From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Evensky Subject: Re: [PATCH] kvm tools: adds a PCI device that exports a host shared segment as a PCI BAR in the guest Date: Sun, 28 Aug 2011 21:55:26 -0700 Message-ID: <20110829045526.GA23759@dancer.ca.sandia.gov> References: <1314248794.32391.60.camel@jaguar> <1314271548.3692.22.camel@lappy> <30669_1314285268_p7PFESZN013126_20110825150806.GF24996@dancer.ca.sandia.gov> <20110825210018.GA27983@dancer.ca.sandia.gov> <4E56BA67.9000901@redhat.com> <20110825220353.GC27983@dancer.ca.sandia.gov> <4E59EF95.5000806@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Sasha Levin , Stefan Hajnoczi , Pekka Enberg , Alexander Graf , kvm@vger.kernel.org, cam@cs.ualberta.ca To: Avi Kivity Return-path: Received: from sentry-two.sandia.gov ([132.175.109.14]:41064 "EHLO sentry-two.sandia.gov" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751040Ab1H2Ezl (ORCPT ); Mon, 29 Aug 2011 00:55:41 -0400 Content-Disposition: inline In-Reply-To: <4E59EF95.5000806@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On Sun, Aug 28, 2011 at 10:34:45AM +0300, Avi Kivity wrote: > On 08/26/2011 01:03 AM, David Evensky wrote: > >I need to specify the physical address because I need to ioremap the > >memory during boot. > > Did you consider pci_ioremap_bar()? No, the code needs a physical memory address, not a PCI device. I suppose we could do something special for PCI devices, but that wasn't our intent. That said, this was also one of those things you learn every day. :-) > >The production issue I think is a memory limitation. We certainly do > >use QEMU a lot; but for this the kvm tool is a better fit. > > > > What is this memory limitation? It isn't a hard and fast number; just trying to maximize the number of guests we can have. > -- > I have a truly marvellous patch that fixes the bug which this > signature is too narrow to contain. >