block/nbd-client.c

   1 /*
   2  * QEMU Block driver for  NBD
   3  *
   4  * Copyright (C) 2016 Red Hat, Inc.
   5  * Copyright (C) 2008 Bull S.A.S.
   6  *     Author: Laurent Vivier <Laurent.Vivier@bull.net>
   7  *
   8  * Some parts:
   9  *    Copyright (C) 2007 Anthony Liguori <anthony@codemonkey.ws>
  10  *
  11  * Permission is hereby granted, free of charge, to any person obtaining a copy
  12  * of this software and associated documentation files (the "Software"), to deal
  13  * in the Software without restriction, including without limitation the rights
  14  * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  15  * copies of the Software, and to permit persons to whom the Software is
  16  * furnished to do so, subject to the following conditions:
  17  *
  18  * The above copyright notice and this permission notice shall be included in
  19  * all copies or substantial portions of the Software.
  20  *
  21  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  22  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  23  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
  24  * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  25  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  26  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  27  * THE SOFTWARE.
  28  */
  29
  30 #include "qemu/osdep.h"
  31 #include "qapi/error.h"
  32 #include "nbd-client.h"
  33
  34 #define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
  35 #define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
  36
  37 static void nbd_recv_coroutines_wake_all(NBDClientSession *s)
  38 {
  39     int i;
  40
  41     for (i = 0; i < MAX_NBD_REQUESTS; i++) {
  42         NBDClientRequest *req = &s->requests[i];
  43
  44         if (req->coroutine && req->receiving) {
  45             aio_co_wake(req->coroutine);
  46         }
  47     }
  48 }
  49
  50 static void nbd_teardown_connection(BlockDriverState *bs)
  51 {
  52     NBDClientSession *client = nbd_get_client_session(bs);
  53
  54     if (!client->ioc) { /* Already closed */
  55         return;
  56     }
  57
  58     /* finish any pending coroutines */
  59     qio_channel_shutdown(client->ioc,
  60                          QIO_CHANNEL_SHUTDOWN_BOTH,
  61                          NULL);
  62     BDRV_POLL_WHILE(bs, client->read_reply_co);
  63
  64     nbd_client_detach_aio_context(bs);
  65     object_unref(OBJECT(client->sioc));
  66     client->sioc = NULL;
  67     object_unref(OBJECT(client->ioc));
  68     client->ioc = NULL;
  69 }
  70
  71 static coroutine_fn void nbd_read_reply_entry(void *opaque)
  72 {
  73     NBDClientSession *s = opaque;
  74     uint64_t i;
  75     int ret = 0;
  76     Error *local_err = NULL;
  77
  78     while (!s->quit) {
  79         assert(s->reply.handle == 0);
  80         ret = nbd_receive_reply(s->ioc, &s->reply, &local_err);
  81         if (ret < 0) {
  82             error_report_err(local_err);
  83         }
  84         if (ret <= 0) {
  85             break;
  86         }
  87
  88         /* There's no need for a mutex on the receive side, because the
  89          * handler acts as a synchronization point and ensures that only
  90          * one coroutine is called until the reply finishes.
  91          */
  92         i = HANDLE_TO_INDEX(s, s->reply.handle);
  93         if (i >= MAX_NBD_REQUESTS ||
  94             !s->requests[i].coroutine ||
  95             !s->requests[i].receiving) {
  96             break;
  97         }
  98
  99         /* We're woken up again by the request itself.  Note that there
 100          * is no race between yielding and reentering read_reply_co.  This
 101          * is because:
 102          *
 103          * - if the request runs on the same AioContext, it is only
 104          *   entered after we yield
 105          *
 106          * - if the request runs on a different AioContext, reentering
 107          *   read_reply_co happens through a bottom half, which can only
 108          *   run after we yield.
