> On Mon, Feb 27, 2012 at 2:22 PM, Hefty, Sean <sean.he...@intel.com> wrote:
> >> > --- a/include/linux/socket.h
> >> > +++ b/include/linux/socket.h
> >> > @@ -184,6 +184,7 @@ struct ucred {
> >> > #define AF_PPPOX 24 /* PPPoX sockets */
> >> > #define AF_WANPIPE 25 /* Wanpipe API Sockets */
> >> > #define AF_LLC 26 /* Linux LLC */
> >> > +#define AF_IB 27 /* Native InfiniBand address */
> >> > #define AF_CAN 29 /* Controller Area Network */
> >> > #define AF_TIPC 30 /* TIPC
> sockets */
> >> > #define AF_BLUETOOTH 31 /* Bluetooth sockets */
> >> > @@ -227,6 +228,7 @@ struct ucred {
> >> > #define PF_PPPOX AF_PPPOX
> >> > #define PF_WANPIPE AF_WANPIPE
> >> > #define PF_LLC AF_LLC
> >> > +#define PF_IB AF_IB
> >> > #define PF_CAN AF_CAN
> >> > #define PF_TIPC AF_TIPC
> >> > #define PF_BLUETOOTH AF_BLUETOOTH
> >>
> >> Has this been run by the networking community? Are they OK with this
> >> assignment?
> >
> > I did copy netdev on the original submissions, but I don't remember any
> explicit ack or nack.
>
> David, any feeling yay or nay about adding these?
>
> Is the kernel the final arbiter of AF_xxx / PF_xxx assignments, or
> is there anything else we have to worry about?
To clarify the intent of this change:
The RDMA CM allows users to specify addresses using struct sockaddr. Today,
only INET/6 are supported. The intent is to allow a user to specify native
InfiniBand addresses through that interface. In the more immediate, this helps
to solve InfiniBand scaling issues. Longer term, this can also be used to
control path failover.
- Sean
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
