xref: /openbmc/qemu/block/rbd.c (revision 19f4ed36)
1 /*
2  * QEMU Block driver for RADOS (Ceph)
3  *
4  * Copyright (C) 2010-2011 Christian Brunner <chb@muc.de>,
5  *                         Josh Durgin <josh.durgin@dreamhost.com>
6  *
7  * This work is licensed under the terms of the GNU GPL, version 2.  See
8  * the COPYING file in the top-level directory.
9  *
10  * Contributions after 2012-01-13 are licensed under the terms of the
11  * GNU GPL, version 2 or (at your option) any later version.
12  */
13 
14 #include "qemu/osdep.h"
15 
16 #include <rbd/librbd.h>
17 #include "qapi/error.h"
18 #include "qemu/error-report.h"
19 #include "qemu/module.h"
20 #include "qemu/option.h"
21 #include "block/block_int.h"
22 #include "block/qdict.h"
23 #include "crypto/secret.h"
24 #include "qemu/cutils.h"
25 #include "sysemu/replay.h"
26 #include "qapi/qmp/qstring.h"
27 #include "qapi/qmp/qdict.h"
28 #include "qapi/qmp/qjson.h"
29 #include "qapi/qmp/qlist.h"
30 #include "qapi/qobject-input-visitor.h"
31 #include "qapi/qapi-visit-block-core.h"
32 
33 /*
34  * When specifying the image filename use:
35  *
36  * rbd:poolname/devicename[@snapshotname][:option1=value1[:option2=value2...]]
37  *
38  * poolname must be the name of an existing rados pool.
39  *
40  * devicename is the name of the rbd image.
41  *
42  * Each option given is used to configure rados, and may be any valid
43  * Ceph option, "id", or "conf".
44  *
45  * The "id" option indicates what user we should authenticate as to
46  * the Ceph cluster.  If it is excluded we will use the Ceph default
47  * (normally 'admin').
48  *
49  * The "conf" option specifies a Ceph configuration file to read.  If
50  * it is not specified, we will read from the default Ceph locations
51  * (e.g., /etc/ceph/ceph.conf).  To avoid reading _any_ configuration
52  * file, specify conf=/dev/null.
53  *
54  * Configuration values containing :, @, or = can be escaped with a
55  * leading "\".
56  */
57 
58 /* rbd_aio_discard added in 0.1.2 */
59 #if LIBRBD_VERSION_CODE >= LIBRBD_VERSION(0, 1, 2)
60 #define LIBRBD_SUPPORTS_DISCARD
61 #else
62 #undef LIBRBD_SUPPORTS_DISCARD
63 #endif
64 
65 #define OBJ_MAX_SIZE (1UL << OBJ_DEFAULT_OBJ_ORDER)
66 
67 #define RBD_MAX_SNAPS 100
68 
69 /* The LIBRBD_SUPPORTS_IOVEC is defined in librbd.h */
70 #ifdef LIBRBD_SUPPORTS_IOVEC
71 #define LIBRBD_USE_IOVEC 1
72 #else
73 #define LIBRBD_USE_IOVEC 0
74 #endif
75 
76 typedef enum {
77     RBD_AIO_READ,
78     RBD_AIO_WRITE,
79     RBD_AIO_DISCARD,
80     RBD_AIO_FLUSH
81 } RBDAIOCmd;
82 
83 typedef struct RBDAIOCB {
84     BlockAIOCB common;
85     int64_t ret;
86     QEMUIOVector *qiov;
87     char *bounce;
88     RBDAIOCmd cmd;
89     int error;
90     struct BDRVRBDState *s;
91 } RBDAIOCB;
92 
93 typedef struct RADOSCB {
94     RBDAIOCB *acb;
95     struct BDRVRBDState *s;
96     int64_t size;
97     char *buf;
98     int64_t ret;
99 } RADOSCB;
100 
101 typedef struct BDRVRBDState {
102     rados_t cluster;
103     rados_ioctx_t io_ctx;
104     rbd_image_t image;
105     char *image_name;
106     char *snap;
107     char *namespace;
108     uint64_t image_size;
109 } BDRVRBDState;
110 
111 static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
112                             BlockdevOptionsRbd *opts, bool cache,
113                             const char *keypairs, const char *secretid,
114                             Error **errp);
115 
116 static char *qemu_rbd_next_tok(char *src, char delim, char **p)
117 {
118     char *end;
119 
120     *p = NULL;
121 
122     for (end = src; *end; ++end) {
123         if (*end == delim) {
124             break;
125         }
126         if (*end == '\\' && end[1] != '\0') {
127             end++;
128         }
129     }
130     if (*end == delim) {
131         *p = end + 1;
132         *end = '\0';
133     }
134     return src;
135 }
136 
137 static void qemu_rbd_unescape(char *src)
138 {
139     char *p;
140 
141     for (p = src; *src; ++src, ++p) {
142         if (*src == '\\' && src[1] != '\0') {
143             src++;
144         }
145         *p = *src;
146     }
147     *p = '\0';
148 }
149 
150 static void qemu_rbd_parse_filename(const char *filename, QDict *options,
151                                     Error **errp)
152 {
153     const char *start;
154     char *p, *buf;
155     QList *keypairs = NULL;
156     char *found_str, *image_name;
157 
158     if (!strstart(filename, "rbd:", &start)) {
159         error_setg(errp, "File name must start with 'rbd:'");
160         return;
161     }
162 
163     buf = g_strdup(start);
164     p = buf;
165 
166     found_str = qemu_rbd_next_tok(p, '/', &p);
167     if (!p) {
168         error_setg(errp, "Pool name is required");
169         goto done;
170     }
171     qemu_rbd_unescape(found_str);
172     qdict_put_str(options, "pool", found_str);
173 
174     if (strchr(p, '@')) {
175         image_name = qemu_rbd_next_tok(p, '@', &p);
176 
177         found_str = qemu_rbd_next_tok(p, ':', &p);
178         qemu_rbd_unescape(found_str);
