Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!ukma!mailrus!cornell!rochester!udel!mmdf From: limonce%pilot.njin.net@UDEL.EDU Newsgroups: comp.sys.amiga Subject: Re: OS/2 vs. Pournelle Message-ID: <4446@louie.udel.EDU> Date: 3 Oct 88 22:53:49 GMT Sender: mmdf@udel.EDU Lines: 53 Received: from CUNYVM by CUNYVM.BITNET (Mailer X2.00) with BSMTP id 4482; Sun, 02 Oct 88 00:00:55 EDT Received: from UDEL.EDU by CUNYVM.CUNY.EDU (IBM VM SMTP R1.1) with TCP; Sun, 02 Oct 88 00:00:52 EDT Received: from Louie.UDEL.EDU by Louie.udel.EDU id ad03247; 1 Oct 88 22:09 EDT Received: from USENET by Louie.UDEL.EDU id aa03242; 1 Oct 88 22:07 EDT From: Tom Limoncelli Subject: Re: OS/2 vs. Pournelle Message-ID: Date: 2 Oct 88 01:29:37 GMT Organization: NJ InterCampus Network, New Brunswick, N.J. To: amiga-relay@UDEL.EDU Sender: amiga-relay-request@UDEL.EDU In article <800@super.ORG> rminnich@super.ORG (Ronald G Minnich) writes: > [dicussing OS/2 & J.P. and life in general with kurt@tc.fluke.COM ] > [ (Kurt Guntheroth) ... much deleted ] > But how does the message passing work? Is it 'copy the message' or > 'map it into your space'? If it is the latter, where do they get mapped > in? To the same virtual address per process? Or to a different virtual > address? If to a different virtual address, then all the pointers > in the message are invalid. If to the same, how do they guarantee > that there is not something already there? > [much deleted] > ron The book "Inside OS/2" really explains things well. Basically, at every corner and every decision the designers made sure that not only could resources be tracked/recovered but there'd be a way to kill the process without ending up in a deadlock. (The book in general is good too.) Basically, when you allocate shared memory it gives you a virtual address that is unused on other processes unique. In other words, it sets aside a range of virtual addresses that will not be allocated with the same virtual address to any other process. This means that shared memory must be explictly allocated. The one thing about OS/2 that kills me is that all the theory looks like it's going to work. Every design choice is the one that I would have picked (except one small one) if I had to implement an OS on a segmented CPU. Almost makes me want to but a machine with an Intel chip... naaaah. I'll just wait for a MMU and a complete redesign of the AmigaOS to handle resource tracking. :-) Tom -- Tom Limoncelli -- Drew University, Box 1060, Madison, NJ 07940 TLimonce@Drew.Bitnet -- limonce@pilot.njin.net -- VoiceMail (201)408-5389 Drew College of Liberal Arts: male/female ratio: 2:3 student/pc ratio: 1:1 "The opinions expressed are mine... just mine."