From: Peter Krempa <pkrempa(a)redhat.com&gt; Detect that the 'stream' netdev backend supports reconnecting. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com&gt; Signed-off-by: Laine Stump <laine(a)redhat.com&gt; --- src/qemu/qemu_capabilities.c | 4 ++++ src/qemu/qemu_capabilities.h | 3 +++ tests/qemucapabilitiesdata/caps_8.0.0.x86_64.xml | 1 + 3 files changed, 8 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index d92d5a62dd..3cb5785baa 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -687,6 +687,9 @@ VIR_ENUM_IMPL(virQEMUCaps, "virtio-crypto", /* QEMU_CAPS_DEVICE_VIRTIO_CRYPTO */ "cryptodev-backend-lkcf", /* QEMU_CAPS_OBJECT_CRYPTO_LKCF */ "pvpanic-pci", /* QEMU_CAPS_DEVICE_PANIC_PCI */ + + /* 445 */ + "netdev.stream.reconnect", /* QEMU_CAPS_NETDEV_STREAM_RECONNECT */ ); @@ -1559,6 +1562,7 @@ static struct virQEMUCapsStringFlags virQEMUCapsQMPSchemaQueries[] = { { "nbd-server-start/arg-type/tls-creds", QEMU_CAPS_NBD_TLS }, { "nbd-server-add/arg-type/bitmap", QEMU_CAPS_NBD_BITMAP }, { "netdev_add/arg-type/+stream", QEMU_CAPS_NETDEV_STREAM }, + { "netdev_add/arg-type/+stream/reconnect", QEMU_CAPS_NETDEV_STREAM_RECONNECT }, { "netdev_add/arg-type/+vhost-vdpa", QEMU_CAPS_NETDEV_VHOST_VDPA }, /* JSON support for -netdev was introduced for the 'dgram' netdev type */ { "netdev_add/arg-type/type/^dgram", QEMU_CAPS_NETDEV_JSON }, diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index b72348cf88..d049f79dd9 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -667,6 +667,9 @@ typedef enum { /* virQEMUCapsFlags grouping marker for syntax-check */ QEMU_CAPS_OBJECT_CRYPTO_LKCF, /* -object cryptodev-backend-lkcf */ QEMU_CAPS_DEVICE_PANIC_PCI, /* -device pvpanic-pci */ + /* 445 */ + QEMU_CAPS_NETDEV_STREAM_RECONNECT, /* -netdev stream supports reconnect */ + QEMU_CAPS_LAST /* this must always be the last item */ } virQEMUCapsFlags; diff --git a/tests/qemucapabilitiesdata/caps_8.0.0.x86_64.xml b/tests/qemucapabilitiesdata/caps_8.0.0.x86_64.xml index bcad3db057..ce051d3f1c 100644 --- a/tests/qemucapabilitiesdata/caps_8.0.0.x86_64.xml +++ b/tests/qemucapabilitiesdata/caps_8.0.0.x86_64.xml @@ -205,6 +205,7 @@ <flag name='virtio-crypto'/> <flag name='cryptodev-backend-lkcf'/> <flag name='pvpanic-pci'/> + <flag name='netdev.stream.reconnect'/> <version>7002050</version> <kvmVersion>0</kvmVersion> <microcodeVersion>43100244</microcodeVersion> -- 2.39.2

When a QEMU netdev is of type "stream", if the socket it uses for connectivity to the host network gets closed, then QEMU will send a NETDEV_STREAM_DISCONNECTED event. We know that any stream netdev we've created is backed by a passt process, and if the socket was closed, that means the passt process has disappeared. When we receive this event, we can respond by starting a new passt process with the same options (including socket path) we originally used. If we have previously created the stream netdev device with a "reconnect" option, then QEMU will automatically reconnect to this new passt process. (If we hadn't used "reconnect", then QEMU will never try to reconnect to the new passt process, so there's no point in starting it.) Note that NETDEV_STREAM_DISCONNECTED is an event sent for the netdev (ie "host side") of the network device, and so it sends the "netdev-id" to specify which device was disconnected. But libvirt's virDomainNetDef (the object used to keep track of network devices) is the internal representation of both the host-side "netdev", and the guest side device, and virDomainNetDef doesn't directly keep track of the netdev-id, only of the device's "alias" (which is the "id" parameter of the *guest* side of the device). Fortunately, by convention libvirt always names the host-side of devices as "host" + alias, so in order to search for the affected NetDef, all we need to do is trim the 1st 4 characters from the netdev-id and look for the NetDef having that resulting trimmed string as its alias. (Contrast this to NIC_RX_FILTER_CHANGED, which is an event received for the guest side of the device, and so directly contains the device alias.) Resolves: https://bugzilla.redhat.com/2172098 Signed-off-by: Laine Stump <laine(a)redhat.com&gt; --- src/qemu/qemu_domain.c | 1 + src/qemu/qemu_domain.h | 1 + src/qemu/qemu_driver.c | 82 ++++++++++++++++++++++++++++++++++++ src/qemu/qemu_monitor.c | 11 +++++ src/qemu/qemu_monitor.h | 6 +++ src/qemu/qemu_monitor_json.c | 16 +++++++ src/qemu/qemu_process.c | 18 ++++++++ 7 files changed, 135 insertions(+) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index e9bc0f375d..4cf9a259ea 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -11238,6 +11238,7 @@ qemuProcessEventFree(struct qemuProcessEvent *event) break; case QEMU_PROCESS_EVENT_WATCHDOG: case QEMU_PROCESS_EVENT_DEVICE_DELETED: + case QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED: case QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED: case QEMU_PROCESS_EVENT_SERIAL_CHANGED: case QEMU_PROCESS_EVENT_MONITOR_EOF: diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 1053d1d4cb..6adc067681 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -447,6 +447,7 @@ typedef enum { QEMU_PROCESS_EVENT_WATCHDOG = 0, QEMU_PROCESS_EVENT_GUESTPANIC, QEMU_PROCESS_EVENT_DEVICE_DELETED, + QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED, QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED, QEMU_PROCESS_EVENT_SERIAL_CHANGED, QEMU_PROCESS_EVENT_JOB_STATUS_CHANGE, diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 6154fe9bfe..47d6a0dd95 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -40,6 +40,7 @@ #include "qemu_hostdev.h" #include "qemu_hotplug.h" #include "qemu_monitor.h" +#include "qemu_passt.h" #include "qemu_process.h" #include "qemu_migration.h" #include "qemu_migration_params.h" @@ -3622,6 +3623,84 @@ processDeviceDeletedEvent(virQEMUDriver *driver, } +static void +processNetdevStreamDisconnectedEvent(virDomainObj *vm, + const char *netdevId) +{ + virDomainDeviceDef dev; + virDomainNetDef *def; + qemuDomainObjPrivate *priv; + virQEMUCaps *qemuCaps; + const char *devAlias = NULL; + + /* The event sends us the "netdev-id", but we don't store the + * netdev-id in the NetDef and thus can't use it to find the + * correct NetDef. We *do* keep the device alias in the NetDef. + * By definition, the netdev-id is "host" + devAlias, so we just + * need to remove "host" from the front of netdev-id to get + * something we can use to find the proper NetDef. + */ + if (STREQLEN(netdevId, "host", 4)) + devAlias = &netdevId[4]; + + if (!devAlias) { + VIR_WARN("Received NETDEV_STREAM_DISCONNECTED event for unrecognized netdev %s from domain %p %s", + netdevId, vm, vm->def->name); + return; + } + + VIR_DEBUG("Received NETDEV_STREAM_DISCONNECTED event for device %s from domain %p %s", + devAlias, vm, vm->def->name); + + if (virDomainObjBeginJob(vm, VIR_JOB_QUERY) < 0) + return; + + if (!virDomainObjIsActive(vm)) { + VIR_DEBUG("Domain