179         qdict_put_str(options, "snapshot", found_str);
180     } else {
181         image_name = qemu_rbd_next_tok(p, ':', &p);
182     }
183     /* Check for namespace in the image_name */
184     if (strchr(image_name, '/')) {
185         found_str = qemu_rbd_next_tok(image_name, '/', &image_name);
186         qemu_rbd_unescape(found_str);
187         qdict_put_str(options, "namespace", found_str);
188     } else {
189         qdict_put_str(options, "namespace", "");
190     }
191     qemu_rbd_unescape(image_name);
192     qdict_put_str(options, "image", image_name);
193     if (!p) {
194         goto done;
195     }
196 
197     /* The following are essentially all key/value pairs, and we treat
198      * 'id' and 'conf' a bit special.  Key/value pairs may be in any order. */
199     while (p) {
200         char *name, *value;
201         name = qemu_rbd_next_tok(p, '=', &p);
202         if (!p) {
203             error_setg(errp, "conf option %s has no value", name);
204             break;
205         }
206 
207         qemu_rbd_unescape(name);
208 
209         value = qemu_rbd_next_tok(p, ':', &p);
210         qemu_rbd_unescape(value);
211 
212         if (!strcmp(name, "conf")) {
213             qdict_put_str(options, "conf", value);
214         } else if (!strcmp(name, "id")) {
215             qdict_put_str(options, "user", value);
216         } else {
217             /*
218              * We pass these internally to qemu_rbd_set_keypairs(), so
219              * we can get away with the simpler list of [ "key1",
220              * "value1", "key2", "value2" ] rather than a raw dict
221              * { "key1": "value1", "key2": "value2" } where we can't
222              * guarantee order, or even a more correct but complex
223              * [ { "key1": "value1" }, { "key2": "value2" } ]
224              */
225             if (!keypairs) {
226                 keypairs = qlist_new();
227             }
228             qlist_append_str(keypairs, name);
229             qlist_append_str(keypairs, value);
230         }
231     }
232 
233     if (keypairs) {
234         qdict_put(options, "=keyvalue-pairs",
235                   qstring_from_gstring(qobject_to_json(QOBJECT(keypairs))));
236     }
237 
238 done:
239     g_free(buf);
240     qobject_unref(keypairs);
241     return;
242 }
243 
244 
245 static void qemu_rbd_refresh_limits(BlockDriverState *bs, Error **errp)
246 {
247     /* XXX Does RBD support AIO on less than 512-byte alignment? */
248     bs->bl.request_alignment = 512;
249 }
250 
251 
252 static int qemu_rbd_set_auth(rados_t cluster, BlockdevOptionsRbd *opts,
253                              Error **errp)
254 {
255     char *key, *acr;
256     int r;
257     GString *accu;
258     RbdAuthModeList *auth;
259 
260     if (opts->key_secret) {
261         key = qcrypto_secret_lookup_as_base64(opts->key_secret, errp);
262         if (!key) {
263             return -EIO;
264         }
265         r = rados_conf_set(cluster, "key", key);
266         g_free(key);
267         if (r < 0) {
268             error_setg_errno(errp, -r, "Could not set 'key'");
269             return r;
270         }
271     }
272 
273     if (opts->has_auth_client_required) {
274         accu = g_string_new("");
275         for (auth = opts->auth_client_required; auth; auth = auth->next) {
276             if (accu->str[0]) {
277                 g_string_append_c(accu, ';');
278             }
279             g_string_append(accu, RbdAuthMode_str(auth->value));
280         }
281         acr = g_string_free(accu, FALSE);
282         r = rados_conf_set(cluster, "auth_client_required", acr);
283         g_free(acr);
284         if (r < 0) {
285             error_setg_errno(errp, -r,
286                              "Could not set 'auth_client_required'");
287             return r;
288         }
289     }
290 
291     return 0;
292 }
293 
294 static int qemu_rbd_set_keypairs(rados_t cluster, const char *keypairs_json,
295                                  Error **errp)
296 {
297     QList *keypairs;
298     QString *name;
299     QString *value;
300     const char *key;
301     size_t remaining;
302     int ret = 0;
303 
304     if (!keypairs_json) {
305         return ret;
306     }
307     keypairs = qobject_to(QList,
308                           qobject_from_json(keypairs_json, &error_abort));
309     remaining = qlist_size(keypairs) / 2;
310     assert(remaining);
311 
312     while (remaining--) {
313         name = qobject_to(QString, qlist_pop(keypairs));
314         value = qobject_to(QString, qlist_pop(keypairs));
315         assert(name && value);
316         key = qstring_get_str(name);
317 
318         ret = rados_conf_set(cluster, key, qstring_get_str(value));
319         qobject_unref(value);
320         if (ret < 0) {
321             error_setg_errno(errp, -ret, "invalid conf option %s", key);
322             qobject_unref(name);
323             ret = -EINVAL;
324             break;
325         }
