From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:35864) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UHGYU-0000ZI-Ii for qemu-devel@nongnu.org; Sun, 17 Mar 2013 12:36:36 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UHGPO-0003KW-Un for qemu-devel@nongnu.org; Sun, 17 Mar 2013 12:29:52 -0400 Received: from mx1.redhat.com ([209.132.183.28]:53896) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UHCY1-000674-NK for qemu-devel@nongnu.org; Sun, 17 Mar 2013 08:18:45 -0400 Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id r2HCIi1D014717 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Sun, 17 Mar 2013 08:18:44 -0400 Date: Sun, 17 Mar 2013 14:19:08 +0200 From: "Michael S. Tsirkin" Message-ID: <20130317121908.GA22279@redhat.com> References: <20130314215150.10566.37912.stgit@bling.home> <20130314220135.10566.6582.stgit@bling.home> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130314220135.10566.6582.stgit@bling.home> Subject: Re: [Qemu-devel] [PATCH 7/7] pcie: Mangle types to match topology List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Williamson Cc: qemu-devel@nongnu.org On Thu, Mar 14, 2013 at 04:01:35PM -0600, Alex Williamson wrote: > Windows will fail to start drivers for devices with an Endpoint type > PCIe capability attached to a Root Complex (code 10 - Device cannot > start). The proper type for such a device is Root Complex Integrated > Endpoint. Devices don't care which they are, so do this conversion > automatically. > > This allows the Windows driver to load for nec-usb-xhci when attached > to pcie.0 of a q35 machine. > > Signed-off-by: Alex Williamson I think it's a bit ugly from the API perspective, in that an integrated endpoint is not converted to a regular one. I think it would be cleaner to have pcie_cap_init do exactly what it's told to do, maybe failing if you give it an incorrect configuration. On top of this add pcie_endpoint_cap_init which sets the type explicitly. Hmm? Not critical, so applied as is for now. > --- > hw/pci/pcie.c | 13 +++++++++++++ > 1 file changed, 13 insertions(+) > > diff --git a/hw/pci/pcie.c b/hw/pci/pcie.c > index 485c94c..bcfbae4 100644 > --- a/hw/pci/pcie.c > +++ b/hw/pci/pcie.c > @@ -48,6 +48,19 @@ int pcie_cap_init(PCIDevice *dev, uint8_t offset, uint8_t type, uint8_t port) > > assert(pci_is_express(dev)); > > + /* > + * Mangle type to convert Endpoints to Root Complex Integrated Endpoints. > + * Windows will report Code 10 (device cannot start) for regular Endpoints > + * on the Root Complex. > + */ > + if (pci_bus_is_express(dev->bus) && pci_bus_is_root(dev->bus)) { > + switch (type) { > + case PCI_EXP_TYPE_ENDPOINT: > + type = PCI_EXP_TYPE_RC_END; > + break; > + } > + } > + > pos = pci_add_capability(dev, PCI_CAP_ID_EXP, offset, > PCI_EXP_VER2_SIZEOF); > if (pos < 0) {