This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: segfault in sysdep/i386/backtrace.h
- From: Andrew Haley <aph at redhat dot com>
- To: Danny Smith <dannysmith at clear dot net dot nz>
- Cc: "'GCJ'" <java at gcc dot gnu dot org>
- Date: Wed, 21 Feb 2007 10:00:41 +0000
- Subject: Re: segfault in sysdep/i386/backtrace.h
- References: <000001c7559c$4b401ae0$8c6d65da@anykey>
Danny Smith writes:
> Reference: http://gcc.gnu.org/ml/java/2007-02/msg00146.html
>
> Andrew Haley wrote
>
> > It's going to be hard. If you want to use DWARF and you want to
> > unwind through (and throw execptions though) libraries compiled with
> > MSVC then you have to teach gcc how to unwind through Structured
> > Exception Handling blocks. Perhaps there are Windows library calls to
> > do this.
>
> That part has been done, See attached patch, specifically the
> w32-unwind.h contribution from Pascal Obry. For an example of how it
> could be used see ada/seh_init.c.
>
> Note this patch was extracted from a much larger local patchset.
> I haven't tested independently of the other bits.
>
> The part that hasn't been done is the ability to unwind from
> callback functions passed to win32api functions. Such callbacks are
> common in windows GUI world. mingw users -- at least those who have
> expressed an opinion -- want them to work in C++. SJLJ works. DW2
> does not and will not until we can convince MS (or ReactOS) to
> compile the w32api with DW2 tables.
What for? Why can't we unwind through Win32 code blocks?
Andrew.