[ceph.git] / ceph / src / seastar / dpdk / drivers / net / virtio / virtio_user / vhost_user.c

/*-
 *   BSD LICENSE
 *
 *   Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
 *   All rights reserved.
 *
 *   Redistribution and use in source and binary forms, with or without
 *   modification, are permitted provided that the following conditions
 *   are met:
 *
 *     * Redistributions of source code must retain the above copyright
 *       notice, this list of conditions and the following disclaimer.
 *     * Redistributions in binary form must reproduce the above copyright
 *       notice, this list of conditions and the following disclaimer in
 *       the documentation and/or other materials provided with the
 *       distribution.
 *     * Neither the name of Intel Corporation nor the names of its
 *       contributors may be used to endorse or promote products derived
 *       from this software without specific prior written permission.
 *
 *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
 *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <sys/socket.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/un.h>
#include <string.h>
#include <errno.h>

#include "vhost.h"
#include "virtio_user_dev.h"

/* The version of the protocol we support */
#define VHOST_USER_VERSION    0x1

#define VHOST_MEMORY_MAX_NREGIONS 8
struct vhost_memory {
	uint32_t nregions;
	uint32_t padding;
	struct vhost_memory_region regions[VHOST_MEMORY_MAX_NREGIONS];
};

struct vhost_user_msg {
	enum vhost_user_request request;

#define VHOST_USER_VERSION_MASK     0x3
#define VHOST_USER_REPLY_MASK       (0x1 << 2)
	uint32_t flags;
	uint32_t size; /* the following payload size */
	union {
#define VHOST_USER_VRING_IDX_MASK   0xff
#define VHOST_USER_VRING_NOFD_MASK  (0x1 << 8)
		uint64_t u64;
		struct vhost_vring_state state;
		struct vhost_vring_addr addr;
		struct vhost_memory memory;
	} payload;
	int fds[VHOST_MEMORY_MAX_NREGIONS];
} __attribute((packed));

#define VHOST_USER_HDR_SIZE offsetof(struct vhost_user_msg, payload.u64)
#define VHOST_USER_PAYLOAD_SIZE \
	(sizeof(struct vhost_user_msg) - VHOST_USER_HDR_SIZE)

static int
vhost_user_write(int fd, void *buf, int len, int *fds, int fd_num)
{
	int r;
	struct msghdr msgh;
	struct iovec iov;
	size_t fd_size = fd_num * sizeof(int);
	char control[CMSG_SPACE(fd_size)];
	struct cmsghdr *cmsg;

	memset(&msgh, 0, sizeof(msgh));
	memset(control, 0, sizeof(control));

	iov.iov_base = (uint8_t *)buf;
	iov.iov_len = len;

	msgh.msg_iov = &iov;
	msgh.msg_iovlen = 1;
	msgh.msg_control = control;
	msgh.msg_controllen = sizeof(control);

	cmsg = CMSG_FIRSTHDR(&msgh);
	cmsg->cmsg_len = CMSG_LEN(fd_size);
	cmsg->cmsg_level = SOL_SOCKET;
	cmsg->cmsg_type = SCM_RIGHTS;
	memcpy(CMSG_DATA(cmsg), fds, fd_size);

	do {
		r = sendmsg(fd, &msgh, 0);
	} while (r < 0 && errno == EINTR);

	return r;
}

static int
vhost_user_read(int fd, struct vhost_user_msg *msg)
{
	uint32_t valid_flags = VHOST_USER_REPLY_MASK | VHOST_USER_VERSION;
	int ret, sz_hdr = VHOST_USER_HDR_SIZE, sz_payload;

	ret = recv(fd, (void *)msg, sz_hdr, 0);
	if (ret < sz_hdr) {
		PMD_DRV_LOG(ERR, "Failed to recv msg hdr: %d instead of %d.",
			    ret, sz_hdr);
		goto fail;
	}