is not running"); + goto endjob; + } + + if (virDomainDefFindDevice(vm->def, devAlias, &dev, true) < 0) { + VIR_WARN("NETDEV_STREAM_DISCONNECTED event received for non-existent device %s in domain %s", + devAlias, vm->def->name); + goto endjob; + } + if (dev.type != VIR_DOMAIN_DEVICE_NET) { + VIR_WARN("NETDEV_STREAM_DISCONNECTED event received for non-network device %s in domain %s", + devAlias, vm->def->name); + goto endjob; + } + def = dev.data.net; + + if (def->backend.type != VIR_DOMAIN_NET_BACKEND_PASST) { + VIR_DEBUG("ignore NETDEV_STREAM_DISCONNECTED event for non-passt network device %s in domain %s", + def->info.alias, vm->def->name); + goto endjob; + } + + priv = vm->privateData; + qemuCaps = priv->qemuCaps; + + if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_NETDEV_STREAM_RECONNECT)) { + VIR_WARN("ignore NETDEV_STREAM_DISCONNECTED event for passt network device %s in domain %s - QEMU binary does not support reconnect", + def->info.alias, vm->def->name); + goto endjob; + } + + /* handle the event - restart the passt process with its original + * parameters + */ + VIR_DEBUG("process NETDEV_STREAM_DISCONNECTED event for network device %s in domain %s", + def->info.alias, vm->def->name); + + if (qemuPasstStart(vm, def) < 0) + goto endjob; + + endjob: + virDomainObjEndJob(vm); +} + + static void processNicRxFilterChangedEvent(virDomainObj *vm, const char *devAlias) @@ -3971,6 +4050,9 @@ static void qemuProcessEventHandler(void *data, void *opaque) case QEMU_PROCESS_EVENT_DEVICE_DELETED: processDeviceDeletedEvent(driver, vm, processEvent->data); break; + case QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED: + processNetdevStreamDisconnectedEvent(vm, processEvent->data); + break; case QEMU_PROCESS_EVENT_NIC_RX_FILTER_CHANGED: processNicRxFilterChangedEvent(vm, processEvent->data); break; diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 38f89167e0..1fa35f03cc 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -1265,6 +1265,17 @@ qemuMonitorEmitNicRxFilterChanged(qemuMonitor *mon, } +void +qemuMonitorEmitNetdevStreamDisconnected(qemuMonitor *mon, + const char *devAlias) +{ + VIR_DEBUG("mon=%p", mon); + + QEMU_MONITOR_CALLBACK(mon, domainNetdevStreamDisconnected, + mon->vm, devAlias); +} + + void qemuMonitorEmitSerialChange(qemuMonitor *mon, const char *devAlias, diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index 2d16214ba2..2fa06b99a3 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -250,6 +250,9 @@ typedef void (*qemuMonitorDomainDeviceUnplugErrCallback)(qemuMonitor *mon, virDomainObj *vm, const char *devPath, const char *devAlias); +typedef void (*qemuMonitorDomainNetdevStreamDisconnectedCallback)(qemuMonitor *mon, + virDomainObj *vm, + const char *devAlias); typedef void (*qemuMonitorDomainNicRxFilterChangedCallback)(qemuMonitor *mon, virDomainObj *vm, const char *devAlias); @@ -397,6 +400,7 @@ struct _qemuMonitorCallbacks { qemuMonitorDomainMemoryFailureCallback domainMemoryFailure; qemuMonitorDomainMemoryDeviceSizeChange domainMemoryDeviceSizeChange; qemuMonitorDomainDeviceUnplugErrCallback domainDeviceUnplugError; + qemuMonitorDomainNetdevStreamDisconnectedCallback domainNetdevStreamDisconnected; }; qemuMonitor *qemuMonitorOpen(virDomainObj *vm, @@ -480,6 +484,8 @@ void qemuMonitorEmitDeviceDeleted(qemuMonitor *mon, void qemuMonitorEmitDeviceUnplugErr(qemuMonitor *mon, const char *devPath, const char *devAlias); +void