-05 01:16 t0 File Added: infolog.txt
2010-12-05 11:45 hoijui Note Added: 0006027
2010-12-05 18:13 t0 File Added: infolog.txt_crashed
2010-12-05 18:15 t0 Note Added: 0006030
2010-12-05 22:50 hoijui Note Added: 0006031
2010-12-05 22:51 hoijui Note Edited: 0006031
2010-12-06 20:22 abma Note Added: 0006032
2010-12-06 20:24 abma Summary crashed at exit => crashed at exit in 3DModel.cpp:144
2010-12-08 03:08 t0 Note Added: 0006044
2010-12-09 11:14 abma Category General => Gfx driver issues
2010-12-09 11:14 abma Note Added: 0006048
2010-12-09 11:14 abma Status new => resolved
2010-12-09 11:14 abma Resolution open => won't fix
2010-12-09 11:14 abma Assigned To => abma
+Issue History