From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org (gate.crashing.org [63.228.1.57]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id C9A69DDE1A for ; Tue, 15 May 2007 06:53:21 +1000 (EST) Subject: Re: [RFC/PATCH] powerpc: Rewrite IO allocation & mapping on powerpc64 From: Benjamin Herrenschmidt To: Segher Boessenkool In-Reply-To: <246be0d2277753c0c0b61d0f4131c092@kernel.crashing.org> References: <20070514075415.7999BDDEC8@ozlabs.org> <246be0d2277753c0c0b61d0f4131c092@kernel.crashing.org> Content-Type: text/plain Date: Tue, 15 May 2007 06:53:14 +1000 Message-Id: <1179175994.32247.91.camel@localhost.localdomain> Mime-Version: 1.0 Cc: linuxppc-dev@ozlabs.org, Paul Mackerras , Anton Blanchard List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Mon, 2007-05-14 at 15:50 +0200, Segher Boessenkool wrote: > > - Cleanup and separate tracking of the ISA space in the reserved > > low 64K of IO space. No ISA -> Nothing mapped there. > > > I now always reserve the first 64K of IO space for the bridge with > > the ISA bus on it. > > This is wrong -- there are systems with ISA that > also have PCI devices that require 16-bit legacy > I/O windows (e.g., systems with the AMD8111 chip, > like Maple). > > Or do you mean you reserve the low 64k for the PHB? > That would work in this case, but still not in > general. Not sure if we care though. Yes, if you have an "isa" node, you get a 64k mapping of IO's for your PHB in the first 64K of IO space (in addition to a full mapping elsewhere for devices with proper BARs). If you don't, you don't get that mapping. I might add a call for platforms to explicitely map those low 64K to a given PCI host though for platforms who really want it. Ben.