net/vmw_vsock/virtio_transport_common.c

   1 // SPDX-License-Identifier: GPL-2.0-only
   2 /*
   3  * common code for virtio vsock
   4  *
   5  * Copyright (C) 2013-2015 Red Hat, Inc.
   6  * Author: Asias He <asias@redhat.com>
   7  *         Stefan Hajnoczi <stefanha@redhat.com>
   8  */
   9 #include <linux/spinlock.h>
  10 #include <linux/module.h>
  11 #include <linux/sched/signal.h>
  12 #include <linux/ctype.h>
  13 #include <linux/list.h>
  14 #include <linux/virtio.h>
  15 #include <linux/virtio_ids.h>
  16 #include <linux/virtio_config.h>
  17 #include <linux/virtio_vsock.h>
  18 #include <uapi/linux/vsockmon.h>
  19
  20 #include <net/sock.h>
  21 #include <net/af_vsock.h>
  22
  23 #define CREATE_TRACE_POINTS
  24 #include <trace/events/vsock_virtio_transport_common.h>
  25
  26 /* How long to wait for graceful shutdown of a connection */
  27 #define VSOCK_CLOSE_TIMEOUT (8 * HZ)
  28
  29 /* Threshold for detecting small packets to copy */
  30 #define GOOD_COPY_LEN  128
  31
  32 static const struct virtio_transport *virtio_transport_get_ops(void)
  33 {
  34         const struct vsock_transport *t = vsock_core_get_transport();
  35
  36         return container_of(t, struct virtio_transport, transport);
  37 }
  38
  39 static struct virtio_vsock_pkt *
  40 virtio_transport_alloc_pkt(struct virtio_vsock_pkt_info *info,
  41                            size_t len,
  42                            u32 src_cid,
  43                            u32 src_port,
  44                            u32 dst_cid,
  45                            u32 dst_port)
  46 {
  47         struct virtio_vsock_pkt *pkt;
  48         int err;
  49
  50         pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
  51         if (!pkt)
  52                 return NULL;
  53
  54         pkt->hdr.type           = cpu_to_le16(info->type);
  55         pkt->hdr.op             = cpu_to_le16(info->op);
  56         pkt->hdr.src_cid        = cpu_to_le64(src_cid);
  57         pkt->hdr.dst_cid        = cpu_to_le64(dst_cid);
  58         pkt->hdr.src_port       = cpu_to_le32(src_port);
  59         pkt->hdr.dst_port       = cpu_to_le32(dst_port);
  60         pkt->hdr.flags          = cpu_to_le32(info->flags);
  61         pkt->len                = len;
  62         pkt->hdr.len            = cpu_to_le32(len);
  63         pkt->reply              = info->reply;
  64         pkt->vsk                = info->vsk;
  65
  66         if (info->msg && len > 0) {
  67                 pkt->buf = kmalloc(len, GFP_KERNEL);
  68                 if (!pkt->buf)
  69                         goto out_pkt;
  70
  71                 pkt->buf_len = len;
  72
  73                 err = memcpy_from_msg(pkt->buf, info->msg, len);
  74                 if (err)
  75                         goto out;
  76         }
  77
  78         trace_virtio_transport_alloc_pkt(src_cid, src_port,
  79                                          dst_cid, dst_port,
  80                                          len,
  81                                          info->type,
  82                                          info->op,
  83                                          info->flags);
  84
  85         return pkt;
  86
  87 out:
  88         kfree(pkt->buf);
  89 out_pkt:
  90         kfree(pkt);
  91         return NULL;
  92 }
  93
  94 /* Packet capture */
  95 static struct sk_buff *virtio_transport_build_skb(void *opaque)
  96 {
  97         struct virtio_vsock_pkt *pkt = opaque;
  98         struct af_vsockmon_hdr *hdr;
  99         struct sk_buff *skb;
 100         size_t payload_len;
 101         void *payload_buf;
 102
 103         /* A packet could be split to fit the RX buffer, so we can retrieve
 104          * the payload length from the header and the buffer pointer taking
 105          * care of the offset in the original packet.