	/* validate msg flags */
	if (msg->flags != (valid_flags)) {
		PMD_DRV_LOG(ERR, "Failed to recv msg: flags %x instead of %x.",
			    msg->flags, valid_flags);
		goto fail;
	}

	sz_payload = msg->size;
	if (sz_payload) {
		ret = recv(fd, (void *)((char *)msg + sz_hdr), sz_payload, 0);
		if (ret < sz_payload) {
			PMD_DRV_LOG(ERR,
				"Failed to recv msg payload: %d instead of %d.",
				ret, msg->size);
			goto fail;
		}
	}

	return 0;

fail:
	return -1;
}

struct hugepage_file_info {
	uint64_t addr;            /**< virtual addr */
	size_t   size;            /**< the file size */
	char     path[PATH_MAX];  /**< path to backing file */
};

/* Two possible options:
 * 1. Match HUGEPAGE_INFO_FMT to find the file storing struct hugepage_file
 * array. This is simple but cannot be used in secondary process because
 * secondary process will close and munmap that file.
 * 2. Match HUGEFILE_FMT to find hugepage files directly.
 *
 * We choose option 2.
 */
static int
get_hugepage_file_info(struct hugepage_file_info huges[], int max)
{
	int idx;
	FILE *f;
	char buf[BUFSIZ], *tmp, *tail;
	char *str_underline, *str_start;
	int huge_index;
	uint64_t v_start, v_end;

	f = fopen("/proc/self/maps", "r");
	if (!f) {
		PMD_DRV_LOG(ERR, "cannot open /proc/self/maps");
		return -1;
	}

	idx = 0;
	while (fgets(buf, sizeof(buf), f) != NULL) {
		if (sscanf(buf, "%" PRIx64 "-%" PRIx64, &v_start, &v_end) < 2) {
			PMD_DRV_LOG(ERR, "Failed to parse address");
			goto error;
		}

		tmp = strchr(buf, ' ') + 1; /** skip address */
		tmp = strchr(tmp, ' ') + 1; /** skip perm */
		tmp = strchr(tmp, ' ') + 1; /** skip offset */
		tmp = strchr(tmp, ' ') + 1; /** skip dev */
		tmp = strchr(tmp, ' ') + 1; /** skip inode */
		while (*tmp == ' ')         /** skip spaces */
			tmp++;
		tail = strrchr(tmp, '\n');  /** remove newline if exists */
		if (tail)
			*tail = '\0';

		/* Match HUGEFILE_FMT, aka "%s/%smap_%d",
		 * which is defined in eal_filesystem.h
		 */
		str_underline = strrchr(tmp, '_');
		if (!str_underline)
			continue;

		str_start = str_underline - strlen("map");
		if (str_start < tmp)
			continue;

		if (sscanf(str_start, "map_%d", &huge_index) != 1)
			continue;

		if (idx >= max) {
			PMD_DRV_LOG(ERR, "Exceed maximum of %d", max);
			goto error;
		}
		huges[idx].addr = v_start;
		huges[idx].size = v_end - v_start;
		snprintf(huges[idx].path, PATH_MAX, "%s", tmp);
		idx++;
	}

	fclose(f);
	return idx;

error:
	fclose(f);
	return -1;
}

static int
prepare_vhost_memory_user(struct vhost_user_msg *msg, int fds[])
{
	int i, num;
	struct hugepage_file_info huges[VHOST_MEMORY_MAX_NREGIONS];
	struct vhost_memory_region *mr;

	num = get_hugepage_file_info(huges, VHOST_MEMORY_MAX_NREGIONS);
	if (num < 0) {
		PMD_INIT_LOG(ERR, "Failed to prepare memory for vhost-user");
		return -1;
	}

	for (i = 0; i < num; ++i) {
		mr = &msg->payload.memory.regions[i];
		mr->guest_phys_addr = huges[i].addr; /* use vaddr! */
		mr->userspace_addr = huges[i].addr;
		mr->memory_size = huges[i].size;
		mr->mmap_offset = 0;
		fds[i] = open(huges[i].path, O_RDWR);
	}

	msg->payload.memory.nregions = num;
	msg->payload.memory.padding = 0;

	return 0;
}

static struct vhost_user_msg m;

