vhost-user.c 26.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10
/*
 * vhost-user
 *
 * Copyright (c) 2013 Virtual Open Systems Sarl.
 *
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 *
 */

P
Peter Maydell 已提交
11
#include "qemu/osdep.h"
12
#include "qapi/error.h"
13 14
#include "hw/virtio/vhost.h"
#include "hw/virtio/vhost-backend.h"
15
#include "hw/virtio/virtio-net.h"
16
#include "chardev/char-fe.h"
17 18 19 20 21 22 23 24 25 26
#include "sysemu/kvm.h"
#include "qemu/error-report.h"
#include "qemu/sockets.h"

#include <sys/ioctl.h>
#include <sys/socket.h>
#include <sys/un.h>
#include <linux/vhost.h>

#define VHOST_MEMORY_MAX_NREGIONS    8
27
#define VHOST_USER_F_PROTOCOL_FEATURES 30
28

29 30 31 32
enum VhostUserProtocolFeature {
    VHOST_USER_PROTOCOL_F_MQ = 0,
    VHOST_USER_PROTOCOL_F_LOG_SHMFD = 1,
    VHOST_USER_PROTOCOL_F_RARP = 2,
33
    VHOST_USER_PROTOCOL_F_REPLY_ACK = 3,
34
    VHOST_USER_PROTOCOL_F_NET_MTU = 4,
35
    VHOST_USER_PROTOCOL_F_SLAVE_REQ = 5,
36
    VHOST_USER_PROTOCOL_F_CROSS_ENDIAN = 6,
37 38 39 40 41

    VHOST_USER_PROTOCOL_F_MAX
};

#define VHOST_USER_PROTOCOL_FEATURE_MASK ((1 << VHOST_USER_PROTOCOL_F_MAX) - 1)
42 43 44 45 46 47

typedef enum VhostUserRequest {
    VHOST_USER_NONE = 0,
    VHOST_USER_GET_FEATURES = 1,
    VHOST_USER_SET_FEATURES = 2,
    VHOST_USER_SET_OWNER = 3,
48
    VHOST_USER_RESET_OWNER = 4,
49 50 51 52 53 54 55 56 57 58
    VHOST_USER_SET_MEM_TABLE = 5,
    VHOST_USER_SET_LOG_BASE = 6,
    VHOST_USER_SET_LOG_FD = 7,
    VHOST_USER_SET_VRING_NUM = 8,
    VHOST_USER_SET_VRING_ADDR = 9,
    VHOST_USER_SET_VRING_BASE = 10,
    VHOST_USER_GET_VRING_BASE = 11,
    VHOST_USER_SET_VRING_KICK = 12,
    VHOST_USER_SET_VRING_CALL = 13,
    VHOST_USER_SET_VRING_ERR = 14,
59 60
    VHOST_USER_GET_PROTOCOL_FEATURES = 15,
    VHOST_USER_SET_PROTOCOL_FEATURES = 16,
61
    VHOST_USER_GET_QUEUE_NUM = 17,
62
    VHOST_USER_SET_VRING_ENABLE = 18,
63
    VHOST_USER_SEND_RARP = 19,
64
    VHOST_USER_NET_SET_MTU = 20,
65
    VHOST_USER_SET_SLAVE_REQ_FD = 21,
66
    VHOST_USER_IOTLB_MSG = 22,
67
    VHOST_USER_SET_VRING_ENDIAN = 23,
68 69 70
    VHOST_USER_MAX
} VhostUserRequest;

71 72
typedef enum VhostUserSlaveRequest {
    VHOST_USER_SLAVE_NONE = 0,
73
    VHOST_USER_SLAVE_IOTLB_MSG = 1,
74 75 76
    VHOST_USER_SLAVE_MAX
}  VhostUserSlaveRequest;

77 78 79 80
typedef struct VhostUserMemoryRegion {
    uint64_t guest_phys_addr;
    uint64_t memory_size;
    uint64_t userspace_addr;
81
    uint64_t mmap_offset;
82 83 84 85 86 87 88 89
} VhostUserMemoryRegion;

typedef struct VhostUserMemory {
    uint32_t nregions;
    uint32_t padding;
    VhostUserMemoryRegion regions[VHOST_MEMORY_MAX_NREGIONS];
} VhostUserMemory;

