cm36+@andrew.cmu.edu (C. V. R. Murthy) (04/04/88)
Hello,
PREFACE:
It is well known that the stream mechanism in System V is
extremely modular in that it enables flexible PUSHing of protocol layers onto
the stream.
I am trying to port an RPC package developed (on BSD environment) at CMU to the
System V.
The RPC package uses DATAGRAM sockets supported by UDP/IP prtocols.
QUESTION:
Is there any System V out there (there must be one !) on the
Internet using UDP/IP protocols
to communicate with the outside world ?
If so, thank god! I need the following information on the same :
1. What sort of interface to the device it employs : old-styled or stream ?
2. System Release Version.
3. References to the relavant documentation and technical details.
Thanks Very much to any body who cares to reply.
Murthy (cm36@andrew.cmu.edu)lm@arizona.edu (Larry McVoy) (04/04/88)
In article <12778@brl-adm.ARPA> cm36+@andrew.cmu.edu (C. V. R. Murthy) writes: > Is there any System V out there (there must be one !) on the >Internet using UDP/IP protocols to communicate with the outside world ? Lachman Associates, Inc. 1901 North Naper Blvd. Naperville, Illinois 60540-1031 1-800-LAI-UNIX They provide a streams based set of internet protocols. IP, TCP, UDP, rlogin, etc., for System V boxes. They are the only source for NFS for System V that I have ever heard of. Disclaimer: I work for them. But I would still be happy with the product even if I didn't (you get rlogin et al and System V starts to look OK). -- "These aren't my thoughts, they're my cat walking on the keyboard." Larry McVoy lm@arizona.edu or ...!{uwvax,sun}!arizona.edu!lm
ron@topaz.rutgers.edu (Ron Natalie) (04/04/88)
Some System V implementations use Wollongong code (such as it was), this includes most of AT&T's so called computers. The CRAY UNICOS (mostly system V) used Wollongong as well, but they got over it and have redone it themselves. None of the above use streams, they use the traditional Berkeley interfaces hacked into the Sys V kernel. -Ron
ekrell@hector.UUCP (Eduardo Krell) (04/05/88)
In article <Apr.4.10.54.29.1988.6558@topaz.rutgers.edu> ron@topaz.rutgers.edu (Ron Natalie) writes: >Some System V implementations use Wollongong code (such as it was), this >includes most of AT&T's so called computers. The CRAY UNICOS (mostly >system V) used Wollongong as well, but they got over it and have redone >it themselves. None of the above use streams, they use the traditional ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >Berkeley interfaces hacked into the Sys V kernel. Where did you hear that?. All the Wollongong code that runs on SVR3 is streams-based. They don't use "traditional Berkeley interfaces hacked into the sys V kernel"; they have a user-level socket library which implements the BSD system calls using streams. Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ UUCP: {ihnp4,ucbvax}!ulysses!ekrell ARPA: ekrell@ulysses.att.com
jeff@polyslo.UUCP (Jeff Weinstein) (04/05/88)
In article <10209@ulysses.homer.nj.att.com> ekrell@hector (Eduardo Krell) writes: >Where did you hear that?. All the Wollongong code that runs on SVR3 >is streams-based. They don't use "traditional Berkeley interfaces >hacked into the sys V kernel"; they have a user-level socket library >which implements the BSD system calls using streams. WRONG, WRONG, WRONG First of all the original note mentioned System V, not SVR3, so it could have been Release 2 based. Secondly, there are two Wollongong implementations for SVR3, one based on streams, one not based on streams. The one not based on streams is the one that we run here at Cal Poly, which AT&T gave us to run on our 3b2's. The code does provide the "traditional Berkeley Interfaces", as well as the AT&T TLI interface, but no streams. Jeff Weinstein Computer Systems Lab Cal Poly State Univ. ucbvax!voder!polyslo!jeff jeff@polyslo.uucp
daveb@llama.rtech.UUCP (Crack? No thanks, I've got a new CD player) (04/06/88)
In <Apr.4.10.54.29.1988.6558@topaz.rutgers.edu> Ron Natalie writes: >Some System V implementations use Wollongong code (such as it was), this >includes most of AT&T's so called computers. The CRAY UNICOS (mostly >system V) used Wollongong as well, but they got over it and have redone >it themselves. None of the above use streams, they use the traditional >Berkeley interfaces hacked into the Sys V kernel. But, curiously, the only interface TWG supports for their SV.3 tcp/ip is a socket emulation library. Perhaps I haven't looked far enough, or the manuals are poorly indexed, but I did not see a STREAMS/TLI interface documented. Is Cray using a TLI interface for tcp/ip? How about any of the UNIX/386 ports? Is there a standard TLI interface for tcp/ip, or is everyone rolling their own in incompatible ways? bye now, -dB "Do you programmers do sh*t like this a lot?" "Every g*d*mn day." {amdahl, cpsc6a, mtxinu, ptsfa, sun, hoptoad}!rtech!daveb daveb@rtech.uucp