326         qobject_unref(name);
327     }
328 
329     qobject_unref(keypairs);
330     return ret;
331 }
332 
333 static void qemu_rbd_memset(RADOSCB *rcb, int64_t offs)
334 {
335     if (LIBRBD_USE_IOVEC) {
336         RBDAIOCB *acb = rcb->acb;
337         iov_memset(acb->qiov->iov, acb->qiov->niov, offs, 0,
338                    acb->qiov->size - offs);
339     } else {
340         memset(rcb->buf + offs, 0, rcb->size - offs);
341     }
342 }
343 
344 /* FIXME Deprecate and remove keypairs or make it available in QMP. */
345 static int qemu_rbd_do_create(BlockdevCreateOptions *options,
346                               const char *keypairs, const char *password_secret,
347                               Error **errp)
348 {
349     BlockdevCreateOptionsRbd *opts = &options->u.rbd;
350     rados_t cluster;
351     rados_ioctx_t io_ctx;
352     int obj_order = 0;
353     int ret;
354 
355     assert(options->driver == BLOCKDEV_DRIVER_RBD);
356     if (opts->location->has_snapshot) {
357         error_setg(errp, "Can't use snapshot name for image creation");
358         return -EINVAL;
359     }
360 
361     if (opts->has_cluster_size) {
362         int64_t objsize = opts->cluster_size;
363         if ((objsize - 1) & objsize) {    /* not a power of 2? */
364             error_setg(errp, "obj size needs to be power of 2");
365             return -EINVAL;
366         }
367         if (objsize < 4096) {
368             error_setg(errp, "obj size too small");
369             return -EINVAL;
370         }
371         obj_order = ctz32(objsize);
372     }
373 
374     ret = qemu_rbd_connect(&cluster, &io_ctx, opts->location, false, keypairs,
375                            password_secret, errp);
376     if (ret < 0) {
377         return ret;
378     }
379 
380     ret = rbd_create(io_ctx, opts->location->image, opts->size, &obj_order);
381     if (ret < 0) {
382         error_setg_errno(errp, -ret, "error rbd create");
383         goto out;
384     }
385 
386     ret = 0;
387 out:
388     rados_ioctx_destroy(io_ctx);
389     rados_shutdown(cluster);
390     return ret;
391 }
392 
393 static int qemu_rbd_co_create(BlockdevCreateOptions *options, Error **errp)
394 {
395     return qemu_rbd_do_create(options, NULL, NULL, errp);
396 }
397 
398 static int coroutine_fn qemu_rbd_co_create_opts(BlockDriver *drv,
399                                                 const char *filename,
400                                                 QemuOpts *opts,
401                                                 Error **errp)
402 {
403     BlockdevCreateOptions *create_options;
404     BlockdevCreateOptionsRbd *rbd_opts;
405     BlockdevOptionsRbd *loc;
406     Error *local_err = NULL;
407     const char *keypairs, *password_secret;
408     QDict *options = NULL;
409     int ret = 0;
410 
411     create_options = g_new0(BlockdevCreateOptions, 1);
412     create_options->driver = BLOCKDEV_DRIVER_RBD;
413     rbd_opts = &create_options->u.rbd;
414 
415     rbd_opts->location = g_new0(BlockdevOptionsRbd, 1);
416 
417     password_secret = qemu_opt_get(opts, "password-secret");
418 
419     /* Read out options */
420     rbd_opts->size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
421                               BDRV_SECTOR_SIZE);
422     rbd_opts->cluster_size = qemu_opt_get_size_del(opts,
423                                                    BLOCK_OPT_CLUSTER_SIZE, 0);
424     rbd_opts->has_cluster_size = (rbd_opts->cluster_size != 0);
425 
426     options = qdict_new();
427     qemu_rbd_parse_filename(filename, options, &local_err);
428     if (local_err) {
429         ret = -EINVAL;
430         error_propagate(errp, local_err);
431         goto exit;
432     }
433 
434     /*
435      * Caution: while qdict_get_try_str() is fine, getting non-string
436      * types would require more care.  When @options come from -blockdev
437      * or blockdev_add, its members are typed according to the QAPI
438      * schema, but when they come from -drive, they're all QString.
439      */
440     loc = rbd_opts->location;
441     loc->pool        = g_strdup(qdict_get_try_str(options, "pool"));
442     loc->conf        = g_strdup(qdict_get_try_str(options, "conf"));
443     loc->has_conf    = !!loc->conf;
444     loc->user        = g_strdup(qdict_get_try_str(options, "user"));
445     loc->has_user    = !!loc->user;
446     loc->q_namespace = g_strdup(qdict_get_try_str(options, "namespace"));
447     loc->has_q_namespace = !!loc->q_namespace;
448     loc->image       = g_strdup(qdict_get_try_str(options, "image"));
449     keypairs         = qdict_get_try_str(options, "=keyvalue-pairs");
450 
451     ret = qemu_rbd_do_create(create_options, keypairs, password_secret, errp);
452     if (ret < 0) {
453         goto exit;
454     }
455 
456 exit:
457     qobject_unref(options);
458     qapi_free_BlockdevCreateOptions(create_options);
459     return ret;
460 }
461 
462 /*
463  * This aio completion is being called from rbd_finish_bh() and runs in qemu
464  * BH context.