 106          */
 107         payload_len = le32_to_cpu(pkt->hdr.len);
 108         payload_buf = pkt->buf + pkt->off;
 109
 110         skb = alloc_skb(sizeof(*hdr) + sizeof(pkt->hdr) + payload_len,
 111                         GFP_ATOMIC);
 112         if (!skb)
 113                 return NULL;
 114
 115         hdr = skb_put(skb, sizeof(*hdr));
 116
 117         /* pkt->hdr is little-endian so no need to byteswap here */
 118         hdr->src_cid = pkt->hdr.src_cid;
 119         hdr->src_port = pkt->hdr.src_port;
 120         hdr->dst_cid = pkt->hdr.dst_cid;
 121         hdr->dst_port = pkt->hdr.dst_port;
 122
 123         hdr->transport = cpu_to_le16(AF_VSOCK_TRANSPORT_VIRTIO);
 124         hdr->len = cpu_to_le16(sizeof(pkt->hdr));
 125         memset(hdr->reserved, 0, sizeof(hdr->reserved));
 126
 127         switch (le16_to_cpu(pkt->hdr.op)) {
 128         case VIRTIO_VSOCK_OP_REQUEST:
 129         case VIRTIO_VSOCK_OP_RESPONSE:
 130                 hdr->op = cpu_to_le16(AF_VSOCK_OP_CONNECT);
 131                 break;
 132         case VIRTIO_VSOCK_OP_RST:
 133         case VIRTIO_VSOCK_OP_SHUTDOWN:
 134                 hdr->op = cpu_to_le16(AF_VSOCK_OP_DISCONNECT);
 135                 break;
 136         case VIRTIO_VSOCK_OP_RW:
 137                 hdr->op = cpu_to_le16(AF_VSOCK_OP_PAYLOAD);
 138                 break;
 139         case VIRTIO_VSOCK_OP_CREDIT_UPDATE:
 140         case VIRTIO_VSOCK_OP_CREDIT_REQUEST:
 141                 hdr->op = cpu_to_le16(AF_VSOCK_OP_CONTROL);
 142                 break;
 143         default:
 144                 hdr->op = cpu_to_le16(AF_VSOCK_OP_UNKNOWN);
 145                 break;
 146         }
 147
 148         skb_put_data(skb, &pkt->hdr, sizeof(pkt->hdr));
 149
 150         if (payload_len) {
 151                 skb_put_data(skb, payload_buf, payload_len);
 152         }
 153
 154         return skb;
 155 }
 156
 157 void virtio_transport_deliver_tap_pkt(struct virtio_vsock_pkt *pkt)
 158 {
 159         vsock_deliver_tap(virtio_transport_build_skb, pkt);
 160 }
 161 EXPORT_SYMBOL_GPL(virtio_transport_deliver_tap_pkt);
 162
 163 static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
 164                                           struct virtio_vsock_pkt_info *info)
 165 {
 166         u32 src_cid, src_port, dst_cid, dst_port;
 167         struct virtio_vsock_sock *vvs;
 168         struct virtio_vsock_pkt *pkt;
 169         u32 pkt_len = info->pkt_len;
 170
 171         src_cid = vm_sockets_get_local_cid();
 172         src_port = vsk->local_addr.svm_port;
 173         if (!info->remote_cid) {
 174                 dst_cid = vsk->remote_addr.svm_cid;
 175                 dst_port = vsk->remote_addr.svm_port;
 176         } else {
 177                 dst_cid = info->remote_cid;
 178                 dst_port = info->remote_port;
 179         }
 180
 181         vvs = vsk->trans;
 182
 183         /* we can send less than pkt_len bytes */
 184         if (pkt_len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE)
 185                 pkt_len = VIRTIO_VSOCK_MAX_PKT_BUF_SIZE;
 186
 187         /* virtio_transport_get_credit might return less than pkt_len credit */
 188         pkt_len = virtio_transport_get_credit(vvs, pkt_len);
 189
 190         /* Do not send zero length OP_RW pkt */
 191         if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
 192                 return pkt_len;
 193
 194         pkt = virtio_transport_alloc_pkt(info, pkt_len,
 195                                          src_cid, src_port,
 196                                          dst_cid, dst_port);
 197         if (!pkt) {
 198                 virtio_transport_put_credit(vvs, pkt_len);
 199                 return -ENOMEM;
 200         }
 201
 202         virtio_transport_inc_tx_pkt(vvs, pkt);
 203
 204         return virtio_transport_get_ops()->send_pkt(pkt);
 205 }
 206
 207 static void virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs,
 208                                         struct virtio_vsock_pkt *pkt)
 209 {
 210         vvs->rx_bytes += pkt->len;
 211 }
 212
 213 static void virtio_transport_dec_rx_pkt(struct virtio_vsock_sock *vvs,
 214                                         struct virtio_vsock_pkt *pkt)
 215 {
 216         vvs->rx_bytes -= pkt->len;
 217         vvs->fwd_cnt += pkt->len;
 218 }
 219
 220 void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct virtio_vsock_pkt *pkt)
 221 {
 222         spin_lock_bh(&vvs->rx_lock);
 223         vvs->last_fwd_cnt = vvs->fwd_cnt;
 224         pkt->hdr.fwd_cnt = cpu_to_le32(vvs->fwd_cnt);
 225         pkt->hdr.buf_alloc = cpu_to_le32(vvs->buf_alloc);
 226         spin_unlock_bh(&vvs->rx_lock);
 227 }
 228 EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt);
 229
 230 u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit)
 231 {
 232         u32 ret;
 233
 234         spin_lock_bh(&vvs->tx_lock);
 235         ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt);
 236         if (ret > credit)
 237                 ret = credit;
 238         vvs->tx_cnt += ret;
 239         spin_unlock_bh(&vvs->tx_lock);
 240
 241         return ret;
 242 }
 243 EXPORT_SYMBOL_GPL(virtio_transport_get_credit);
 244
 245 void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit)
 246 {
 247         spin_lock_bh(&vvs->tx_lock);
 248         vvs->tx_cnt -= credit;
 249         spin_unlock_bh(&vvs->tx_lock);
 250 }
 251 EXPORT_SYMBOL_GPL(virtio_transport_put_credit);
 252
 253 static int virtio_transport_send_credit_update(struct vsock_sock *vsk,
 254                                                int type,
 255                                                struct virtio_vsock_hdr *hdr)
 256 {
 257         struct virtio_vsock_pkt_info info = {
 258                 .op = VIRTIO_VSOCK_OP_CREDIT_UPDATE,
 259                 .type = type,
 260                 .vsk = vsk,
 261         };
 262
 263         return virtio_transport_send_pkt_info(vsk, &info);
 264 }
 265
 266 static ssize_t
 267 virtio_transport_stream_do_dequeue(struct vsock_sock *vsk,
 268                                    struct msghdr *msg,
 269                                    size_t len)
 270 {
 271         struct virtio_vsock_sock *vvs = vsk->trans;
 272         struct virtio_vsock_pkt *pkt;
 273         size_t bytes, total = 0;
 274         u32 free_space;
 275         int err = -EFAULT;
 276
 277         spin_lock_bh(&vvs->rx_lock);
 278         while (total < len && !list_empty(&vvs->rx_queue)) {
 279                 pkt = list_first_entry(&vvs->rx_queue,
 280                                        struct virtio_vsock_pkt, list);
 281
 282                 bytes = len - total;
 283                 if (bytes > pkt->len - pkt->off)
 284                         bytes = pkt->len - pkt->off;
 285
 286                 /* sk_lock is held by caller so no one else can dequeue.
