This has been reported before, but I haven't seen any indication
that things might get fixed --
Basically stack unwinding is totally broken. This is pretty obvious when
you try to do backtrace. It also seems to relate to seeing incorrect
values
for variables -- if you don't interpret the current stack frame
correctly, you
can't determine the location of variables or their locations.
Breakpoints do the wrong thing. They stop alright, but they don't act
as breakpoints -- just exceptions. This means none of the mechanisms for
tying commands to breakpoints work.
I'd like to know if these problems have always been there or are symptoms of
changes to the gdb generic sources that didn't get reflected in the msp
specific
sources. Also, is anybody likely to fix these ?
Geoffrey Brown