465  */
466 static void qemu_rbd_complete_aio(RADOSCB *rcb)
467 {
468     RBDAIOCB *acb = rcb->acb;
469     int64_t r;
470 
471     r = rcb->ret;
472 
473     if (acb->cmd != RBD_AIO_READ) {
474         if (r < 0) {
475             acb->ret = r;
476             acb->error = 1;
477         } else if (!acb->error) {
478             acb->ret = rcb->size;
479         }
480     } else {
481         if (r < 0) {
482             qemu_rbd_memset(rcb, 0);
483             acb->ret = r;
484             acb->error = 1;
485         } else if (r < rcb->size) {
486             qemu_rbd_memset(rcb, r);
487             if (!acb->error) {
488                 acb->ret = rcb->size;
489             }
490         } else if (!acb->error) {
491             acb->ret = r;
492         }
493     }
494 
495     g_free(rcb);
496 
497     if (!LIBRBD_USE_IOVEC) {
498         if (acb->cmd == RBD_AIO_READ) {
499             qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
500         }
501         qemu_vfree(acb->bounce);
502     }
503 
504     acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
505 
506     qemu_aio_unref(acb);
507 }
508 
509 static char *qemu_rbd_mon_host(BlockdevOptionsRbd *opts, Error **errp)
510 {
511     const char **vals;
512     const char *host, *port;
513     char *rados_str;
514     InetSocketAddressBaseList *p;
515     int i, cnt;
516 
517     if (!opts->has_server) {
518         return NULL;
519     }
520 
521     for (cnt = 0, p = opts->server; p; p = p->next) {
522         cnt++;
523     }
524 
525     vals = g_new(const char *, cnt + 1);
526 
527     for (i = 0, p = opts->server; p; p = p->next, i++) {
528         host = p->value->host;
529         port = p->value->port;
530 
531         if (strchr(host, ':')) {
532             vals[i] = g_strdup_printf("[%s]:%s", host, port);
533         } else {
534             vals[i] = g_strdup_printf("%s:%s", host, port);
535         }
536     }
537     vals[i] = NULL;
538 
539     rados_str = i ? g_strjoinv(";", (char **)vals) : NULL;
540     g_strfreev((char **)vals);
541     return rados_str;
542 }
543 
544 static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
545                             BlockdevOptionsRbd *opts, bool cache,
546                             const char *keypairs, const char *secretid,
547                             Error **errp)
548 {
549     char *mon_host = NULL;
550     Error *local_err = NULL;
551     int r;
552 
553     if (secretid) {
554         if (opts->key_secret) {
555             error_setg(errp,
556                        "Legacy 'password-secret' clashes with 'key-secret'");
557             return -EINVAL;
558         }
559         opts->key_secret = g_strdup(secretid);
560         opts->has_key_secret = true;
561     }
562 
563     mon_host = qemu_rbd_mon_host(opts, &local_err);
564     if (local_err) {
565         error_propagate(errp, local_err);
566         r = -EINVAL;
567         goto out;
568     }
569 
570     r = rados_create(cluster, opts->user);
571     if (r < 0) {
572         error_setg_errno(errp, -r, "error initializing");
573         goto out;
574     }
575 
576     /* try default location when conf=NULL, but ignore failure */
577     r = rados_conf_read_file(*cluster, opts->conf);
578     if (opts->has_conf && r < 0) {
579         error_setg_errno(errp, -r, "error reading conf file %s", opts->conf);
580         goto failed_shutdown;
581     }
582 
583     r = qemu_rbd_set_keypairs(*cluster, keypairs, errp);
584     if (r < 0) {
585         goto failed_shutdown;
586     }
587 
588     if (mon_host) {
589         r = rados_conf_set(*cluster, "mon_host", mon_host);
590         if (r < 0) {
591             goto failed_shutdown;
592         }
593     }
594 
595     r = qemu_rbd_set_auth(*cluster, opts, errp);
596     if (r < 0) {
597         goto failed_shutdown;
598     }
599 
600     /*
601      * Fallback to more conservative semantics if setting cache
602      * options fails. Ignore errors from setting rbd_cache because the
603      * only possible error is that the option does not exist, and
604      * librbd defaults to no caching. If write through caching cannot
605      * be set up, fall back to no caching.
606      */
607     if (cache) {
608         rados_conf_set(*cluster, "rbd_cache", "true");
609     } else {
610         rados_conf_set(*cluster, "rbd_cache", "false");
611     }
612 
613     r = rados_connect(*cluster);
614     if (r < 0) {
615         error_setg_errno(errp, -r, "error connecting");
616         goto failed_shutdown;
617     }
618 
619     r = rados_ioctx_create(*cluster, opts->pool, io_ctx);
620     if (r < 0) {
621         error_setg_errno(errp, -r, "error opening pool %s", opts->pool);
622         goto failed_shutdown;
623     }
624     /*
625      * Set the namespace after opening the io context on the pool,
626      * if nspace == NULL or if nspace == "", it is just as we did nothing
627      */
628     rados_ioctx_set_namespace(*io_ctx, opts->q_namespace);
629 
630     r = 0;
631     goto out;
632 
633 failed_shutdown:
634     rados_shutdown(*cluster);
635 out:
636     g_free(mon_host);
637     return r;
638 }
639 
640 static int qemu_rbd_convert_options(QDict *options, BlockdevOptionsRbd **opts,
641                                     Error **errp)
642 {
643     Visitor *v;
644 
645     /* Convert the remaining options into a QAPI object */
646     v = qobject_input_visitor_new_flat_confused(options, errp);
647     if (!v) {
648         return -EINVAL;
649     }
650 
651     visit_type_BlockdevOptionsRbd(v, NULL, opts, errp);
652     visit_free(v);
653     if (!opts) {
654         return -EINVAL;
655     }
656 
657     return 0;
658 }
659 
660 static int qemu_rbd_attempt_legacy_options(QDict *options,
661                                            BlockdevOptionsRbd **opts,
662                                            char **keypairs)
663 {
664     char *filename;
665     int r;
666 
667     filename = g_strdup(qdict_get_try_str(options, "filename"));
668     if (!filename) {
669         return -EINVAL;
670     }
671     qdict_del(options, "filename");
672 
673     qemu_rbd_parse_filename(filename, options, NULL);
674 
675     /* keypairs freed by caller */
