4 * Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34 #ifndef _VIRTIO_NET_H_
35 #define _VIRTIO_NET_H_
39 * Interface to vhost net
43 #include <linux/virtio_ring.h>
44 #include <linux/virtio_net.h>
45 #include <sys/eventfd.h>
46 #include <sys/socket.h>
49 #include <rte_memory.h>
50 #include <rte_mempool.h>
54 #define VHOST_MEMORY_MAX_NREGIONS 8
56 /* Used to indicate that the device is running on a data core */
57 #define VIRTIO_DEV_RUNNING 1
59 /* Backend value set by guest. */
60 #define VIRTIO_DEV_STOPPED -1
63 /* Enum for virtqueue management. */
64 enum {VIRTIO_RXQ, VIRTIO_TXQ, VIRTIO_QNUM};
66 #define BUF_VECTOR_MAX 256
69 * Structure contains buffer address, length and descriptor index
70 * from vring to do scatter RX.
79 * Structure contains variables relevant to RX/TX virtqueues.
81 struct vhost_virtqueue {
82 struct vring_desc *desc; /**< Virtqueue descriptor ring. */
83 struct vring_avail *avail; /**< Virtqueue available ring. */
84 struct vring_used *used; /**< Virtqueue used ring. */
85 uint32_t size; /**< Size of descriptor ring. */
86 uint32_t backend; /**< Backend value to determine if device should started/stopped. */
87 uint16_t vhost_hlen; /**< Vhost header length (varies depending on RX merge buffers. */
88 volatile uint16_t last_used_idx; /**< Last index used on the available ring */
89 volatile uint16_t last_used_idx_res; /**< Used for multiple devices reserving buffers. */
90 int callfd; /**< Used to notify the guest (trigger interrupt). */
91 int kickfd; /**< Currently unused as polling mode is enabled. */
93 uint64_t reserved[16]; /**< Reserve some spaces for future extension. */
94 struct buf_vector buf_vec[BUF_VECTOR_MAX]; /**< for scatter RX. */
95 } __rte_cache_aligned;
99 * Make an extra wrapper for VIRTIO_NET_F_MQ and
100 * VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX as they are
101 * introduced since kernel v3.8. This makes our
102 * code buildable for older kernel.
104 #ifdef VIRTIO_NET_F_MQ
105 #define VHOST_MAX_QUEUE_PAIRS VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX
106 #define VHOST_SUPPORTS_MQ (1ULL << VIRTIO_NET_F_MQ)
108 #define VHOST_MAX_QUEUE_PAIRS 1
109 #define VHOST_SUPPORTS_MQ 0
113 * Define virtio 1.0 for older kernels
115 #ifndef VIRTIO_F_VERSION_1
116 #define VIRTIO_F_VERSION_1 32
120 * Device structure contains all configuration information relating to the device.
123 struct virtio_memory *mem; /**< QEMU memory and memory region information. */
124 uint64_t features; /**< Negotiated feature set. */
125 uint64_t protocol_features; /**< Negotiated protocol feature set. */
126 uint64_t device_fh; /**< device identifier. */
127 uint32_t flags; /**< Device flags. Only used to check if device is running on data core. */
128 #define IF_NAME_SZ (PATH_MAX > IFNAMSIZ ? PATH_MAX : IFNAMSIZ)
129 char ifname[IF_NAME_SZ]; /**< Name of the tap device or socket path. */
130 uint32_t virt_qp_nb; /**< number of queue pair we have allocated */
131 void *priv; /**< private context */
132 uint64_t reserved[64]; /**< Reserve some spaces for future extension. */
133 struct vhost_virtqueue *virtqueue[VHOST_MAX_QUEUE_PAIRS * 2]; /**< Contains all virtqueue information. */
134 } __rte_cache_aligned;
137 * Information relating to memory regions including offsets to addresses in QEMUs memory file.
139 struct virtio_memory_regions {
140 uint64_t guest_phys_address; /**< Base guest physical address of region. */
141 uint64_t guest_phys_address_end; /**< End guest physical address of region. */
142 uint64_t memory_size; /**< Size of region. */
143 uint64_t userspace_address; /**< Base userspace address of region. */
144 uint64_t address_offset; /**< Offset of region for address translation. */
149 * Memory structure includes region and mapping information.