90 91 92 93 94
typedef struct VhostUserLog {
    uint64_t mmap_size;
    uint64_t mmap_offset;
} VhostUserLog;

95 96 97 98 99
typedef struct VhostUserMsg {
    VhostUserRequest request;

#define VHOST_USER_VERSION_MASK     (0x3)
#define VHOST_USER_REPLY_MASK       (0x1<<2)
100
#define VHOST_USER_NEED_REPLY_MASK  (0x1 << 3)
101 102 103 104 105 106 107 108 109
    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;
        VhostUserMemory memory;
110
        VhostUserLog log;
111
        struct vhost_iotlb_msg iotlb;
112
    } payload;
113 114 115 116 117 118 119 120 121 122 123 124
} QEMU_PACKED VhostUserMsg;

static VhostUserMsg m __attribute__ ((unused));
#define VHOST_USER_HDR_SIZE (sizeof(m.request) \
                            + sizeof(m.flags) \
                            + sizeof(m.size))

#define VHOST_USER_PAYLOAD_SIZE (sizeof(m) - VHOST_USER_HDR_SIZE)

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

125 126
struct vhost_user {
    CharBackend *chr;
127
    int slave_fd;
128 129
};

130 131 132 133 134 135 136
static bool ioeventfd_enabled(void)
{
    return kvm_enabled() && kvm_eventfds_enabled();
}

static int vhost_user_read(struct vhost_dev *dev, VhostUserMsg *msg)
{
137 138
    struct vhost_user *u = dev->opaque;
    CharBackend *chr = u->chr;
139 140 141 142 143
    uint8_t *p = (uint8_t *) msg;
    int r, size = VHOST_USER_HDR_SIZE;

    r = qemu_chr_fe_read_all(chr, p, size);
    if (r != size) {
144 145
        error_report("Failed to read msg header. Read %d instead of %d."
                     " Original request %d.", r, size, msg->request);
146 147 148 149 150 151
        goto fail;
    }

    /* validate received flags */
    if (msg->flags != (VHOST_USER_REPLY_MASK | VHOST_USER_VERSION)) {
        error_report("Failed to read msg header."
152
                " Flags 0x%x instead of 0x%x.", msg->flags,
153 154 155 156 157 158 159
                VHOST_USER_REPLY_MASK | VHOST_USER_VERSION);
        goto fail;
    }

    /* validate message size is sane */
    if (msg->size > VHOST_USER_PAYLOAD_SIZE) {
        error_report("Failed to read msg header."
160
                " Size %d exceeds the maximum %zu.", msg->size,
161 162 163 164 165 166 167 168 169 170
                VHOST_USER_PAYLOAD_SIZE);
        goto fail;
    }

    if (msg->size) {
        p += VHOST_USER_HDR_SIZE;
        size = msg->size;
        r = qemu_chr_fe_read_all(chr, p, size);
        if (r != size) {
            error_report("Failed to read msg payload."
171
                         " Read %d instead of %d.", r, msg->size);
172 173 174 175 176 177 178 179 180 181
            goto fail;
        }
    }

    return 0;

fail:
    return -1;
}

182
static int process_message_reply(struct vhost_dev *dev,
183
                                 const VhostUserMsg *msg)
184
{
185
    VhostUserMsg msg_reply;
186

187
    if ((msg->flags & VHOST_USER_NEED_REPLY_MASK) == 0) {
188 189 190 191
        return 0;
    }

    if (vhost_user_read(dev, &msg_reply) < 0) {
192 193 194
        return -1;
    }

195
    if (msg_reply.request != msg->request) {
196 197
        error_report("Received unexpected msg type."
                     "Expected %d received %d",
198
                     msg->request, msg_reply.request);
199 200 201
        return -1;
    }

202
    return msg_reply.payload.u64 ? -1 : 0;
203 204
}

205 206 207 208
static bool vhost_user_one_time_request(VhostUserRequest request)
{
    switch (request) {
    case VHOST_USER_SET_OWNER:
209
    case VHOST_USER_RESET_OWNER:
210 211
    case VHOST_USER_SET_MEM_TABLE:
    case VHOST_USER_GET_QUEUE_NUM:
212
    case VHOST_USER_NET_SET_MTU:
213 214 215 216 217 218 219
        return true;
    default:
        return false;
    }
}

