[Bug ada/82141] [8 Regression] raised RTSFIND.RE_NOT_AVAILABLE : rtsfind.adb:851 on darwin

["iains at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82141

Iain Sandoe <iains at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |iains at gcc dot gnu.org

--- Comment #14 from Iain Sandoe <iains at gcc dot gnu.org> ---
(In reply to Eric Botcazou from comment #8)
> > The exception handling on darwin is indeed the same as in linux, but does it
> > mean that all its mishandling leads to duplicate?
> 
> The issue is what I described in my previous comment.  Into how many
> sub-issues it must be split in order to be fully addressed is IMO rather
> uninteresting.

Hi Eric,

I am trying to catch up on these two issue (80556/82141).

So far, I've (hacked out) removal of the use of libgcc's unwinder and I'm
working on 10.11 (so that the version mismatch is irrelevant).

I'm checking which libraries are loaded (DYLD_PRINT_LIBARIES) and which symbols
are dynamically bound (DYLD_PRINT_BINDINGS). AFAICT the unwinder symbols used
by gnat1 are coming from libunwind.dylib (as is correct on Darwin15/OS X
10.11).

The 'raised RTSFIND.RE_NOT_AVAILABLE : rtsfind.adb:851' is still present.

So.. I wonder if we have (a) a bug in libunwind that's not dealing correctly
with some expected behaviour or (b) some extension to DWARF unwinding that's
needed to support Ada, which is not yet present in libunwind (libunwind is Open
Sourced as part of the LLVM project).

Are you aware of any special needs for the unwinder to support Ada?