From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH 2/4] pci init: fail qemu if devfn exceeding the max function number supported on bus Date: Tue, 7 Sep 2010 20:19:24 +0300 Message-ID: <20100907171924.GC18374@redhat.com> References: <20100824132525.GA19469@kt> <20100824132710.GB19469@kt> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: yamahata@valinux.co.jp, avi@redhat.com, kvm@vger.kernel.org, qemu-devel@nongnu.org To: Ken CC Return-path: Received: from mx1.redhat.com ([209.132.183.28]:1025 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932142Ab0IGRZ0 (ORCPT ); Tue, 7 Sep 2010 13:25:26 -0400 Content-Disposition: inline In-Reply-To: <20100824132710.GB19469@kt> Sender: kvm-owner@vger.kernel.org List-ID: On Tue, Aug 24, 2010 at 09:27:10PM +0800, Ken CC wrote: > Check if devfn < PCIBUS_MAX_DEVICES * PCI_FUNCTIONS_PER_DEVICE > > Signed-off-by: Ken CC > --- > hw/pci.c | 1 + > 1 files changed, 1 insertions(+), 0 deletions(-) > > diff --git a/hw/pci.c b/hw/pci.c > index 9234fe3..fc4becd 100644 > --- a/hw/pci.c > +++ b/hw/pci.c > @@ -747,6 +747,7 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus, > PCIConfigWriteFunc *config_write, > bool is_bridge) > { > + assert(devfn / PCI_FUNCTIONS_PER_DEVICE < PCIBUS_MAX_DEVICES); I guess it'll happen to work even for < 0, but it might be obvious to only do this in >= 0 case. Just remove the 'else' and stick the assert there. > if (devfn < 0) { > for(devfn = bus->devfn_min ; devfn < ARRAY_SIZE(bus->devices); > devfn += PCI_FUNC_MAX) {