/* SPDX-License-Identifier: GPL-2.0-or-later
 * Copyright Red Hat
 * Author: Laurent Vivier <lvivier@redhat.com>
 *
 * vhost-user API, command management and virtio interface
 */
/* some parts from QEMU subprojects/libvhost-user/libvhost-user.c */

#include <errno.h>
#include <fcntl.h>
#include <stdlib.h>
#include <stdio.h>
#include <stdint.h>
#include <stddef.h>
#include <string.h>
#include <assert.h>
#include <stdbool.h>
#include <inttypes.h>
#include <time.h>
#include <net/ethernet.h>
#include <netinet/in.h>
#include <sys/epoll.h>
#include <sys/eventfd.h>
#include <sys/mman.h>
#include <linux/vhost_types.h>
#include <linux/virtio_net.h>

#include "util.h"
#include "passt.h"
#include "tap.h"
#include "vhost_user.h"

/* vhost-user version we are compatible with */
#define VHOST_USER_VERSION 1

/**
 * vu_print_capabilities() - print vhost-user capabilities
 * 			     this is part of the vhost-user backend
 * 			     convention.
 */
void vu_print_capabilities(void)
{
	printf("{\n");
	printf("  \"type\": \"net\"\n");
	printf("}\n");
	exit(EXIT_SUCCESS);
}

/**
 * vu_request_to_string() - convert a vhost-user request number to its name
 * @req:	request number
 *
 * Return: the name of request number
 */
static const char *vu_request_to_string(unsigned int req)
{
	if (req < VHOST_USER_MAX) {
#define REQ(req) [req] = #req
		static const char * const vu_request_str[] = {
			REQ(VHOST_USER_NONE),
			REQ(VHOST_USER_GET_FEATURES),
			REQ(VHOST_USER_SET_FEATURES),
			REQ(VHOST_USER_SET_OWNER),
			REQ(VHOST_USER_RESET_OWNER),
			REQ(VHOST_USER_SET_MEM_TABLE),
			REQ(VHOST_USER_SET_LOG_BASE),
			REQ(VHOST_USER_SET_LOG_FD),
			REQ(VHOST_USER_SET_VRING_NUM),
			REQ(VHOST_USER_SET_VRING_ADDR),
			REQ(VHOST_USER_SET_VRING_BASE),
			REQ(VHOST_USER_GET_VRING_BASE),
			REQ(VHOST_USER_SET_VRING_KICK),
			REQ(VHOST_USER_SET_VRING_CALL),
			REQ(VHOST_USER_SET_VRING_ERR),
			REQ(VHOST_USER_GET_PROTOCOL_FEATURES),
			REQ(VHOST_USER_SET_PROTOCOL_FEATURES),
			REQ(VHOST_USER_GET_QUEUE_NUM),
			REQ(VHOST_USER_SET_VRING_ENABLE),
			REQ(VHOST_USER_SEND_RARP),
			REQ(VHOST_USER_NET_SET_MTU),
			REQ(VHOST_USER_SET_BACKEND_REQ_FD),
			REQ(VHOST_USER_IOTLB_MSG),
			REQ(VHOST_USER_SET_VRING_ENDIAN),
			REQ(VHOST_USER_GET_CONFIG),
			REQ(VHOST_USER_SET_CONFIG),
			REQ(VHOST_USER_POSTCOPY_ADVISE),
			REQ(VHOST_USER_POSTCOPY_LISTEN),
			REQ(VHOST_USER_POSTCOPY_END),
			REQ(VHOST_USER_GET_INFLIGHT_FD),
			REQ(VHOST_USER_SET_INFLIGHT_FD),
			REQ(VHOST_USER_GPU_SET_SOCKET),
			REQ(VHOST_USER_VRING_KICK),
			REQ(VHOST_USER_GET_MAX_MEM_SLOTS),
			REQ(VHOST_USER_ADD_MEM_REG),
			REQ(VHOST_USER_REM_MEM_REG),
			REQ(VHOST_USER_MAX),
		};
#undef REQ
		return vu_request_str[req];
	}

	return "unknown";
}

/**
 * qva_to_va() -  Translate front-end (QEMU) virtual address to our virtual
 * 		  address.
 * @dev:		Vhost-user device
 * @qemu_addr:		front-end userspace address
 *
 * Return: the memory address in our process virtual address space.
 */
static void *qva_to_va(struct vu_dev *dev, uint64_t qemu_addr)
{
	unsigned int i;

	/* Find matching memory region.  */
	for (i = 0; i < dev->nregions; i++) {
		const struct vu_dev_region *r = &dev->regions[i];

		if ((qemu_addr >= r->qva) && (qemu_addr < (r->qva + r->size))) {
			/* NOLINTNEXTLINE(performance-no-int-to-ptr) */
			return (void *)(qemu_addr - r->qva + r->mmap_addr +
					r->mmap_offset);
		}
	}

	return NULL;
}

/**
 * vmsg_close_fds() - Close all file descriptors of a given message
 * @vmsg:	Vhost-user message with the list of the file descriptors
 */
static void vmsg_close_fds(const struct vhost_user_msg *vmsg)
{
	int i;

	for (i = 0; i < vmsg->fd_num; i++)
		close(vmsg->fds[i]);
}

/**
 * vu_remove_watch() - Remove a file descriptor from an our passt epoll
 * 		       file descriptor
 * @vdev:	Vhost-user device
 * @fd:		file descriptor to remove
 */
static void vu_remove_watch(const struct vu_dev *vdev, int fd)
{
	epoll_ctl(vdev->context->epollfd, EPOLL_CTL_DEL, fd, NULL);
}

