Return to BSD News archive
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!cs.utexas.edu!geraldo.cc.utexas.edu!vax From: vax@no.permanent.address (VaX#n8) Newsgroups: comp.os.386bsd.questions Subject: Re: Networking code stability Date: 13 Oct 1994 01:28:33 GMT Organization: The University of Texas at Austin, Austin, Texas Lines: 17 Message-ID: <37i2g1$8tj@geraldo.cc.utexas.edu> References: <tporczykCxJ40r.1qJ@netcom.com> <MICHAELV.94Oct12002237@MindBender.HeadCandy.com> NNTP-Posting-Host: slip-1-25.ots.utexas.edu X-Newsreader: TIN [version 1.2 PL2] Michael L. VanLoon (michaelv@MindBender.HeadCandy.com) wrote: : Linux networking code isn't *bad* -- it's just a from-scratch rewrite : that still has quite a way to go to be production quality. On the I have heard rumors (i.e. I haven't seen this first-hand!) that the latest version of "slap" (I've heard that's the new name for term?), which uses UDP and TCP packets both, has a tendency to panic Linux systems, due to "less than robust" programming practices. The consensus, right or wrong, is that Linux networking has a ways to go. They say this is so because a normal user process can reliably crash the system. I'm sure you will find several people who think it's just fine, though; for most purposes it is (just "don't move your arm like that" the doctor said :) Then again, why reinvent a wheel with a broken spoke? -- The above address may be incorrect, in case you hadn't noticed. Try vax@ccwf.cc.utexas.edu Victimless crimes are a figment of the state's collective imagination.