/* most non-init callers ignore the error */
220 221 222
static int vhost_user_write(struct vhost_dev *dev, VhostUserMsg *msg,
                            int *fds, int fd_num)
{
223 224
    struct vhost_user *u = dev->opaque;
    CharBackend *chr = u->chr;
225
    int ret, size = VHOST_USER_HDR_SIZE + msg->size;
226

227 228 229 230 231 232
    /*
     * For non-vring specific requests, like VHOST_USER_SET_MEM_TABLE,
     * we just need send it once in the first time. For later such
     * request, we just ignore it.
     */
    if (vhost_user_one_time_request(msg->request) && dev->vq_index != 0) {
233
        msg->flags &= ~VHOST_USER_NEED_REPLY_MASK;
234 235 236
        return 0;
    }

237
    if (qemu_chr_fe_set_msgfds(chr, fds, fd_num) < 0) {
238
        error_report("Failed to set msg fds.");
239 240
        return -1;
    }
241

242 243 244 245 246 247 248 249
    ret = qemu_chr_fe_write_all(chr, (const uint8_t *) msg, size);
    if (ret != size) {
        error_report("Failed to write msg."
                     " Wrote %d instead of %d.", ret, size);
        return -1;
    }

    return 0;
250 251
}

252 253
static int vhost_user_set_log_base(struct vhost_dev *dev, uint64_t base,
                                   struct vhost_log *log)
254
{
255 256 257 258 259 260 261
    int fds[VHOST_MEMORY_MAX_NREGIONS];
    size_t fd_num = 0;
    bool shmfd = virtio_has_feature(dev->protocol_features,
                                    VHOST_USER_PROTOCOL_F_LOG_SHMFD);
    VhostUserMsg msg = {
        .request = VHOST_USER_SET_LOG_BASE,
        .flags = VHOST_USER_VERSION,
M
Michael S. Tsirkin 已提交
262
        .payload.log.mmap_size = log->size * sizeof(*(log->log)),
263 264
        .payload.log.mmap_offset = 0,
        .size = sizeof(msg.payload.log),
265 266 267 268 269 270
    };

    if (shmfd && log->fd != -1) {
        fds[fd_num++] = log->fd;
    }

271 272 273
    if (vhost_user_write(dev, &msg, fds, fd_num) < 0) {
        return -1;
    }
274 275 276 277

    if (shmfd) {
        msg.size = 0;
        if (vhost_user_read(dev, &msg) < 0) {
278
            return -1;
279 280 281 282 283 284 285 286
        }

        if (msg.request != VHOST_USER_SET_LOG_BASE) {
            error_report("Received unexpected msg type. "
                         "Expected %d received %d",
                         VHOST_USER_SET_LOG_BASE, msg.request);
            return -1;
        }
287
    }
288 289

    return 0;
290 291
}

292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345
static int vhost_user_set_mem_table(struct vhost_dev *dev,
                                    struct vhost_memory *mem)
{
    int fds[VHOST_MEMORY_MAX_NREGIONS];
    int i, fd;
    size_t fd_num = 0;
    bool reply_supported = virtio_has_feature(dev->protocol_features,
                                              VHOST_USER_PROTOCOL_F_REPLY_ACK);

    VhostUserMsg msg = {
        .request = VHOST_USER_SET_MEM_TABLE,
        .flags = VHOST_USER_VERSION,
    };

    if (reply_supported) {
        msg.flags |= VHOST_USER_NEED_REPLY_MASK;
    }

    for (i = 0; i < dev->mem->nregions; ++i) {
        struct vhost_memory_region *reg = dev->mem->regions + i;
        ram_addr_t offset;
        MemoryRegion *mr;

        assert((uintptr_t)reg->userspace_addr == reg->userspace_addr);
        mr = memory_region_from_host((void *)(uintptr_t)reg->userspace_addr,
                                     &offset);
        fd = memory_region_get_fd(mr);
        if (fd > 0) {
            msg.payload.memory.regions[fd_num].userspace_addr = reg->userspace_addr;
            msg.payload.memory.regions[fd_num].memory_size  = reg->memory_size;
            msg.payload.memory.regions[fd_num].guest_phys_addr = reg->guest_phys_addr;
            msg.payload.memory.regions[fd_num].mmap_offset = offset;
            assert(fd_num < VHOST_MEMORY_MAX_NREGIONS);
            fds[fd_num++] = fd;
        }
    }

