From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from va3outboundpool.messaging.microsoft.com (va3ehsobe005.messaging.microsoft.com [216.32.180.31]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (Client CN "mail.global.frontbridge.com", Issuer "Microsoft Secure Server Authority" (not verified)) by ozlabs.org (Postfix) with ESMTPS id 7B6742C007F for ; Tue, 24 Jul 2012 11:49:31 +1000 (EST) Message-ID: <500DFF21.2060804@freescale.com> Date: Mon, 23 Jul 2012 20:49:21 -0500 From: Scott Wood MIME-Version: 1.0 To: Benjamin Herrenschmidt Subject: Re: [PATCH] powerpc/mm: add ZONE_NORMAL zone for 64 bit kernel References: <1342786906-12634-1-git-send-email-Shaohui.Xie@freescale.com> <1343023569.2957.19.camel@pasglop> <500D7921.6060804@freescale.com> <1343082030.2957.38.camel@pasglop> In-Reply-To: <1343082030.2957.38.camel@pasglop> Content-Type: text/plain; charset="UTF-8" Cc: Mingkai Hu , linuxppc-dev@lists.ozlabs.org, Shaohui Xie , Chen Yuanquan List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 07/23/2012 05:20 PM, Benjamin Herrenschmidt wrote: > On Mon, 2012-07-23 at 11:17 -0500, Scott Wood wrote: >>> This is wrong. >> >> How so? >> >>> Don't you have an iommu do deal with those devices anyway ? >> >> Yes, but we don't yet have DMA API support for it, it would lower >> performance because we'd have to use a lot of subwindows which are >> poorly cached (and even then we wouldn't be able to map more than 256 >> pages at once on a given device), and the IOMMU may not be available at >> all if we're being virtualized. > > Ugh ? You mean some designers need to be fired urgently and wasted > everybody's time implementing an unusable iommu ? Nice one ... Yeah, that's old news. It's somewhat usable for specific purposes, using very large pages, but not for arbitrary use. >>> But even then, I'm dubious this is really needed. >> >> We'd like our drivers to stop crashing with more than 4GiB of RAM on 64-bit. > > Fix your HW :-) I wish... Some hardware people may solicit input from us every now and again, but ultimately they do what they want. Returning addresses in excess of a device's declared DMA mask is something that needs fixing too, though. -Scott