From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33473) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WXADu-000113-H9 for qemu-devel@nongnu.org; Mon, 07 Apr 2014 10:08:36 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WXADo-0005is-BB for qemu-devel@nongnu.org; Mon, 07 Apr 2014 10:08:30 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51072) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WXADo-0005io-3B for qemu-devel@nongnu.org; Mon, 07 Apr 2014 10:08:24 -0400 Date: Mon, 7 Apr 2014 17:09:01 +0300 From: "Michael S. Tsirkin" Message-ID: <20140407140901.GB16907@redhat.com> References: <1396868342-12005-1-git-send-email-marcel.a@redhat.com> <20140407121508.GD16369@redhat.com> <1396874646.5001.76.camel@nilsson.home.kraxel.org> <20140407133415.GB16541@redhat.com> <1396878714.10570.18.camel@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1396878714.10570.18.camel@localhost.localdomain> Subject: Re: [Qemu-devel] [SeaBIOS] [PATCH] hw/pci: reserve IO and mem for pci-2-pci bridges with no devices attached List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Marcel Apfelbaum Cc: kevin@koconnor.net, seabios@seabios.org, Gerd Hoffmann , qemu-devel@nongnu.org On Mon, Apr 07, 2014 at 04:51:54PM +0300, Marcel Apfelbaum wrote: > On Mon, 2014-04-07 at 16:34 +0300, Michael S. Tsirkin wrote: > > On Mon, Apr 07, 2014 at 02:44:06PM +0200, Gerd Hoffmann wrote: > > > Hi, > > > > > > > > + u8 shpc_cap = pci_find_capability(s->bus_dev, PCI_CAP_ID_SHPC); > > > > > > > One thing I'd do is maybe check that the relevant memory type is > > > > enabled in the bridge (probably just by writing fff to base and reading > > > > it back). > > > > > > > This will give hypervisors an option to avoid wasting resources: > > > > e.g. it's uncommon for express devices to claim IO. > > > > > > I don't think we'll need that for the SHPC bridge. > > > > Why not? > Because "has shpc" => not an PCIe port. (as far as I know) > Anyway, why have shpc capability but no I/O or mem to support it? > > Thanks, > Marcel > > > I'm referring to this text in the bridge specification: > > > > The I/O Base and I/O Limit registers are optional and define an address > > range that is used > > by the bridge to determine when to forward I/O transactions from one > > interface to the > > other. > > If a bridge does not implement an I/O address range, then both the I/O > > Base and I/O > > Limit registers must be implemented as read-only registers that return > > zero when read. If a > > bridge supports an I/O address range, then these registers must be > > initialized by > > configuration software so default states are not specified. > > > > So we should probe bridge for I/O support before wasting I/O resources on it. > > The spec does not provide a way to detect this, but we can do it like this: > > > > - write value ffffffff to I/O base register > Why write? A simple read would be enough. > It will never be 0(if I/O or mem is required) > because of the "Base Address" > part of the register which represents the address range, right? > > Thanks, > Marcel AFAIK the spec does not list reset value for this register. IIRC QEMU resets both to 0. > > - read back value > > > > value 0 means bridge does not support I/O. > > > > > > A similar trick should work for other optional resources. > > > > > > > For express it indeed makes sense to avoid claiming IO address space. > > > I'd try to find something more automatic though, where you don't need > > > some kind of "disable io for this express port" config option. > > > > Won't same trick as above work? > > > > > For express ports which can only have a single device underneath we can > > > check whenever we have a device and if one is present already don't > > > bother claiming extra resources for hotplug. > > > > > > > > + for (cap = pci_config_readb(pci->bdf, PCI_CAPABILITY_LIST); cap; > > > > > + cap = pci_config_readb(pci->bdf, cap + PCI_CAP_LIST_NEXT)) > > > > > + if (pci_config_readb(pci->bdf, cap + PCI_CAP_LIST_ID) == cap_id) > > > > > + return cap; > > > > > > > > I would also limit this to 256 iterations, to make sure > > > > we dont' get into an infinite loop with a broken device. > > > > > > Good point. > > > > > > cheers, > > > Gerd > > > > >