This must be a windows issue (I get crashes when changing the threads with the gmt version fo your executables). On linux I don't see this behavior and I just tried it here. The last set of changes had a thorough workout (excluding my last commit), so it is puzzling. I assume the last fix I made didn't break it?
The random analysis issue - I see that occasionally too on Linux. In fact that's what I am trying to hunt down (Was why I was poking my head in eval.c last night when I committed the one fix). On 13/09/09 8:30 AM, "Massimiliano Maini" <[email protected]> wrote: > There's something worse than before. > > I still get the random analysis issue in nogui exes with >1 thread, > but now I also get crashes in the sse2 g threads exes (gui and nogui) > as soon as i enter the command "set threads 2" ... > The same with win threads seems to work fine. > > MaX. > > P.S. > I've included M.Petch last commit (eval.c 1.384, Properly initialize > all thread specific data to NULL). > > _______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