676     *keypairs = g_strdup(qdict_get_try_str(options, "=keyvalue-pairs"));
677     if (*keypairs) {
678         qdict_del(options, "=keyvalue-pairs");
679     }
680 
681     r = qemu_rbd_convert_options(options, opts, NULL);
682 
683     g_free(filename);
684     return r;
685 }
686 
687 static int qemu_rbd_open(BlockDriverState *bs, QDict *options, int flags,
688                          Error **errp)
689 {
690     BDRVRBDState *s = bs->opaque;
691     BlockdevOptionsRbd *opts = NULL;
692     const QDictEntry *e;
693     Error *local_err = NULL;
694     char *keypairs, *secretid;
695     int r;
696 
697     keypairs = g_strdup(qdict_get_try_str(options, "=keyvalue-pairs"));
698     if (keypairs) {
699         qdict_del(options, "=keyvalue-pairs");
700     }
701 
702     secretid = g_strdup(qdict_get_try_str(options, "password-secret"));
703     if (secretid) {
704         qdict_del(options, "password-secret");
705     }
706 
707     r = qemu_rbd_convert_options(options, &opts, &local_err);
708     if (local_err) {
709         /* If keypairs are present, that means some options are present in
710          * the modern option format.  Don't attempt to parse legacy option
711          * formats, as we won't support mixed usage. */
712         if (keypairs) {
713             error_propagate(errp, local_err);
714             goto out;
715         }
716 
717         /* If the initial attempt to convert and process the options failed,
718          * we may be attempting to open an image file that has the rbd options
719          * specified in the older format consisting of all key/value pairs
720          * encoded in the filename.  Go ahead and attempt to parse the
721          * filename, and see if we can pull out the required options. */
722         r = qemu_rbd_attempt_legacy_options(options, &opts, &keypairs);
723         if (r < 0) {
724             /* Propagate the original error, not the legacy parsing fallback
725              * error, as the latter was just a best-effort attempt. */
726             error_propagate(errp, local_err);
727             goto out;
728         }
729         /* Take care whenever deciding to actually deprecate; once this ability
730          * is removed, we will not be able to open any images with legacy-styled
731          * backing image strings. */
732         warn_report("RBD options encoded in the filename as keyvalue pairs "
733                     "is deprecated");
734     }
735 
736     /* Remove the processed options from the QDict (the visitor processes
737      * _all_ options in the QDict) */
738     while ((e = qdict_first(options))) {
739         qdict_del(options, e->key);
740     }
741 
742     r = qemu_rbd_connect(&s->cluster, &s->io_ctx, opts,
743                          !(flags & BDRV_O_NOCACHE), keypairs, secretid, errp);
744     if (r < 0) {
745         goto out;
746     }
747 
748     s->snap = g_strdup(opts->snapshot);
749     s->image_name = g_strdup(opts->image);
750 
751     /* rbd_open is always r/w */
752     r = rbd_open(s->io_ctx, s->image_name, &s->image, s->snap);
753     if (r < 0) {
754         error_setg_errno(errp, -r, "error reading header from %s",
755                          s->image_name);
756         goto failed_open;
757     }
758 
759     r = rbd_get_size(s->image, &s->image_size);
760     if (r < 0) {
761         error_setg_errno(errp, -r, "error getting image size from %s",
762                          s->image_name);
763         rbd_close(s->image);
764         goto failed_open;
765     }
766 
767     /* If we are using an rbd snapshot, we must be r/o, otherwise
768      * leave as-is */
769     if (s->snap != NULL) {
770         r = bdrv_apply_auto_read_only(bs, "rbd snapshots are read-only", errp);
771         if (r < 0) {
772             rbd_close(s->image);
773             goto failed_open;
774         }
775     }
776 
777     /* When extending regular files, we get zeros from the OS */
778     bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE;
779 
780     r = 0;
781     goto out;
782 
783 failed_open:
784     rados_ioctx_destroy(s->io_ctx);
785     g_free(s->snap);
786     g_free(s->image_name);
787     rados_shutdown(s->cluster);
788 out:
789     qapi_free_BlockdevOptionsRbd(opts);
790     g_free(keypairs);
791     g_free(secretid);
792     return r;
793 }
794 
795 
796 /* Since RBD is currently always opened R/W via the API,
797  * we just need to check if we are using a snapshot or not, in
798  * order to determine if we will allow it to be R/W */
799 static int qemu_rbd_reopen_prepare(BDRVReopenState *state,
800                                    BlockReopenQueue *queue, Error **errp)
801 {
802     BDRVRBDState *s = state->bs->opaque;
803     int ret = 0;
804 
805     if (s->snap && state->flags & BDRV_O_RDWR) {
806         error_setg(errp,
807                    "Cannot change node '%s' to r/w when using RBD snapshot",
808                    bdrv_get_device_or_node_name(state->bs));
809         ret = -EINVAL;
810     }
811 
812     return ret;
813 }
814 
815 static void qemu_rbd_close(BlockDriverState *bs)
816 {
817     BDRVRBDState *s = bs->opaque;
818 
819     rbd_close(s->image);
820     rados_ioctx_destroy(s->io_ctx);
821     g_free(s->snap);
822     g_free(s->image_name);
823     rados_shutdown(s->cluster);
824 }
825 
826 /* Resize the RBD image and update the 'image_size' with the current size */
827 static int qemu_rbd_resize(BlockDriverState *bs, uint64_t size)
828 {
829     BDRVRBDState *s = bs->opaque;
830     int r;
831 
832     r = rbd_resize(s->image, size);
833     if (r < 0) {
834         return r;
835     }
836 
837     s->image_size = size;
838 
839     return 0;
840 }
841 
842 static const AIOCBInfo rbd_aiocb_info = {
843     .aiocb_size = sizeof(RBDAIOCB),
844 };
845 
846 static void rbd_finish_bh(void *opaque)
847 {
848     RADOSCB *rcb = opaque;
849     qemu_rbd_complete_aio(rcb);
850 }
851 
852 /*
853  * This is the callback function for rbd_aio_read and _write
854  *
855  * Note: this function is being called from a non qemu thread so
856  * we need to be careful about what we do here. Generally we only
857  * schedule a BH, and do the rest of the io completion handling
858  * from rbd_finish_bh() which runs in a qemu context.
