home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.misc
- Path: sparky!uunet!snorkelwacker.mit.edu!linus!linus.mitre.org!virtual.mitre.org!user
- From: pathak@mitre.org (Heeren Pathak)
- Subject: Re: Request for info about kernel pre-emption and priorities
- Message-ID: <pathak-150892161741@virtual.mitre.org>
- Followup-To: comp.os.os2.misc,comp.os.os2.programmer
- Sender: news@linus.mitre.org (News Service)
- Nntp-Posting-Host: virtual.mitre.org
- Organization: Mitre Corporation
- References: <1992Aug13.221616.7894@ttinews.tti.com> <1992Aug14.054245.25657@nuscc.nus.sg> <19920814.005557.382@almaden.ibm.com> <mg.713847158@elan>
- Date: Sat, 15 Aug 1992 20:27:54 GMT
- Lines: 40
-
- In article <mg.713847158@elan>, mg@elan (Michael Golan) wrote:
- >
- > Now what was this nonsense about making the kernel 'completely re-entrant
- > and preemptive'? Did I hear that IBM is redesigning the whole thing, or
- > simply looking into some slow sys-calls to make them give control,
- > voluntarily more often/in more cases (which, as a result, requires more
- > semaphores to be put in to make those calls re-entrant)? Which one is it:
- > 1) A bad rumor (not the post I followed up, BTW)
- > 2) A total new amazing design of a truly preempting kernel
- > 3) An improvement of some of the more sluggish syscalls to put this 4ms figure
- > down to 1ms or so?
-
- I believe you are referring to the symmetric multi-processing version of
- OS/2. IBM has stated that the next major version of OS/2 (3.0) will
- support symmetric multi-processing.
-
- I don't fully understand the changes needed, but an OS that supports
- symmetric multi-processing needs to be structured differently than a
- "standard" OS kernel. The reason for this is that the kernel may be be
- running on different processors at same time. Thus, the kernel needs to be
- re-entrant. I don't know if the kernel needs to be "completely" re-entrant
- (or if it even possible to write a completely re-entrant kernel) but you do
- want to make the kernel as re-entrant as possible.
-
- As for needing a "completely" preemptive kernel.... I really can't see any
- reason why this is necessary. Maybe someone with a bit more OS experience
- can explain this.
-
- BTW: I believe that IBM is looking into using the Mach micro-kernel for
- the next version. This is the same micro-kernel used in the NeXT computer
- and in OSF/1.
-
-
- -------------------------------------------------------------------------
- Heeren Pathak | Millions long for immortality who do
- pathak@mitre.org | not know what to do with themselves
- Mitre Corporation | on a rainy Sunday afternoon.
- (617) 271-7465 | -- Susan Ertz
- -------------------------------------------------------------------------
- Disclaimer: Mine not Mitre's.
-