 287                  * Unlock rx_lock since memcpy_to_msg() may sleep.
 288                  */
 289                 spin_unlock_bh(&vvs->rx_lock);
 290
 291                 err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes);
 292                 if (err)
 293                         goto out;
 294
 295                 spin_lock_bh(&vvs->rx_lock);
 296
 297                 total += bytes;
 298                 pkt->off += bytes;
 299                 if (pkt->off == pkt->len) {
 300                         virtio_transport_dec_rx_pkt(vvs, pkt);
 301                         list_del(&pkt->list);
 302                         virtio_transport_free_pkt(pkt);
 303                 }
 304         }
 305
 306         free_space = vvs->buf_alloc - (vvs->fwd_cnt - vvs->last_fwd_cnt);
 307
 308         spin_unlock_bh(&vvs->rx_lock);
 309
 310         /* To reduce the number of credit update messages,
 311          * don't update credits as long as lots of space is available.
 312          * Note: the limit chosen here is arbitrary. Setting the limit
 313          * too high causes extra messages. Too low causes transmitter
 314          * stalls. As stalls are in theory more expensive than extra
 315          * messages, we set the limit to a high value. TODO: experiment
 316          * with different values.
 317          */
 318         if (free_space < VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) {
 319                 virtio_transport_send_credit_update(vsk,
 320                                                     VIRTIO_VSOCK_TYPE_STREAM,
 321                                                     NULL);
 322         }
 323
 324         return total;
 325
 326 out:
 327         if (total)
 328                 err = total;
 329         return err;
 330 }
 331
 332 ssize_t
 333 virtio_transport_stream_dequeue(struct vsock_sock *vsk,
 334                                 struct msghdr *msg,
 335                                 size_t len, int flags)
 336 {
 337         if (flags & MSG_PEEK)
 338                 return -EOPNOTSUPP;
 339
 340         return virtio_transport_stream_do_dequeue(vsk, msg, len);
 341 }
 342 EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue);
 343
 344 int
 345 virtio_transport_dgram_dequeue(struct vsock_sock *vsk,
 346                                struct msghdr *msg,
 347                                size_t len, int flags)
 348 {
 349         return -EOPNOTSUPP;
 350 }
 351 EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue);
 352
 353 s64 virtio_transport_stream_has_data(struct vsock_sock *vsk)
 354 {
 355         struct virtio_vsock_sock *vvs = vsk->trans;
 356         s64 bytes;
 357
 358         spin_lock_bh(&vvs->rx_lock);
 359         bytes = vvs->rx_bytes;
 360         spin_unlock_bh(&vvs->rx_lock);
 361
 362         return bytes;
 363 }
 364 EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data);
 365
 366 static s64 virtio_transport_has_space(struct vsock_sock *vsk)
 367 {
 368         struct virtio_vsock_sock *vvs = vsk->trans;
 369         s64 bytes;
 370
 371         bytes = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt);
 372         if (bytes < 0)
 373                 bytes = 0;
 374
 375         return bytes;
 376 }
 377
 378 s64 virtio_transport_stream_has_space(struct vsock_sock *vsk)
 379 {
 380         struct virtio_vsock_sock *vvs = vsk->trans;
 381         s64 bytes;
 382
 383         spin_lock_bh(&vvs->tx_lock);
 384         bytes = virtio_transport_has_space(vsk);
 385         spin_unlock_bh(&vvs->tx_lock);
 386
 387         return bytes;
 388 }
 389 EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space);
 390
 391 int virtio_transport_do_socket_init(struct vsock_sock *vsk,
 392                                     struct vsock_sock *psk)
 393 {
 394         struct virtio_vsock_sock *vvs;
 395
 396         vvs = kzalloc(sizeof(*vvs), GFP_KERNEL);
 397         if (!vvs)
 398                 return -ENOMEM;
 399
 400         vsk->trans = vvs;
 401         vvs->vsk = vsk;
 402         if (psk) {
 403                 struct virtio_vsock_sock *ptrans = psk->trans;
 404
 405                 vvs->buf_size   = ptrans->buf_size;
 406                 vvs->buf_size_min = ptrans->buf_size_min;
 407                 vvs->buf_size_max = ptrans->buf_size_max;
 408                 vvs->peer_buf_alloc = ptrans->peer_buf_alloc;
 409         } else {
 410                 vvs->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE;
 411                 vvs->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE;
 412                 vvs->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE;
 413         }
 414
 415         vvs->buf_alloc = vvs->buf_size;
 416
 417         spin_lock_init(&vvs->rx_lock);
 418         spin_lock_init(&vvs->tx_lock);
 419         INIT_LIST_HEAD(&vvs->rx_queue);
 420
 421         return 0;
 422 }
 423 EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init);
 424
 425 u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk)
 426 {
 427         struct virtio_vsock_sock *vvs = vsk->trans;
 428
 429         return vvs->buf_size;
 430 }
 431 EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size);
 432
 433 u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk)
 434 {
 435         struct virtio_vsock_sock *vvs = vsk->trans;
 436
 437         return vvs->buf_size_min;
 438 }
 439 EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size);
 440
 441 u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk)
 442 {
 443         struct virtio_vsock_sock *vvs = vsk->trans;
 444
 445         return vvs->buf_size_max;
 446 }
 447 EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size);
 448
 449 void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val)
 450 {
 451         struct virtio_vsock_sock *vvs = vsk->trans;
 452
 453         if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
