On 12/18/23 16:57, Nathan Sidwell wrote:
On 12/18/23 16:31, Jason Merrill wrote:
Tested x86_64-pc-linux-gnu.  Does this make sense?  Did you have another theory
about how to merge these?

Why isn't push_abi_namespace doing the right setup here? (and I think get_global_binding might be similarly problematic?)

What would the right setup be? It pushes into the global module, but before this change lookup doesn't find things imported into the global module, and so we get two independent (and so non-equivalent) declarations.

The comment for get_namespace_binding says "Users of this who, having found nothing, push a new decl must be prepared for that pushing to match an existing decl." But if lookup_elaborated_type fails, so we pushtag a new type, check_module_override doesn't try to merge them because TREE_PUBLIC isn't set on the TYPE_DECL yet at that point, and they coexist until we complain about redeclaring __dynamic_cast with non-matching parameter types.

I tried setting TREE_PUBLIC on the TYPE_DECL, and then check_module_override called duplicate_decls, and rejected the redeclaration as a different type.

-- 8< --

Doing a dynamic_cast in both TUs broke because we were declaring a new
__class_type_info in _b that conflicted with the one imported in the global module from _a.  lookup_elaborated_type has a comment that we probably don't want to find such imports in general, but in this case it seems necessary to
make the artificial lazy declarations of RTTI types work.

gcc/cp/ChangeLog:

    * name-lookup.cc (lookup_elaborated_type): Look for bindings
    in the global namespace in the ABI namespace.

gcc/testsuite/ChangeLog:

    * g++.dg/modules/pr106304_b.C: Add dynamic_cast.
---
  gcc/cp/name-lookup.cc                     | 10 ++++++++++
  gcc/testsuite/g++.dg/modules/pr106304_b.C |  1 +
  2 files changed, 11 insertions(+)

diff --git a/gcc/cp/name-lookup.cc b/gcc/cp/name-lookup.cc
index 09dc6ef3e5a..f15b338025d 100644
--- a/gcc/cp/name-lookup.cc
+++ b/gcc/cp/name-lookup.cc
@@ -8092,6 +8092,16 @@ lookup_elaborated_type (tree name, TAG_how how)
            // FIXME: This isn't quite right, if we find something
            // here, from the language PoV we're not supposed to
            // know it?
+          // We at least need to do this in __cxxabiv1 to unify lazy
+          // declarations of __class_type_info in build_dynamic_cast_1.
+          if (current_namespace == abi_node)
+        {
+          tree g = (BINDING_VECTOR_CLUSTER (*slot, 0)
+                .slots[BINDING_SLOT_GLOBAL]);
+          for (ovl_iterator iter (g); iter; ++iter)
+            if (qualify_lookup (*iter, LOOK_want::TYPE))
+              return *iter;
+        }
          }
      }
      }
diff --git a/gcc/testsuite/g++.dg/modules/pr106304_b.C b/gcc/testsuite/g++.dg/modules/pr106304_b.C
index e8333909c8d..0d1da086176 100644
--- a/gcc/testsuite/g++.dg/modules/pr106304_b.C
+++ b/gcc/testsuite/g++.dg/modules/pr106304_b.C
@@ -5,4 +5,5 @@ module pr106304;
  void f(A& a) {
    as_b(a);
+  dynamic_cast<B*>(&a);
  }

base-commit: 5347263b347d02e875879ca40ca6e289ac178919
prerequisite-patch-id: 66735c0c7beb22586ed4b632d10ec9094bb9920c


Reply via email to