859  */
860 static void rbd_finish_aiocb(rbd_completion_t c, RADOSCB *rcb)
861 {
862     RBDAIOCB *acb = rcb->acb;
863 
864     rcb->ret = rbd_aio_get_return_value(c);
865     rbd_aio_release(c);
866 
867     replay_bh_schedule_oneshot_event(bdrv_get_aio_context(acb->common.bs),
868                                      rbd_finish_bh, rcb);
869 }
870 
871 static int rbd_aio_discard_wrapper(rbd_image_t image,
872                                    uint64_t off,
873                                    uint64_t len,
874                                    rbd_completion_t comp)
875 {
876 #ifdef LIBRBD_SUPPORTS_DISCARD
877     return rbd_aio_discard(image, off, len, comp);
878 #else
879     return -ENOTSUP;
880 #endif
881 }
882 
883 static int rbd_aio_flush_wrapper(rbd_image_t image,
884                                  rbd_completion_t comp)
885 {
886 #ifdef LIBRBD_SUPPORTS_AIO_FLUSH
887     return rbd_aio_flush(image, comp);
888 #else
889     return -ENOTSUP;
890 #endif
891 }
892 
893 static BlockAIOCB *rbd_start_aio(BlockDriverState *bs,
894                                  int64_t off,
895                                  QEMUIOVector *qiov,
896                                  int64_t size,
897                                  BlockCompletionFunc *cb,
898                                  void *opaque,
899                                  RBDAIOCmd cmd)
900 {
901     RBDAIOCB *acb;
902     RADOSCB *rcb = NULL;
903     rbd_completion_t c;
904     int r;
905 
906     BDRVRBDState *s = bs->opaque;
907 
908     acb = qemu_aio_get(&rbd_aiocb_info, bs, cb, opaque);
909     acb->cmd = cmd;
910     acb->qiov = qiov;
911     assert(!qiov || qiov->size == size);
912 
913     rcb = g_new(RADOSCB, 1);
914 
915     if (!LIBRBD_USE_IOVEC) {
916         if (cmd == RBD_AIO_DISCARD || cmd == RBD_AIO_FLUSH) {
917             acb->bounce = NULL;
918         } else {
919             acb->bounce = qemu_try_blockalign(bs, qiov->size);
920             if (acb->bounce == NULL) {
921                 goto failed;
922             }
923         }
924         if (cmd == RBD_AIO_WRITE) {
925             qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
926         }
927         rcb->buf = acb->bounce;
928     }
929 
930     acb->ret = 0;
931     acb->error = 0;
932     acb->s = s;
933 
934     rcb->acb = acb;
935     rcb->s = acb->s;
936     rcb->size = size;
937     r = rbd_aio_create_completion(rcb, (rbd_callback_t) rbd_finish_aiocb, &c);
938     if (r < 0) {
939         goto failed;
940     }
941 
942     switch (cmd) {
943     case RBD_AIO_WRITE: {
944         /*
945          * RBD APIs don't allow us to write more than actual size, so in order
946          * to support growing images, we resize the image before write
947          * operations that exceed the current size.
948          */
949         if (off + size > s->image_size) {
950             r = qemu_rbd_resize(bs, off + size);
951             if (r < 0) {
952                 goto failed_completion;
953             }
954         }
955 #ifdef LIBRBD_SUPPORTS_IOVEC
956             r = rbd_aio_writev(s->image, qiov->iov, qiov->niov, off, c);
957 #else
958             r = rbd_aio_write(s->image, off, size, rcb->buf, c);
959 #endif
960         break;
961     }
962     case RBD_AIO_READ:
963 #ifdef LIBRBD_SUPPORTS_IOVEC
964             r = rbd_aio_readv(s->image, qiov->iov, qiov->niov, off, c);
965 #else
966             r = rbd_aio_read(s->image, off, size, rcb->buf, c);
967 #endif
968         break;
969     case RBD_AIO_DISCARD:
970         r = rbd_aio_discard_wrapper(s->image, off, size, c);
971         break;
972     case RBD_AIO_FLUSH:
973         r = rbd_aio_flush_wrapper(s->image, c);
974         break;
975     default:
976         r = -EINVAL;
977     }
978 
979     if (r < 0) {
980         goto failed_completion;
981     }
982     return &acb->common;
983 
984 failed_completion:
985     rbd_aio_release(c);
986 failed:
987     g_free(rcb);
988     if (!LIBRBD_USE_IOVEC) {
989         qemu_vfree(acb->bounce);
990     }
991 
992     qemu_aio_unref(acb);
993     return NULL;
994 }
995 
996 static BlockAIOCB *qemu_rbd_aio_preadv(BlockDriverState *bs,
997                                        uint64_t offset, uint64_t bytes,
998                                        QEMUIOVector *qiov, int flags,
999                                        BlockCompletionFunc *cb,
1000                                        void *opaque)
1001 {
1002     return rbd_start_aio(bs, offset, qiov, bytes, cb, opaque,
1003                          RBD_AIO_READ);
1004 }
1005 
1006 static BlockAIOCB *qemu_rbd_aio_pwritev(BlockDriverState *bs,
1007                                         uint64_t offset, uint64_t bytes,
1008                                         QEMUIOVector *qiov, int flags,
1009                                         BlockCompletionFunc *cb,