/**
 * vmsg_set_reply_u64() - Set reply payload.u64 and clear request flags
 * 			  and fd_num
 * @vmsg:	Vhost-user message
 * @val:	64bit value to reply
 */
static void vmsg_set_reply_u64(struct vhost_user_msg *vmsg, uint64_t val)
{
	vmsg->hdr.flags = 0; /* defaults will be set by vu_send_reply() */
	vmsg->hdr.size = sizeof(vmsg->payload.u64);
	vmsg->payload.u64 = val;
	vmsg->fd_num = 0;
}

/**
 * vu_message_read_default() - Read incoming vhost-user message from the
 * 			       front-end
 * @conn_fd:	Vhost-user command socket
 * @vmsg:	Vhost-user message
 *
 * Return: -1 there is an error,
 *          0 if recvmsg() has been interrupted,
 *          1 if a message has been received
 */
static int vu_message_read_default(int conn_fd, struct vhost_user_msg *vmsg)
{
	char control[CMSG_SPACE(VHOST_MEMORY_BASELINE_NREGIONS *
		     sizeof(int))] = { 0 };
	struct iovec iov = {
		.iov_base = (char *)vmsg,
		.iov_len = VHOST_USER_HDR_SIZE,
	};
	struct msghdr msg = {
		.msg_iov = &iov,
		.msg_iovlen = 1,
		.msg_control = control,
		.msg_controllen = sizeof(control),
	};
	size_t fd_size;
	struct cmsghdr *cmsg;
	ssize_t ret, sz_payload;

	ret = recvmsg(conn_fd, &msg, MSG_DONTWAIT);
	if (ret < 0) {
		if (errno == EINTR || errno == EAGAIN || errno == EWOULDBLOCK)
			return 0;
		return -1;
	}

	vmsg->fd_num = 0;
	for (cmsg = CMSG_FIRSTHDR(&msg); cmsg != NULL;
	     cmsg = CMSG_NXTHDR(&msg, cmsg)) {
		if (cmsg->cmsg_level == SOL_SOCKET &&
		    cmsg->cmsg_type == SCM_RIGHTS) {
			fd_size = cmsg->cmsg_len - CMSG_LEN(0);
			vmsg->fd_num = fd_size / sizeof(int);
			memcpy(vmsg->fds, CMSG_DATA(cmsg), fd_size);
			break;
		}
	}

	sz_payload = vmsg->hdr.size;
	if ((size_t)sz_payload > sizeof(vmsg->payload)) {
		vu_panic("Error: too big message request: %d,"
			 " size: vmsg->size: %zd, "
			 "while sizeof(vmsg->payload) = %zu",
			 vmsg->hdr.request, sz_payload, sizeof(vmsg->payload));
	}

	if (sz_payload) {
		do {
			ret = recv(conn_fd, &vmsg->payload, sz_payload, 0);
		} while (ret < 0 && (errno == EINTR || errno == EAGAIN));

		if (ret < sz_payload)
			vu_panic("Error while reading: %s",
				 strerror(errno));
	}

	return 1;
}

/**
 * vu_message_write() - send a message to the front-end
 * @conn_fd:	Vhost-user command socket
 * @vmsg:	Vhost-user message
 *
 * #syscalls:vu sendmsg
 */
static void vu_message_write(int conn_fd, struct vhost_user_msg *vmsg)
{
	int rc;
	const uint8_t *p = (uint8_t *)vmsg;
	char control[CMSG_SPACE(VHOST_MEMORY_BASELINE_NREGIONS * sizeof(int))] = { 0 };
	struct iovec iov = {
		.iov_base = (char *)vmsg,
		.iov_len = VHOST_USER_HDR_SIZE,
	};
	struct msghdr msg = {
		.msg_iov = &iov,
		.msg_iovlen = 1,
		.msg_control = control,
	};

	memset(control, 0, sizeof(control));
	assert(vmsg->fd_num <= VHOST_MEMORY_BASELINE_NREGIONS);
	if (vmsg->fd_num > 0) {
		size_t fdsize = vmsg->fd_num * sizeof(int);
		struct cmsghdr *cmsg;

		msg.msg_controllen = CMSG_SPACE(fdsize);
		cmsg = CMSG_FIRSTHDR(&msg);
		cmsg->cmsg_len = CMSG_LEN(fdsize);
		cmsg->cmsg_level = SOL_SOCKET;
		cmsg->cmsg_type = SCM_RIGHTS;
		memcpy(CMSG_DATA(cmsg), vmsg->fds, fdsize);
	} else {
		msg.msg_controllen = 0;
	}

	do {
		rc = sendmsg(conn_fd, &msg, 0);
	} while (rc < 0 && (errno == EINTR || errno == EAGAIN));

	if (vmsg->hdr.size) {
		do {
			rc = write(conn_fd, p + VHOST_USER_HDR_SIZE,
				   vmsg->hdr.size);
		} while (rc < 0 && (errno == EINTR || errno == EAGAIN));
	}

