From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42189) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZIYov-00070r-6g for qemu-devel@nongnu.org; Fri, 24 Jul 2015 04:59:10 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZIYoq-0000zJ-0K for qemu-devel@nongnu.org; Fri, 24 Jul 2015 04:59:09 -0400 Message-ID: <55B1FE51.1090001@redhat.com> Date: Fri, 24 Jul 2015 10:58:57 +0200 From: Laurent Vivier MIME-Version: 1.0 References: <1437566099-10004-1-git-send-email-lvivier@redhat.com> <1437675858-14070-1-git-send-email-lvivier@redhat.com> <1437684406.7562.62.camel@kernel.crashing.org> <20150724000923-mutt-send-email-mst@redhat.com> <20150723213819.15254.14516@loki> <20150723214902.15254.51200@loki> In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [Qemu-ppc] [PATCH v2] pci: allow 0 address for PCI IO/MEM regions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , Michael Roth Cc: "Michael S. Tsirkin" , Mark Cave-Ayland , QEMU Developers , Blue Swirl , "qemu-ppc@nongnu.org" , David Gibson On 24/07/2015 10:46, Peter Maydell wrote: > On 23 July 2015 at 22:49, Michael Roth wrote: >> I seem to recall Michael suggesting it may have already been >> fixed on x86. I think we had a TODO to figure out all >> the architectures that don't use IO windows and figure out >> if they need a fix as well. > > If we think we've fixed x86, then I think for 2.5 I would > be happy with "allow 0 addresses globally and see if anything > breaks". I'm pretty confident none of the ARM platforms will I agree. It was exactly what I was thinking... > have issues -- they don't do any kind of overlapping of > non-PCI memory regions with the MMIO window(s). So, after a "grep PCI" in defconfigs, we can have issue with powerpc, s390x and sparc64 (I guess it is already fixed for x86_64 and i386 ?). Laurent