Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!cs.utexas.edu!sun-barr!apple!usc!randvax!edhall@rand.org From: edhall@rand.org Newsgroups: comp.os.minix Subject: Re: SLIP and MINIX? (Was Re: Bruce Evan's Opus) Summary: Well, at least I'm not confused... Keywords: SLIP, MINIX, KA9Q Message-ID: <2110@randvax.UUCP> Date: 23 Jun 89 19:27:38 GMT References: <16698@louie.udel.EDU> <336@blenheim.nsc.com> <7139@xyzzy.UUCP> <2107@randvax.UUCP> <105@altos86.Altos.COM> Sender: edhall@randvax.UUCP Reply-To: edhall@rand.org (Ed Hall) Followup-To: comp.os.minix Organization: The RAND Corporation, Santa Monica, CA 90406 Lines: 67 In article <105@altos86.Altos.COM> dtynan@altos86.Altos.COM (Dermot Tynan) writes: >In article <2107@randvax.UUCP>, edhall@randvax.UUCP (Ed Hall) writes: >> > Any chance of that making it to the net??? >> > - Dave Rivers - >> >> However, I am willing to send anyone the NET program (sources and >> well). Just send me a pre-packaged, postage-paid 1.2MB floppy, and I'll >> send you just what I sent Desmond Young. Note that this version does >> NOT support anything other than SLIP, and that its support of FTP, >> TELNET and SMTP is pretty basic. That said, its TCP/IP implementation >> seems quite robust; I've used it for several megabytes worth of FTP >> transfers and numerous TELNET sessions without a hitch. >> >> -Ed Hall > >Ooops! I think you're somewhat confused, Ed. The original offer was from >Des. That is what they were asking for. Des took the KA9Q code and ported >it to Minix. He got slip working months ago. His latest version supports >Ethernet on a National Semiconductor chipset (the 3-Com board, I believe). >Des is out of town at the moment, so don't expect any replies from him. I >don't know what this 'NET' program you refer to is, but Des' implementation >is very solid, and very complete. He has added a lot of the standard >Berkeley commands to FTP, etc (such as mkdir, chdir, etc). He spent a *lot* >of time fixing things in the Minix kernel, so that it would do what it is >supposed to do. He has been running it on his own machines for a while now, >and by all accounts, it is pretty good. Of course, the final word is his. >If you want to contact him, about distribution or whatever, try: > des@logic.nsc.com or des@musashi.UUCP > >Either should work. I think he said he'd be back next week... > - Der We are talking about variants of the same program, here. Des did not say whether his version of NET (yup, that's what it's called) still supports SLIP (though it may well); much of his work had to do with getting it to function with the ethernet interface, and I suspect that much of the modification he had to do to MINIX was related to this as well. My version runs SLIP--and only SLIP--with minimal (~25 lines) modification to MINIX; I've added most of the features you mention (chdir, etc.). Some of us are minimalists when it comes to straying from the ``published'' version of MINIX. This was my approach. Des did offer, in his original posting, to send his version on a 1.2MB floppy. I assume that this offer still stands. His version is no doubt solid--probably more solid than mine, though I've yet to have any real problems--but it is hardly ``complete'' by his own admission (you DID read his posting, didn't you?). The current version of Phil Karn's NET program is about twice as big as the largest program that fits under MINIX (that durn 64K limit). Just so no one gets confused: Des has added ethernet support and a moderate amount of kernel reworking for his version of NET. I did just the minimum necessary to get SLIP going under MINIX, and improved FTP and SMTP support a bit (though probably not as much as Des). Neither Des' nor mine is near to being a complete version of NET. In any case, a poster asked for my version, so I offered it. Nine people have requested it. If any of those people would rather have Des' version, just let me--and Des--know. -Ed Hall edhall@rand.org ..!uunet!edhall@rand.org ..!decvax!randvax!edhall