	if (rc <= 0)
		vu_panic("Error while writing: %s", strerror(errno));
}

/**
 * vu_send_reply() - Update message flags and send it to front-end
 * @conn_fd:	Vhost-user command socket
 * @vmsg:	Vhost-user message
 */
static void vu_send_reply(int conn_fd, struct vhost_user_msg *msg)
{
	msg->hdr.flags &= ~VHOST_USER_VERSION_MASK;
	msg->hdr.flags |= VHOST_USER_VERSION;
	msg->hdr.flags |= VHOST_USER_REPLY_MASK;

	vu_message_write(conn_fd, msg);
}

/**
 * vu_get_features_exec() - Provide back-end features bitmask to front-end
 * @vmsg:	Vhost-user message
 *
 * Return: true as a reply is requested
 */
static bool vu_get_features_exec(struct vhost_user_msg *msg)
{
	uint64_t features =
		1ULL << VIRTIO_F_VERSION_1 |
		1ULL << VIRTIO_NET_F_MRG_RXBUF |
		1ULL << VHOST_USER_F_PROTOCOL_FEATURES;

	vmsg_set_reply_u64(msg, features);

	debug("Sending back to guest u64: 0x%016"PRIx64, msg->payload.u64);

	return true;
}

/**
 * vu_set_enable_all_rings() - Enable/disable all the virqueues
 * @vdev:	Vhost-user device
 * @enabled:	New virtqueues state
 */
static void vu_set_enable_all_rings(struct vu_dev *vdev, bool enabled)
{
	uint16_t i;

	for (i = 0; i < VHOST_USER_MAX_QUEUES; i++)
		vdev->vq[i].enable = enabled;
}

/**
 * vu_set_features_exec() - Enable features of the back-end
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_features_exec(struct vu_dev *vdev,
				 struct vhost_user_msg *msg)
{
	debug("u64: 0x%016"PRIx64, msg->payload.u64);

	vdev->features = msg->payload.u64;
	/*
	 * We only support devices conforming to VIRTIO 1.0 or
	 * later
	 */
	if (!vu_has_feature(vdev, VIRTIO_F_VERSION_1))
		vu_panic("virtio legacy devices aren't supported by passt");

	if (!vu_has_feature(vdev, VHOST_USER_F_PROTOCOL_FEATURES))
		vu_set_enable_all_rings(vdev, true);

	/* virtio-net features */

	if (vu_has_feature(vdev, VIRTIO_F_VERSION_1) ||
	    vu_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF)) {
		vdev->hdrlen = sizeof(struct virtio_net_hdr_mrg_rxbuf);
	} else {
		vdev->hdrlen = sizeof(struct virtio_net_hdr);
	}

	return false;
}

/**
 * vu_set_owner_exec() - Session start flag, do nothing in our case
 *
 * Return: false as no reply is requested
 */
static bool vu_set_owner_exec(void)
{
	return false;
}

/**
 * map_ring() - Convert ring front-end (QEMU) addresses to our process
 * 		virtual address space.
 * @vdev:	Vhost-user device
 * @vq:		Virtqueue
 *
 * Return: true if ring cannot be mapped to our address space
 */
static bool map_ring(struct vu_dev *vdev, struct vu_virtq *vq)
{
	vq->vring.desc = qva_to_va(vdev, vq->vra.desc_user_addr);
	vq->vring.used = qva_to_va(vdev, vq->vra.used_user_addr);
	vq->vring.avail = qva_to_va(vdev, vq->vra.avail_user_addr);

	debug("Setting virtq addresses:");
	debug("    vring_desc  at %p", (void *)vq->vring.desc);
	debug("    vring_used  at %p", (void *)vq->vring.used);
	debug("    vring_avail at %p", (void *)vq->vring.avail);

	return !(vq->vring.desc && vq->vring.used && vq->vring.avail);
}

/**
 * vu_packet_check_range() - Check if a given memory zone is contained in
 * 			     a mapped guest memory region
 * @buf:	Array of the available memory regions
 * @offset:	Offset of data range in packet descriptor
 * @size:	Length of desired data range
 * @start:	Start of the packet descriptor
 * 
 * Return: 0 if the zone in a mapped memory region, -1 otherwise
 */
int vu_packet_check_range(void *buf, size_t offset, size_t len,
			  const char *start)
{
	struct vu_dev_region *dev_region;

	for (dev_region = buf; dev_region->mmap_addr; dev_region++) {
		/* NOLINTNEXTLINE(performance-no-int-to-ptr) */
		char *m = (char *)dev_region->mmap_addr;

		if (m <= start &&
		    start + offset + len < m + dev_region->mmap_offset +
					       dev_region->size)
			return 0;
	}

	return -1;
}

/**
 * vu_set_mem_table_exec() - Sets the memory map regions to be able to
 * 			     translate the vring addresses. 
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 *
 * #syscalls:vu mmap munmap
 */
static bool vu_set_mem_table_exec(struct vu_dev *vdev,
				  struct vhost_user_msg *msg)
{
	unsigned int i;
	struct vhost_user_memory m = msg->payload.memory, *memory = &m;

