home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!shravani!rajat
- From: rajat@watson.ibm.com (Rajat Datta)
- Subject: Re: GNU kids on the block? (sorry... couldn't resist)
- Sender: news@watson.ibm.com (NNTP News Poster)
- Message-ID: <1992Aug31.180438.34997@watson.ibm.com>
- Date: Mon, 31 Aug 1992 18:04:38 GMT
- Disclaimer: This posting represents the poster's views, not necessarily those of IBM
- References: <1992Aug27.135703.9312@crd.ge.com> <17m6sbINN64o@darkstar.UCSC.EDU> <1992Aug29.115034.17334@hacker.UUCP>
- Nntp-Posting-Host: shravani.watson.ibm.com
- Organization: IBM T.J. Watson Research Center
- Keywords: Linux, Mach, CISC, RISC
- Lines: 31
-
- In article <1992Aug29.115034.17334@hacker.UUCP> steve@hacker.UUCP (Stephen M. Youndt) writes:
- >kernel as opposed to a "micro" kernel. It just seems that the micro kernel
- >people are the only ones doing it (for Unix anyway; VMS can hardly be
- >considered a micro kernel). Well, that's my $.02 worth.
- >
-
-
- Actually, AIX has loadable device-drivers and loadable kernel
- extensions. And AIX is certainly not anything like a micro-kernel.
-
- I think you're absolutely right about loadable device-drivers, and in
- fact, they make more sense on the PC class machines than on ones that
- run VMS. There is a far greater variety of devices available for PCs
- and the resources available (memory) is usually much less. A fat
- kernel with every likely device driver can be tolerated better on
- systems with 64Mbyte plus systems than on our "poor" machines with
- 8Mbytes and often less.
-
- I've spent the last few years of my life working on micro-kernels,
- Mach in particular, and came to the conclusion that there's no magic
- here. It's like the C++ vs. C debate. There's a lot of stuff that is
- elegant to do in C++, but you can also approximate them in C and maybe
- the performance tradeoff is serious. As machines get faster, the
- performance tradeoff issue becomes less important and the elegance
- becomes much more important. Perhaps similar issues hold true with
- micro-kernels vs. the traditional approach.
-
- Basically, there's no magic way to get neatness, and neatness is
- possible with the traditional macro-kernel.
- --
- rajat (rajat@watson.ibm.com)
-