49. Vhost Library
The vhost library implements a user space virtio net server allowing the user to manipulate the virtio ring directly. In another words, it allows the user to fetch/put packets from/to the VM virtio net device. To achieve this, a vhost library should be able to:
Access the guest memory:
For QEMU, this is done by using the
-object memory-backend-file,share=on,...option. Which means QEMU will create a file to serve as the guest RAM. The
share=onoption allows another process to map that file, which means it can access the guest RAM.
Know all the necessary information about the vring:
Information such as where the available ring is stored. Vhost defines some messages (passed through a Unix domain socket file) to tell the backend all the information it needs to know how to manipulate the vring.
49.1. Vhost API Overview
The following is an overview of some key Vhost API functions:
This function registers a vhost driver into the system.
pathspecifies the Unix domain socket file path.
Currently supported flags are:
DPDK vhost-user will act as the client when this flag is given. See below for an explanation.
When DPDK vhost-user acts as the client it will keep trying to reconnect to the server (QEMU) until it succeeds. This is useful in two cases:
- When QEMU is not started yet.
- When QEMU restarts (for example due to a guest OS reboot).
This reconnect option is enabled by default. However, it can be turned off by setting this flag.
IOMMU support will be enabled when this flag is set. It is disabled by default.
Enabling this flag makes possible to use guest vIOMMU to protect vhost from accessing memory the virtio device isn’t allowed to, when the feature is negotiated and an IOMMU device is declared.
Postcopy live-migration support will be enabled when this flag is set. It is disabled by default.
Enabling this flag should only be done when the calling application does not pre-fault the guest shared memory, otherwise migration would fail.
Enabling this flag forces vhost dequeue function to only provide linear pktmbuf (no multi-segmented pktmbuf).
The vhost library by default provides a single pktmbuf for given a packet, but if for some reason the data doesn’t fit into a single pktmbuf (e.g., TSO is enabled), the library will allocate additional pktmbufs from the same mempool and chain them together to create a multi-segmented pktmbuf.
However, the vhost application needs to support multi-segmented format. If the vhost application does not support that format and requires large buffers to be dequeue, this flag should be enabled to force only linear buffers (see RTE_VHOST_USER_EXTBUF_SUPPORT) or drop the packet.
It is disabled by default.
Enabling this flag allows vhost dequeue function to allocate and attach an external buffer to a pktmbuf if the pkmbuf doesn’t provide enough space to store all data.
This is useful when the vhost application wants to support large packets but doesn’t want to increase the default mempool object size nor to support multi-segmented mbufs (non-linear). In this case, a fresh buffer is allocated using rte_malloc() which gets attached to a pktmbuf using rte_pktmbuf_attach_extbuf().
See RTE_VHOST_USER_LINEARBUF_SUPPORT as well to disable multi-segmented mbufs for application that doesn’t support chained mbufs.
It is disabled by default.
Asynchronous data path will be enabled when this flag is set. Async data path allows applications to register async copy devices (typically hardware DMA channels) to the vhost queues. Vhost leverages the copy device registered to free CPU from memory copy operations. A set of async data path APIs are defined for DPDK applications to make use of the async capability. Only packets enqueued/dequeued by async APIs are processed through the async data path.
Currently this feature is only implemented on split ring enqueue data path.
It is disabled by default.
Since v16.04, the vhost library forwards checksum and gso requests for packets received from a virtio driver by filling Tx offload metadata in the mbuf. This behavior is inconsistent with other drivers but it is left untouched for existing applications that might rely on it.
This flag disables the legacy behavior and instead ask vhost to simply populate Rx offload metadata in the mbuf.
It is disabled by default.
This function sets the feature bits the vhost-user driver supports. The vhost-user driver could be vhost-user net, yet it could be something else, say, vhost-user SCSI.
This function registers a set of callbacks, to let DPDK applications take the appropriate action when some events happen. The following events are currently supported:
This callback is invoked when a virtio device becomes ready.
vidis the vhost device ID.
This callback is invoked when a virtio device is paused or shut down.
vring_state_changed(int vid, uint16_t queue_id, int enable)
This callback is invoked when a specific queue’s state is changed, for example to enabled or disabled.
features_changed(int vid, uint64_t features)
This callback is invoked when the features is changed. For example,
VHOST_F_LOG_ALLwill be set/cleared at the start/end of live migration, respectively.
This callback is invoked on new vhost-user socket connection. If DPDK acts as the server the device should not be deleted before
destroy_connectioncallback is received.
This callback is invoked when vhost-user socket connection is closed. It indicates that device with id
vidis no longer in use and can be safely deleted.
This function disables/enables some features. For example, it can be used to disable mergeable buffers and TSO features, which both are enabled by default.
This function triggers the vhost-user negotiation. It should be invoked at the end of initializing a vhost-user driver.
rte_vhost_enqueue_burst(vid, queue_id, pkts, count)
countpackets from host to guest.
rte_vhost_dequeue_burst(vid, queue_id, mbuf_pool, pkts, count)
countpackets from guest, and stored them at
rte_vhost_crypto_create(vid, cryptodev_id, sess_mempool, socket_id)
As an extension of new_device(), this function adds virtio-crypto workload acceleration capability to the device. All crypto workload is processed by DPDK cryptodev with the device ID of
Frees the memory and vhost-user message handlers created in rte_vhost_crypto_create().
rte_vhost_crypto_fetch_requests(vid, queue_id, ops, nb_ops)
nb_opsvirtio-crypto requests from guest, parses them to DPDK Crypto Operations, and fills the
opswith parsing results.
rte_vhost_crypto_finalize_requests(queue_id, ops, nb_ops)
opsare dequeued from Cryptodev, finalizes the jobs and notifies the guest(s).