	for (i = 0; i < vdev->nregions; i++) {
		struct vu_dev_region *r = &vdev->regions[i];
		/* NOLINTNEXTLINE(performance-no-int-to-ptr) */
		void *mm = (void *)r->mmap_addr;

		if (mm)
			munmap(mm, r->size + r->mmap_offset);
	}
	vdev->nregions = memory->nregions;

	debug("Nregions: %u", memory->nregions);
	for (i = 0; i < vdev->nregions; i++) {
		void *mmap_addr;
		struct vhost_user_memory_region *msg_region = &memory->regions[i];
		struct vu_dev_region *dev_region = &vdev->regions[i];

		debug("Region %d", i);
		debug("    guest_phys_addr: 0x%016"PRIx64,
		      msg_region->guest_phys_addr);
		debug("    memory_size:     0x%016"PRIx64,
		      msg_region->memory_size);
		debug("    userspace_addr   0x%016"PRIx64,
		      msg_region->userspace_addr);
		debug("    mmap_offset      0x%016"PRIx64,
		      msg_region->mmap_offset);

		dev_region->gpa = msg_region->guest_phys_addr;
		dev_region->size = msg_region->memory_size;
		dev_region->qva = msg_region->userspace_addr;
		dev_region->mmap_offset = msg_region->mmap_offset;

		/* We don't use offset argument of mmap() since the
		 * mapped address has to be page aligned, and we use huge
		 * pages.
		 */
		mmap_addr = mmap(0, dev_region->size + dev_region->mmap_offset,
				 PROT_READ | PROT_WRITE, MAP_SHARED |
				 MAP_NORESERVE, msg->fds[i], 0);

		if (mmap_addr == MAP_FAILED)
			vu_panic("region mmap error: %s", strerror(errno));

		dev_region->mmap_addr = (uint64_t)(uintptr_t)mmap_addr;
		debug("    mmap_addr:       0x%016"PRIx64,
		      dev_region->mmap_addr);

		close(msg->fds[i]);
	}

	for (i = 0; i < VHOST_USER_MAX_QUEUES; i++) {
		if (vdev->vq[i].vring.desc) {
			if (map_ring(vdev, &vdev->vq[i]))
				vu_panic("remapping queue %d during setmemtable", i);
		}
	}

	/* As vu_packet_check_range() has no access to the number of
	 * memory regions, mark the end of the array with mmap_addr = 0
	 */
	ASSERT(vdev->nregions < VHOST_USER_MAX_RAM_SLOTS - 1);
	vdev->regions[vdev->nregions].mmap_addr = 0;

	tap_sock_update_buf(vdev->regions, 0);

	return false;
}

/**
 * vu_set_vring_num_exec() - Set the size of the queue (vring size)
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_num_exec(struct vu_dev *vdev,
				  struct vhost_user_msg *msg)
{
	unsigned int idx = msg->payload.state.index;
	unsigned int num = msg->payload.state.num;

	debug("State.index: %u", idx);
	debug("State.num:   %u", num);
	vdev->vq[idx].vring.num = num;

	return false;
}

/**
 * vu_set_vring_addr_exec() - Set the addresses of the vring
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_addr_exec(struct vu_dev *vdev,
				   struct vhost_user_msg *msg)
{
	struct vhost_vring_addr addr = msg->payload.addr, *vra = &addr;
	struct vu_virtq *vq = &vdev->vq[vra->index];

	debug("vhost_vring_addr:");
	debug("    index:  %d", vra->index);
	debug("    flags:  %d", vra->flags);
	debug("    desc_user_addr:   0x%016" PRIx64, (uint64_t)vra->desc_user_addr);
	debug("    used_user_addr:   0x%016" PRIx64, (uint64_t)vra->used_user_addr);
	debug("    avail_user_addr:  0x%016" PRIx64, (uint64_t)vra->avail_user_addr);
	debug("    log_guest_addr:   0x%016" PRIx64, (uint64_t)vra->log_guest_addr);

	vq->vra = *vra;
	vq->vring.flags = vra->flags;
	vq->vring.log_guest_addr = vra->log_guest_addr;

	if (map_ring(vdev, vq))
		vu_panic("Invalid vring_addr message");

	vq->used_idx = le16toh(vq->vring.used->idx);

	if (vq->last_avail_idx != vq->used_idx) {
		debug("Last avail index != used index: %u != %u",
		      vq->last_avail_idx, vq->used_idx);
	}

	return false;
}
/**
 * vu_set_vring_base_exec() - Sets the next index to use for descriptors
 * 			      in this vring
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_base_exec(struct vu_dev *vdev,
				   struct vhost_user_msg *msg)
{
	unsigned int idx = msg->payload.state.index;
	unsigned int num = msg->payload.state.num;

	debug("State.index: %u", idx);
	debug("State.num:   %u", num);
	vdev->vq[idx].shadow_avail_idx = vdev->vq[idx].last_avail_idx = num;

	return false;
}

/**
 * vu_get_vring_base_exec() - Stops the vring and returns the current
 * 			      descriptor index or indices
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as a reply is requested
 */
static bool vu_get_vring_base_exec(struct vu_dev *vdev,
				   struct vhost_user_msg *msg)
{
	unsigned int idx = msg->payload.state.index;

