Re: RFC: C++ PATCH to adjust empty class parameter passing ABI

[Kyrill Tkachov <kyrylo dot tkachov at foss dot arm dot com>]

On 14/04/16 09:02, Christophe Lyon wrote:
> On 13 April 2016 at 22:12, Jason Merrill <jason@redhat.com> wrote:
> > On 04/13/2016 03:18 PM, Jakub Jelinek wrote:
> > > On Wed, Apr 13, 2016 at 03:11:34PM -0400, Jason Merrill wrote:
> > > > commit 761983a023b5217ef831a43f423779940c788ecf
> > > > Author: Jason Merrill <jason@redhat.com>
> > > > Date:   Tue Apr 12 13:16:50 2016 -0400
> > > >
> > > >       gcc/
> > > >          * cfgexpand.c (pass_expand::execute): Handle attribute
> > > > abi_warning.
> > > >          * expr.c (expand_expr_real_1): Likewise.
> > > >       gcc/cp/
> > > >          * call.c (empty_class_msg, mark_for_abi_warning): New.
> > > >          (build_call_a): Use them.
> > > >          * decl.c (store_parm_decls): Use mark_for_abi_warning.
> > > >          * error.c (pp_format_to_string): New.
> > >
> > > I think you should put a space into the attribute name instead of _
> > > to make it clear that it is not an attribute users can use directly in
> > > their
> > > code through __attribute__.
> > >
> > > Otherwise it looks reasonable to me.
> >
> > Thanks, applied with that change.
> Hi,
>
> I'm seeing g++.dg/abi/empty13.C failing at execution on
> aarch64-none-linux-gnu (using qemu):
> qemu: uncaught target signal 11 (Segmentation fault) - core dumped
>
> and aarch64_be-none-elf (using the Foundation Model):
> Terminated by exception.
>
> But it passes on aarch64-none-elf (using the Foundation Model too)
>
> Am I the only one?
Hi Christophe,

I see the test failing on aarch64-none-linux-gnu (native)
with no output, just:
spawn [open ...]
FAIL: g++.dg/abi/empty13.C  -std=gnu++98 execution test

And I see it passing on aarch64-none-elf.

I haven't tried aarch64_be-none-elf yet.

Kyrill


> Christophe.
>
>
>
> > Jason