From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-15?Q?Andreas_F=E4rber?= Subject: Re: [PATCH 00/14] Remove old_portio users for memory region PIO mapping Date: Mon, 08 Oct 2012 18:18:01 +0200 Message-ID: <5072FCB9.80606@suse.de> References: <1349699033-6703-1-git-send-email-agraf@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <1349699033-6703-1-git-send-email-agraf@suse.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org Sender: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org To: Alexander Graf Cc: Anthony Liguori , xen-devel@lists.xensource.com, Stefano Stabellini , qemu-devel qemu-devel , "qemu-ppc@nongnu.org List" , Avi Kivity , Scott Wood , David Gibson List-Id: xen-devel@lists.xenproject.org Am 08.10.2012 14:23, schrieb Alexander Graf: > When running on PowerPC, we don't have native PIO support. There are a = few hacks > around to enable PIO access on PowerPC nevertheless. >=20 > The most typical one is the isa-mmio device. It takes MMIO requests and= converts > them to PIO requests on the (QEMU internal) PIO bus. >=20 > This however is not how real hardware works and it limits us in the abi= lity to > spawn eventfd's on PIO ports which doesn't work with this approach. In= stead, > let's model it more easily. Let's just map the PIO memory region into M= MIO > space. >=20 > For this to work, we need to get rid of all old_portio struct users, as= they > break with this approach. This is what this patch set does. Looks sensible to me as far as reviewed, but I'm missing a patch 13 that actually rips out old_portio afterwards. :) Andreas > It also converts > the e500 machines and sPAPR to the new memory region model. >=20 >=20 > Alex --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrnbe= rg