Re: GCJ/minGW produced executables and linux/wine

["Ranjit Mathew" <rmathew at hotmail dot com>]

>  > > Indeed.  Has gcj on that target ever worked with DWARF2 exceptions?
>  >
>  > Yes, statically linked executables do work fine. Only interpreted
>  > programs have problems due to the reason (I strongly believe) cited
>  > above.
>
> What, even unwinding through invoke() ?

No, not at all, since that ultimately uses libffi as well via
_Jv_CallAnyMethodA( ). Ok, so the "work fine" should be taken
with this grain of salt. :-P

BTW, I now understand what the original question from you
and the one from Jeff meant (sorry, 'am a little thick):

Win32 uses win32_exception_handler( ) in win32.cc via
Win32 SetUnhandledExceptionFilter( ) to catch
EXCEPTION_ACCESS_VIOLATION (throw NullPointerException), etc.
and since SJLJ uses a longjmp to implement "throw", would this
create problems? This is similar to catch_segv( ), etc. in
prims.cc.

Well honestly, I don't know and I'm surprised that it works
on Linux (so signal delivery must be the "last mile" in that
the OS doesn't really care if the handler returns or not).

I can imagine that there would be problems on Win32 because
this handler is supposed to return a value to the OS telling
it whether to execute the default handler (abort program)
or whether the exception has been handled successfully. See
MSDN docs for SetUnhandledExceptionFilter( ) or this article
for more details:

http://www.microsoft.com/msj/0197/exception/exception.htm

However, right now I don't know what better methods there
would be to generate a proper NullPointerException,
ArithmeticException, etc.

Ranjit.

_________________________________________________________________
Cricket World Cup 2003- News, Views and Match Reports. 
http://server1.msn.co.in/msnspecials/worldcup03/