 109          */
 110         aio_co_wake(s->requests[i].coroutine);
 111         qemu_coroutine_yield();
 112     }
 113
 114     s->quit = true;
 115     nbd_recv_coroutines_wake_all(s);
 116     s->read_reply_co = NULL;
 117 }
 118
 119 static int nbd_co_send_request(BlockDriverState *bs,
 120                                NBDRequest *request,
 121                                QEMUIOVector *qiov)
 122 {
 123     NBDClientSession *s = nbd_get_client_session(bs);
 124     int rc, ret, i;
 125
 126     qemu_co_mutex_lock(&s->send_mutex);
 127     while (s->in_flight == MAX_NBD_REQUESTS) {
 128         qemu_co_queue_wait(&s->free_sema, &s->send_mutex);
 129     }
 130     s->in_flight++;
 131
 132     for (i = 0; i < MAX_NBD_REQUESTS; i++) {
 133         if (s->requests[i].coroutine == NULL) {
 134             break;
 135         }
 136     }
 137
 138     g_assert(qemu_in_coroutine());
 139     assert(i < MAX_NBD_REQUESTS);
 140
 141     s->requests[i].coroutine = qemu_coroutine_self();
 142     s->requests[i].receiving = false;
 143
 144     request->handle = INDEX_TO_HANDLE(s, i);
 145
 146     if (s->quit) {
 147         rc = -EIO;
 148         goto err;
 149     }
 150     if (!s->ioc) {
 151         rc = -EPIPE;
 152         goto err;
 153     }
 154
 155     if (qiov) {
 156         qio_channel_set_cork(s->ioc, true);
 157         rc = nbd_send_request(s->ioc, request);
 158         if (rc >= 0 && !s->quit) {
 159             ret = nbd_rwv(s->ioc, qiov->iov, qiov->niov, request->len, false,
 160                           NULL);
 161             if (ret != request->len) {
 162                 rc = -EIO;
 163             }
 164         }
 165         qio_channel_set_cork(s->ioc, false);
 166     } else {
 167         rc = nbd_send_request(s->ioc, request);
 168     }
 169
 170 err:
 171     if (rc < 0) {
 172         s->quit = true;
 173         s->requests[i].coroutine = NULL;
 174         s->in_flight--;
 175         qemu_co_queue_next(&s->free_sema);
 176     }
 177     qemu_co_mutex_unlock(&s->send_mutex);
 178     return rc;
 179 }
 180
 181 static void nbd_co_receive_reply(NBDClientSession *s,
 182                                  NBDRequest *request,
 183                                  NBDReply *reply,
 184                                  QEMUIOVector *qiov)
 185 {
 186     int i = HANDLE_TO_INDEX(s, request->handle);
 187     int ret;
 188
 189     /* Wait until we're woken up by nbd_read_reply_entry.  */
 190     s->requests[i].receiving = true;
 191     qemu_coroutine_yield();
 192     s->requests[i].receiving = false;
 193     *reply = s->reply;
 194     if (reply->handle != request->handle || !s->ioc || s->quit) {
 195         reply->error = EIO;
 196     } else {
 197         if (qiov && reply->error == 0) {
 198             ret = nbd_rwv(s->ioc, qiov->iov, qiov->niov, request->len, true,
 199                           NULL);
 200             if (ret != request->len) {
 201                 reply->error = EIO;
 202                 s->quit = true;
 203             }
 204         }
 205
 206         /* Tell the read handler to read another header.  */
 207         s->reply.handle = 0;
 208     }
 209
 210     s->requests[i].coroutine = NULL;
 211
 212     /* Kick the read_reply_co to get the next reply.  */
 213     if (s->read_reply_co) {
 214         aio_co_wake(s->read_reply_co);
 215     }
 216
 217     qemu_co_mutex_lock(&s->send_mutex);
 218     s->in_flight--;
 219     qemu_co_queue_next(&s->free_sema);
 220     qemu_co_mutex_unlock(&s->send_mutex);
 221 }
 222
 223 static int nbd_co_request(BlockDriverState *bs,
 224                           NBDRequest *request,
 225                           QEMUIOVector *qiov)
 226 {
 227     NBDClientSession *client = nbd_get_client_session(bs);
 228     NBDReply reply;
 229     int ret;
 230
 231     assert(!qiov || request->type == NBD_CMD_WRITE ||
 232            request->type == NBD_CMD_READ);
 233     ret = nbd_co_send_request(bs, request,
 234                               request->type == NBD_CMD_WRITE ? qiov : NULL);
 235     if (ret < 0) {
 236         reply.error = -ret;
 237     } else {
 238         nbd_co_receive_reply(client, request, &reply,
 239                              request->type == NBD_CMD_READ ? qiov : NULL);
 240     }
 241     return -reply.error;
 242 }
 243
 244 int nbd_client_co_preadv(BlockDriverState *bs, uint64_t offset,
 245                          uint64_t bytes, QEMUIOVector *qiov, int flags)
 246 {
 247     NBDRequest request = {
 248         .type = NBD_CMD_READ,
 249         .from = offset,
 250         .len = bytes,
 251     };
 252
 253     assert(bytes <= NBD_MAX_BUFFER_SIZE);
 254     assert(!flags);
 255
 256     return nbd_co_request(bs, &request, qiov);
 257 }
 258
 259 int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,
 260                           uint64_t bytes, QEMUIOVector *qiov, int flags)
 261 {
 262     NBDClientSession *client = nbd_get_client_session(bs);
 263     NBDRequest request = {
 264         .type = NBD_CMD_WRITE,
