Message ID | 20210820142002.152994-7-lvivier@redhat.com |
---|---|
State | New |
Headers | show |
Series | virtio-net failover cleanup and new features | expand |
Laurent Vivier <lvivier@redhat.com> wrote: > The intend of failover is to allow a VM with a VFIO networking card to > be migrated without disrupting the network operation by switching > to a virtio-net device during the migration. > > This simple change allows a simulated device like e1000e to be tested > rather than a vfio device, even if it's useless in real life it can help > to debug failover. > > This is interesting to developers that want to test failover on > a system with no vfio device. Moreover it simplifies host networking > configuration as we can use the same bridge for virtio-net and > the other failover networking device. > > Without this change the migration of a system configured with failover > fails with: > > ... > -device virtio-net-pci,id=virtionet0,failover=on,... \ > -device e1000,failover_pair_id=virtionet0,... \ > ... > > (qemu) migrate ... > > Unknown ramblock "0000:00:01.1:00.0/e1000e.rom", cannot accept migration > error while loading state for instance 0x0 of device 'ram' > load of migration failed: Invalid argument > > This happens because QEMU correctly unregisters the interface vmstate but > not the ROM one. This patch fixes that. > > Signed-off-by: Laurent Vivier <lvivier@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> One could even defend that unpluging the device and *NOT* unpluging the ROM is a bug, independently of failover, no? Later, Juan.
diff --git a/hw/pci/pci.c b/hw/pci/pci.c index dd5e95216abf..10b875fdfad1 100644 --- a/hw/pci/pci.c +++ b/hw/pci/pci.c @@ -2208,6 +2208,7 @@ static void pci_dev_handle_migration(PCIDevice *pci_dev, MigrationState *s) if (migration_in_setup(s)) { if (pci_dev_migration_unplug(pci_dev)) { vmstate_unregister(VMSTATE_IF(dev), qdev_get_vmsd(dev), dev); + pci_del_option_rom(pci_dev); qapi_event_send_unplug_primary(dev->id); } else { warn_report("couldn't unplug primary device");
The intend of failover is to allow a VM with a VFIO networking card to be migrated without disrupting the network operation by switching to a virtio-net device during the migration. This simple change allows a simulated device like e1000e to be tested rather than a vfio device, even if it's useless in real life it can help to debug failover. This is interesting to developers that want to test failover on a system with no vfio device. Moreover it simplifies host networking configuration as we can use the same bridge for virtio-net and the other failover networking device. Without this change the migration of a system configured with failover fails with: ... -device virtio-net-pci,id=virtionet0,failover=on,... \ -device e1000,failover_pair_id=virtionet0,... \ ... (qemu) migrate ... Unknown ramblock "0000:00:01.1:00.0/e1000e.rom", cannot accept migration error while loading state for instance 0x0 of device 'ram' load of migration failed: Invalid argument This happens because QEMU correctly unregisters the interface vmstate but not the ROM one. This patch fixes that. Signed-off-by: Laurent Vivier <lvivier@redhat.com> --- hw/pci/pci.c | 1 + 1 file changed, 1 insertion(+)