From owner-freebsd-current@FreeBSD.ORG Mon Nov 21 17:05:57 2005 Return-Path: X-Original-To: current@freebsd.org Delivered-To: freebsd-current@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7D38416A41F for ; Mon, 21 Nov 2005 17:05:57 +0000 (GMT) (envelope-from freebsd@deadcafe.de) Received: from deadcafe.de (deadcafe.de [81.169.162.144]) by mx1.FreeBSD.org (Postfix) with ESMTP id CC45443D49 for ; Mon, 21 Nov 2005 17:05:56 +0000 (GMT) (envelope-from freebsd@deadcafe.de) Received: from dialin.t-online.de (p54A5DBF0.dip.t-dialin.net [84.165.219.240]) by deadcafe.de (8.13.4+Sun/8.13.4/Rock) with ESMTP id jALH5mAo023958 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Nov 2005 18:05:52 +0100 (CET) Received: from [172.23.7.254] (doom.rock.net [172.23.7.254]) by dialin.t-online.de (8.13.4+Sun/8.13.4/Rock) with ESMTP id jALH5ed0014869; Mon, 21 Nov 2005 18:05:40 +0100 (CET) Message-ID: <4381FE62.5040208@deadcafe.de> Date: Mon, 21 Nov 2005 18:05:38 +0100 From: Daniel Rock User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923) X-Accept-Language: de-DE, de, en-us, en MIME-Version: 1.0 To: "Chad Leigh -- Shire.Net LLC" References: <20051120183309.ysgk624asksckwow@mail.bafirst.com> <20051121004749.GA39061@xor.obsecurity.org> <43812AC7.3020506@rogers.com> <20051121020816.GA19252@xor.obsecurity.org> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=0.1 required=5.5 tests=FORGED_RCVD_HELO autolearn=disabled version=3.0.4 X-Spam-Checker-Version: SpamAssassin 3.0.4 (2005-06-05) on deadcafe.de X-Mailman-Approved-At: Tue, 22 Nov 2005 03:00:19 +0000 Cc: current@freebsd.org, Kris Kennaway Subject: Re: avail memory is short by 1G on my FreeBSD 6.0-STABLE MP Dell X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Nov 2005 17:05:57 -0000 Chad Leigh -- Shire.Net LLC schrieb: > I assume you are talking about the "phenomenon" of RAM "missing" due > to PCI and other I/O mapping into the 4GB address space? What struck > me about the OP is that the amount of RAM "missing" is more than I have > ever seen due to this. My systems are usually 3.4-3.6GB of RAM with > 4GB installed, not 2.86GB (3005MB)... So are we sure that the PCI > space mapping is the problem? This is nothing unusual. I have a machine on which over 1.25GB is mapped "out of the way". It is a Tyan Thunder K8WE (S2895). In BIOS I can choose between Hardware/Software/Disabled mapping: Hardware (requires E-stepping CPUs): real memory = 5637144576 (5376 MB) Physical memory chunk(s): 0x0000000000001000 - 0x0000000000090fff, 589824 bytes (144 pages) 0x0000000000985000 - 0x00000000a616ffff, 2776543232 bytes (677867 pages) 0x0000000100000000 - 0x000000014ffdffff, 1342046208 bytes (327648 pages) avail memory = 4109500416 (3919 MB) Software (should work regardless of stepping): real memory = 6442450944 (6144 MB) Physical memory chunk(s): 0x0000000000001000 - 0x0000000000090fff, 589824 bytes (144 pages) 0x0000000000985000 - 0x000000007ff1ffff, 2136584192 bytes (521627 pages) 0x0000000100000000 - 0x0000000174baffff, 1958412288 bytes (478128 pages) avail memory = 4085907456 (3896 MB) Disabled: real memory = 2951872512 (2815 MB) Physical memory chunk(s): 0x0000000000001000 - 0x0000000000090fff, 589824 bytes (144 pages) 0x0000000000983000 - 0x00000000aac56fff, 2855092224 bytes (697044 pages) avail memory = 2846003200 (2714 MB) Daniel