 454                 val = VIRTIO_VSOCK_MAX_BUF_SIZE;
 455         if (val < vvs->buf_size_min)
 456                 vvs->buf_size_min = val;
 457         if (val > vvs->buf_size_max)
 458                 vvs->buf_size_max = val;
 459         vvs->buf_size = val;
 460         vvs->buf_alloc = val;
 461 }
 462 EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size);
 463
 464 void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val)
 465 {
 466         struct virtio_vsock_sock *vvs = vsk->trans;
 467
 468         if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
 469                 val = VIRTIO_VSOCK_MAX_BUF_SIZE;
 470         if (val > vvs->buf_size)
 471                 vvs->buf_size = val;
 472         vvs->buf_size_min = val;
 473 }
 474 EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size);
 475
 476 void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val)
 477 {
 478         struct virtio_vsock_sock *vvs = vsk->trans;
 479
 480         if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
 481                 val = VIRTIO_VSOCK_MAX_BUF_SIZE;
 482         if (val < vvs->buf_size)
 483                 vvs->buf_size = val;
 484         vvs->buf_size_max = val;
 485 }
 486 EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size);
 487
 488 int
 489 virtio_transport_notify_poll_in(struct vsock_sock *vsk,
 490                                 size_t target,
 491                                 bool *data_ready_now)
 492 {
 493         if (vsock_stream_has_data(vsk))
 494                 *data_ready_now = true;
 495         else
 496                 *data_ready_now = false;
 497
 498         return 0;
 499 }
 500 EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in);
 501
 502 int
 503 virtio_transport_notify_poll_out(struct vsock_sock *vsk,
 504                                  size_t target,
 505                                  bool *space_avail_now)
 506 {
 507         s64 free_space;
 508
 509         free_space = vsock_stream_has_space(vsk);
 510         if (free_space > 0)
 511                 *space_avail_now = true;
 512         else if (free_space == 0)
 513                 *space_avail_now = false;
 514
 515         return 0;
 516 }
 517 EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out);
 518
 519 int virtio_transport_notify_recv_init(struct vsock_sock *vsk,
 520         size_t target, struct vsock_transport_recv_notify_data *data)
 521 {
 522         return 0;
 523 }
 524 EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init);
 525
 526 int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk,
 527         size_t target, struct vsock_transport_recv_notify_data *data)
 528 {
 529         return 0;
 530 }
 531 EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block);
 532
 533 int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk,
 534         size_t target, struct vsock_transport_recv_notify_data *data)
 535 {
 536         return 0;
 537 }
 538 EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue);
 539
 540 int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk,
 541         size_t target, ssize_t copied, bool data_read,
 542         struct vsock_transport_recv_notify_data *data)
 543 {
 544         return 0;
 545 }
 546 EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue);
 547
 548 int virtio_transport_notify_send_init(struct vsock_sock *vsk,
 549         struct vsock_transport_send_notify_data *data)
 550 {
 551         return 0;
 552 }
 553 EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init);
 554
 555 int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk,
 556         struct vsock_transport_send_notify_data *data)
 557 {
 558         return 0;
 559 }
 560 EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block);
 561
 562 int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk,
 563         struct vsock_transport_send_notify_data *data)
 564 {
 565         return 0;
 566 }
 567 EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue);
 568
 569 int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk,
 570         ssize_t written, struct vsock_transport_send_notify_data *data)
 571 {
 572         return 0;
 573 }
 574 EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue);
 575
 576 u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk)
 577 {
 578         struct virtio_vsock_sock *vvs = vsk->trans;
 579
 580         return vvs->buf_size;
 581 }
 582 EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat);
 583
 584 bool virtio_transport_stream_is_active(struct vsock_sock *vsk)
 585 {
 586         return true;
 587 }
 588 EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active);
 589
 590 bool virtio_transport_stream_allow(u32 cid, u32 port)
 591 {
 592         return true;
 593 }
 594 EXPORT_SYMBOL_GPL(virtio_transport_stream_allow);
 595
 596 int virtio_transport_dgram_bind(struct vsock_sock *vsk,
 597                                 struct sockaddr_vm *addr)
 598 {
 599         return -EOPNOTSUPP;
 600 }
 601 EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind);
 602
 603 bool virtio_transport_dgram_allow(u32 cid, u32 port)
 604 {
 605         return false;
 606 }
 607 EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow);
 608
 609 int virtio_transport_connect(struct vsock_sock *vsk)
 610 {
 611         struct virtio_vsock_pkt_info info = {
 612                 .op = VIRTIO_VSOCK_OP_REQUEST,
 613                 .type = VIRTIO_VSOCK_TYPE_STREAM,
 614                 .vsk = vsk,
 615         };
 616
 617         return virtio_transport_send_pkt_info(vsk, &info);
 618 }
 619 EXPORT_SYMBOL_GPL(virtio_transport_connect);
 620
 621 int virtio_transport_shutdown(struct vsock_sock *vsk, int mode)
 622 {
 623         struct virtio_vsock_pkt_info info = {
 624                 .op = VIRTIO_VSOCK_OP_SHUTDOWN,
 625                 .type = VIRTIO_VSOCK_TYPE_STREAM,
 626                 .flags = (mode & RCV_SHUTDOWN ?
