Date: Fri, 14 May 2004 16:14:33 -0500 From: Mark Linimon <linimon@lonesome.com> To: freebsd-current@freebsd.org Cc: Kris Kennaway <kris@obsecurity.org> Subject: Re: but withoutI WITNESS and INVARIANTS ? Message-ID: <200405141614.33302.linimon@lonesome.com> In-Reply-To: <20040514210157.GA96254@xor.obsecurity.org> References: <40A3F1D5.5010209@bellsouth.net> <40A4CCB1.8010700@cronyx.ru> <20040514210157.GA96254@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 14 May 2004 04:01 pm, Kris Kennaway wrote: > On Fri, May 14, 2004 at 05:42:09PM +0400, Roman Kurakin wrote: > > The possibility that by disabling this debug stuff you will get some > > instability is > > very very .... very low :-) > > Not so low..it changes the timings of operations, which can expose > race conditions. I'll echo what Kris said. The difference between "should have no effect" and "can be proven to have no effect" is basically what Software Engineering (capitals intended) is all about. And after doing it for decades, lemme tell you, it's much more fun to just fool around and code than it is to do all that stuff. Especially given the, oh, say, 1:10 ratio of the one to the other ... mcl
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200405141614.33302.linimon>