1010                                         void *opaque)
1011 {
1012     return rbd_start_aio(bs, offset, qiov, bytes, cb, opaque,
1013                          RBD_AIO_WRITE);
1014 }
1015 
1016 #ifdef LIBRBD_SUPPORTS_AIO_FLUSH
1017 static BlockAIOCB *qemu_rbd_aio_flush(BlockDriverState *bs,
1018                                       BlockCompletionFunc *cb,
1019                                       void *opaque)
1020 {
1021     return rbd_start_aio(bs, 0, NULL, 0, cb, opaque, RBD_AIO_FLUSH);
1022 }
1023 
1024 #else
1025 
1026 static int qemu_rbd_co_flush(BlockDriverState *bs)
1027 {
1028 #if LIBRBD_VERSION_CODE >= LIBRBD_VERSION(0, 1, 1)
1029     /* rbd_flush added in 0.1.1 */
1030     BDRVRBDState *s = bs->opaque;
1031     return rbd_flush(s->image);
1032 #else
1033     return 0;
1034 #endif
1035 }
1036 #endif
1037 
1038 static int qemu_rbd_getinfo(BlockDriverState *bs, BlockDriverInfo *bdi)
1039 {
1040     BDRVRBDState *s = bs->opaque;
1041     rbd_image_info_t info;
1042     int r;
1043 
1044     r = rbd_stat(s->image, &info, sizeof(info));
1045     if (r < 0) {
1046         return r;
1047     }
1048 
1049     bdi->cluster_size = info.obj_size;
1050     return 0;
1051 }
1052 
1053 static int64_t qemu_rbd_getlength(BlockDriverState *bs)
1054 {
1055     BDRVRBDState *s = bs->opaque;
1056     rbd_image_info_t info;
1057     int r;
1058 
1059     r = rbd_stat(s->image, &info, sizeof(info));
1060     if (r < 0) {
1061         return r;
1062     }
1063 
1064     return info.size;
1065 }
1066 
1067 static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
1068                                              int64_t offset,
1069                                              bool exact,
1070                                              PreallocMode prealloc,
1071                                              BdrvRequestFlags flags,
1072                                              Error **errp)
1073 {
1074     int r;
1075 
1076     if (prealloc != PREALLOC_MODE_OFF) {
1077         error_setg(errp, "Unsupported preallocation mode '%s'",
1078                    PreallocMode_str(prealloc));
1079         return -ENOTSUP;
1080     }
1081 
1082     r = qemu_rbd_resize(bs, offset);
1083     if (r < 0) {
1084         error_setg_errno(errp, -r, "Failed to resize file");
1085         return r;
1086     }
1087 
1088     return 0;
1089 }
1090 
1091 static int qemu_rbd_snap_create(BlockDriverState *bs,
1092                                 QEMUSnapshotInfo *sn_info)
1093 {
1094     BDRVRBDState *s = bs->opaque;
1095     int r;
1096 
1097     if (sn_info->name[0] == '\0') {
1098         return -EINVAL; /* we need a name for rbd snapshots */
1099     }
1100 
1101     /*
1102      * rbd snapshots are using the name as the user controlled unique identifier
1103      * we can't use the rbd snapid for that purpose, as it can't be set
1104      */
1105     if (sn_info->id_str[0] != '\0' &&
1106         strcmp(sn_info->id_str, sn_info->name) != 0) {
1107         return -EINVAL;
1108     }
1109 
1110     if (strlen(sn_info->name) >= sizeof(sn_info->id_str)) {
1111         return -ERANGE;
1112     }
1113 
1114     r = rbd_snap_create(s->image, sn_info->name);
1115     if (r < 0) {
1116         error_report("failed to create snap: %s", strerror(-r));
1117         return r;
1118     }
1119 
1120     return 0;
1121 }
1122 
1123 static int qemu_rbd_snap_remove(BlockDriverState *bs,
1124                                 const char *snapshot_id,
1125                                 const char *snapshot_name,
1126                                 Error **errp)
1127 {
1128     BDRVRBDState *s = bs->opaque;
1129     int r;
1130 
1131     if (!snapshot_name) {
1132         error_setg(errp, "rbd need a valid snapshot name");
1133         return -EINVAL;
1134     }
1135 
1136     /* If snapshot_id is specified, it must be equal to name, see
1137        qemu_rbd_snap_list() */
1138     if (snapshot_id && strcmp(snapshot_id, snapshot_name)) {
1139         error_setg(errp,
1140                    "rbd do not support snapshot id, it should be NULL or "
1141                    "equal to snapshot name");
1142         return -EINVAL;
1143     }
1144 
1145     r = rbd_snap_remove(s->image, snapshot_name);
1146     if (r < 0) {
1147         error_setg_errno(errp, -r, "Failed to remove the snapshot");
1148     }
1149     return r;
1150 }
1151 
1152 static int qemu_rbd_snap_rollback(BlockDriverState *bs,
1153                                   const char *snapshot_name)
1154 {
1155     BDRVRBDState *s = bs->opaque;
1156 
1157     return rbd_snap_rollback(s->image, snapshot_name);
1158 }
1159 
1160 static int qemu_rbd_snap_list(BlockDriverState *bs,
1161                               QEMUSnapshotInfo **psn_tab)
1162 {
1163     BDRVRBDState *s = bs->opaque;
1164     QEMUSnapshotInfo *sn_info, *sn_tab = NULL;