	debug("State.index: %u", idx);
	msg->payload.state.num = vdev->vq[idx].last_avail_idx;
	msg->hdr.size = sizeof(msg->payload.state);

	vdev->vq[idx].started = false;

	if (vdev->vq[idx].call_fd != -1) {
		close(vdev->vq[idx].call_fd);
		vdev->vq[idx].call_fd = -1;
	}
	if (vdev->vq[idx].kick_fd != -1) {
		vu_remove_watch(vdev,  vdev->vq[idx].kick_fd);
		close(vdev->vq[idx].kick_fd);
		vdev->vq[idx].kick_fd = -1;
	}

	return true;
}

/**
 * vu_set_watch() - Add a file descriptor to the passt epoll file descriptor
 * @vdev:	vhost-user device
 * @fd:		file descriptor to add
 */
static void vu_set_watch(const struct vu_dev *vdev, int fd)
{
	union epoll_ref ref = { .type = EPOLL_TYPE_VHOST_KICK, .fd = fd };
	struct epoll_event ev = { 0 };

	ev.data.u64 = ref.u64;
	ev.events = EPOLLIN;
	epoll_ctl(vdev->context->epollfd, EPOLL_CTL_ADD, fd, &ev);
}

/**
 * vu_wait_queue() - wait new free entries in the virtqueue
 * @vq:		virtqueue to wait on
 */
static int vu_wait_queue(const struct vu_virtq *vq)
{
	eventfd_t kick_data;
	ssize_t rc;
	int status;

	/* wait the kernel to put new entries in the queue */

	status = fcntl(vq->kick_fd, F_GETFL);
	if (status == -1)
		return -1;

	fcntl(vq->kick_fd, F_SETFL, status & ~O_NONBLOCK);
	rc =  eventfd_read(vq->kick_fd, &kick_data);
	fcntl(vq->kick_fd, F_SETFL, status);
	if (rc == -1)
		return -1;

	return 0;
}

/**
 * vu_send() - Send a buffer to the front-end using the RX virtqueue
 * @vdev:	vhost-user device
 * @buf:	address of the buffer
 * @size:	size of the buffer
 *
 * Return: number of bytes sent, -1 if there is an error
 */
int vu_send(struct vu_dev *vdev, const void *buf, size_t size)
{
	size_t hdrlen = vdev->hdrlen;
	struct vu_virtq *vq = &vdev->vq[VHOST_USER_RX_QUEUE];
	struct vu_virtq_element elem[VIRTQUEUE_MAX_SIZE];
	struct iovec in_sg[VIRTQUEUE_MAX_SIZE];
	size_t lens[VIRTQUEUE_MAX_SIZE];
	size_t offset;
	int i, j;
	__virtio16 *num_buffers_ptr;
	int in_sg_count;

	debug("vu_send size %zu hdrlen %zu", size, hdrlen);

	if (!vu_queue_enabled(vq) || !vu_queue_started(vq)) {
		err("Got packet, but no available descriptors on RX virtq.");
		return 0;
	}

	offset = 0;
	i = 0;
	num_buffers_ptr = NULL;
	in_sg_count = 0;
	while (offset < size) {
		size_t len;
		int total;
		int ret;

		total = 0;

		if (i == ARRAY_SIZE(elem) ||
		    in_sg_count == ARRAY_SIZE(in_sg)) {
			err("virtio-net unexpected long buffer chain");
			goto err;
		}

		elem[i].out_num = 0;
		elem[i].out_sg = NULL;
		elem[i].in_num = ARRAY_SIZE(in_sg) - in_sg_count;
		elem[i].in_sg = &in_sg[in_sg_count];

		ret = vu_queue_pop(vdev, vq, &elem[i]);
		if (ret < 0) {
			if (vu_wait_queue(vq) != -1)
				continue;
			if (i) {
				err("virtio-net unexpected empty queue: "
				    "i %d mergeable %d offset %zd, size %zd, "
				    "features 0x%" PRIx64,
				    i, vu_has_feature(vdev,
						      VIRTIO_NET_F_MRG_RXBUF),
				    offset, size, vdev->features);
			}
			offset = -1;
			goto err;
		}
		in_sg_count += elem[i].in_num;

		if (elem[i].in_num < 1) {
			err("virtio-net receive queue contains no in buffers");
			vu_queue_detach_element(vdev, vq, elem[i].index, 0);
			offset = -1;
			goto err;
		}

		if (i == 0) {
			struct virtio_net_hdr hdr = {
				.flags = VIRTIO_NET_HDR_F_DATA_VALID,
				.gso_type = VIRTIO_NET_HDR_GSO_NONE,
			};

			ASSERT(offset == 0);
			ASSERT(elem[i].in_sg[0].iov_len >= hdrlen);

			len = iov_from_buf(elem[i].in_sg, elem[i].in_num, 0,
					   &hdr, sizeof(hdr));

			num_buffers_ptr = (__virtio16 *)((char *)elem[i].in_sg[0].iov_base +
							 len);

			total += hdrlen;
		}

		len = iov_from_buf(elem[i].in_sg, elem[i].in_num, total,
				   (char *)buf + offset, size - offset);

		total += len;
		offset += len;