 627                           VIRTIO_VSOCK_SHUTDOWN_RCV : 0) |
 628                          (mode & SEND_SHUTDOWN ?
 629                           VIRTIO_VSOCK_SHUTDOWN_SEND : 0),
 630                 .vsk = vsk,
 631         };
 632
 633         return virtio_transport_send_pkt_info(vsk, &info);
 634 }
 635 EXPORT_SYMBOL_GPL(virtio_transport_shutdown);
 636
 637 int
 638 virtio_transport_dgram_enqueue(struct vsock_sock *vsk,
 639                                struct sockaddr_vm *remote_addr,
 640                                struct msghdr *msg,
 641                                size_t dgram_len)
 642 {
 643         return -EOPNOTSUPP;
 644 }
 645 EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue);
 646
 647 ssize_t
 648 virtio_transport_stream_enqueue(struct vsock_sock *vsk,
 649                                 struct msghdr *msg,
 650                                 size_t len)
 651 {
 652         struct virtio_vsock_pkt_info info = {
 653                 .op = VIRTIO_VSOCK_OP_RW,
 654                 .type = VIRTIO_VSOCK_TYPE_STREAM,
 655                 .msg = msg,
 656                 .pkt_len = len,
 657                 .vsk = vsk,
 658         };
 659
 660         return virtio_transport_send_pkt_info(vsk, &info);
 661 }
 662 EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue);
 663
 664 void virtio_transport_destruct(struct vsock_sock *vsk)
 665 {
 666         struct virtio_vsock_sock *vvs = vsk->trans;
 667
 668         kfree(vvs);
 669 }
 670 EXPORT_SYMBOL_GPL(virtio_transport_destruct);
 671
 672 static int virtio_transport_reset(struct vsock_sock *vsk,
 673                                   struct virtio_vsock_pkt *pkt)
 674 {
 675         struct virtio_vsock_pkt_info info = {
 676                 .op = VIRTIO_VSOCK_OP_RST,
 677                 .type = VIRTIO_VSOCK_TYPE_STREAM,
 678                 .reply = !!pkt,
 679                 .vsk = vsk,
 680         };
 681
 682         /* Send RST only if the original pkt is not a RST pkt */
 683         if (pkt && le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST)
 684                 return 0;
 685
 686         return virtio_transport_send_pkt_info(vsk, &info);
 687 }
 688
 689 /* Normally packets are associated with a socket.  There may be no socket if an
 690  * attempt was made to connect to a socket that does not exist.
 691  */
 692 static int virtio_transport_reset_no_sock(struct virtio_vsock_pkt *pkt)
 693 {
 694         const struct virtio_transport *t;
 695         struct virtio_vsock_pkt *reply;
 696         struct virtio_vsock_pkt_info info = {
 697                 .op = VIRTIO_VSOCK_OP_RST,
 698                 .type = le16_to_cpu(pkt->hdr.type),
 699                 .reply = true,
 700         };
 701
 702         /* Send RST only if the original pkt is not a RST pkt */
 703         if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST)
 704                 return 0;
 705
 706         reply = virtio_transport_alloc_pkt(&info, 0,
 707                                            le64_to_cpu(pkt->hdr.dst_cid),
 708                                            le32_to_cpu(pkt->hdr.dst_port),
 709                                            le64_to_cpu(pkt->hdr.src_cid),
 710                                            le32_to_cpu(pkt->hdr.src_port));
 711         if (!reply)
 712                 return -ENOMEM;
 713
 714         t = virtio_transport_get_ops();
 715         if (!t) {
 716                 virtio_transport_free_pkt(reply);
 717                 return -ENOTCONN;
 718         }
 719
 720         return t->send_pkt(reply);
 721 }
 722
 723 static void virtio_transport_wait_close(struct sock *sk, long timeout)
 724 {
 725         if (timeout) {
 726                 DEFINE_WAIT_FUNC(wait, woken_wake_function);
 727
 728                 add_wait_queue(sk_sleep(sk), &wait);
 729
 730                 do {
 731                         if (sk_wait_event(sk, &timeout,
 732                                           sock_flag(sk, SOCK_DONE), &wait))
 733                                 break;
 734                 } while (!signal_pending(current) && timeout);
 735
 736                 remove_wait_queue(sk_sleep(sk), &wait);
 737         }
 738 }
 739
 740 static void virtio_transport_do_close(struct vsock_sock *vsk,
 741                                       bool cancel_timeout)
 742 {
 743         struct sock *sk = sk_vsock(vsk);
 744
 745         sock_set_flag(sk, SOCK_DONE);
 746         vsk->peer_shutdown = SHUTDOWN_MASK;
 747         if (vsock_stream_has_data(vsk) <= 0)
 748                 sk->sk_state = TCP_CLOSING;
 749         sk->sk_state_change(sk);
 750
 751         if (vsk->close_work_scheduled &&
 752             (!cancel_timeout || cancel_delayed_work(&vsk->close_work))) {
