home *** CD-ROM | disk | FTP | other *** search
- Path: hearst.acc.Virginia.EDU!adastra!mbs
- Newsgroups: comp.sys.amiga.misc
- From: mbs@adastra.cvl.va.us (Michael B. Smith)
- Subject: Re: OS features
- References: <92747544038@PAPA.NORTH.DE> <4b3h9s$1st@alterdial.UU.NET> <2152.6561T63T2136@cycor.ca> <4b7i18$si1@vixen.cso.uiuc.edu> <oj6raxxrr0o.fsf@hpsrk.fc.hp.com> <13213431@sourcery.han.de> <4cp0un$cve@serpens.rhein.de> <4cs4ji$rou@news.mpd.tandem.com> <4csfkl$oco@serpens.rhein.de>
- X-NewsReader: GRn 3.0b24 December 30, 1995
- MIME-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: 8bit
- Message-ID: <mbs.4720@adastra.cvl.va.us>
- Date: Tue, 9 Jan 96 22:18:57 EDT
- Organization: Only if you insist...
-
- In article <4csfkl$oco@serpens.rhein.de> mlelstv@serpens.rhein.de (Michael van Elst) writes:
- > Joseph Crowe <jcrowe> writes:
- > > I'm struggling to understand what you mean by the above paragraph. I don't
- > >find it at all desirable to have a system where an interface library can
- > >arbitrarily corrupt data structures in user task space. Could you relate this
- > >concept to a more concrete example?
-
- > So the design for memory protection doesn't need to protect arbitrary
- > objects. You just have to protect the good guys (well debugged libraries)
- > against the bad guys (arbitrary user code).
-
- "Because there be dragons...."
-
- A fairly well-known mainframe OS has a capability for user-mode callbacks
- to occur from some operating system functions.
-
- It goes through a great deal of effort to protect the OS from the user
- possible faults.
-
- > > With demand paged VM, there's often multiple sizes of allocation units, at
- > >least, once you leave user space.
- >
- > Huh ? The "allocation" unit is a page. And a page is usually of constant
- > size for the whole system.
-
- On Unix, that's typically been true. On mainframe OS'es page is not normally
- a constant.
-
- The same OS I talked about above maintains a page table (never swapped)
- that defines the size of all memory blocks; and whether they are swapped
- out or not (plus quite a bit of other status data). Only the page size
- is swapped (and I _think_, but am not sure, that the limitation is only
- that the page size must be quad-word aligned (in this case, a word is
- six bytes) so you may "waste" 23 bytes but no more).
- --
- // Michael B. Smith
- \X/ mbs@adastra.cvl.va.us
-