 265         .from = offset,
 266         .len = bytes,
 267     };
 268
 269     if (flags & BDRV_REQ_FUA) {
 270         assert(client->info.flags & NBD_FLAG_SEND_FUA);
 271         request.flags |= NBD_CMD_FLAG_FUA;
 272     }
 273
 274     assert(bytes <= NBD_MAX_BUFFER_SIZE);
 275
 276     return nbd_co_request(bs, &request, qiov);
 277 }
 278
 279 int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
 280                                 int bytes, BdrvRequestFlags flags)
 281 {
 282     NBDClientSession *client = nbd_get_client_session(bs);
 283     NBDRequest request = {
 284         .type = NBD_CMD_WRITE_ZEROES,
 285         .from = offset,
 286         .len = bytes,
 287     };
 288
 289     if (!(client->info.flags & NBD_FLAG_SEND_WRITE_ZEROES)) {
 290         return -ENOTSUP;
 291     }
 292
 293     if (flags & BDRV_REQ_FUA) {
 294         assert(client->info.flags & NBD_FLAG_SEND_FUA);
 295         request.flags |= NBD_CMD_FLAG_FUA;
 296     }
 297     if (!(flags & BDRV_REQ_MAY_UNMAP)) {
 298         request.flags |= NBD_CMD_FLAG_NO_HOLE;
 299     }
 300
 301     return nbd_co_request(bs, &request, NULL);
 302 }
 303
 304 int nbd_client_co_flush(BlockDriverState *bs)
 305 {
 306     NBDClientSession *client = nbd_get_client_session(bs);
 307     NBDRequest request = { .type = NBD_CMD_FLUSH };
 308
 309     if (!(client->info.flags & NBD_FLAG_SEND_FLUSH)) {
 310         return 0;
 311     }
 312
 313     request.from = 0;
 314     request.len = 0;
 315
 316     return nbd_co_request(bs, &request, NULL);
 317 }
 318
 319 int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int bytes)
 320 {
 321     NBDClientSession *client = nbd_get_client_session(bs);
 322     NBDRequest request = {
 323         .type = NBD_CMD_TRIM,
 324         .from = offset,
 325         .len = bytes,
 326     };
 327
 328     if (!(client->info.flags & NBD_FLAG_SEND_TRIM)) {
 329         return 0;
 330     }
 331
 332     return nbd_co_request(bs, &request, NULL);
 333 }
 334
 335 void nbd_client_detach_aio_context(BlockDriverState *bs)
 336 {
 337     NBDClientSession *client = nbd_get_client_session(bs);
 338     qio_channel_detach_aio_context(QIO_CHANNEL(client->ioc));
 339 }
 340
 341 void nbd_client_attach_aio_context(BlockDriverState *bs,
 342                                    AioContext *new_context)
 343 {
 344     NBDClientSession *client = nbd_get_client_session(bs);
 345     qio_channel_attach_aio_context(QIO_CHANNEL(client->ioc), new_context);
 346     aio_co_schedule(new_context, client->read_reply_co);
 347 }
 348
 349 void nbd_client_close(BlockDriverState *bs)
 350 {
 351     NBDClientSession *client = nbd_get_client_session(bs);
 352     NBDRequest request = { .type = NBD_CMD_DISC };
 353
 354     if (client->ioc == NULL) {
 355         return;
 356     }
 357
 358     nbd_send_request(client->ioc, &request);
 359
 360     nbd_teardown_connection(bs);
 361 }
 362
 363 int nbd_client_init(BlockDriverState *bs,
 364                     QIOChannelSocket *sioc,
 365                     const char *export,
 366                     QCryptoTLSCreds *tlscreds,
 367                     const char *hostname,
 368                     Error **errp)
 369 {
 370     NBDClientSession *client = nbd_get_client_session(bs);
 371     int ret;
 372
 373     /* NBD handshake */
 374     logout("session init %s\n", export);
 375     qio_channel_set_blocking(QIO_CHANNEL(sioc), true, NULL);
 376
 377     client->info.request_sizes = true;
 378     ret = nbd_receive_negotiate(QIO_CHANNEL(sioc), export,
 379                                 tlscreds, hostname,
 380                                 &client->ioc, &client->info, errp);
 381     if (ret < 0) {
 382         logout("Failed to negotiate with the NBD server\n");
 383         return ret;
 384     }
 385     if (client->info.flags & NBD_FLAG_SEND_FUA) {
 386         bs->supported_write_flags = BDRV_REQ_FUA;
 387         bs->supported_zero_flags |= BDRV_REQ_FUA;
 388     }
 389     if (client->info.flags & NBD_FLAG_SEND_WRITE_ZEROES) {
 390         bs->supported_zero_flags |= BDRV_REQ_MAY_UNMAP;
 391     }
 392     if (client->info.min_block > bs->bl.request_alignment) {
 393         bs->bl.request_alignment = client->info.min_block;
 394     }
 395
 396     qemu_co_mutex_init(&client->send_mutex);
 397     qemu_co_queue_init(&client->free_sema);
 398     client->sioc = sioc;
 399     object_ref(OBJECT(client->sioc));
 400
 401     if (!client->ioc) {
 402         client->ioc = QIO_CHANNEL(sioc);
 403         object_ref(OBJECT(client->ioc));
 404     }
 405
 406     /* Now that we're connected, set the socket to be non-blocking and
 407      * kick the reply mechanism.  */
 408     qio_channel_set_blocking(QIO_CHANNEL(sioc), false, NULL);
 409     client->read_reply_co = qemu_coroutine_create(nbd_read_reply_entry, client);
 410     nbd_client_attach_aio_context(bs, bdrv_get_aio_context(bs));
 411
 412     logout("Established connection with NBD server\n");
 413     return 0;
 414 }