[flang-commits] [flang] [flang] Lower special bind(c) cases without binding labels (PR #65758)

[Peter Klausler via flang-commits]

================
@@ -99,11 +99,14 @@ Fortran::lower::mangle::mangleName(const Fortran::semantics::Symbol &symbol,
 
   // TODO: A procedure that inherits BIND(C) through another interface
   // (procedure(iface)) should be dealt with in GetBindName() or some wrapper.
-  if (!Fortran::semantics::IsPointer(ultimateSymbol) &&
-      Fortran::semantics::IsBindCProcedure(ultimateSymbol) &&
-      Fortran::semantics::ClassifyProcedure(symbol) !=
-          Fortran::semantics::ProcedureDefinitionClass::Internal)
-    return ultimateSymbol.name().ToString();
+  if (const auto *procDetails{
----------------
klausler wrote:

> Do you mean that the following should be forbidden?
> 
> ```
>  interface
>    subroutine iface() bind(c, "some_c_name")
>    end subroutine
>  end interface
>  procedure(iface) :: foo
>  call foo()
> end
> ```

What would it mean?  Does it make 'foo' an alias for 'iface', using the same binding label?  Does it mean that `foo` has the `BIND(C)` attribute with no explicit binding label?  The standard doesn't say.

>  > You could call SetBindNameOn(symbol) on foo as part of procedure-declaration-stmt name resolution
> 
> So you think we can already call `Fortran::semantics::IsBindCProcedure(symbol)`instead of using `attrs_->test(Attr::BIND_C)` in SetBindNameOn?

Why would you want to make that change?

Also - a procedure-declaration-stmt *does* allow a `BIND(C[,NAME=...])` attribute -- I didn't notice it in the syntax because it was indirected.  So now we have to worry about the interpretation of having `BIND(C)` on the explicit interface declaration and on the procedure-declaration-stmt.




https://github.com/llvm/llvm-project/pull/65758