Re: [PATCH] c++, v4: Retry the aliasing of base/complete cdtor optimization at import_export_decl time [PR113208]

[Jason Merrill <jason@redhat.com>]

On 4/25/24 07:22, Jakub Jelinek wrote:
> On Thu, Apr 25, 2024 at 02:02:32PM +0200, Jakub Jelinek wrote:
>> I've tried the following patch, but unfortunately that lead to large
>> number of regressions:
>> +FAIL: g++.dg/cpp0x/initlist25.C  -std=c++17 (test for excess errors)
> 
> So the reduced testcase for this is
> template <typename T, typename U> struct A {
>    T a1;
>    U a2;
>    template <typename V, typename W, bool = true>
>    constexpr A(V &&x, W &&y) : a1(x), a2(y) {}
> };
> template <typename> struct B;
> namespace std {
> template <class> struct initializer_list {
>    int *_M_array;
>    decltype (sizeof 0) _M_len;
> };
> }
> template <typename T, typename U> struct C {
>    void foo (std::initializer_list<A<const T, U>>);
> };
> template <class> struct D;
> template <typename T, typename = D<T>, typename = B<T>>
> struct E { E (const char *); ~E (); };
> int
> main ()
> {
>    C<E<char>, E<char>> m;
>    m.foo ({{"t", "t"}, {"y", "y"}});
> }
> Without the patch I've just posted or even with the earlier version
> of the patch the
> _ZN1AIK1EIc1DIcE1BIcEES5_EC[12]IRA2_KcSB_Lb1EEEOT_OT0_
> ctors were emitted, but with this patch they are unresolved externals.
> 
> The reason is that the code actually uses (calls) the
> _ZN1AIK1EIc1DIcE1BIcEES5_EC1IRA2_KcSB_Lb1EEEOT_OT0_
> __ct_comp constructor, that one has TREE_USED, while the
> _ZN1AIK1EIc1DIcE1BIcEES5_EC2IRA2_KcSB_Lb1EEEOT_OT0_
> __ct_base constructor is not TREE_USED.
> 
> But the c_parse_final_cleanups loop over
> FOR_EACH_VEC_SAFE_ELT (deferred_fns, i, decl)
> will ignore the TREE_USED __ct_comp because it is an alias
> and so has !DECL_SAVED_TREE:
> 5273		  if (!DECL_SAVED_TREE (decl))
> 5274		    continue;

Hmm, maybe maybe_clone_body shouldn't clear DECL_SAVED_TREE for aliases, 
but rather set it to some stub like void_node?

Though with all these changes, it's probably better to go with your 
first patch for GCC 14 and delay this approach to 15.  Your v1 patch is 
OK for 14.

Jason