const char * const vhost_msg_strings[] = {
	[VHOST_USER_SET_OWNER] = "VHOST_SET_OWNER",
	[VHOST_USER_RESET_OWNER] = "VHOST_RESET_OWNER",
	[VHOST_USER_SET_FEATURES] = "VHOST_SET_FEATURES",
	[VHOST_USER_GET_FEATURES] = "VHOST_GET_FEATURES",
	[VHOST_USER_SET_VRING_CALL] = "VHOST_SET_VRING_CALL",
	[VHOST_USER_SET_VRING_NUM] = "VHOST_SET_VRING_NUM",
	[VHOST_USER_SET_VRING_BASE] = "VHOST_SET_VRING_BASE",
	[VHOST_USER_GET_VRING_BASE] = "VHOST_GET_VRING_BASE",
	[VHOST_USER_SET_VRING_ADDR] = "VHOST_SET_VRING_ADDR",
	[VHOST_USER_SET_VRING_KICK] = "VHOST_SET_VRING_KICK",
	[VHOST_USER_SET_MEM_TABLE] = "VHOST_SET_MEM_TABLE",
	[VHOST_USER_SET_VRING_ENABLE] = "VHOST_SET_VRING_ENABLE",
};

static int
vhost_user_sock(struct virtio_user_dev *dev,
		enum vhost_user_request req,
		void *arg)
{
	struct vhost_user_msg msg;
	struct vhost_vring_file *file = 0;
	int need_reply = 0;
	int fds[VHOST_MEMORY_MAX_NREGIONS];
	int fd_num = 0;
	int i, len;
	int vhostfd = dev->vhostfd;

	RTE_SET_USED(m);

	PMD_DRV_LOG(INFO, "%s", vhost_msg_strings[req]);

	msg.request = req;
	msg.flags = VHOST_USER_VERSION;
	msg.size = 0;

	switch (req) {
	case VHOST_USER_GET_FEATURES:
		need_reply = 1;
		break;

	case VHOST_USER_SET_FEATURES:
	case VHOST_USER_SET_LOG_BASE:
		msg.payload.u64 = *((__u64 *)arg);
		msg.size = sizeof(m.payload.u64);
		break;

	case VHOST_USER_SET_OWNER:
	case VHOST_USER_RESET_OWNER:
		break;

	case VHOST_USER_SET_MEM_TABLE:
		if (prepare_vhost_memory_user(&msg, fds) < 0)
			return -1;
		fd_num = msg.payload.memory.nregions;
		msg.size = sizeof(m.payload.memory.nregions);
		msg.size += sizeof(m.payload.memory.padding);
		msg.size += fd_num * sizeof(struct vhost_memory_region);
		break;

	case VHOST_USER_SET_LOG_FD:
		fds[fd_num++] = *((int *)arg);
		break;

	case VHOST_USER_SET_VRING_NUM:
	case VHOST_USER_SET_VRING_BASE:
	case VHOST_USER_SET_VRING_ENABLE:
		memcpy(&msg.payload.state, arg, sizeof(msg.payload.state));
		msg.size = sizeof(m.payload.state);
		break;

	case VHOST_USER_GET_VRING_BASE:
		memcpy(&msg.payload.state, arg, sizeof(msg.payload.state));
		msg.size = sizeof(m.payload.state);
		need_reply = 1;
		break;

	case VHOST_USER_SET_VRING_ADDR:
		memcpy(&msg.payload.addr, arg, sizeof(msg.payload.addr));
		msg.size = sizeof(m.payload.addr);
		break;

	case VHOST_USER_SET_VRING_KICK:
	case VHOST_USER_SET_VRING_CALL:
	case VHOST_USER_SET_VRING_ERR:
		file = arg;
		msg.payload.u64 = file->index & VHOST_USER_VRING_IDX_MASK;
		msg.size = sizeof(m.payload.u64);
		if (file->fd > 0)
			fds[fd_num++] = file->fd;
		else
			msg.payload.u64 |= VHOST_USER_VRING_NOFD_MASK;
		break;

	default:
		PMD_DRV_LOG(ERR, "trying to send unhandled msg type");
		return -1;
	}

	len = VHOST_USER_HDR_SIZE + msg.size;
	if (vhost_user_write(vhostfd, &msg, len, fds, fd_num) < 0) {
		PMD_DRV_LOG(ERR, "%s failed: %s",
			    vhost_msg_strings[req], strerror(errno));
		return -1;
	}