 753                 vsk->close_work_scheduled = false;
 754
 755                 vsock_remove_sock(vsk);
 756
 757                 /* Release refcnt obtained when we scheduled the timeout */
 758                 sock_put(sk);
 759         }
 760 }
 761
 762 static void virtio_transport_close_timeout(struct work_struct *work)
 763 {
 764         struct vsock_sock *vsk =
 765                 container_of(work, struct vsock_sock, close_work.work);
 766         struct sock *sk = sk_vsock(vsk);
 767
 768         sock_hold(sk);
 769         lock_sock(sk);
 770
 771         if (!sock_flag(sk, SOCK_DONE)) {
 772                 (void)virtio_transport_reset(vsk, NULL);
 773
 774                 virtio_transport_do_close(vsk, false);
 775         }
 776
 777         vsk->close_work_scheduled = false;
 778
 779         release_sock(sk);
 780         sock_put(sk);
 781 }
 782
 783 /* User context, vsk->sk is locked */
 784 static bool virtio_transport_close(struct vsock_sock *vsk)
 785 {
 786         struct sock *sk = &vsk->sk;
 787
 788         if (!(sk->sk_state == TCP_ESTABLISHED ||
 789               sk->sk_state == TCP_CLOSING))
 790                 return true;
 791
 792         /* Already received SHUTDOWN from peer, reply with RST */
 793         if ((vsk->peer_shutdown & SHUTDOWN_MASK) == SHUTDOWN_MASK) {
 794                 (void)virtio_transport_reset(vsk, NULL);
 795                 return true;
 796         }
 797
 798         if ((sk->sk_shutdown & SHUTDOWN_MASK) != SHUTDOWN_MASK)
 799                 (void)virtio_transport_shutdown(vsk, SHUTDOWN_MASK);
 800
 801         if (sock_flag(sk, SOCK_LINGER) && !(current->flags & PF_EXITING))
 802                 virtio_transport_wait_close(sk, sk->sk_lingertime);
 803
 804         if (sock_flag(sk, SOCK_DONE)) {
 805                 return true;
 806         }
 807
 808         sock_hold(sk);
 809         INIT_DELAYED_WORK(&vsk->close_work,
 810                           virtio_transport_close_timeout);
 811         vsk->close_work_scheduled = true;
 812         schedule_delayed_work(&vsk->close_work, VSOCK_CLOSE_TIMEOUT);
 813         return false;
 814 }
 815
 816 void virtio_transport_release(struct vsock_sock *vsk)
 817 {
 818         struct virtio_vsock_sock *vvs = vsk->trans;
 819         struct virtio_vsock_pkt *pkt, *tmp;
 820         struct sock *sk = &vsk->sk;
 821         bool remove_sock = true;
 822
 823         lock_sock_nested(sk, SINGLE_DEPTH_NESTING);
 824         if (sk->sk_type == SOCK_STREAM)
 825                 remove_sock = virtio_transport_close(vsk);
 826
 827         list_for_each_entry_safe(pkt, tmp, &vvs->rx_queue, list) {
 828                 list_del(&pkt->list);
 829                 virtio_transport_free_pkt(pkt);
 830         }
 831         release_sock(sk);
 832
 833         if (remove_sock)
 834                 vsock_remove_sock(vsk);
 835 }
 836 EXPORT_SYMBOL_GPL(virtio_transport_release);
 837
 838 static int
 839 virtio_transport_recv_connecting(struct sock *sk,
 840                                  struct virtio_vsock_pkt *pkt)
 841 {
 842         struct vsock_sock *vsk = vsock_sk(sk);
 843         int err;
 844         int skerr;
 845
 846         switch (le16_to_cpu(pkt->hdr.op)) {
 847         case VIRTIO_VSOCK_OP_RESPONSE:
 848                 sk->sk_state = TCP_ESTABLISHED;
 849                 sk->sk_socket->state = SS_CONNECTED;
 850                 vsock_insert_connected(vsk);
 851                 sk->sk_state_change(sk);
 852                 break;
 853         case VIRTIO_VSOCK_OP_INVALID:
 854                 break;
 855         case VIRTIO_VSOCK_OP_RST:
 856                 skerr = ECONNRESET;
 857                 err = 0;
 858                 goto destroy;
 859         default:
 860                 skerr = EPROTO;
 861                 err = -EINVAL;
 862                 goto destroy;
 863         }
 864         return 0;
 865
 866 destroy:
 867         virtio_transport_reset(vsk, pkt);
 868         sk->sk_state = TCP_CLOSE;
 869         sk->sk_err = skerr;
 870         sk->sk_error_report(sk);
 871         return err;
 872 }
 873
 874 static void
 875 virtio_transport_recv_enqueue(struct vsock_sock *vsk,
 876                               struct virtio_vsock_pkt *pkt)
 877 {
 878         struct virtio_vsock_sock *vvs = vsk->trans;
 879         bool free_pkt = false;
 880
 881         pkt->len = le32_to_cpu(pkt->hdr.len);
 882         pkt->off = 0;
 883
 884         spin_lock_bh(&vvs->rx_lock);
 885
 886         virtio_transport_inc_rx_pkt(vvs, pkt);
 887
 888         /* Try to copy small packets into the buffer of last packet queued,
 889          * to avoid wasting memory queueing the entire buffer with a small
 890          * payload.
