Newsgroups: comp.realtime,comp.os.qnx,comp.os.ms-windows.advocacy,comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!cs.utexas.edu!utnut!nott!cunews!revcan!quantum!danh
From: danh@qnx.com (Dan Hildebrand)
Subject: Re: Real-time systems: Windows-NT or QNX
Message-ID: <cpz0rj@qnx.com>
Date: Mon, 17 Oct 94 17:04:04 GMT
Organization: QNX Software Systems
References: <FriOct14102309EST1994@eric> <37kql1$e4d@vixen.cso.uiuc.edu> <37ltcm$ged@sol.sun.csd.unb.ca> <37msqd$5ds@vixen.cso.uiuc.edu>
Lines: 18
Xref: glinda.oz.cs.cmu.edu comp.realtime:7195 comp.os.qnx:2251 comp.os.ms-windows.advocacy:40669 comp.robotics:14463

In article <37msqd$5ds@vixen.cso.uiuc.edu>,
Todd Walk <walk@mrcnext.cso.uiuc.edu> wrote:
>
>Sorry Chris, but NT doesn't have real-time scheduling either (despite
>what the priority settings say).  Neither OS/2 or NT is well suited
>for something like this, but OS/2 isn't any worse at it than NT
>(actually it's a little better in the fact that there are many
>more real world examples to look at).

I heard a rumor that NT didn't pre-empt the execution of a process on the 
occurance of the hardware interrupt a higher priority process was waiting 
for, but rather, waits for the next timer tick.  This resulting in a 
pre-emption scheduling latency one timer tick.  Can anyone confirm or deny 
this rumor?
-- 
Dan Hildebrand      danh@qnx.com         QNX Software Systems, Ltd.   
phone: (613) 591-0931 x204 (voice)       175 Terence Matthews          
       (613) 591-3579      (fax)         Kanata, Ontario, Canada K2M 1W8
