Re: [libvirt] [PATCHv10 3/6] clean up UpdateDevice for consolidation

At 04/19/2011 03:44 PM, KAMEZAWA Hiroyuki Write: > This patch strips reusable part of qemudDomainUpdateDeviceFlags() > and consolidate it to qemudDomainModifyDeviceFlags(). > No functional changes. > Based on Eric's and Hu's work. > > > Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu(a)jp.fujitsu.com&gt; > > * src/qemu/qemu_driver.c > (qemudDomainUpdateDeviceLive) : core of UpdateDevice, extracted from > UpdateDeviceFlags() > (qemudDomainUpdateDeviceFlags): reworked as a wrapper function of > ModifyDeviceFlags() > --- > src/qemu/qemu_driver.c | 209 +++++++++++++++++++++--------------------------- > 1 files changed, 90 insertions(+), 119 deletions(-) > > diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c > index f33a7f4..2bdf42e 100644 > --- a/src/qemu/qemu_driver.c > +++ b/src/qemu/qemu_driver.c > @@ -3991,6 +3991,74 @@ static int qemudDomainDetachDeviceLive(virDomainObjPtr vm, > return ret; > } > > +static int > +qemudDomainChangeDiskMediaLive(virDomainObjPtr vm, > + virDomainDeviceDefPtr dev, > + struct qemud_driver *driver, > + virBitmapPtr qemuCaps, > + bool force) > +{ > + virDomainDiskDefPtr disk = dev->data.disk; > + virCgroupPtr cgroup = NULL; > + int ret; > + > + if (qemuCgroupControllerActive(driver, VIR_CGROUP_CONTROLLER_DEVICES)) { > + if (virCgroupForDomain(driver->cgroup, > + vm->def->name, &cgroup, 0) !=0 ) { > + qemuReportError(VIR_ERR_INTERNAL_ERROR, > + _("Unable to find cgroup for %s"), > + vm->def->name); > + goto end; > + } > + if (qemuSetupDiskCgroup(driver, vm, cgroup, disk) < 0) > + goto end; > + } > + > + switch (disk->device) { > + case VIR_DOMAIN_DISK_DEVICE_CDROM: > + case VIR_DOMAIN_DISK_DEVICE_FLOPPY: > + ret = qemuDomainChangeEjectableMedia(driver, vm, disk, qemuCaps, force); > + if (ret == 0) > + dev->data.disk = NULL; > + break; > + default: > + qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED, > + _("disk bus '%s' cannot be updated."), > + virDomainDiskBusTypeToString(disk->bus)); > + break; > + } > +end: If ret is not 0 and cgroup is not NULL, you should call qemuTeardownDiskCgroup() to do some cleanup.