> I've been shown this text from the cygwin mailing list: > > --- start --- > Searching the mailing list archives would show that this question has > already come up (several times this week, in fact). The fault is > EXPECTED, and HARMLESS - just continue on debugging from that point. > Basically, inside cygwin functions, cygwin installs its own fault > detection handler, and pthread_key_create intentionally faults on the > first time the key is used. It is just a matter of teaching gdb that this > fault is expected and will be covered seamlessly by cygwin. > --- end --- > > There is this message that is more thorough: > > http://www.cygwin.com/ml/cygwin-patches/2006-q1/msg00039.html >
Running flightgear in the debugger and continuing with execution when the pthreads "errors" are raised finally ends in this: Program received signal SIGILL, Illegal instruction. 0x69596fff in gluPickMatrix () (gdb) where #0 0x69596fff in gluPickMatrix () #1 0x000086c8 in ?? () #2 0x020ab628 in ?? () #3 0x6973815b in nvoglnt!DrvCopyContext () #4 0x020ab628 in ?? () #5 0x00000000 in ?? () from I just recently installed a new video card - a GeForce 6800. Line #3, above, looks like an nvidia opengl call? And then, there's a call to a glu routine? Suspicious? Jon _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