		/* If buffers can't be merged, at this point we
		 * must have consumed the complete packet.
		 * Otherwise, drop it.
		 */
		if (!vu_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF) &&
		    offset < size) {
			vu_queue_unpop(vdev, vq, elem[i].index, total);
			goto err;
		}

		lens[i] = total;
		i++;
	}

	if (num_buffers_ptr && vu_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF))
		*num_buffers_ptr = htole16(i);

	for (j = 0; j < i; j++) {
		debug("filling total %zd idx %d", lens[j], j);
		vu_queue_fill(vq, &elem[j], lens[j], j);
	}

	vu_queue_flush(vq, i);
	vu_queue_notify(vdev, vq);

	debug("sent %zu", offset);

	return offset;
err:
	for (j = 0; j < i; j++)
		vu_queue_detach_element(vdev, vq, elem[j].index, lens[j]);

	return offset;
}

/**
 * vu_handle_tx() - Receive data from the TX virqueue
 * @vdev:	vhost-user device
 * @index:	index of the virtqueue
 */
static void vu_handle_tx(struct vu_dev *vdev, int index)
{
	struct vu_virtq *vq = &vdev->vq[index];
	int hdrlen = vdev->hdrlen;
	struct timespec now;
	struct vu_virtq_element  elem[VIRTQUEUE_MAX_SIZE];
	struct iovec out_sg[VIRTQUEUE_MAX_SIZE];
	int out_sg_count;

	int count;

	if (index % 2 != VHOST_USER_TX_QUEUE) {
		debug("index %d is not a TX queue", index);
		return;
	}

	clock_gettime(CLOCK_MONOTONIC, &now);

	tap_flush_pools();

	count = 0;
	out_sg_count = 0;
	while (1) {
		int ret;

		ASSERT(index == VHOST_USER_TX_QUEUE);

		elem[count].out_num = 1;
		elem[count].out_sg = &out_sg[out_sg_count];
		elem[count].in_num = 0;
		elem[count].in_sg = NULL;
		ret = vu_queue_pop(vdev, vq, &elem[count]);
		if (ret < 0)
			break;
		out_sg_count += elem[count].out_num;

		if (elem[count].out_num < 1) {
			debug("virtio-net header not in first element");
			break;
		}
		ASSERT(elem[count].out_num == 1);

		tap_add_packet(vdev->context,
			       elem[count].out_sg[0].iov_len - hdrlen,
			       (char *)elem[count].out_sg[0].iov_base + hdrlen);
		count++;
	}
	tap_handler(vdev->context, &now);

	if (count) {
		int i;

		for (i = 0; i < count; i++)
			vu_queue_fill(vq, &elem[i], 0, i);
		vu_queue_flush(vq, count);
		vu_queue_notify(vdev, vq);
	}
}

/**
 * vu_kick_cb() - Called on a kick event to start to receive data
 * @vdev:	vhost-user device
 * @ref:	epoll reference information
 */
void vu_kick_cb(struct vu_dev *vdev, union epoll_ref ref)
{
	eventfd_t kick_data;
	ssize_t rc;
	int idx;

	for (idx = 0; idx < VHOST_USER_MAX_QUEUES; idx++)
		if (vdev->vq[idx].kick_fd == ref.fd)
			break;

	if (idx == VHOST_USER_MAX_QUEUES)
		return;

	rc =  eventfd_read(ref.fd, &kick_data);
	if (rc == -1)
		vu_panic("kick eventfd_read(): %s", strerror(errno));

	debug("Got kick_data: %016"PRIx64" idx:%d",
	      kick_data, idx);
	if (idx % 2 == VHOST_USER_TX_QUEUE)
		vu_handle_tx(vdev, idx);
}

/**
 * vu_check_queue_msg_file() - Check if a message is valid,
 * 			       close fds if NOFD bit is set
 * @vmsg:	Vhost-user message
 */
static void vu_check_queue_msg_file(struct vhost_user_msg *msg)
{
	int idx = msg->payload.u64 & VHOST_USER_VRING_IDX_MASK;
	bool nofd = msg->payload.u64 & VHOST_USER_VRING_NOFD_MASK;

	if (idx >= VHOST_USER_MAX_QUEUES)
		vu_panic("Invalid queue index: %u", idx);

	if (nofd) {
		vmsg_close_fds(msg);
		return;
	}

	if (msg->fd_num != 1)
		vu_panic("Invalid fds in request: %d", msg->hdr.request);
}

/**
 * vu_set_vring_kick_exec() - Set the event file descriptor for adding buffers
 * 			      to the vring
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_kick_exec(struct vu_dev *vdev,
				   struct vhost_user_msg *msg)
{
	int idx = msg->payload.u64 & VHOST_USER_VRING_IDX_MASK;
	bool nofd = msg->payload.u64 & VHOST_USER_VRING_NOFD_MASK;

	debug("u64: 0x%016"PRIx64, msg->payload.u64);

	vu_check_queue_msg_file(msg);

	if (vdev->vq[idx].kick_fd != -1) {
		vu_remove_watch(vdev, vdev->vq[idx].kick_fd);
		close(vdev->vq[idx].kick_fd);
		vdev->vq[idx].kick_fd = -1;
	}

