From owner-freebsd-current Wed Aug 28 02:24:51 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA14648 for current-outgoing; Wed, 28 Aug 1996 02:24:51 -0700 (PDT) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id CAA14641 for ; Wed, 28 Aug 1996 02:24:47 -0700 (PDT) Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.12/8.6.9) id SAA14146; Wed, 28 Aug 1996 18:49:28 +0930 From: Michael Smith Message-Id: <199608280919.SAA14146@genesis.atrad.adelaide.edu.au> Subject: Re: FreeBSD malloc.c, -lmalloc, and squid. To: dg@root.com Date: Wed, 28 Aug 1996 18:49:27 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, stesin@gu.kiev.ua, angio@aros.net, squid-users@nlanr.net, current@freebsd.org In-Reply-To: <199608280736.AAA18958@root.com> from "David Greenman" at Aug 28, 96 00:36:36 am MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk David Greenman stands accused of saying: > > > I > >seem to recall that you could create logicals on a system-wide basis as > >well as per-session (or was that per-user?) > > Early versions of VMS had a simple {process, job, group, system} hierarchy. > Later versions of VMS created a much more generic system that allowed for > the creation of arbitrary logical name tables that allowed for custimized > access/use. The old hierarchy was preserved using the new mechanism, however. > Logical name tables have access permissions (system, owner, group, world) > and individual entries have various flags which affect how the logical name > is translated. What's the search order on logicals, or are the seperate spaces completely isolated? I have to look at phk's stuff I guess; for now doscmd gets most of my time, but this is a good one to think about while I'm drooling out documentation. Ramble ramble: - there are three sets of namespaces; the kernel namespace, process-group namespaces and process namespaces. - within at least the kernel namespace, seperate subspaces can be allocated (eg. for tunables, drivers, etc...) - init gets a default namespace (read from disk, perhaps?), which is the master inherited by other process groups, and thus other processes. This is perhaps bad; maybe when a process becomes a group leader it should inherit a default process-group namespace previously loaded into the kernel. What should happen to its process namespace? - a process' visible namespace is the result of overlapping its own namespace, that of its process group, and the kernel, with logicals capable of being marked such that they can't be occluded by the lower layers. (ie. the search order is process, group, kernel (or some kernel subspace?)). - logicals are byte arrays of arb. length (limits?), with textual names. Syntax? Nesting limits? Any thoughts, "I/we/they did it like this.."'s, would be educational... > David Greenman -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] Collector of old Unix hardware. "Where are your PEZ?" The Tick [[