	if (req == VHOST_USER_SET_MEM_TABLE)
		for (i = 0; i < fd_num; ++i)
			close(fds[i]);

	if (need_reply) {
		if (vhost_user_read(vhostfd, &msg) < 0) {
			PMD_DRV_LOG(ERR, "Received msg failed: %s",
				    strerror(errno));
			return -1;
		}

		if (req != msg.request) {
			PMD_DRV_LOG(ERR, "Received unexpected msg type");
			return -1;
		}

		switch (req) {
		case VHOST_USER_GET_FEATURES:
			if (msg.size != sizeof(m.payload.u64)) {
				PMD_DRV_LOG(ERR, "Received bad msg size");
				return -1;
			}
			*((__u64 *)arg) = msg.payload.u64;
			break;
		case VHOST_USER_GET_VRING_BASE:
			if (msg.size != sizeof(m.payload.state)) {
				PMD_DRV_LOG(ERR, "Received bad msg size");
				return -1;
			}
			memcpy(arg, &msg.payload.state,
			       sizeof(struct vhost_vring_state));
			break;
		default:
			PMD_DRV_LOG(ERR, "Received unexpected msg type");
			return -1;
		}
	}

	return 0;
}

/**
 * Set up environment to talk with a vhost user backend.
 *
 * @return
 *   - (-1) if fail;
 *   - (0) if succeed.
 */
static int
vhost_user_setup(struct virtio_user_dev *dev)
{
	int fd;
	int flag;
	struct sockaddr_un un;

	fd = socket(AF_UNIX, SOCK_STREAM, 0);
	if (fd < 0) {
		PMD_DRV_LOG(ERR, "socket() error, %s", strerror(errno));
		return -1;
	}

	flag = fcntl(fd, F_GETFD);
	if (fcntl(fd, F_SETFD, flag | FD_CLOEXEC) < 0)
		PMD_DRV_LOG(WARNING, "fcntl failed, %s", strerror(errno));

	memset(&un, 0, sizeof(un));
	un.sun_family = AF_UNIX;
	snprintf(un.sun_path, sizeof(un.sun_path), "%s", dev->path);
	if (connect(fd, (struct sockaddr *)&un, sizeof(un)) < 0) {
		PMD_DRV_LOG(ERR, "connect error, %s", strerror(errno));
		close(fd);
		return -1;
	}

	dev->vhostfd = fd;
	return 0;
}

static int
vhost_user_enable_queue_pair(struct virtio_user_dev *dev,
			     uint16_t pair_idx,
			     int enable)
{
	int i;

	for (i = 0; i < 2; ++i) {
		struct vhost_vring_state state = {
			.index = pair_idx * 2 + i,
			.num   = enable,
		};

