[C++ PATCH] For -gdwarf-5 emit DW_TAG_variable instead of DW_TAG_member for C++ static data members (take 2)

[Jason Merrill]

OK.

On Wed, Feb 22, 2017 at 8:57 AM, Jakub Jelinek <jakub@redhat.com> wrote:
> On Tue, Feb 21, 2017 at 12:37:32PM -0800, Jason Merrill wrote:
>> On Sat, Feb 18, 2017 at 12:17 AM, Jakub Jelinek <jakub@redhat.com> wrote:
>> > On Fri, Feb 17, 2017 at 09:37:09PM -0500, Jason Merrill wrote:
>> >> On Fri, Feb 17, 2017 at 1:52 PM, Jakub Jelinek <jakub@redhat.com> wrote:
>> >> > -      && die->die_tag != DW_TAG_member)
>> >> > +      && die->die_tag != DW_TAG_member
>> >> > +      && (die->die_tag != DW_TAG_variable || !class_scope_p (die->die_parent)))
>> >>
>> >> How about we only check class_scope_p (die->die_parent), and don't
>> >> consider the TAG at all?  DW_TAG_member should only appear at class
>> >> scope.
>> >
>> > That wouldn't work, because that would avoid adding linkage attributes to
>> > methods.  I could use:
>> >       && (die->die_tag == DW_TAG_subprogram || !class_scope_p (die->die_parent))
>> > (not 100% sure if some other tag than those can make it here)
>> > or
>> >       && ((die->die_tag != DW_TAG_member || die->die_tag != DW_TAG_variable)
>> >           || !class_scope_p (die->die_parent))
>> > or just use
>> >       && (!VAR_P (decl) || !class_scope_p (die->die_parent))
>> > or similar.
>> >
>> >> > -         if (old_die->die_tag == DW_TAG_member)
>> >> > +         if (old_die->die_tag == DW_TAG_member
>> >> > +             || (dwarf_version >= 5 && class_scope_p (old_die->die_parent)))
>> >>
>> >> Likewise here.
>> >
>> > This spot probably can be changed as you wrote, it is in gen_variable_die,
>> > so methods shouldn't appear there.
>>
>> Hmm, let's think about the behavior we want here.  I don't see any
>> reason not to put AT_linkage_name on a member DW_TAG_variable; I think
>> the old behavior avoided putting it on DW_TAG_member just because it
>> isn't defined for DW_TAG_member.  So it's not clear to me that we need
>> any changes in add_linkage_name or its call site.
>
> Ah, ok.  But then we need to make sure gen_variable_die doesn't set
> no_linkage_name for -gdwarf-5, because otherwise on the attached testcase
> we don't emit linkage_name on b, c, d anywhere (we used to emit that
> before).
>
> Starting bootstrap/regtest now, does this look ok if that passes?
>
> 2017-02-22  Jakub Jelinek  <jakub@redhat.com>
>
>         * dwarf2out.c (gen_variable_die): For -gdwarf-5, use DW_TAG_variable
>         instead of DW_TAG_member for static data member declarations and don't
>         set no_linkage_name for static inline data members.
>         (gen_member_die): For -gdwarf-5 don't change DW_TAG_variable
>         to DW_TAG_member.
>
>         * g++.dg/debug/dwarf2/inline-var-2.C: New test.
>
> --- gcc/dwarf2out.c.jj  2017-02-18 17:11:18.759821871 +0100
> +++ gcc/dwarf2out.c     2017-02-22 17:48:27.378223877 +0100
> @@ -22669,7 +22669,8 @@ gen_variable_die (tree decl, tree origin
>        && lang_hooks.decls.decl_dwarf_attribute (decl, DW_AT_inline) != -1)
>      {
>        declaration = true;
> -      no_linkage_name = true;
> +      if (dwarf_version < 5)
> +       no_linkage_name = true;
>      }
>
>    ultimate_origin = decl_ultimate_origin (decl_or_origin);
> @@ -22820,9 +22821,10 @@ gen_variable_die (tree decl, tree origin
>      }
>
>    /* For static data members, the declaration in the class is supposed
> -     to have DW_TAG_member tag; the specification should still be
> -     DW_TAG_variable referencing the DW_TAG_member DIE.  */
> -  if (declaration && class_scope_p (context_die))
> +     to have DW_TAG_member tag in DWARF{3,4} and we emit it for compatibility
> +     also in DWARF2; the specification should still be DW_TAG_variable
> +     referencing the DW_TAG_member DIE.  */
> +  if (declaration && class_scope_p (context_die) && dwarf_version < 5)
>      var_die = new_die (DW_TAG_member, context_die, decl);
>    else
>      var_die = new_die (DW_TAG_variable, context_die, decl);
> @@ -24091,7 +24093,8 @@ gen_member_die (tree type, dw_die_ref co
>               && get_AT (child, DW_AT_specification) == NULL)
>             {
>               reparent_child (child, context_die);
> -             child->die_tag = DW_TAG_member;
> +             if (dwarf_version < 5)
> +               child->die_tag = DW_TAG_member;
>             }
>           else
>             splice_child_die (context_die, child);
> @@ -24113,7 +24116,7 @@ gen_member_die (tree type, dw_die_ref co
>         }
>
>        /* For C++ inline static data members emit immediately a DW_TAG_variable
> -        DIE that will refer to that DW_TAG_member through
> +        DIE that will refer to that DW_TAG_member/DW_TAG_variable through
>          DW_AT_specification.  */
>        if (TREE_STATIC (member)
>           && (lang_hooks.decls.decl_dwarf_attribute (member, DW_AT_inline)
> --- gcc/testsuite/g++.dg/debug/dwarf2/inline-var-2.C.jj 2017-02-22 17:54:28.019597440 +0100
> +++ gcc/testsuite/g++.dg/debug/dwarf2/inline-var-2.C    2017-02-22 17:54:10.000000000 +0100
> @@ -0,0 +1,35 @@
> +// { dg-do compile }
> +// { dg-options "-O -std=c++1z -gdwarf-5 -dA -gno-strict-dwarf" }
> +// { dg-require-weak "" }
> +// { dg-final { scan-assembler-not "DW_TAG_member" { xfail *-*-aix* } } }
> +
> +inline int a;
> +struct S
> +{
> +  static inline double b = 4.0;
> +  static constexpr int c = 2;
> +  static constexpr inline char d = 3;
> +  static const int j = 7;
> +  static int k;
> +  static double l;
> +} s;
> +const int S::j;
> +int S::k = 8;
> +template <int N>
> +inline int e = N;
> +int &f = e<2>;
> +template <int N>
> +struct T
> +{
> +  static inline double g = 4.0;
> +  static constexpr int h = 2;
> +  static inline constexpr char i = 3;
> +  static const int m = 8;
> +  static int n;
> +  static double o;
> +};
> +T<5> t;
> +template <>
> +const int T<5>::m;
> +template <>
> +int T<5>::n = 9;
>
>
>         Jakub