On Tue, Apr 25, 2023 at 5:34 PM Edward Catmur via Std-Discussion <std-discussion@lists.isocpp.org> wrote:
On Tue, 25 Apr 2023 at 14:39, Thiago Macieira via Std-Discussion <std-discussion@lists.isocpp.org> wrote:
On Tuesday, 25 April 2023 11:53:46 PDT Lukas Barth via Std-Discussion wrote:
> At this point I'm at my wit's end. How is that rule to be interpreted?
Why are you trying to interpret it? Can't you just assume that if the type
isn't a trivial scalar, it can't be passed?
An enum is a trivial scalar.
> Could it be that this use of "compatible" in `[stdarg.syn]/1.2` wasgrandfathered in from C, but is not defined anymore?
Yes.
It seems that we need to open an LWG issue about this. "Compatible" is indeed not defined and we cannot simply assume that it means what it means in the C standard, because not every C++ type is a valid C type, meaning that it would not be clear how to apply the definition from C.
The damage is limited by https://eel.is/c++draft/expr.call#11.sentence-7 :
Passing a potentially-evaluated argument of a scoped enumeration type or of a class type ([class]) having an eligible non-trivial copy constructor, an eligible non-trivial move constructor, or a non-trivial destructor ([special]), with no corresponding parameter, is conditionally-supported with implementation-defined semantics.
-- Janet Cobb