		if (vhost_user_sock(dev, VHOST_USER_SET_VRING_ENABLE, &state))
			return -1;
	}

	return 0;
}

struct virtio_user_backend_ops ops_user = {
	.setup = vhost_user_setup,
	.send_request = vhost_user_sock,
	.enable_qp = vhost_user_enable_queue_pair
};
Commit	Line	Data
7c673cae FG	1	/*-
	2	* BSD LICENSE
	3	*
	4	* Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
	5	* All rights reserved.
	6	*
	7	* Redistribution and use in source and binary forms, with or without
	8	* modification, are permitted provided that the following conditions
	9	* are met:
	10	*
	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
	16	* distribution.
	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.
	20	*
	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.
	32	*/
	33
	34	#include <sys/socket.h>
	35	#include <sys/types.h>
	36	#include <sys/stat.h>
	37	#include <unistd.h>
	38	#include <fcntl.h>
	39	#include <sys/un.h>
	40	#include <string.h>
	41	#include <errno.h>
	42
	43	#include "vhost.h"
11fdf7f2 TL	44	#include "virtio_user_dev.h"
	45
	46	/* The version of the protocol we support */
	47	#define VHOST_USER_VERSION 0x1
	48
	49	#define VHOST_MEMORY_MAX_NREGIONS 8
	50	struct vhost_memory {
	51	uint32_t nregions;
	52	uint32_t padding;
	53	struct vhost_memory_region regions[VHOST_MEMORY_MAX_NREGIONS];
	54	};
	55
	56	struct vhost_user_msg {
	57	enum vhost_user_request request;
	58
	59	#define VHOST_USER_VERSION_MASK 0x3
	60	#define VHOST_USER_REPLY_MASK (0x1 << 2)
	61	uint32_t flags;
	62	uint32_t size; /* the following payload size */
	63	union {
	64	#define VHOST_USER_VRING_IDX_MASK 0xff
	65	#define VHOST_USER_VRING_NOFD_MASK (0x1 << 8)
	66	uint64_t u64;
	67	struct vhost_vring_state state;
	68	struct vhost_vring_addr addr;
	69	struct vhost_memory memory;
	70	} payload;
	71	int fds[VHOST_MEMORY_MAX_NREGIONS];
	72	} __attribute((packed));
	73
	74	#define VHOST_USER_HDR_SIZE offsetof(struct vhost_user_msg, payload.u64)
	75	#define VHOST_USER_PAYLOAD_SIZE \
	76	(sizeof(struct vhost_user_msg) - VHOST_USER_HDR_SIZE)
7c673cae FG	77
	78	static int
	79	vhost_user_write(int fd, void buf, int len, int fds, int fd_num)
	80	{
	81	int r;
	82	struct msghdr msgh;
	83	struct iovec iov;
	84	size_t fd_size = fd_num * sizeof(int);
	85	char control[CMSG_SPACE(fd_size)];
	86	struct cmsghdr *cmsg;
	87
	88	memset(&msgh, 0, sizeof(msgh));
	89	memset(control, 0, sizeof(control));
	90
	91	iov.iov_base = (uint8_t *)buf;
	92	iov.iov_len = len;
	93
	94	msgh.msg_iov = &iov;
	95	msgh.msg_iovlen = 1;
	96	msgh.msg_control = control;
	97	msgh.msg_controllen = sizeof(control);
	98
	99	cmsg = CMSG_FIRSTHDR(&msgh);
	100	cmsg->cmsg_len = CMSG_LEN(fd_size);
	101	cmsg->cmsg_level = SOL_SOCKET;
	102	cmsg->cmsg_type = SCM_RIGHTS;
	103	memcpy(CMSG_DATA(cmsg), fds, fd_size);
	104
	105	do {
	106	r = sendmsg(fd, &msgh, 0);
	107	} while (r < 0 && errno == EINTR);
	108
	109	return r;
	110	}
	111
	112	static int
	113	vhost_user_read(int fd, struct vhost_user_msg *msg)
	114	{
	115	uint32_t valid_flags = VHOST_USER_REPLY_MASK \| VHOST_USER_VERSION;
	116	int ret, sz_hdr = VHOST_USER_HDR_SIZE, sz_payload;
	117
	118	ret = recv(fd, (void *)msg, sz_hdr, 0);
	119	if (ret < sz_hdr) {