qemuMonitorEmitNetdevStreamDisconnected(qemuMonitor *mon, + const char *devAlias); void qemuMonitorEmitNicRxFilterChanged(qemuMonitor *mon, const char *devAlias); void qemuMonitorEmitSerialChange(qemuMonitor *mon, diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index ba6276ec8e..e81b464eea 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -84,6 +84,7 @@ static void qemuMonitorJSONHandleRdmaGidStatusChanged(qemuMonitor *mon, virJSONV static void qemuMonitorJSONHandleMemoryFailure(qemuMonitor *mon, virJSONValue *data); static void qemuMonitorJSONHandleMemoryDeviceSizeChange(qemuMonitor *mon, virJSONValue *data); static void qemuMonitorJSONHandleDeviceUnplugErr(qemuMonitor *mon, virJSONValue *data); +static void qemuMonitorJSONHandleNetdevStreamDisconnected(qemuMonitor *mon, virJSONValue *data); typedef struct { const char *type; @@ -106,6 +107,7 @@ static qemuEventHandler eventHandlers[] = { { "MEMORY_FAILURE", qemuMonitorJSONHandleMemoryFailure, }, { "MIGRATION", qemuMonitorJSONHandleMigrationStatus, }, { "MIGRATION_PASS", qemuMonitorJSONHandleMigrationPass, }, + { "NETDEV_STREAM_DISCONNECTED", qemuMonitorJSONHandleNetdevStreamDisconnected, }, { "NIC_RX_FILTER_CHANGED", qemuMonitorJSONHandleNicRxFilterChanged, }, { "PR_MANAGER_STATUS_CHANGED", qemuMonitorJSONHandlePRManagerStatusChanged, }, { "RDMA_GID_STATUS_CHANGED", qemuMonitorJSONHandleRdmaGidStatusChanged, }, @@ -1021,6 +1023,20 @@ qemuMonitorJSONHandleDeviceUnplugErr(qemuMonitor *mon, virJSONValue *data) } +static void +qemuMonitorJSONHandleNetdevStreamDisconnected(qemuMonitor *mon, virJSONValue *data) +{ + const char *name; + + if (!(name = virJSONValueObjectGetString(data, "netdev-id"))) { + VIR_WARN("missing device in NETDEV_STREAM_DISCONNECTED event"); + return; + } + + qemuMonitorEmitNetdevStreamDisconnected(mon, name); +} + + static void qemuMonitorJSONHandleNicRxFilterChanged(qemuMonitor *mon, virJSONValue *data) { diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index d4499c6f84..63d7e1138d 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -1361,6 +1361,23 @@ qemuProcessHandleBlockThreshold(qemuMonitor *mon G_GNUC_UNUSED, } +static void +qemuProcessHandleNetdevStreamDisconnected(qemuMonitor *mon G_GNUC_UNUSED, + virDomainObj *vm, + const char *devAlias) +{ + virObjectLock(vm); + + VIR_DEBUG("Device %s Netdev Stream Disconnected in domain %p %s", + devAlias, vm, vm->def->name); + + qemuProcessEventSubmit(vm, QEMU_PROCESS_EVENT_NETDEV_STREAM_DISCONNECTED, + 0, 0, g_strdup(devAlias)); + + virObjectUnlock(vm); +} + + static void qemuProcessHandleNicRxFilterChanged(qemuMonitor *mon G_GNUC_UNUSED, virDomainObj *vm, @@ -1802,6 +1819,7 @@ static qemuMonitorCallbacks monitorCallbacks = { .domainMemoryFailure = qemuProcessHandleMemoryFailure, .domainMemoryDeviceSizeChange = qemuProcessHandleMemoryDeviceSizeChange, .domainDeviceUnplugError = qemuProcessHandleDeviceUnplugErr, + .domainNetdevStreamDisconnected = qemuProcessHandleNetdevStreamDisconnected, }; static void -- 2.39.2

On 2/22/23 5:21 AM, Michal Prívozník wrote: Ah yes, right you are! Changed. I had originally had them initialized, but moved the assignment down until it was actually needed because of some warped idea about saving cycles, or avoiding a double de-reference in case one of them might be NULL or something. But in retrospect, just delaying wouldn't have solved that problem anyway. I'm changing it to use QEMU_DOMAIN_PRIVATE before pushing. Thanks for the review!