1165     int i, snap_count;
1166     rbd_snap_info_t *snaps;
1167     int max_snaps = RBD_MAX_SNAPS;
1168 
1169     do {
1170         snaps = g_new(rbd_snap_info_t, max_snaps);
1171         snap_count = rbd_snap_list(s->image, snaps, &max_snaps);
1172         if (snap_count <= 0) {
1173             g_free(snaps);
1174         }
1175     } while (snap_count == -ERANGE);
1176 
1177     if (snap_count <= 0) {
1178         goto done;
1179     }
1180 
1181     sn_tab = g_new0(QEMUSnapshotInfo, snap_count);
1182 
1183     for (i = 0; i < snap_count; i++) {
1184         const char *snap_name = snaps[i].name;
1185 
1186         sn_info = sn_tab + i;
1187         pstrcpy(sn_info->id_str, sizeof(sn_info->id_str), snap_name);
1188         pstrcpy(sn_info->name, sizeof(sn_info->name), snap_name);
1189 
1190         sn_info->vm_state_size = snaps[i].size;
1191         sn_info->date_sec = 0;
1192         sn_info->date_nsec = 0;
1193         sn_info->vm_clock_nsec = 0;
1194     }
1195     rbd_snap_list_end(snaps);
1196     g_free(snaps);
1197 
1198  done:
1199     *psn_tab = sn_tab;
1200     return snap_count;
1201 }
1202 
1203 #ifdef LIBRBD_SUPPORTS_DISCARD
1204 static BlockAIOCB *qemu_rbd_aio_pdiscard(BlockDriverState *bs,
1205                                          int64_t offset,
1206                                          int bytes,
1207                                          BlockCompletionFunc *cb,
1208                                          void *opaque)
1209 {
1210     return rbd_start_aio(bs, offset, NULL, bytes, cb, opaque,
1211                          RBD_AIO_DISCARD);
1212 }
1213 #endif
1214 
1215 #ifdef LIBRBD_SUPPORTS_INVALIDATE
1216 static void coroutine_fn qemu_rbd_co_invalidate_cache(BlockDriverState *bs,
1217                                                       Error **errp)
1218 {
1219     BDRVRBDState *s = bs->opaque;
1220     int r = rbd_invalidate_cache(s->image);
1221     if (r < 0) {
1222         error_setg_errno(errp, -r, "Failed to invalidate the cache");
1223     }
1224 }
1225 #endif
1226 
1227 static QemuOptsList qemu_rbd_create_opts = {
1228     .name = "rbd-create-opts",
1229     .head = QTAILQ_HEAD_INITIALIZER(qemu_rbd_create_opts.head),
1230     .desc = {
1231         {
1232             .name = BLOCK_OPT_SIZE,
1233             .type = QEMU_OPT_SIZE,
1234             .help = "Virtual disk size"
1235         },
1236         {
1237             .name = BLOCK_OPT_CLUSTER_SIZE,
1238             .type = QEMU_OPT_SIZE,
1239             .help = "RBD object size"
1240         },
1241         {
1242             .name = "password-secret",
1243             .type = QEMU_OPT_STRING,
1244             .help = "ID of secret providing the password",
1245         },
1246         { /* end of list */ }
1247     }
1248 };
1249 
1250 static const char *const qemu_rbd_strong_runtime_opts[] = {
1251     "pool",
1252     "namespace",
1253     "image",
1254     "conf",
1255     "snapshot",
1256     "user",
1257     "server.",
1258     "password-secret",
1259 
1260     NULL
1261 };
1262 
1263 static BlockDriver bdrv_rbd = {
1264     .format_name            = "rbd",
1265     .instance_size          = sizeof(BDRVRBDState),
1266     .bdrv_parse_filename    = qemu_rbd_parse_filename,
1267     .bdrv_refresh_limits    = qemu_rbd_refresh_limits,
1268     .bdrv_file_open         = qemu_rbd_open,
1269     .bdrv_close             = qemu_rbd_close,
1270     .bdrv_reopen_prepare    = qemu_rbd_reopen_prepare,
1271     .bdrv_co_create         = qemu_rbd_co_create,
1272     .bdrv_co_create_opts    = qemu_rbd_co_create_opts,
1273     .bdrv_has_zero_init     = bdrv_has_zero_init_1,
1274     .bdrv_get_info          = qemu_rbd_getinfo,
1275     .create_opts            = &qemu_rbd_create_opts,
1276     .bdrv_getlength         = qemu_rbd_getlength,
1277     .bdrv_co_truncate       = qemu_rbd_co_truncate,
1278     .protocol_name          = "rbd",
1279 
1280     .bdrv_aio_preadv        = qemu_rbd_aio_preadv,
1281     .bdrv_aio_pwritev       = qemu_rbd_aio_pwritev,
1282 
1283 #ifdef LIBRBD_SUPPORTS_AIO_FLUSH
1284     .bdrv_aio_flush         = qemu_rbd_aio_flush,
1285 #else
1286     .bdrv_co_flush_to_disk  = qemu_rbd_co_flush,
1287 #endif
1288 
1289 #ifdef LIBRBD_SUPPORTS_DISCARD
1290     .bdrv_aio_pdiscard      = qemu_rbd_aio_pdiscard,
1291 #endif
1292 
1293     .bdrv_snapshot_create   = qemu_rbd_snap_create,
1294     .bdrv_snapshot_delete   = qemu_rbd_snap_remove,
1295     .bdrv_snapshot_list     = qemu_rbd_snap_list,
1296     .bdrv_snapshot_goto     = qemu_rbd_snap_rollback,
1297 #ifdef LIBRBD_SUPPORTS_INVALIDATE
1298     .bdrv_co_invalidate_cache = qemu_rbd_co_invalidate_cache,
1299 #endif
1300 
1301     .strong_runtime_opts    = qemu_rbd_strong_runtime_opts,
1302 };
1303 
1304 static void bdrv_rbd_init(void)
1305 {
1306     bdrv_register(&bdrv_rbd);
1307 }
1308 
1309 block_init(bdrv_rbd_init);
1310