Xref: utzoo comp.lang.c:11529 comp.arch:5735
Path: utzoo!utgpu!attcan!uunet!lll-winken!lll-tis!ames!amdahl!nsc!voder!apple!baum
From: baum@Apple.COM (Allen J. Baum)
Newsgroups: comp.lang.c,comp.arch
Subject: Re: Self-modifying code
Message-ID: <14795@apple.Apple.COM>
Date: 27 Jul 88 17:50:12 GMT
References: <5262@june.cs.washington.edu> <260@thor.wright.EDU> <479@m3.mfci.UUCP> <61459@sun.uucp>
Reply-To: baum@apple.UUCP (Allen Baum)
Organization: Apple Computer, Inc.
Lines: 18

[]
>In article <61459@sun.uucp> guy@gorodish.Sun.COM (Guy Harris) writes:
>I don't see that, for example, doing these checks in microcode, as opposed to
>regular code, is *ipso facto* the only way to do things....
 Presumably, hardware (done in parallel with fetches) is the way to go
>... the questions that capability machines pose are:
>
>	1) How much "safer" are they than "conventional" machines plus
>	   compilers that generate run-time checks?
>
>and
>
>	2) If this extra safety is expensive, is it worth the cost?

In some sense, the cost is very little, probably not much more than your 
standard segmentation scheme. The hardware cost is usually a tag bit, which
is fast, pretty cheap on chip, but possibly expensive in your memory system.

--
{decwrl,hplabs,ihnp4}!nsc!apple!baum		(408)973-3385