From owner-freebsd-hackers  Wed May 15 10:47:00 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id KAA08362
          for hackers-outgoing; Wed, 15 May 1996 10:47:00 -0700 (PDT)
Received: from brasil.moneng.mei.com (brasil.moneng.mei.com [151.186.109.160])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id KAA08356;
          Wed, 15 May 1996 10:46:55 -0700 (PDT)
Received: (from jgreco@localhost) by brasil.moneng.mei.com (8.7.Beta.1/8.7.Beta.1) id MAA12362; Wed, 15 May 1996 12:46:08 -0500
From: Joe Greco <jgreco@brasil.moneng.mei.com>
Message-Id: <199605151746.MAA12362@brasil.moneng.mei.com>
Subject: Re: A question for the VM gurus..!
To: toor@dyson.iquest.net (John S. Dyson)
Date: Wed, 15 May 1996 12:46:07 -0500 (CDT)
Cc: dyson@freebsd.org, hackers@freebsd.org
In-Reply-To: <199605151714.MAA22723@dyson.iquest.net> from "John S. Dyson" at May 15, 96 12:14:47 pm
X-Mailer: ELM [version 2.4 PL24]
Content-Type: text
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> > Not portable.  Solaris, at least, implements the madvise() stuff..
> 
> Okay..
> 
> > What it seems to do:
> > 
> > If you set MADV_SEQUENTIAL on a region, if it has to fault a page "n" in, it
> > looks like it discards all pages from 0 to n-1 in that region..  notably it
> > does NOT seem to do anything if it doesn't have to fault a page in.
> 
> Is there only one process accessing the region?  What happens if there
> are other processes using that region also?  Otherwise, that is easy to
> do in our VM system.

Well, the situations under which I consider this to be useful might be where
you are moving vast quantities of data in a linear fashion...  so multiple
processes accessing it _probably_ would not be an issue.

As a larger issue, I don't know what the ramifications of mixing madvise()'s
on a region - either in the same process or in different processes - should
be.  Maybe something to ponder for a while...

Then again, maybe wise to remember that an application smart enough to
provide "hints" as to optimal VM usage is probably smart enough to work out
the side effects too.  Maybe the VM implementation _should_ be relatively
simplistic, so that the behavior can be predicted.

... Joe

-------------------------------------------------------------------------------
Joe Greco - Systems Administrator			      jgreco@ns.sol.net
Solaria Public Access UNIX - Milwaukee, WI			   414/546-7968