	120	PMD_DRV_LOG(ERR, "Failed to recv msg hdr: %d instead of %d.",
	121	ret, sz_hdr);
	122	goto fail;
	123	}
	124
	125	/* validate msg flags */
	126	if (msg->flags != (valid_flags)) {
	127	PMD_DRV_LOG(ERR, "Failed to recv msg: flags %x instead of %x.",
	128	msg->flags, valid_flags);
	129	goto fail;
	130	}
	131
	132	sz_payload = msg->size;
	133	if (sz_payload) {
	134	ret = recv(fd, (void )((char )msg + sz_hdr), sz_payload, 0);
	135	if (ret < sz_payload) {
	136	PMD_DRV_LOG(ERR,
	137	"Failed to recv msg payload: %d instead of %d.",
	138	ret, msg->size);
	139	goto fail;
	140	}
141	}
142
143	return 0;
144
145	fail:
146	return -1;
147	}
148
149	struct hugepage_file_info {
150	uint64_t addr; /*< virtual addr /
151	size_t size; /*< the file size /
152	char path[PATH_MAX]; /*< path to backing file /
153	};
154
155	/* Two possible options:
156	* 1. Match HUGEPAGE_INFO_FMT to find the file storing struct hugepage_file
157	* array. This is simple but cannot be used in secondary process because
158	* secondary process will close and munmap that file.
159	* 2. Match HUGEFILE_FMT to find hugepage files directly.
160	*
161	* We choose option 2.
162	*/
163	static int
164	get_hugepage_file_info(struct hugepage_file_info huges[], int max)
165	{
166	int idx;
167	FILE *f;
168	char buf[BUFSIZ], tmp, tail;
169	char str_underline, str_start;
170	int huge_index;
171	uint64_t v_start, v_end;
172
173	f = fopen("/proc/self/maps", "r");
174	if (!f) {
175	PMD_DRV_LOG(ERR, "cannot open /proc/self/maps");
176	return -1;
177	}
178
179	idx = 0;
180	while (fgets(buf, sizeof(buf), f) != NULL) {
181	if (sscanf(buf, "%" PRIx64 "-%" PRIx64, &v_start, &v_end) < 2) {
182	PMD_DRV_LOG(ERR, "Failed to parse address");
183	goto error;
184	}
185
186	tmp = strchr(buf, ' ') + 1; /** skip address */
187	tmp = strchr(tmp, ' ') + 1; /** skip perm */
188	tmp = strchr(tmp, ' ') + 1; /** skip offset */
189	tmp = strchr(tmp, ' ') + 1; /** skip dev */
190	tmp = strchr(tmp, ' ') + 1; /** skip inode */
191	while (tmp == ' ') /* skip spaces */
192	tmp++;
193	tail = strrchr(tmp, '\n'); /** remove newline if exists */
194	if (tail)
195	*tail = '\0';
196
197	/* Match HUGEFILE_FMT, aka "%s/%smap_%d",
198	* which is defined in eal_filesystem.h
199	*/
200	str_underline = strrchr(tmp, '_');
201	if (!str_underline)
202	continue;
203
204	str_start = str_underline - strlen("map");
205	if (str_start < tmp)
206	continue;
207
208	if (sscanf(str_start, "map_%d", &huge_index) != 1)
209	continue;
210
211	if (idx >= max) {
212	PMD_DRV_LOG(ERR, "Exceed maximum of %d", max);
213	goto error;
214	}
215	huges[idx].addr = v_start;
216	huges[idx].size = v_end - v_start;
217	snprintf(huges[idx].path, PATH_MAX, "%s", tmp);
218	idx++;
219	}
220
221	fclose(f);
222	return idx;
223
224	error:
225	fclose(f);
226	return -1;
227	}
228
229	static int
230	prepare_vhost_memory_user(struct vhost_user_msg *msg, int fds[])
231	{
232	int i, num;
233	struct hugepage_file_info huges[VHOST_MEMORY_MAX_NREGIONS];
234	struct vhost_memory_region *mr;
235
236	num = get_hugepage_file_info(huges, VHOST_MEMORY_MAX_NREGIONS);
237	if (num < 0) {
238	PMD_INIT_LOG(ERR, "Failed to prepare memory for vhost-user");
239	return -1;
240	}
241
242	for (i = 0; i < num; ++i) {