    msg.payload.memory.nregions = fd_num;

    if (!fd_num) {
        error_report("Failed initializing vhost-user memory map, "
                     "consider using -object memory-backend-file share=on");
        return -1;
    }

    msg.size = sizeof(msg.payload.memory.nregions);
    msg.size += sizeof(msg.payload.memory.padding);
    msg.size += fd_num * sizeof(VhostUserMemoryRegion);

    if (vhost_user_write(dev, &msg, fds, fd_num) < 0) {
        return -1;
    }

    if (reply_supported) {
346
        return process_message_reply(dev, &msg);
347 348 349 350 351
    }

    return 0;
}

352 353 354 355 356 357
static int vhost_user_set_vring_addr(struct vhost_dev *dev,
                                     struct vhost_vring_addr *addr)
{
    VhostUserMsg msg = {
        .request = VHOST_USER_SET_VRING_ADDR,
        .flags = VHOST_USER_VERSION,
358
        .payload.addr = *addr,
359
        .size = sizeof(msg.payload.addr),
360
    };
361

362 363 364
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
365

366 367
    return 0;
}
368

369 370 371
static int vhost_user_set_vring_endian(struct vhost_dev *dev,
                                       struct vhost_vring_state *ring)
{
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390
    bool cross_endian = virtio_has_feature(dev->protocol_features,
                                           VHOST_USER_PROTOCOL_F_CROSS_ENDIAN);
    VhostUserMsg msg = {
        .request = VHOST_USER_SET_VRING_ENDIAN,
        .flags = VHOST_USER_VERSION,
        .payload.state = *ring,
        .size = sizeof(msg.payload.state),
    };

    if (!cross_endian) {
        error_report("vhost-user trying to send unhandled ioctl");
        return -1;
    }

    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }

    return 0;
391
}
392

393 394 395 396 397 398 399
static int vhost_set_vring(struct vhost_dev *dev,
                           unsigned long int request,
                           struct vhost_vring_state *ring)
{
    VhostUserMsg msg = {
        .request = request,
        .flags = VHOST_USER_VERSION,
400
        .payload.state = *ring,
401
        .size = sizeof(msg.payload.state),
402 403
    };

404 405 406
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424

    return 0;
}

static int vhost_user_set_vring_num(struct vhost_dev *dev,
                                    struct vhost_vring_state *ring)
{
    return vhost_set_vring(dev, VHOST_USER_SET_VRING_NUM, ring);
}

static int vhost_user_set_vring_base(struct vhost_dev *dev,
                                     struct vhost_vring_state *ring)
{
    return vhost_set_vring(dev, VHOST_USER_SET_VRING_BASE, ring);
}

static int vhost_user_set_vring_enable(struct vhost_dev *dev, int enable)
{
425
    int i;
426

427
    if (!virtio_has_feature(dev->features, VHOST_USER_F_PROTOCOL_FEATURES)) {
428 429 430
        return -1;
    }

431 432 433 434 435 436 437 438
    for (i = 0; i < dev->nvqs; ++i) {
        struct vhost_vring_state state = {
            .index = dev->vq_index + i,
            .num   = enable,
        };

        vhost_set_vring(dev, VHOST_USER_SET_VRING_ENABLE, &state);
    }
439

440 441
    return 0;
}
442 443 444 445 446 447 448

static int vhost_user_get_vring_base(struct vhost_dev *dev,
                                     struct vhost_vring_state *ring)
{
    VhostUserMsg msg = {
        .request = VHOST_USER_GET_VRING_BASE,
        .flags = VHOST_USER_VERSION,
449
        .payload.state = *ring,
450
        .size = sizeof(msg.payload.state),
451 452
    };

453 454 455
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
456 457

    if (vhost_user_read(dev, &msg) < 0) {
458
        return -1;
459 460
    }

461 462 463 464 465
    if (msg.request != VHOST_USER_GET_VRING_BASE) {
        error_report("Received unexpected msg type. Expected %d received %d",
                     VHOST_USER_GET_VRING_BASE, msg.request);
        return -1;
    }
466

