Re: Another backtrace for that crash....

2009-03-28 Thread rgheck
Vincent van Ravesteijn wrote: Note that if you put the call to LFUN_RECENTER/LFUN_SCREEN_SHOW_CURSOR before the first call to textMetrics(), the crash disappears in all cases. So it looks like a caching problem, initialization problem. That is also supported by the fact that I don't see this at

Re: Another backtrace for that crash....

2009-03-28 Thread Vincent van Ravesteijn
Note that if you put the call to LFUN_RECENTER/LFUN_SCREEN_SHOW_CURSOR before the first call to textMetrics(), the crash disappears in all cases. So it looks like a caching problem, initialization problem. That is also supported by the fact that I don't see this at Windows. How was it again, Wi

Re: Another backtrace for that crash....

2009-03-27 Thread Richard heck
...and another. Not sure why I get different ones from time to time. Note that this one is at the assignment added at r28342. gdb /cvs/lyx16/src/lyx --interpreter=mi2 -quiet (gdb) bt bt #0 std::_Rb_tree, std::_Select1st >, std::less, std::allocatorlyx::ParagraphMetrics> > >::_M_erase (this=