Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.mira.net.au!inquo!nntp.uio.no!news.apfel.de!news.maxwell.syr.edu!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!panix!news.panix.com!not-for-mail From: tls@panix.com (Thor Lancelot Simon) Newsgroups: comp.unix.bsd.netbsd.misc,comp.security.unix Subject: Re: Careless integration of others' code (WAS Re: Why no addusr?) Date: 16 Feb 1997 02:48:57 -0500 Organization: Panix Lines: 41 Message-ID: <5e6e59$nmq@panix2.panix.com> References: <none-ya023480001912962244220001@news.infi.net> <DERAADT.97Feb15155022@zeus.pacifier.com> <5e5vkb$d89@panix2.panix.com> <DERAADT.97Feb15212032@zeus.pacifier.com> Reply-To: tls@rek.tjls.com NNTP-Posting-Host: panix2.panix.com Xref: euryale.cc.adfa.oz.au comp.unix.bsd.netbsd.misc:5389 comp.security.unix:31777 In article <DERAADT.97Feb15212032@zeus.pacifier.com>, Theo de Raadt <deraadt@theos.com> wrote: >In article <5e5vkb$d89@panix2.panix.com> tls@panix.com (Thor Lancelot Simon) writes: > > You hold OpenBSD up as a paragon of security, and yet integrate critical > pieces of code such as locore.s without even reading them?! > >Really, security has very little to do with a kernel that cold-resets >the machine at boot. The code wasn't commited until it worked. That >took a while, of course. You are not telling the truth. Perhaps you should begin doing so, before you discredit yourself further. That code was in the OpenBSD tree for *eight weeks*, evidently without anyone ever looking at it, a period of time which included another revision blindly checked in. I just confirmed that using your AnonCVS server. I mean, my God, I don't know Alpha assembler either, but it's pretty obvious what's likely to happpen if you don't return from that function. >That #ifndef change was designed to cause difficulties. It succeeded. I can't speak to that, since I didn't make that change. It "succeeded", in your terminology, only because OpenBSD developers check things into your source tree -- and leave them there for eight weeks! -- without even looking at them first. I seriously doubt that the person who made that change even expected it to get merged into OpenBSD at all. I mean, the assumption is that you look at code before you merge it. I mean, how could one *miss* that? The only reasonable conclusion is that external code gets merged into OpenBSD without even being examined at all for at least eight weeks. Again: this is _prima facie_ evidence that OpenBSD developers import foreign code into the tree without even looking at it first. You think that has "very little" to do with security? Sleep on that. -- This space not left unintentionally unblank. tls@rek.tjls.com $OpenBSD: locore.s,v 1.5 1996/10/30: Blindly integrating source code, $OpenBSD: locore.s,v 1.7 1997/01/24: so you can lose for 8 weeks. "Sleep tight."