243	mr = &msg->payload.memory.regions[i];
244	mr->guest_phys_addr = huges[i].addr; /* use vaddr! */
245	mr->userspace_addr = huges[i].addr;
246	mr->memory_size = huges[i].size;
247	mr->mmap_offset = 0;
248	fds[i] = open(huges[i].path, O_RDWR);
249	}
250
251	msg->payload.memory.nregions = num;
252	msg->payload.memory.padding = 0;
253
254	return 0;
255	}
256
257	static struct vhost_user_msg m;
258
11fdf7f2 TL	259	const char * const vhost_msg_strings[] = {
	260	[VHOST_USER_SET_OWNER] = "VHOST_SET_OWNER",
	261	[VHOST_USER_RESET_OWNER] = "VHOST_RESET_OWNER",
	262	[VHOST_USER_SET_FEATURES] = "VHOST_SET_FEATURES",
	263	[VHOST_USER_GET_FEATURES] = "VHOST_GET_FEATURES",
	264	[VHOST_USER_SET_VRING_CALL] = "VHOST_SET_VRING_CALL",
	265	[VHOST_USER_SET_VRING_NUM] = "VHOST_SET_VRING_NUM",
	266	[VHOST_USER_SET_VRING_BASE] = "VHOST_SET_VRING_BASE",
	267	[VHOST_USER_GET_VRING_BASE] = "VHOST_GET_VRING_BASE",
	268	[VHOST_USER_SET_VRING_ADDR] = "VHOST_SET_VRING_ADDR",
	269	[VHOST_USER_SET_VRING_KICK] = "VHOST_SET_VRING_KICK",
	270	[VHOST_USER_SET_MEM_TABLE] = "VHOST_SET_MEM_TABLE",
	271	[VHOST_USER_SET_VRING_ENABLE] = "VHOST_SET_VRING_ENABLE",
7c673cae FG	272	};
7c673cae FG	273
11fdf7f2 TL	274	static int
	275	vhost_user_sock(struct virtio_user_dev *dev,
	276	enum vhost_user_request req,
	277	void *arg)
7c673cae FG	278	{
	279	struct vhost_user_msg msg;
	280	struct vhost_vring_file *file = 0;
	281	int need_reply = 0;
	282	int fds[VHOST_MEMORY_MAX_NREGIONS];
	283	int fd_num = 0;
	284	int i, len;
11fdf7f2	285	int vhostfd = dev->vhostfd;
7c673cae FG	286
7c673cae FG	287	RTE_SET_USED(m);
7c673cae FG	288
	289	PMD_DRV_LOG(INFO, "%s", vhost_msg_strings[req]);
	290
	291	msg.request = req;
	292	msg.flags = VHOST_USER_VERSION;
	293	msg.size = 0;
	294
	295	switch (req) {
	296	case VHOST_USER_GET_FEATURES:
	297	need_reply = 1;
	298	break;
	299
	300	case VHOST_USER_SET_FEATURES:
	301	case VHOST_USER_SET_LOG_BASE:
	302	msg.payload.u64 = ((__u64 )arg);
	303	msg.size = sizeof(m.payload.u64);
	304	break;
	305
	306	case VHOST_USER_SET_OWNER:
	307	case VHOST_USER_RESET_OWNER:
	308	break;
	309
	310	case VHOST_USER_SET_MEM_TABLE:
	311	if (prepare_vhost_memory_user(&msg, fds) < 0)
	312	return -1;
	313	fd_num = msg.payload.memory.nregions;
	314	msg.size = sizeof(m.payload.memory.nregions);
	315	msg.size += sizeof(m.payload.memory.padding);
	316	msg.size += fd_num * sizeof(struct vhost_memory_region);
	317	break;
	318
	319	case VHOST_USER_SET_LOG_FD:
	320	fds[fd_num++] = ((int )arg);
	321	break;
	322
	323	case VHOST_USER_SET_VRING_NUM:
	324	case VHOST_USER_SET_VRING_BASE:
	325	case VHOST_USER_SET_VRING_ENABLE:
	326	memcpy(&msg.payload.state, arg, sizeof(msg.payload.state));
	327	msg.size = sizeof(m.payload.state);
	328	break;
	329
	330	case VHOST_USER_GET_VRING_BASE:
	331	memcpy(&msg.payload.state, arg, sizeof(msg.payload.state));
	332	msg.size = sizeof(m.payload.state);
	333	need_reply = 1;
	334	break;
	335
	336	case VHOST_USER_SET_VRING_ADDR:
	337	memcpy(&msg.payload.addr, arg, sizeof(msg.payload.addr));
	338	msg.size = sizeof(m.payload.addr);
	339	break;
	340
	341	case VHOST_USER_SET_VRING_KICK:
	342	case VHOST_USER_SET_VRING_CALL:
	343	case VHOST_USER_SET_VRING_ERR:
	344	file = arg;
