Uh, probably not. The first VAX Unix implementation was
produced by BBN, under ARPA contract. I think ARPA's contract with
Berkeley was primarilly to add shared-memory/VM features to BSD Unix
(umm... this work was later abandoned in favor of the SysV extensions
in that area?). It is unclear to me, at this point in time, whether
the BSD developers incorporated portions of BBN's work into BSD Unix,
or whether they created a new IP stack from scratch.
>*The* first implementation was on PDP-10. Either TENEX or
>TOPS-20. With completely different API, BTW. Non-UNIX stack on VAXen is
>seriously younger - ask Cutler for details. -11... Dunno. 2.10BSD seems to
>be the first there. IIRC RT-11 never got that. Maybe RSX did, but I never
>heard of such event.
Although the Tenex IP stack was, I think, the primary
experimental and early production platform, PDP-11's running various
operating systems were also important. I wrote an IP/UDP (no TCP)
stack for the EPOS operating system at ISI in 1980, which was used for
network speech and video (ST, IPv5) research, among other functions.
At the time, I had access to listings of two or three earlier PDP-11
IP stacks (minimal) for reference. There may even have been a
(user-mode?) PDP-11 Unix implementation.
RT-11 had an early, and actually quite significant, role in IP
implementation. The "Fuzzball" RT-11's implemented the network time
protocols and a number of other useful utilities; I think they could
even be used as routers ("gateways", in the original IP parlance).
The EPOS IP implementation also served as a bridging router (ARPANet
to 3 Mbit Ethernet), but it was installed at only two locations,
whereas Fuzzballs were widely distributed.
Craig Milo Rogers
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/