Re: [Patch net-next 4/7] sunrpc: use generic union inet_addr

["J. Bruce Fields" <bfields@fieldses.org>]

On Thu, Jul 25, 2013 at 08:34:43PM +0800, Cong Wang wrote:
> On Tue, 2013-07-23 at 12:40 -0400, J. Bruce Fields wrote:
> > On Mon, Jul 22, 2013 at 03:05:10PM +0800, Cong Wang wrote: 
> > >  static inline unsigned short rpc_get_port(const struct sockaddr *sap)
> > >  {
> > > -	switch (sap->sa_family) {
> > > -	case AF_INET:
> > > -		return ntohs(((struct sockaddr_in *)sap)->sin_port);
> > > -	case AF_INET6:
> > > -		return ntohs(((struct sockaddr_in6 *)sap)->sin6_port);
> > > -	}
> > > -	return 0;
> > > +	return inet_addr_get_port((const union inet_addr *)sap);
> > >  }
> > 
> > Is there any reason to keep the rpc_get_port wrapper at all after this?
> > Or if its still useful to have the convenience of not having to do the
> > cast, maybe the wrapper should move to common code to?  (Is there some
> > reason only the rpc code needs this?)
> > 
> 
> Yeah, the only reason why I keep it is that I don't want to touch all
> the callers of rpc_get_port().

Makes sense, thanks.

--b.

> I agree that we can make it a generic
> function.
> 
> Thanks.
>