 891          */
 892         if (pkt->len <= GOOD_COPY_LEN && !list_empty(&vvs->rx_queue)) {
 893                 struct virtio_vsock_pkt *last_pkt;
 894
 895                 last_pkt = list_last_entry(&vvs->rx_queue,
 896                                            struct virtio_vsock_pkt, list);
 897
 898                 /* If there is space in the last packet queued, we copy the
 899                  * new packet in its buffer.
 900                  */
 901                 if (pkt->len <= last_pkt->buf_len - last_pkt->len) {
 902                         memcpy(last_pkt->buf + last_pkt->len, pkt->buf,
 903                                pkt->len);
 904                         last_pkt->len += pkt->len;
 905                         free_pkt = true;
 906                         goto out;
 907                 }
 908         }
 909
 910         list_add_tail(&pkt->list, &vvs->rx_queue);
 911
 912 out:
 913         spin_unlock_bh(&vvs->rx_lock);
 914         if (free_pkt)
 915                 virtio_transport_free_pkt(pkt);
 916 }
 917
 918 static int
 919 virtio_transport_recv_connected(struct sock *sk,
 920                                 struct virtio_vsock_pkt *pkt)
 921 {
 922         struct vsock_sock *vsk = vsock_sk(sk);
 923         int err = 0;
 924
 925         switch (le16_to_cpu(pkt->hdr.op)) {
 926         case VIRTIO_VSOCK_OP_RW:
 927                 virtio_transport_recv_enqueue(vsk, pkt);
 928                 sk->sk_data_ready(sk);
 929                 return err;
 930         case VIRTIO_VSOCK_OP_CREDIT_UPDATE:
 931                 sk->sk_write_space(sk);
 932                 break;
 933         case VIRTIO_VSOCK_OP_SHUTDOWN:
 934                 if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV)
 935                         vsk->peer_shutdown |= RCV_SHUTDOWN;
 936                 if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND)
 937                         vsk->peer_shutdown |= SEND_SHUTDOWN;
 938                 if (vsk->peer_shutdown == SHUTDOWN_MASK &&
 939                     vsock_stream_has_data(vsk) <= 0) {
 940                         sock_set_flag(sk, SOCK_DONE);
 941                         sk->sk_state = TCP_CLOSING;
 942                 }
 943                 if (le32_to_cpu(pkt->hdr.flags))
 944                         sk->sk_state_change(sk);
 945                 break;
 946         case VIRTIO_VSOCK_OP_RST:
 947                 virtio_transport_do_close(vsk, true);
 948                 break;
 949         default:
 950                 err = -EINVAL;
 951                 break;
 952         }
 953
 954         virtio_transport_free_pkt(pkt);
 955         return err;
 956 }
 957
 958 static void
 959 virtio_transport_recv_disconnecting(struct sock *sk,
 960                                     struct virtio_vsock_pkt *pkt)
 961 {
 962         struct vsock_sock *vsk = vsock_sk(sk);
 963
 964         if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST)
 965                 virtio_transport_do_close(vsk, true);
 966 }
 967
 968 static int
 969 virtio_transport_send_response(struct vsock_sock *vsk,
 970                                struct virtio_vsock_pkt *pkt)
 971 {
 972         struct virtio_vsock_pkt_info info = {
 973                 .op = VIRTIO_VSOCK_OP_RESPONSE,
 974                 .type = VIRTIO_VSOCK_TYPE_STREAM,
 975                 .remote_cid = le64_to_cpu(pkt->hdr.src_cid),
 976                 .remote_port = le32_to_cpu(pkt->hdr.src_port),
 977                 .reply = true,
 978                 .vsk = vsk,
 979         };
 980
 981         return virtio_transport_send_pkt_info(vsk, &info);
 982 }
 983
 984 /* Handle server socket */
 985 static int
 986 virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt)
 987 {
 988         struct vsock_sock *vsk = vsock_sk(sk);
 989         struct vsock_sock *vchild;
 990         struct sock *child;
 991
 992         if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_REQUEST) {
 993                 virtio_transport_reset(vsk, pkt);
 994                 return -EINVAL;
 995         }
 996
 997         if (sk_acceptq_is_full(sk)) {
 998                 virtio_transport_reset(vsk, pkt);
 999                 return -ENOMEM;
1000         }
1001
1002         child = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL,
1003                                sk->sk_type, 0);
1004         if (!child) {
1005                 virtio_transport_reset(vsk, pkt);
1006                 return -ENOMEM;
1007         }
1008
1009         sk->sk_ack_backlog++;
1010
1011         lock_sock_nested(child, SINGLE_DEPTH_NESTING);
1012
1013         child->sk_state = TCP_ESTABLISHED;
1014
1015         vchild = vsock_sk(child);
1016         vsock_addr_init(&vchild->local_addr, le64_to_cpu(pkt->hdr.dst_cid),
1017                         le32_to_cpu(pkt->hdr.dst_port));