467
    if (msg.size != sizeof(msg.payload.state)) {
468 469
        error_report("Received bad msg size.");
        return -1;
470 471
    }

472
    *ring = msg.payload.state;
473

474 475 476
    return 0;
}

477 478 479
static int vhost_set_vring_file(struct vhost_dev *dev,
                                VhostUserRequest request,
                                struct vhost_vring_file *file)
480
{
481 482
    int fds[VHOST_MEMORY_MAX_NREGIONS];
    size_t fd_num = 0;
483
    VhostUserMsg msg = {
484
        .request = request,
485
        .flags = VHOST_USER_VERSION,
486
        .payload.u64 = file->index & VHOST_USER_VRING_IDX_MASK,
487
        .size = sizeof(msg.payload.u64),
488 489
    };

490 491 492
    if (ioeventfd_enabled() && file->fd > 0) {
        fds[fd_num++] = file->fd;
    } else {
493
        msg.payload.u64 |= VHOST_USER_VRING_NOFD_MASK;
494 495
    }

496 497 498
    if (vhost_user_write(dev, &msg, fds, fd_num) < 0) {
        return -1;
    }
499

500 501
    return 0;
}
502

503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519
static int vhost_user_set_vring_kick(struct vhost_dev *dev,
                                     struct vhost_vring_file *file)
{
    return vhost_set_vring_file(dev, VHOST_USER_SET_VRING_KICK, file);
}

static int vhost_user_set_vring_call(struct vhost_dev *dev,
                                     struct vhost_vring_file *file)
{
    return vhost_set_vring_file(dev, VHOST_USER_SET_VRING_CALL, file);
}

static int vhost_user_set_u64(struct vhost_dev *dev, int request, uint64_t u64)
{
    VhostUserMsg msg = {
        .request = request,
        .flags = VHOST_USER_VERSION,
520
        .payload.u64 = u64,
521
        .size = sizeof(msg.payload.u64),
522 523
    };

524 525 526
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551

    return 0;
}

static int vhost_user_set_features(struct vhost_dev *dev,
                                   uint64_t features)
{
    return vhost_user_set_u64(dev, VHOST_USER_SET_FEATURES, features);
}

static int vhost_user_set_protocol_features(struct vhost_dev *dev,
                                            uint64_t features)
{
    return vhost_user_set_u64(dev, VHOST_USER_SET_PROTOCOL_FEATURES, features);
}

static int vhost_user_get_u64(struct vhost_dev *dev, int request, uint64_t *u64)
{
    VhostUserMsg msg = {
        .request = request,
        .flags = VHOST_USER_VERSION,
    };

    if (vhost_user_one_time_request(request) && dev->vq_index != 0) {
        return 0;
552
    }
553

554 555 556
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
557 558

    if (vhost_user_read(dev, &msg) < 0) {
559
        return -1;
560 561 562 563 564 565 566 567
    }

    if (msg.request != request) {
        error_report("Received unexpected msg type. Expected %d received %d",
                     request, msg.request);
        return -1;
    }

568
    if (msg.size != sizeof(msg.payload.u64)) {
569 570 571 572
        error_report("Received bad msg size.");
        return -1;
    }

573
    *u64 = msg.payload.u64;
574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589

    return 0;
}

static int vhost_user_get_features(struct vhost_dev *dev, uint64_t *features)
{
    return vhost_user_get_u64(dev, VHOST_USER_GET_FEATURES, features);
}

static int vhost_user_set_owner(struct vhost_dev *dev)
{
    VhostUserMsg msg = {
        .request = VHOST_USER_SET_OWNER,
        .flags = VHOST_USER_VERSION,
    };

590 591 592
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
593 594 595 596 597 598 599

    return 0;
}

static int vhost_user_reset_device(struct vhost_dev *dev)
{
    VhostUserMsg msg = {
600
        .request = VHOST_USER_RESET_OWNER,
601 602 603
        .flags = VHOST_USER_VERSION,
    };

