Re: [libvirt] [PATCH 2/7] prevent hot unplugging multi function PCI device

Hi I applied these pacches to libvirt tree 8077d64f964705c1034555abeea38773532b762f And built a qemu-kvm from the upstream qemu-kvm tree version 0.14. I tested these new binaries on my Redhat Enterprise Linux 6.1 GA. Now I can have multiple function devices after I changed the VM XML config file manually, please see the 'lspci' output below. Also, I can hotplug one device, but it seems that unplugging it failed. ------------------------------------------------------------------------------------- [root@kvm-rhel-6 qemu]# virsh qemu-monitor-command kvm-rhel-6.1-multifunction-1 --hmp 'device_add lsi,id=scsi10,bus=pci.0,addr=0x7.0x00' VM: **In the VM, 'lspci" can see one line added: VM: **00:07.0 SCSI storage controller: LSI logic / Symbios Logic 5sc895a [root@kvm-rhel-6 qemu]# virsh qemu-monitor-command kvm-rhel-6.1-multifunction-1 --hmp 'device_del lsi,id=scsi10,bus=pci.0,addr=0x7.0x00'

Device 'lsi,id=scsi10,bus=pci.0,addr=0x7.0x00' not found --------------------------------------------------------------------------------------------------------------- #lspci 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] (rev 02) 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 00:01.1 IDE interface: Intel Coporation 82371SB PIIX3 IDE [Natoma/Triton II] 00:01.2 USB Controller: Intel Corporation 82371SB PIIX3 USB Natoma/Triton II] 00:01.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 03) 00:02.0 VGA compatible controller: Cirrus Logic GD 5446 00:03.0 Ethernet controller: Red Hat, Inc Virtio network device ---> function 0 00:03.1 Ethernet controller: Red Hat, Inc Virtio network device 00:03.2 Ethernet controller: Red Hat, Inc Virtio network device 00:03.3 Ethernet controller: Red Hat, Inc Virtio network device 00:03.4 Ethernet controller: Red Hat, Inc Virtio network device 00:03.5 Ethernet controller: Red Hat, Inc Virtio network device 00:03.6 Ethernet controller: Red Hat, Inc Virtio network device 00:03.7 Ethernet controller: Red Hat, Inc Virtio network device ---> function 7 00:04.0 Audio device: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) High 00:05.0 SCSI storage controller: Red Hat, Inc Virtio bloc device 00:06.0 RAM memory: Red Hat, Inc Virtio memory balloon Daniel P. Berrange: > On Fri, May 27, 2011 at 06:20:10PM +0800, Wen Congyang wrote: >> We do not support to hot unplug multi function PCI device now. If the >> device is >> one function of multi function PCI device, we shoul not allow to hot >> unplugg >> it. >> --- >> src/qemu/qemu_hotplug.c | 52 >> +++++++++++++++++++++++++++++++++++++++++++++++ >> 1 files changed, 52 insertions(+), 0 deletions(-) >> >> diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c >> index 3cf7d35..e98c677 100644 >> --- a/src/qemu/qemu_hotplug.c >> +++ b/src/qemu/qemu_hotplug.c >> @@ -1100,6 +1100,30 @@ static inline int qemuFindDisk(virDomainDefPtr >> def, const char *dst) >> return -1; >> } >> >> +static int qemuComparePCIDevice(virDomainDefPtr def ATTRIBUTE_UNUSED, >> + virDomainDeviceInfoPtr dev1, >> + void *opaque) >> +{ >> + virDomainDeviceInfoPtr dev2 = opaque; >> + >> + if (dev1->type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI || >> + dev2->type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) >> + return 0; >> + >> + if (dev1->addr.pci.slot == dev2->addr.pci.slot&& >> + dev1->addr.pci.function != dev2->addr.pci.function) >> + return -1; >> + return 0; >> +} >> + >> +static bool qemuIsMultiFunctionDevice(virDomainDefPtr def, >> + virDomainDeviceInfoPtr dev) >> +{ >> + if (virDomainDeviceInfoIterate(def, qemuComparePCIDevice, dev)< 0) >> + return true; >> + return false; >> +} >> + >> >> int qemuDomainDetachPciDiskDevice(struct qemud_driver *driver, >> virDomainObjPtr vm, >> @@ -1121,6 +1145,13 @@ int qemuDomainDetachPciDiskDevice(struct >> qemud_driver *driver, >> >> detach = vm->def->disks[i]; >> >> + if (qemuIsMultiFunctionDevice(vm->def,&detach->info)) { >> + qemuReportError(VIR_ERR_OPERATION_FAILED, >> + _("cannot hot unplug multifunction PCI >> device: %s"), >> + dev->data.disk->dst); >> + goto cleanup; >> + } >> + >> if (qemuCgroupControllerActive(driver, >> VIR_CGROUP_CONTROLLER_DEVICES)) { >> if (virCgroupForDomain(driver->cgroup, >> vm->def->name,&cgroup, 0) != 0) { >> qemuReportError(VIR_ERR_INTERNAL_ERROR, >> @@ -1351,6 +1382,13 @@ int qemuDomainDetachPciControllerDevice(struct >> qemud_driver *driver, >> goto cleanup; >> } >> >> + if (qemuIsMultiFunctionDevice(vm->def,&detach->info)) { >> + qemuReportError(VIR_ERR_OPERATION_FAILED, >> + _("cannot hot unplug multifunction PCI >> device: %s"), >> + dev->data.disk->dst); >> + goto cleanup; >> + } >> + >> if (qemuDomainControllerIsBusy(vm, detach)) { >> qemuReportError(VIR_ERR_OPERATION_FAILED, "%s", >> _("device cannot be detached: device is >> busy")); >> @@ -1438,6 +1476,13 @@ int qemuDomainDetachNetDevice(struct >> qemud_driver *driver, >> goto cleanup; >> } >> >> + if (qemuIsMultiFunctionDevice(vm->def,&detach->info)) { >> + qemuReportError(VIR_ERR_OPERATION_FAILED, >> + _("cannot hot unplug multifunction PCI >> device :%s"), >> + dev->data.disk->dst); >> + goto cleanup; >> + } >> + >> if ((vlan = qemuDomainNetVLAN(detach))< 0) { >> qemuReportError(VIR_ERR_OPERATION_FAILED, >> "%s", _("unable to determine original VLAN")); >> @@ -1567,6 +1612,13 @@ int qemuDomainDetachHostPciDevice(struct >> qemud_driver *driver, >> return -1; >> } >> >> + if (qemuIsMultiFunctionDevice(vm->def,&detach->info)) { >> + qemuReportError(VIR_ERR_OPERATION_FAILED, >> + _("cannot hot unplug multifunction PCI >> device: %s"), >> + dev->data.disk->dst); >> + return -1; >> + } >> + >> if (!virDomainDeviceAddressIsValid(&detach->info, >> >> VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI)) { >> qemuReportError(VIR_ERR_OPERATION_FAILED, > ACK > > Daniel