This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Adjust empty class parameter passing ABI (PR c++/60336)
On Tue, Nov 07, 2017 at 09:14:58AM +0100, Richard Biener wrote:
> On Mon, 6 Nov 2017, Jason Merrill wrote:
>
> > On Mon, Nov 6, 2017 at 10:37 AM, Marek Polacek <polacek@redhat.com> wrote:
> > > On Fri, Nov 03, 2017 at 12:19:05PM -0400, Jason Merrill wrote:
> > >> On Fri, Nov 3, 2017 at 9:55 AM, Marek Polacek <polacek@redhat.com> wrote:
> > >> > + TYPE_EMPTY_P (t) = targetm.calls.empty_record_p (t);
> > >>
> > >> I think we want to set this in finalize_type_size; since the point of
> > >> all this is becoming compliant with the psABI (and compatible with the
> > >> C front end), I wouldn't think it should be specific to the C++ front
> > >> end.
> > >
> > > Sure, that works. I'd wanted to do the TYPE_EMPTY_P setting in layout_type,
> > > but that wasn't called for the classes in my testcases. I've moved the
> > > setting of TYPE_EMPTY_P to finalize_type_size now.
> > >
> > >> > + TYPE_WARN_EMPTY_P (t) = warn_abi && abi_version_crosses (12);
> > >>
> > >> Can this flag go on the TRANSLATION_UNIT_DECL rather than the type?
> > >
> > > Yeah, that works, too. To avoid needing a lang hook, I'm setting the
> > > flag in cxx_init_decl_processing, hope that's ok.
> >
> > Sure.
> >
> > >> > + if (TREE_CODE (field) == FIELD_DECL
> > >> > + && (DECL_NAME (field)
> > >> > + || RECORD_OR_UNION_TYPE_P (TREE_TYPE (field)))
> > >> > + && !default_is_empty_type (TREE_TYPE (field)))
> > >> > + return false;
> > >> > + return true;
> > >>
> > >> Hmm, this assumes that any unnamed field can be ignored; I'm concerned
> > >> that some front end might clear DECL_NAME for a reason that doesn't
> > >> imply that the field is just padding.
> > >
> > > In that case I guess we need a new lang hook, right? Because in
> > > default_is_empty_type we can't check FE-specific flags such as
> > > DECL_C_BIT_FIELD. For C++, should that lang hook be just
> > > is_really_empty_class? Or is there anything else I can do?
> >
> > Hmm, maybe leave it as it is and just document this assumption about
> > FIELD_DECL with null DECL_NAME, both here and in tree.def.
>
> But are you sure you are not changing the ABI for a language other
> than C++ then?
We might be but I guess we should follow the psABI for all languages,
not just C++.
> I don't think DECL_NAME has any special meaning - why
> not check DECL_ARTIFICIAL or another flag that has appropriate
> semantic - _what_ semantic are you looking for after all?
I believe we just want to see if a field is padding. Just checking
DECL_ARTIFICIAL will not fly, because unnamed bit-field are not artificial, and
we should consider them padding. So I'd think it makes sense to stick with
DECL_NAME.
Marek