604 605 606
    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }
607

608 609 610
    return 0;
}

611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639
static void slave_read(void *opaque)
{
    struct vhost_dev *dev = opaque;
    struct vhost_user *u = dev->opaque;
    VhostUserMsg msg = { 0, };
    int size, ret = 0;

    /* Read header */
    size = read(u->slave_fd, &msg, VHOST_USER_HDR_SIZE);
    if (size != VHOST_USER_HDR_SIZE) {
        error_report("Failed to read from slave.");
        goto err;
    }

    if (msg.size > VHOST_USER_PAYLOAD_SIZE) {
        error_report("Failed to read msg header."
                " Size %d exceeds the maximum %zu.", msg.size,
                VHOST_USER_PAYLOAD_SIZE);
        goto err;
    }

    /* Read payload */
    size = read(u->slave_fd, &msg.payload, msg.size);
    if (size != msg.size) {
        error_report("Failed to read payload from slave.");
        goto err;
    }

    switch (msg.request) {
640 641 642
    case VHOST_USER_SLAVE_IOTLB_MSG:
        ret = vhost_backend_handle_iotlb_msg(dev, &msg.payload.iotlb);
        break;
643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722
    default:
        error_report("Received unexpected msg type.");
        ret = -EINVAL;
    }

    /*
     * REPLY_ACK feature handling. Other reply types has to be managed
     * directly in their request handlers.
     */
    if (msg.flags & VHOST_USER_NEED_REPLY_MASK) {
        msg.flags &= ~VHOST_USER_NEED_REPLY_MASK;
        msg.flags |= VHOST_USER_REPLY_MASK;

        msg.payload.u64 = !!ret;
        msg.size = sizeof(msg.payload.u64);

        size = write(u->slave_fd, &msg, VHOST_USER_HDR_SIZE + msg.size);
        if (size != VHOST_USER_HDR_SIZE + msg.size) {
            error_report("Failed to send msg reply to slave.");
            goto err;
        }
    }

    return;

err:
    qemu_set_fd_handler(u->slave_fd, NULL, NULL, NULL);
    close(u->slave_fd);
    u->slave_fd = -1;
    return;
}

static int vhost_setup_slave_channel(struct vhost_dev *dev)
{
    VhostUserMsg msg = {
        .request = VHOST_USER_SET_SLAVE_REQ_FD,
        .flags = VHOST_USER_VERSION,
    };
    struct vhost_user *u = dev->opaque;
    int sv[2], ret = 0;
    bool reply_supported = virtio_has_feature(dev->protocol_features,
                                              VHOST_USER_PROTOCOL_F_REPLY_ACK);

    if (!virtio_has_feature(dev->protocol_features,
                            VHOST_USER_PROTOCOL_F_SLAVE_REQ)) {
        return 0;
    }

    if (socketpair(PF_UNIX, SOCK_STREAM, 0, sv) == -1) {
        error_report("socketpair() failed");
        return -1;
    }

    u->slave_fd = sv[0];
    qemu_set_fd_handler(u->slave_fd, slave_read, NULL, dev);

    if (reply_supported) {
        msg.flags |= VHOST_USER_NEED_REPLY_MASK;
    }

    ret = vhost_user_write(dev, &msg, &sv[1], 1);
    if (ret) {
        goto out;
    }

    if (reply_supported) {
        ret = process_message_reply(dev, &msg);
    }

out:
    close(sv[1]);
    if (ret) {
        qemu_set_fd_handler(u->slave_fd, NULL, NULL, NULL);
        close(u->slave_fd);
        u->slave_fd = -1;
    }

    return ret;
}

723 724
static int vhost_user_init(struct vhost_dev *dev, void *opaque)
{
725
    uint64_t features, protocol_features;
726
    struct vhost_user *u;
727 728
    int err;

729 730
    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);

731 732
    u = g_new0(struct vhost_user, 1);
    u->chr = opaque;
733
    u->slave_fd = -1;
734
    dev->opaque = u;
735

736
    err = vhost_user_get_features(dev, &features);
737 738 739 740 741 742 743
    if (err < 0) {
        return err;
    }

    if (virtio_has_feature(features, VHOST_USER_F_PROTOCOL_FEATURES)) {
        dev->backend_features |= 1ULL << VHOST_USER_F_PROTOCOL_FEATURES;

