Re: mach_port_t vs task_t (really ipc_space_t) in Mach header files

[Roland McGrath]

> I am doing this right now where feasible (like, I am doing it for host_t,
> but not for mach_port_name_t, esp as there is no typedef for the latter ;)
> It's only used internally to mark a MACH_MSG_TYPE_PORT_NAME parameter).

mach_port_name_t is not a port type at all.  It is the type of port names,
i.e. a 32-bit value that names a port in some task.

> But I am uncertain, as the interface definitions explicitely set the
> "ctype" to mach_port_t (that's what MiGs makes to put this type in the
> header).

But there are typedefs for task_t et al in <mach/mach_types.h>.

> Maybe we should go and change those interface definitions were we
> prefer the specific types without ctype parameter?

There is probably some obscure reason having to do with mig stubs in the
kernel why those ctype declarations are there.  At any rate, I'm
disinclined to fiddle with the .defs files.