Enable or disable zero copy feature of the vhost crypto backend.
rte_vhost_async_channel_register(vid, queue_id, features, ops)
Register a vhost queue with async copy device channel after vring is enabled. Following device
featuresmust be specified together with the registration:
Async copy device can guarantee the ordering of copy completion sequence. Copies are completed in the same order with that at the submission time.
async_inordercapable device is supported by vhost.
The copy length (in bytes) below which CPU copy will be used even if applications call async vhost APIs to enqueue/dequeue data.
Typical value is 512~1024 depending on the async device capability.
Applications must provide following
opscallbacks for vhost lib to work with the async copy devices:
transfer_data(vid, queue_id, descs, opaque_data, count)
vhost invokes this function to submit copy data to the async devices. For non-async_inorder capable devices,
opaque_datacould be used for identifying the completed packets.
check_completed_copies(vid, queue_id, opaque_data, max_packets)
vhost invokes this function to get the copy data completed by async devices.
Unregister the async copy device channel from a vhost queue. Unregistration will fail, if the vhost queue has in-flight packets that are not completed.
Unregister async copy devices in vring_state_changed() may fail, as this API tries to acquire the spinlock of vhost queue. The recommended way is to unregister async copy devices for all vhost queues in destroy_device(), when a virtio device is paused or shut down.
rte_vhost_submit_enqueue_burst(vid, queue_id, pkts, count, comp_pkts, comp_count)
Submit an enqueue request to transmit
countpackets from host to guest by async data path. Successfully enqueued packets can be transfer completed or being occupied by DMA engines; transfer completed packets are returned in
comp_pkts, but others are not guaranteed to finish, when this API call returns.
Applications must not free the packets submitted for enqueue until the packets are completed.
rte_vhost_poll_enqueue_completed(vid, queue_id, pkts, count)
Poll enqueue completion status from async data path. Completed packets are returned to applications through
49.2. Vhost-user Implementations
Vhost-user uses Unix domain sockets for passing messages. This means the DPDK vhost-user implementation has two options:
DPDK vhost-user acts as the server.
DPDK will create a Unix domain socket server file and listen for connections from the frontend.
Note, this is the default mode, and the only mode before DPDK v16.07.
DPDK vhost-user acts as the client.
Unlike the server mode, this mode doesn’t create the socket file; it just tries to connect to the server (which responses to create the file instead).
When the DPDK vhost-user application restarts, DPDK vhost-user will try to connect to the server again. This is how the “reconnect” feature works.
- The “reconnect” feature requires QEMU v2.7 (or above).
- The vhost supported features must be exactly the same before and after the restart. For example, if TSO is disabled and then enabled, nothing will work and issues undefined might happen.
No matter which mode is used, once a connection is established, DPDK vhost-user will start receiving and processing vhost messages from QEMU.
For messages with a file descriptor, the file descriptor can be used directly in the vhost process as it is already installed by the Unix domain socket.
The supported vhost messages are:
VHOST_SET_MEM_TABLE message, QEMU will send information for each
memory region and its file descriptor in the ancillary data of the message.
The file descriptor is used to map that region.
VHOST_SET_VRING_KICK is used as the signal to put the vhost device into
the data plane, and
VHOST_GET_VRING_BASE is used as the signal to remove
the vhost device from the data plane.
When the socket connection is closed, vhost will destroy the device.
49.3. Guest memory requirement
For non-async data path, guest memory pre-allocation is not a must. This can help save of memory. If users really want the guest memory to be pre-allocated (e.g., for performance reason), we can add option
-mem-preallocwhen starting QEMU. Or, we can lock all memory at vhost side which will force memory to be allocated when mmap at vhost side; option –mlockall in ovs-dpdk is an example in hand.
For async data path, we force the VM memory to be pre-allocated at vhost lib when mapping the guest memory; and also we need to lock the memory to prevent pages being swapped out to disk.
share=onQEMU option is given. vhost-user will not work with a QEMU version without shared memory mapping.
49.4. Vhost supported vSwitch reference
For more vhost details and how to support vhost in vSwitch, please refer to the vhost example in the DPDK Sample Applications Guide.
49.5. Vhost data path acceleration (vDPA)
vDPA supports selective datapath in vhost-user lib by enabling virtio ring compatible devices to serve virtio driver directly for datapath acceleration.
rte_vhost_driver_attach_vdpa_device is used to configure the vhost device
with accelerated backend.
Also vhost device capabilities are made configurable to adopt various devices. Such capabilities include supported features, protocol features, queue number.
Finally, a set of device ops is defined for device specific operations:
Called to get supported queue number of the device.
Called to get supported features of the device.
Called to get supported protocol features of the device.
Called to configure the actual device when the virtio device becomes ready.
Called to close the actual device when the virtio device is stopped.
Called to change the state of the vring in the actual device when vring state changes.
Called to set the negotiated features to device.
Called to allow the device to response to RARP sending.
Called to get the VFIO group fd of the device.
Called to get the VFIO device fd of the device.
Called to get the notify area info of the queue.