744
        err = vhost_user_get_u64(dev, VHOST_USER_GET_PROTOCOL_FEATURES,
745
                                 &protocol_features);
746 747 748 749
        if (err < 0) {
            return err;
        }

750 751
        dev->protocol_features =
            protocol_features & VHOST_USER_PROTOCOL_FEATURE_MASK;
752
        err = vhost_user_set_protocol_features(dev, dev->protocol_features);
753 754 755
        if (err < 0) {
            return err;
        }
756 757 758

        /* query the max queues we support if backend supports Multiple Queue */
        if (dev->protocol_features & (1ULL << VHOST_USER_PROTOCOL_F_MQ)) {
759 760
            err = vhost_user_get_u64(dev, VHOST_USER_GET_QUEUE_NUM,
                                     &dev->max_queues);
761 762 763 764
            if (err < 0) {
                return err;
            }
        }
765 766 767 768 769 770 771 772 773 774

        if (virtio_has_feature(features, VIRTIO_F_IOMMU_PLATFORM) &&
                !(virtio_has_feature(dev->protocol_features,
                    VHOST_USER_PROTOCOL_F_SLAVE_REQ) &&
                 virtio_has_feature(dev->protocol_features,
                    VHOST_USER_PROTOCOL_F_REPLY_ACK))) {
            error_report("IOMMU support requires reply-ack and "
                         "slave-req protocol features.");
            return -1;
        }
775 776
    }

777 778 779 780 781 782 783 784
    if (dev->migration_blocker == NULL &&
        !virtio_has_feature(dev->protocol_features,
                            VHOST_USER_PROTOCOL_F_LOG_SHMFD)) {
        error_setg(&dev->migration_blocker,
                   "Migration disabled: vhost-user backend lacks "
                   "VHOST_USER_PROTOCOL_F_LOG_SHMFD feature.");
    }

785 786 787 788 789
    err = vhost_setup_slave_channel(dev);
    if (err < 0) {
        return err;
    }

790 791 792 793 794
    return 0;
}

static int vhost_user_cleanup(struct vhost_dev *dev)
{
795 796
    struct vhost_user *u;

797 798
    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);

799
    u = dev->opaque;
800
    if (u->slave_fd >= 0) {
801
        qemu_set_fd_handler(u->slave_fd, NULL, NULL, NULL);
802 803 804
        close(u->slave_fd);
        u->slave_fd = -1;
    }
805
    g_free(u);
806 807 808 809 810
    dev->opaque = 0;

    return 0;
}

811 812 813 814 815 816 817
static int vhost_user_get_vq_index(struct vhost_dev *dev, int idx)
{
    assert(idx >= dev->vq_index && idx < dev->vq_index + dev->nvqs);

    return idx;
}

818 819 820 821 822
static int vhost_user_memslots_limit(struct vhost_dev *dev)
{
    return VHOST_MEMORY_MAX_NREGIONS;
}

823 824 825 826 827 828 829 830
static bool vhost_user_requires_shm_log(struct vhost_dev *dev)
{
    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);

    return virtio_has_feature(dev->protocol_features,
                              VHOST_USER_PROTOCOL_F_LOG_SHMFD);
}

831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846
static int vhost_user_migration_done(struct vhost_dev *dev, char* mac_addr)
{
    VhostUserMsg msg = { 0 };

    assert(dev->vhost_ops->backend_type == VHOST_BACKEND_TYPE_USER);

    /* If guest supports GUEST_ANNOUNCE do nothing */
    if (virtio_has_feature(dev->acked_features, VIRTIO_NET_F_GUEST_ANNOUNCE)) {
        return 0;
    }

    /* if backend supports VHOST_USER_PROTOCOL_F_RARP ask it to send the RARP */
    if (virtio_has_feature(dev->protocol_features,
                           VHOST_USER_PROTOCOL_F_RARP)) {
        msg.request = VHOST_USER_SEND_RARP;
        msg.flags = VHOST_USER_VERSION;
847
        memcpy((char *)&msg.payload.u64, mac_addr, 6);
848
        msg.size = sizeof(msg.payload.u64);
849

850
        return vhost_user_write(dev, &msg, NULL, 0);
851 852 853 854
    }
    return -1;
}