	/* cppcheck-suppress redundantAssignment */
	vdev->vq[idx].kick_fd = nofd ? -1 : msg->fds[0];
	debug("Got kick_fd: %d for vq: %d", vdev->vq[idx].kick_fd, idx);

	vdev->vq[idx].started = true;

	if (vdev->vq[idx].kick_fd != -1 && idx % 2 == VHOST_USER_TX_QUEUE) {
		vu_set_watch(vdev, vdev->vq[idx].kick_fd);
		debug("Waiting for kicks on fd: %d for vq: %d",
		      vdev->vq[idx].kick_fd, idx);
	}

	return false;
}

/**
 * vu_set_vring_call_exec() - Set the event file descriptor to signal when
 * 			      buffers are used
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_call_exec(struct vu_dev *vdev,
				   struct vhost_user_msg *msg)
{
	int idx = msg->payload.u64 & VHOST_USER_VRING_IDX_MASK;
	bool nofd = msg->payload.u64 & VHOST_USER_VRING_NOFD_MASK;

	debug("u64: 0x%016"PRIx64, msg->payload.u64);

	vu_check_queue_msg_file(msg);

	if (vdev->vq[idx].call_fd != -1) {
		close(vdev->vq[idx].call_fd);
		vdev->vq[idx].call_fd = -1;
	}

	/* cppcheck-suppress redundantAssignment */
	vdev->vq[idx].call_fd = nofd ? -1 : msg->fds[0];

	/* in case of I/O hang after reconnecting */
	if (vdev->vq[idx].call_fd != -1)
		eventfd_write(msg->fds[0], 1);

	debug("Got call_fd: %d for vq: %d", vdev->vq[idx].call_fd, idx);

	return false;
}

/**
 * vu_set_vring_err_exec() - Set the event file descriptor to signal when
 * 			     error occurs
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_err_exec(struct vu_dev *vdev,
				  struct vhost_user_msg *msg)
{
	int idx = msg->payload.u64 & VHOST_USER_VRING_IDX_MASK;
	bool nofd = msg->payload.u64 & VHOST_USER_VRING_NOFD_MASK;

	debug("u64: 0x%016"PRIx64, msg->payload.u64);

	vu_check_queue_msg_file(msg);

	if (vdev->vq[idx].err_fd != -1) {
		close(vdev->vq[idx].err_fd);
		vdev->vq[idx].err_fd = -1;
	}

	/* cppcheck-suppress redundantAssignment */
	vdev->vq[idx].err_fd = nofd ? -1 : msg->fds[0];

	return false;
}

/**
 * vu_get_protocol_features_exec() - Provide the protocol (vhost-user) features
 * 				     to the front-end
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as a reply is requested
 */
static bool vu_get_protocol_features_exec(struct vhost_user_msg *msg)
{
	uint64_t features = 1ULL << VHOST_USER_PROTOCOL_F_REPLY_ACK;

	vmsg_set_reply_u64(msg, features);

	return true;
}

/**
 * vu_set_protocol_features_exec() - Enable protocol (vhost-user) features 
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_protocol_features_exec(struct vu_dev *vdev,
					  struct vhost_user_msg *msg)
{
	uint64_t features = msg->payload.u64;

	debug("u64: 0x%016"PRIx64, features);

	vdev->protocol_features = msg->payload.u64;

	if (vu_has_protocol_feature(vdev,
				    VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS) &&
	    (!vu_has_protocol_feature(vdev, VHOST_USER_PROTOCOL_F_BACKEND_REQ) ||
	     !vu_has_protocol_feature(vdev, VHOST_USER_PROTOCOL_F_REPLY_ACK))) {
	/*
	 * The use case for using messages for kick/call is simulation, to make
	 * the kick and call synchronous. To actually get that behaviour, both
	 * of the other features are required.
	 * Theoretically, one could use only kick messages, or do them without
	 * having F_REPLY_ACK, but too many (possibly pending) messages on the
	 * socket will eventually cause the master to hang, to avoid this in
	 * scenarios where not desired enforce that the settings are in a way
	 * that actually enables the simulation case.
	 */
		vu_panic("F_IN_BAND_NOTIFICATIONS requires F_BACKEND_REQ && F_REPLY_ACK");
		return false;
	}

	return false;
}

/**
 * vu_get_queue_num_exec() - Tell how many queues we support
 * @vmsg:	Vhost-user message
 *
 * Return: true as a reply is requested
 */
static bool vu_get_queue_num_exec(struct vhost_user_msg *msg)
{
	vmsg_set_reply_u64(msg, VHOST_USER_MAX_QUEUES);
	return true;
}

/**
 * vu_set_vring_enable_exec() - Enable or disable corresponding vring
 * @vdev:	Vhost-user device
 * @vmsg:	Vhost-user message
 *
 * Return: false as no reply is requested
 */
static bool vu_set_vring_enable_exec(struct vu_dev *vdev,
				     struct vhost_user_msg *msg)
{
	unsigned int idx = msg->payload.state.index;
	unsigned int enable = msg->payload.state.num;

	debug("State.index:  %u", idx);
	debug("State.enable: %u", enable);

	if (idx >= VHOST_USER_MAX_QUEUES)
		vu_panic("Invalid vring_enable index: %u", idx);

	vdev->vq[idx].enable = enable;
	return false;
}

/**
 * vu_init() - Initialize vhost-user device structure
 * @c:		execution context
 * @vdev:	vhost-user device
 */
void vu_init(struct ctx *c, struct vu_dev *vdev)
{
	int i;

	c->vdev = vdev;
	vdev->context = c;
	vdev->hdrlen = 0;
	for (i = 0; i < VHOST_USER_MAX_QUEUES; i++)
		vdev->vq[i] = (struct vu_virtq){
			.call_fd = -1,
			.kick_fd = -1,
			.err_fd = -1,
			.notification = true,
		};
}