151 struct virtio_memory {
152 uint64_t base_address; /**< Base QEMU userspace address of the memory file. */
153 uint64_t mapped_address; /**< Mapped address of memory file base in our applications memory space. */
154 uint64_t mapped_size; /**< Total size of memory file. */
155 uint32_t nregions; /**< Number of memory regions. */
156 struct virtio_memory_regions regions[0]; /**< Memory region information. */
160 * Device and vring operations.
162 * Make sure to set VIRTIO_DEV_RUNNING to the device flags in new_device and
163 * remove it in destroy_device.
166 struct virtio_net_device_ops {
167 int (*new_device)(struct virtio_net *); /**< Add device. */
168 void (*destroy_device)(volatile struct virtio_net *); /**< Remove device. */
170 int (*vring_state_changed)(struct virtio_net *dev, uint16_t queue_id, int enable); /**< triggered when a vring is enabled or disabled */
173 static inline uint16_t __attribute__((always_inline))
174 rte_vring_available_entries(struct virtio_net *dev, uint16_t queue_id)
176 struct vhost_virtqueue *vq = dev->virtqueue[queue_id];
181 return *(volatile uint16_t *)&vq->avail->idx - vq->last_used_idx_res;
185 * Function to convert guest physical addresses to vhost virtual addresses.
186 * This is used to convert guest virtio buffer addresses.
188 static inline uint64_t __attribute__((always_inline))
189 gpa_to_vva(struct virtio_net *dev, uint64_t guest_pa)
191 struct virtio_memory_regions *region;
193 uint64_t vhost_va = 0;
195 for (regionidx = 0; regionidx < dev->mem->nregions; regionidx++) {
196 region = &dev->mem->regions[regionidx];
197 if ((guest_pa >= region->guest_phys_address) &&
198 (guest_pa <= region->guest_phys_address_end)) {
199 vhost_va = region->address_offset + guest_pa;
207 * Disable features in feature_mask. Returns 0 on success.
209 int rte_vhost_feature_disable(uint64_t feature_mask);
212 * Enable features in feature_mask. Returns 0 on success.
214 int rte_vhost_feature_enable(uint64_t feature_mask);
216 /* Returns currently supported vhost features */
217 uint64_t rte_vhost_feature_get(void);
219 int rte_vhost_enable_guest_notification(struct virtio_net *dev, uint16_t queue_id, int enable);
221 /* Register vhost driver. dev_name could be different for multiple instance support. */
222 int rte_vhost_driver_register(const char *dev_name);
224 /* Unregister vhost driver. This is only meaningful to vhost user. */
225 int rte_vhost_driver_unregister(const char *dev_name);
227 /* Register callbacks. */
228 int rte_vhost_driver_callback_register(struct virtio_net_device_ops const * const);
229 /* Start vhost driver session blocking loop. */
230 int rte_vhost_driver_session_start(void);
233 * This function adds buffers to the virtio devices RX virtqueue. Buffers can
234 * be received from the physical port or from another virtual device. A packet
235 * count is returned to indicate the number of packets that were succesfully
236 * added to the RX queue.
240 * virtio queue index in mq case
242 * array to contain packets to be enqueued
244 * packets num to be enqueued
246 * num of packets enqueued
248 uint16_t rte_vhost_enqueue_burst(struct virtio_net *dev, uint16_t queue_id,
249 struct rte_mbuf **pkts, uint16_t count);
252 * This function gets guest buffers from the virtio device TX virtqueue,
253 * construct host mbufs, copies guest buffer content to host mbufs and
254 * store them in pkts to be processed.
258 * virtio queue index in mq case
260 * mbuf_pool where host mbuf is allocated.
262 * array to contain packets to be dequeued
264 * packets num to be dequeued
266 * num of packets dequeued
268 uint16_t rte_vhost_dequeue_burst(struct virtio_net *dev, uint16_t queue_id,
269 struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count);
271 #endif /* _VIRTIO_NET_H_ */