855 856 857 858
static bool vhost_user_can_merge(struct vhost_dev *dev,
                                 uint64_t start1, uint64_t size1,
                                 uint64_t start2, uint64_t size2)
{
859
    ram_addr_t offset;
860 861 862
    int mfd, rfd;
    MemoryRegion *mr;

863
    mr = memory_region_from_host((void *)(uintptr_t)start1, &offset);
864
    mfd = memory_region_get_fd(mr);
865

866
    mr = memory_region_from_host((void *)(uintptr_t)start2, &offset);
867
    rfd = memory_region_get_fd(mr);
868 869 870 871

    return mfd == rfd;
}

872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895
static int vhost_user_net_set_mtu(struct vhost_dev *dev, uint16_t mtu)
{
    VhostUserMsg msg;
    bool reply_supported = virtio_has_feature(dev->protocol_features,
                                              VHOST_USER_PROTOCOL_F_REPLY_ACK);

    if (!(dev->protocol_features & (1ULL << VHOST_USER_PROTOCOL_F_NET_MTU))) {
        return 0;
    }

    msg.request = VHOST_USER_NET_SET_MTU;
    msg.payload.u64 = mtu;
    msg.size = sizeof(msg.payload.u64);
    msg.flags = VHOST_USER_VERSION;
    if (reply_supported) {
        msg.flags |= VHOST_USER_NEED_REPLY_MASK;
    }

    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -1;
    }

    /* If reply_ack supported, slave has to ack specified MTU is valid */
    if (reply_supported) {
896
        return process_message_reply(dev, &msg);
897 898 899 900 901
    }

    return 0;
}

902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
static int vhost_user_send_device_iotlb_msg(struct vhost_dev *dev,
                                            struct vhost_iotlb_msg *imsg)
{
    VhostUserMsg msg = {
        .request = VHOST_USER_IOTLB_MSG,
        .size = sizeof(msg.payload.iotlb),
        .flags = VHOST_USER_VERSION | VHOST_USER_NEED_REPLY_MASK,
        .payload.iotlb = *imsg,
    };

    if (vhost_user_write(dev, &msg, NULL, 0) < 0) {
        return -EFAULT;
    }

    return process_message_reply(dev, &msg);
}


static void vhost_user_set_iotlb_callback(struct vhost_dev *dev, int enabled)
{
    /* No-op as the receive channel is not dedicated to IOTLB messages. */
}

925 926 927
const VhostOps user_ops = {
        .backend_type = VHOST_BACKEND_TYPE_USER,
        .vhost_backend_init = vhost_user_init,
928
        .vhost_backend_cleanup = vhost_user_cleanup,
929
        .vhost_backend_memslots_limit = vhost_user_memslots_limit,
930 931 932 933 934 935 936 937 938 939 940 941 942 943 944
        .vhost_set_log_base = vhost_user_set_log_base,
        .vhost_set_mem_table = vhost_user_set_mem_table,
        .vhost_set_vring_addr = vhost_user_set_vring_addr,
        .vhost_set_vring_endian = vhost_user_set_vring_endian,
        .vhost_set_vring_num = vhost_user_set_vring_num,
        .vhost_set_vring_base = vhost_user_set_vring_base,
        .vhost_get_vring_base = vhost_user_get_vring_base,
        .vhost_set_vring_kick = vhost_user_set_vring_kick,
        .vhost_set_vring_call = vhost_user_set_vring_call,
        .vhost_set_features = vhost_user_set_features,
        .vhost_get_features = vhost_user_get_features,
        .vhost_set_owner = vhost_user_set_owner,
        .vhost_reset_device = vhost_user_reset_device,
        .vhost_get_vq_index = vhost_user_get_vq_index,
        .vhost_set_vring_enable = vhost_user_set_vring_enable,
945
        .vhost_requires_shm_log = vhost_user_requires_shm_log,
946
        .vhost_migration_done = vhost_user_migration_done,
947
        .vhost_backend_can_merge = vhost_user_can_merge,
948
        .vhost_net_set_mtu = vhost_user_net_set_mtu,
949 950
        .vhost_set_iotlb_callback = vhost_user_set_iotlb_callback,
        .vhost_send_device_iotlb_msg = vhost_user_send_device_iotlb_msg,
951
};