Re: Reminder: Illegal Package-Private Accesses in libgcj

[Bryce McKinlay <bryce at mckinlay dot net dot nz>]

On Nov 3, 2003, at 5:39 PM, Ranjit Mathew wrote:

> Bryce McKinlay wrote:
> > > 1. java.lang.VMThrowable illegally accesses
> > >   gnu.gcj.runtime.StackTrace.stackTraceAddrs( )
> > >
> > > 2. gnu.gcj.runtime.NameFinder illegally accesses
> > >   java.lang.StackTraceElement's constructor
> >
> >
> > I am currently working on some significant changes the stack trace
> > stuff (http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12740) - both of
> > these will go away when I'm done.
>
> Cool! Somewhat off-topic for this thread, I'm curious
> to know how you will be able to avoid calling backtrace( )
> for targets that do not have DWARF-2 EH frame based
> unwinding for an exception. For example, Win32 is an
> SJLJ-EH target (as of now).

I havn't added any support for using backtrace() or platform-specific 
unwinding code at this time. Ideally win32 will switch to using the 
DWARF2 unwinder. It doesn't look like there is much missing - libffi 
unwind information, and an MD_FALLBACK_FRAME_STATE_FOR or equivalent? 
If there are targets that will never support the unwinder (are there?), 
then it might make sense to allow platforms to define some 
platform-specific unwinding functions, but ideally, by far the cleanest 
and most efficient solution is to get the libgcc DWARF2 unwinder 
working.

Regards

Bryce.