Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266497AbUFQOCz (ORCPT ); Thu, 17 Jun 2004 10:02:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S266499AbUFQOCz (ORCPT ); Thu, 17 Jun 2004 10:02:55 -0400 Received: from mx1.redhat.com ([66.187.233.31]:22716 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S266497AbUFQOCw (ORCPT ); Thu, 17 Jun 2004 10:02:52 -0400 Date: Thu, 17 Jun 2004 10:02:29 -0400 From: Alan Cox To: "Salyzyn, Mark" Cc: Alan Cox , Christoph Hellwig , linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org Subject: Re: PATCH: Further aacraid work Message-ID: <20040617140229.GB10138@devserv.devel.redhat.com> References: <547AF3BD0F3F0B4CBDC379BAC7E4189FD23FF9@otce2k03.adaptec.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <547AF3BD0F3F0B4CBDC379BAC7E4189FD23FF9@otce2k03.adaptec.com> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1385 Lines: 28 On Thu, Jun 17, 2004 at 08:53:36AM -0400, Salyzyn, Mark wrote: > available. Since the scsi layer has a propensity to provide sequentially > decreasing pages (sequentially increasing would permit coalescing of SG > elements) for the SG elements, we find that there is an average SG > element size of 4K. That ought to be a case of flipping the way the kernel hands out pages. I've always wondered why we get them often in reverse order but never sat down and worked it out > more than 4G of memory; however we *are* having troubles specifically > with AMD64 systems with more than 4G of memory in 2.6 kernels (the issue > does not occur on 2.4 kernels). I have yet to investigate why this > specific problem exists. The AMD64 is a little unusual in that it has an IO MMU, so requests for mappings that are physically high memory, not easy to merge, etc can be made to appear in a convenient order lower down in memory. Thus asking to map memory at virtual addresses above 4Gb probably hands back a PCI address around 3.5Gb. That mapping will also vanish (gone forever and the PCI address will show other data instead) when you unmap it. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/