Re: 3.0: unwinder changes

[Jeff Sturm <jsturm at one-point dot com>]

On Tue, 12 Jun 2001, Andrew Haley wrote:
>  > So sparc-solaris must have MD_FALLBACK_FRAME_STATE_FOR or there is no
>  > chance of passing the testsuite, I guess... darn.
> 
> Yes, this is a consequence of the new EH structure.  sparc-signal.h is
> way out of date.

Frankly I don't understand how it ever worked:

  register int retaddr = _context->uc_mcontext.gregs[REG_O7];

%o7 has the return address after a call insn, but it doesn't seem to point
to the faulting instruction in a signal handler.

Using this instead almost fixes the problem:

  register int retaddr = _context->uc_mcontext.gregs[REG_PC] - 4;

> You may find that sparc/solaris works in very much the same way as
> Linux.  I think it did the last time I looked.

There is no sigreturn in Solaris.  Certainly there is some equivalent,
but, umm... let me just say I'm learning far more about sparc-solaris
than I bargained for.

Avoidance of signal handlers is partly what made me a java advocate in the
first place ;)

> I don't think that this can possibly happen.  3.0 is near-frozen.

Yep.  Mark is proposing frequent releases from here on, which makes the
3.0 release more tolerable for me.

Jeff