From owner-freebsd-hackers Wed Feb 14 07:01:24 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id HAA13773 for hackers-outgoing; Wed, 14 Feb 1996 07:01:24 -0800 (PST) Received: from terra.Sarnoff.COM (terra.sarnoff.com [130.33.11.203]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA13755 for ; Wed, 14 Feb 1996 07:01:15 -0800 (PST) Received: (from rminnich@localhost) by terra.Sarnoff.COM (8.6.12/8.6.12) id JAA10143; Wed, 14 Feb 1996 09:58:17 -0500 Date: Wed, 14 Feb 1996 09:58:16 -0500 (EST) From: "Ron G. Minnich" To: hackers@FreeBSD.org Subject: Re: VM question In-Reply-To: <3121D613.C680.001@pmail.tepco.co.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org Precedence: bulk > Actually, I'd like to implement a sort of DSM that provides > page-in/out over the network transparently to the user process. well, having done some of these myself I can tell you: the freebsd signal handling code is not quite there and so won't tell you what you need to know when a sigsegv/sigbus happens, to wit: you can't tell what type of fault occurred (read or write). It is now giving you the faulting virtual address, which is an improvement. also, the vm does not quite work right in terms of communicating needed information to the lower-level layers so that information is not available there also. For a view of how to do this right, check out the sunos vm (just look in the include files for protos of getpage/putpage etc.). No bsd unix has done this right for DSM, and neither has linux, which is a shame. IF you want to write a DSM to learn how to do it, go for it. If you just want one you have a couple choices: 1) basic DSM with no coherency support: use NFS shared mapped files over the net 2) Add simple coherency support: many options: 1) use my MNFS, except due to aforementioned VM/VFS problems it does not work completely 2) add a tcp connection to support coherency traffic ron ron