Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!pmafire!dave
From: dave@pmafire.inel.gov (Dave Remien)
Newsgroups: comp.unix.sysv386
Subject: Re: BadHerz VALUE
Keywords: BAD HZ VALUE
Message-ID: <1991Feb11.172800.29591@pmafire.inel.gov>
Date: 11 Feb 91 17:28:00 GMT
References: <1991Feb5.091440.23318@informatik.uni-erlangen.de> <6407@spdcc.SPDCC.COM> <1991Feb11.074644.922@ico.isc.com>
Organization: WINCO
Lines: 20

In article <1991Feb11.074644.922@ico.isc.com> rcd@ico.isc.com (Dick Dunn) writes:
>While process scheduling is initiated by the clock() routine, and clock()
>is invoked every 1/HZ sec, scheduling isn't done on 0.01 sec boundaries;
>it's much less frequent.

Process scheduling is done whenever the currently running process either
blocks itself or reaches MAXSLICE clock-ticks, or a realtime process
(V.4) declares itself ready to run. MAXSLICE is usually set to 100
(i.e., process scheduling runs once a second) so that the maximum amount
of CPU time is devoted to user processes, or so says AT&T. 


>-- 
>Dick Dunn     rcd@ico.isc.com -or- ico!rcd       Boulder, CO   (303)449-2870
>   ...But is it art?


-- 
Dave Remien +*+*+*+*+*+*+*+*+*+*+*+*+*+*+ WINCO Computer Engineering Group 
  dave@pmafire.inel.gov or rzd@inel.gov      "Dave Barry for President"