ekrell@hector.UUCP (Eduardo Krell) (04/06/88)
In article <1772@polyslo.UUCP> jeff@polyslo.UUCP (Jeff Weinstein) writes: >there are two Wollongong implementations >for SVR3, one based on streams, one not based on streams. I was referring to the latest Wollongong release, WIN/3B 2.1, which is streams-based and provides a user-level socket interface library. Eduardo Krell AT&T Bell Laboratories, Murray Hill, NJ UUCP: {ihnp4,ucbvax}!ulysses!ekrell ARPA: ekrell@ulysses.att.com
tgr@picuxa.UUCP (Dr. Emilio Lizardo) (04/07/88)
In article <1772@polyslo.UUCP> jeff@polyslo.UUCP (Jeff Weinstein) writes: :In article <10209@ulysses.homer.nj.att.com> ekrell@hector (Eduardo Krell) writes: :>Where did you hear that?. All the Wollongong code that runs on SVR3 :>is streams-based. They don't use "traditional Berkeley interfaces :>hacked into the sys V kernel"; they have a user-level socket library :>which implements the BSD system calls using streams. : :WRONG, WRONG, WRONG : :First of all the original note mentioned System V, not SVR3, so it could :have been Release 2 based. Secondly, there are two Wollongong implementations :for SVR3, one based on streams, one not based on streams. The one not based :on streams is the one that we run here at Cal Poly, which AT&T gave us to :run on our 3b2's. The code does provide the "traditional Berkeley Interfaces", :as well as the AT&T TLI interface, but no streams. Wollongong WIN/3B Enhanced TCP/IP, Release 2.1, is the STREAMS implementation. It is backwardly compatible, i.e., it has the user-level socket library and does support UDP. -- Tom Gillespie ( ...ihnp4!picuxa!tgr) | (attmail!tgillespie) (201) 952-1178 AT&T/EDS Product Integration Center 299 Jefferson Rd. Parsippany NJ 07054 "Don't take life so serious ... it ain't nohow permanent." -- Walt Kelly
rkh@mtune.ATT.COM (964[jak]-Robert Halloran) (04/07/88)
In article <1919@rtech.UUCP> daveb@rtech.UUCP (Crack? No thanks, I've got a new CD player) writes: >But, curiously, the only interface TWG supports for their SV.3 tcp/ip is >a socket emulation library. Perhaps I haven't looked far enough, or the >manuals are poorly indexed, but I did not see a STREAMS/TLI interface >documented. > >Is Cray using a TLI interface for tcp/ip? How about any of the UNIX/386 >ports? Is there a standard TLI interface for tcp/ip, or is everyone >rolling their own in incompatible ways? Since V.3 has a standard TLI package, TWG uses it rather than the home-rolled version they had in the pre-Streams release for the 3B2. The '386 port on the AT&T box, at least, has a 3N section in the programmers' reference for the TLI calls, and I seem to recall seeing TLI mentioned in the networking extensions for Microport's 386 port as well, though I have no idea about the conformance of calls. I would expect (hope?) that the various TLI implementations are using the SVr3 syntax (I don't know if this stuff has been incorporated as yet into the SVID). Bob Halloran ========================================================================= UUCP: {ATT-ACC, rutgers}!mtune!rkh Internet: rkh@mtune.ATT.COM Disclaimer: People have opinions. Companies have policies. Don't confuse MY opinions with THEIR policies. Quote: "There were incidents & accidents, there were hints & allegations" -- Paul Simon =========================================================================
shor@sphinx.uchicago.edu (Melinda Shore) (04/08/88)
In article <1919@rtech.UUCP> daveb@rtech.UUCP (Crack? No thanks, I've got a new CD player) writes: >In <Apr.4.10.54.29.1988.6558@topaz.rutgers.edu> Ron Natalie writes: >>Some System V implementations use Wollongong code (such as it was), this >>includes most of AT&T's so called computers. The CRAY UNICOS (mostly >>system V) used Wollongong as well, but they got over it and have redone >>it themselves. ... >Is Cray using a TLI interface for tcp/ip? Nope: Cray is still distributing the Wollongong mess with Unicos. My understanding is that their new tcp/ip will be derived from BSD4.3 code, and that they have no plans to make streams available any time in the forseeable future. At the same time, I've heard that the Bell Labs XMP is running streams and that they've implemented a streams-based tcp/ip. -- Melinda Shore ..!uunet!reason.psc.edu!shore Pittsburgh Supercomputing Center shore@reason.psc.edu