/**
 * vu_cleanup() - Reset vhost-user device
 * @vdev:	vhost-user device
 */
void vu_cleanup(struct vu_dev *vdev)
{
	unsigned int i;

	for (i = 0; i < VHOST_USER_MAX_QUEUES; i++) {
		struct vu_virtq *vq = &vdev->vq[i];

		vq->started = false;
		vq->notification = true;

		if (vq->call_fd != -1) {
			close(vq->call_fd);
			vq->call_fd = -1;
		}
		if (vq->err_fd != -1) {
			close(vq->err_fd);
			vq->err_fd = -1;
		}
		if (vq->kick_fd != -1) {
			vu_remove_watch(vdev,  vq->kick_fd);
			close(vq->kick_fd);
			vq->kick_fd = -1;
		}

		vq->vring.desc = 0;
		vq->vring.used = 0;
		vq->vring.avail = 0;
	}
	vdev->hdrlen = 0;

	for (i = 0; i < vdev->nregions; i++) {
		const struct vu_dev_region *r = &vdev->regions[i];
		/* NOLINTNEXTLINE(performance-no-int-to-ptr) */
		void *m = (void *)r->mmap_addr;

		if (m)
			munmap(m, r->size + r->mmap_offset);
	}
	vdev->nregions = 0;
}

/**
 * vu_sock_reset() - Reset connection socket
 * @vdev:	vhost-user device
 */
static void vu_sock_reset(struct vu_dev *vdev)
{
	tap_sock_reset(vdev->context);
}

/**
 * tap_handler_vu() - Packet handler for vhost-user
 * @vdev:	vhost-user device
 * @fd:		vhost-user message socket
 * @events:	epoll events
 */
void tap_handler_vu(struct vu_dev *vdev, int fd, uint32_t events)
{
	struct vhost_user_msg msg = { 0 };
	bool need_reply, reply_requested;
	int ret;

	if (events & (EPOLLRDHUP | EPOLLHUP | EPOLLERR)) {
		vu_sock_reset(vdev);
		return;
	}


	ret = vu_message_read_default(fd, &msg);
	if (ret < 0)
		vu_panic("Error while recvmsg: %s", strerror(errno));
	if (ret == 0) {
		vu_sock_reset(vdev);
		return;
	}
	debug("================ Vhost user message ================");
	debug("Request: %s (%d)", vu_request_to_string(msg.hdr.request),
		msg.hdr.request);
	debug("Flags:   0x%x", msg.hdr.flags);
	debug("Size:    %u", msg.hdr.size);

	need_reply = msg.hdr.flags & VHOST_USER_NEED_REPLY_MASK;
	switch (msg.hdr.request) {
	case VHOST_USER_GET_FEATURES:
		reply_requested = vu_get_features_exec(&msg);
		break;
	case VHOST_USER_SET_FEATURES:
		reply_requested = vu_set_features_exec(vdev, &msg);
		break;
	case VHOST_USER_GET_PROTOCOL_FEATURES:
		reply_requested = vu_get_protocol_features_exec(&msg);
		break;
	case VHOST_USER_SET_PROTOCOL_FEATURES:
		reply_requested = vu_set_protocol_features_exec(vdev, &msg);
		break;
	case VHOST_USER_GET_QUEUE_NUM:
		reply_requested = vu_get_queue_num_exec(&msg);
		break;
	case VHOST_USER_SET_OWNER:
		reply_requested = vu_set_owner_exec();
		break;
	case VHOST_USER_SET_MEM_TABLE:
		reply_requested = vu_set_mem_table_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_NUM:
		reply_requested = vu_set_vring_num_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_ADDR:
		reply_requested = vu_set_vring_addr_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_BASE:
		reply_requested = vu_set_vring_base_exec(vdev, &msg);
		break;
	case VHOST_USER_GET_VRING_BASE:
		reply_requested = vu_get_vring_base_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_KICK:
		reply_requested = vu_set_vring_kick_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_CALL:
		reply_requested = vu_set_vring_call_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_ERR:
		reply_requested = vu_set_vring_err_exec(vdev, &msg);
		break;
	case VHOST_USER_SET_VRING_ENABLE:
		reply_requested = vu_set_vring_enable_exec(vdev, &msg);
		break;
	case VHOST_USER_NONE:
		vu_cleanup(vdev);
		return;
	default:
		vu_panic("Unhandled request: %d", msg.hdr.request);
		return;
	}

	if (!reply_requested && need_reply) {
		msg.payload.u64 = 0;
		msg.hdr.flags = 0;
		msg.hdr.size = sizeof(msg.payload.u64);
		msg.fd_num = 0;
		reply_requested = true;
	}

	if (reply_requested)
		vu_send_reply(fd, &msg);
}