Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rice!sun-spots-request
From: rws@cs.brown.edu (Richard W. Sabourin)
Newsgroups: comp.sys.sun
Subject: Question on 4.1 mlock(3)
Keywords: SunOS
Message-ID: <7506@brazos.Rice.edu>
Date: 8 May 90 20:42:58 GMT
Sender: root@rice.edu
Organization: Sun-Spots
Lines: 16
Approved: Sun-Spots@rice.edu
X-Sun-Spots-Digest: Volume 9, Issue 154, message 6

This simple question is aimed at the Sun engineers whom I KNOW read this;
it deals with SunOS 4.1 on a SPARCstation.

Can I (root) use mlock(3) to wire down blocks of plain, ordinary memory
obtained from malloc (actually valloc), or does it have to be mmap'ed?  I
tried a test program, but "ps" indicates (reading the RSS and %MEM fields)
that the pages get aged out, when they aren't touched by the process in
several seconds.

I imagine this is a case of the usual inaccuracy of ps, vmstat, etc., but
I have no other easy way to tell where the pages lie.  Please email, and I
will summarize.

	Thanks for your wisdom,
	Rick Sabourin, Systems Slave
	<rws@cs.brown.edu>