From owner-freebsd-bugs  Thu Jul 10 02:44:18 1997
Return-Path: <owner-bugs>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id CAA19246
          for bugs-outgoing; Thu, 10 Jul 1997 02:44:18 -0700 (PDT)
Received: from implode.root.com (implode.root.com [198.145.90.17])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA19209
          for <freebsd-bugs@FreeBSD.ORG>; Thu, 10 Jul 1997 02:44:11 -0700 (PDT)
Received: from implode.root.com (localhost [127.0.0.1])
	by implode.root.com (8.8.5/8.8.5) with ESMTP id CAA28234;
	Thu, 10 Jul 1997 02:45:21 -0700 (PDT)
Message-Id: <199707100945.CAA28234@implode.root.com>
To: Poul-Henning Kamp <phk@dk.tfs.com>
cc: freebsd-bugs@FreeBSD.ORG
Subject: Re: kern/4070: cannot access over 64MB memory 
In-reply-to: Your message of "Thu, 10 Jul 1997 08:17:01 +0200."
             <6461.868515421@critter.dk.tfs.com> 
From: David Greenman <dg@root.com>
Reply-To: dg@root.com
Date: Thu, 10 Jul 1997 02:45:21 -0700
Sender: owner-bugs@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>
>>    Malloc is kind of strange when it comes to allocating the virtual address
>> space and will consume up to twice the amount requested.
>
>Well, true, but that is not the case here:
>
>The maximum overallocation for any one call to malloc(bytes) is:
>
>	min(pagesize-1, bytes*2-1)
>
>This is because when "bytes" is > pagesize/2, an integral number of
>pages is allocated, otherwise a 2^n size chunk is allocated.

   Nonetheless, the problem is that he's hitting the process ulimit. As you
know, the failure mode for running out of memory (RAM + swap) is quite
different than the error indicated in this problem report. Perhaps the problem
is caused by VM fragmentation? Or a bug in malloc?

-DG

David Greenman
Core-team/Principal Architect, The FreeBSD Project