Last night (using completely up-to-date CVS) I noticed that fgfs segfaults
reliably if one relocates the a/c to KOAK using the "position a/c on ground"
menu.
After a bit more investigation, it appears that this is caused by a nan in the
yasim turbulence stuff, because it's being fed garbage. Using a JSBSim a/c,
one is unceremoniously dumped into the sea.
Starting the sim at KOAK does not display the problem, one must start
somewhere else (anywhere, as far as I can tell) and relocate for the problem
to show.
AJ
GDB shows;
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread -1225333040 (LWP 32321)]
yasim::Turbulence::getTurbulence (this=0xfc39620, loc=0xa1e0c008,
alt=nan(0x400000), up=0xbfc60cc0, turbOut=0xbfc60d10) at
Turbulence.cpp:100
100 static inline float c2fu(unsigned char c) { return
(c+0.5)*(1.0/256); }
(gdb) bt
#0 yasim::Turbulence::getTurbulence (this=0xfc39620, loc=0xa1e0c008,
alt=nan(0x400000), up=0xbfc60cc0, turbOut=0xbfc60d10) at
Turbulence.cpp:100
#1 0x081d2843 in yasim::Model::localWind (this=0xfc31bac, pos=0xbfc60df0,
s=0xbfc60f10, out=0xbfc60e00, alt=-1.52296607e-18) at Model.cpp:537
#2 0x081d2113 in yasim::Model::calcForces (this=0xfc31bac, s=0xbfc60f10)
at Model.cpp:399
#3 0x081cf70c in yasim::Integrator::calcNewInterval (this=0xfc31bb0)
at Integrator.cpp:176
#4 0x081c79d4 in yasim::FGFDM::iterate (this=0xfc31ba8, dt=0.00833333377)
at FGFDM.cpp:91
#5 0x081c1b70 in YASim::update (this=0xbc03698, dt=0.22352000276247658)
at YASim.cxx:213
#6 0x08051b45 in fgUpdateTimeDepCalcs () at main.cxx:168
#7 0x08052714 in fgMainLoop () at main.cxx:476
#8 0x08089ac9 in fgOSMainLoop () at fg_os_sdl.cxx:232
#9 0x080551b0 in fgMainInit (argc=2, argv=0xbfc61824) at main.cxx:1007
#10 0x080515da in main (argc=-1579106296, argv=0xa1e0c008) at
bootstrap.cxx:193
(gdb)
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@flightgear.org
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
2f585eeea02e2c79d7b1d8c4963bae2d