	345	msg.payload.u64 = file->index & VHOST_USER_VRING_IDX_MASK;
	346	msg.size = sizeof(m.payload.u64);
	347	if (file->fd > 0)
	348	fds[fd_num++] = file->fd;
	349	else
	350	msg.payload.u64 \|= VHOST_USER_VRING_NOFD_MASK;
	351	break;
352
353	default:
354	PMD_DRV_LOG(ERR, "trying to send unhandled msg type");
355	return -1;
356	}
357
358	len = VHOST_USER_HDR_SIZE + msg.size;
359	if (vhost_user_write(vhostfd, &msg, len, fds, fd_num) < 0) {
360	PMD_DRV_LOG(ERR, "%s failed: %s",
361	vhost_msg_strings[req], strerror(errno));
362	return -1;
363	}
364
365	if (req == VHOST_USER_SET_MEM_TABLE)
366	for (i = 0; i < fd_num; ++i)
367	close(fds[i]);
368
369	if (need_reply) {
370	if (vhost_user_read(vhostfd, &msg) < 0) {
371	PMD_DRV_LOG(ERR, "Received msg failed: %s",
372	strerror(errno));
373	return -1;
374	}
375
376	if (req != msg.request) {
377	PMD_DRV_LOG(ERR, "Received unexpected msg type");
378	return -1;
379	}
380
381	switch (req) {
382	case VHOST_USER_GET_FEATURES:
383	if (msg.size != sizeof(m.payload.u64)) {
384	PMD_DRV_LOG(ERR, "Received bad msg size");
385	return -1;
386	}
387	((__u64 )arg) = msg.payload.u64;
388	break;
389	case VHOST_USER_GET_VRING_BASE:
390	if (msg.size != sizeof(m.payload.state)) {
391	PMD_DRV_LOG(ERR, "Received bad msg size");
392	return -1;
393	}
394	memcpy(arg, &msg.payload.state,
395	sizeof(struct vhost_vring_state));
396	break;
397	default:
398	PMD_DRV_LOG(ERR, "Received unexpected msg type");
399	return -1;
400	}
401	}
402
403	return 0;
404	}
405
406	/**
407	* Set up environment to talk with a vhost user backend.
7c673cae FG	408	*
7c673cae FG	409	* @return
11fdf7f2 TL	410	* - (-1) if fail;
11fdf7f2 TL	411	* - (0) if succeed.
7c673cae	412	*/
11fdf7f2 TL	413	static int
11fdf7f2 TL	414	vhost_user_setup(struct virtio_user_dev *dev)
7c673cae FG	415	{
	416	int fd;
	417	int flag;
	418	struct sockaddr_un un;
	419
	420	fd = socket(AF_UNIX, SOCK_STREAM, 0);
	421	if (fd < 0) {
	422	PMD_DRV_LOG(ERR, "socket() error, %s", strerror(errno));
	423	return -1;
	424	}
	425
	426	flag = fcntl(fd, F_GETFD);
	427	if (fcntl(fd, F_SETFD, flag \| FD_CLOEXEC) < 0)
	428	PMD_DRV_LOG(WARNING, "fcntl failed, %s", strerror(errno));
	429
	430	memset(&un, 0, sizeof(un));
	431	un.sun_family = AF_UNIX;
11fdf7f2	432	snprintf(un.sun_path, sizeof(un.sun_path), "%s", dev->path);
7c673cae FG	433	if (connect(fd, (struct sockaddr *)&un, sizeof(un)) < 0) {
	434	PMD_DRV_LOG(ERR, "connect error, %s", strerror(errno));
	435	close(fd);
	436	return -1;
	437	}
	438
11fdf7f2 TL	439	dev->vhostfd = fd;
11fdf7f2 TL	440	return 0;
7c673cae FG	441	}
7c673cae FG	442
11fdf7f2 TL	443	static int
	444	vhost_user_enable_queue_pair(struct virtio_user_dev *dev,
	445	uint16_t pair_idx,
	446	int enable)
7c673cae FG	447	{
	448	int i;
	449
	450	for (i = 0; i < 2; ++i) {
	451	struct vhost_vring_state state = {
	452	.index = pair_idx * 2 + i,
	453	.num = enable,
	454	};
	455
11fdf7f2	456	if (vhost_user_sock(dev, VHOST_USER_SET_VRING_ENABLE, &state))
7c673cae FG	457	return -1;
	458	}
	459
	460	return 0;
	461	}
11fdf7f2 TL	462
	463	struct virtio_user_backend_ops ops_user = {
	464	.setup = vhost_user_setup,
	465	.send_request = vhost_user_sock,
	466	.enable_qp = vhost_user_enable_queue_pair
	467	};