Should there not be a manpage entry in this commit too?

Indeed. Added to manpage.

Why do we have to set c->c2.tuntap->wintun in both do_init_tun and
> do_open_tun? Should one of them not suffice?

Just to be sure that it is assigned :)

Removed unneeded second assignment.

Should this --windows-driver option not be inside the #ifdef _WIN32 ?

In fact it is already in _WIN32, but the next option has #ifdef _WIN32
so I removed it.

> > +#ifdef _WIN32
> > +bool
> > +parse_windows_driver(const char *str, const int msglevel)
> I think this should be a static function. Also a short (doxygen) comment
> explaining what the return value means would be nice.


> > +    bool wintun;
> Did you consider using an enum instead? I think it would make the code
> easier to read.

Honestly I do not see to much value in converting bool to enum,
it is unlikely that we'll have more tun drivers in the near future. Also
changing it here would break follow-up patches.

As verbally agreed, I'll look into it afterwards.

v3 is on its way.

Openvpn-devel mailing list

Reply via email to