[gcc r16-1748] c++: fix decltype_p handling for binary expressions

[Jason Merrill via Gcc-cvs]

https://gcc.gnu.org/g:1f4a917c610f7c199a6de1eab4fcdc878e4fc5ac

commit r16-1748-g1f4a917c610f7c199a6de1eab4fcdc878e4fc5ac
Author: Jason Merrill <ja...@redhat.com>
Date:   Wed Jun 25 16:26:56 2025 -0400

    c++: fix decltype_p handling for binary expressions
    
    With Jakub's constexpr virtual base patch,
    23_containers/vector/bool/cmp_c++20.cc failed the assert I add to
    fixed_type_or_null, meaning that it returned the wrong value.  Let's fix the
    result as well as adding the assert, and fix cp_parser_binary_expression to
    properly wrap any class-type calls in the operands in TARGET_EXPR even
    within a decltype so we don't hit the assert.
    
    gcc/cp/ChangeLog:
    
            * class.cc (fixed_type_or_null): Handle class-type CALL_EXPR.
            * parser.cc (cp_parser_binary_expression): Fix decltype_p handling.

Diff:
---
 gcc/cp/class.cc  | 9 +++++++++
 gcc/cp/parser.cc | 8 ++++++++
 2 files changed, 17 insertions(+)

diff --git a/gcc/cp/class.cc b/gcc/cp/class.cc
index f30cf3fcd095..6cd6e8f1bfcf 100644
--- a/gcc/cp/class.cc
+++ b/gcc/cp/class.cc
@@ -8350,6 +8350,15 @@ fixed_type_or_null (tree instance, int *nonnull, int 
*cdtorp)
            *nonnull = 1;
          return TREE_TYPE (instance);
        }
+      if (CLASS_TYPE_P (TREE_TYPE (instance)))
+       {
+         /* We missed a build_cplus_new somewhere, likely due to tf_decltype
+            mishandling.  */
+         gcc_checking_assert (false);
+         if (nonnull)
+           *nonnull = 1;
+         return TREE_TYPE (instance);
+       }
       return NULL_TREE;
 
     case SAVE_EXPR:
diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc
index 80fd7990bbbb..b1626acb50b7 100644
--- a/gcc/cp/parser.cc
+++ b/gcc/cp/parser.cc
@@ -10835,6 +10835,14 @@ cp_parser_binary_expression (cp_parser* parser, bool 
cast_p,
            goto pop;
        }
 
+      /* If we skipped build_cplus_new in build_cxx_call because of decltype_p,
+        call it now that we know current.lhs is a subexpression.  */
+      if (decltype_p && !processing_template_decl
+         && TREE_CODE (current.lhs) == CALL_EXPR
+         && CLASS_TYPE_P (TREE_TYPE (current.lhs)))
+       current.lhs = build_cplus_new (TREE_TYPE (current.lhs), current.lhs,
+                                      tf_warning_or_error);
+
      get_rhs:
       current.tree_type = binops_by_token[token->type].tree_type;
       current.loc = token->location;