1018         vsock_addr_init(&vchild->remote_addr, le64_to_cpu(pkt->hdr.src_cid),
1019                         le32_to_cpu(pkt->hdr.src_port));
1020
1021         vsock_insert_connected(vchild);
1022         vsock_enqueue_accept(sk, child);
1023         virtio_transport_send_response(vchild, pkt);
1024
1025         release_sock(child);
1026
1027         sk->sk_data_ready(sk);
1028         return 0;
1029 }
1030
1031 static bool virtio_transport_space_update(struct sock *sk,
1032                                           struct virtio_vsock_pkt *pkt)
1033 {
1034         struct vsock_sock *vsk = vsock_sk(sk);
1035         struct virtio_vsock_sock *vvs = vsk->trans;
1036         bool space_available;
1037
1038         /* buf_alloc and fwd_cnt is always included in the hdr */
1039         spin_lock_bh(&vvs->tx_lock);
1040         vvs->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc);
1041         vvs->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt);
1042         space_available = virtio_transport_has_space(vsk);
1043         spin_unlock_bh(&vvs->tx_lock);
1044         return space_available;
1045 }
1046
1047 /* We are under the virtio-vsock's vsock->rx_lock or vhost-vsock's vq->mutex
1048  * lock.
1049  */
1050 void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt)
1051 {
1052         struct sockaddr_vm src, dst;
1053         struct vsock_sock *vsk;
1054         struct sock *sk;
1055         bool space_available;
1056
1057         vsock_addr_init(&src, le64_to_cpu(pkt->hdr.src_cid),
1058                         le32_to_cpu(pkt->hdr.src_port));
1059         vsock_addr_init(&dst, le64_to_cpu(pkt->hdr.dst_cid),
1060                         le32_to_cpu(pkt->hdr.dst_port));
1061
1062         trace_virtio_transport_recv_pkt(src.svm_cid, src.svm_port,
1063                                         dst.svm_cid, dst.svm_port,
1064                                         le32_to_cpu(pkt->hdr.len),
1065                                         le16_to_cpu(pkt->hdr.type),
1066                                         le16_to_cpu(pkt->hdr.op),
1067                                         le32_to_cpu(pkt->hdr.flags),
1068                                         le32_to_cpu(pkt->hdr.buf_alloc),
1069                                         le32_to_cpu(pkt->hdr.fwd_cnt));
1070
1071         if (le16_to_cpu(pkt->hdr.type) != VIRTIO_VSOCK_TYPE_STREAM) {
1072                 (void)virtio_transport_reset_no_sock(pkt);
1073                 goto free_pkt;
1074         }
1075
1076         /* The socket must be in connected or bound table
1077          * otherwise send reset back
1078          */
1079         sk = vsock_find_connected_socket(&src, &dst);
1080         if (!sk) {
1081                 sk = vsock_find_bound_socket(&dst);
1082                 if (!sk) {
1083                         (void)virtio_transport_reset_no_sock(pkt);
1084                         goto free_pkt;
1085                 }
1086         }
1087
1088         vsk = vsock_sk(sk);
1089
1090         space_available = virtio_transport_space_update(sk, pkt);
1091
1092         lock_sock(sk);
1093
1094         /* Update CID in case it has changed after a transport reset event */
1095         vsk->local_addr.svm_cid = dst.svm_cid;
1096
1097         if (space_available)
1098                 sk->sk_write_space(sk);
1099
1100         switch (sk->sk_state) {
1101         case TCP_LISTEN:
1102                 virtio_transport_recv_listen(sk, pkt);
1103                 virtio_transport_free_pkt(pkt);
1104                 break;
1105         case TCP_SYN_SENT:
1106                 virtio_transport_recv_connecting(sk, pkt);
1107                 virtio_transport_free_pkt(pkt);
1108                 break;
1109         case TCP_ESTABLISHED:
1110                 virtio_transport_recv_connected(sk, pkt);
1111                 break;
1112         case TCP_CLOSING:
1113                 virtio_transport_recv_disconnecting(sk, pkt);
1114                 virtio_transport_free_pkt(pkt);
1115                 break;
1116         default:
1117                 virtio_transport_free_pkt(pkt);
1118                 break;
1119         }
1120         release_sock(sk);
1121
1122         /* Release refcnt obtained when we fetched this socket out of the
1123          * bound or connected list.
1124          */
1125         sock_put(sk);
1126         return;
1127
1128 free_pkt:
1129         virtio_transport_free_pkt(pkt);
1130 }
1131 EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt);
1132
1133 void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt)
1134 {
1135         kfree(pkt->buf);
1136         kfree(pkt);
1137 }
1138 EXPORT_SYMBOL_GPL(virtio_transport_free_pkt);
1139
1140 MODULE_LICENSE("GPL v2");
1141 MODULE_AUTHOR("Asias He");
1142 MODULE_DESCRIPTION("common code for virtio vsock");