Path: utzoo!attcan!uunet!samsung!sdd.hp.com!usc!apple!rutgers!njin!limonce From: limonce@pilot.njin.net (Tom Limoncelli) Newsgroups: comp.sys.amiga.tech Subject: Re: Semaphores (Why don't we use 'em?) Message-ID: Date: 1 Jul 90 15:54:20 GMT References: <709.268c893f@desire.wright.edu> <31309@cup.portal.com> Distribution: na Organization: Drew University/NJIN Lines: 34 In article <31309@cup.portal.com> Sullivan@cup.portal.com (sullivan - segall) writes: > Someone asks: > > Why don't people use the Amiga semaphore routines more often? > The Amiga tends to use Forbid() and Permit() to serve this purpose. The > good side of this is that you don't have to know what the name of the > semaphore you want to allocate is. (Sorry about the verb, I've been studying I think you missed two other reasons. First of all, the routines have bugs in them and if you don't have the work-around you might will get discouraged and stop using them. Secondly, if you do read the RKM section on semaphores you'll notice that it explains how to use them, but then says something like "Deadlocks can still occur if you're not careful, find a good book on operating systems (like the Minix book) for more info." Most people without a BA or BS in CS are going to read that and start using Forbid() and Permit(). (Soapbox) This must be a bad week for me because I've come across 3 cases where a software development problem "happened" because the programmers didn't have a good base in xxxx theory. (where xxxx = relational calculus (i.e. database), semaphore and disk scheduling) (Joke mode) Well I only have to take 1 CS class in my last year (next year) of college. I'll try to ignore the theory so I can "fit in" when I get to the job market. -Tom -- tlimonce@drew.edu Tom Limoncelli tlimonce@drew.uucp +1 201 408 5389 tlimonce@drew.Bitnet limonce@pilot.njin.net