blob: 03439565c7383588689d597760a3a8291fecc28d [file] [log] [blame]
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001/*
2 rbd.c -- Export ceph rados objects as a Linux block device
3
4
5 based on drivers/block/osdblk.c:
6
7 Copyright 2009 Red Hat, Inc.
8
9 This program is free software; you can redistribute it and/or modify
10 it under the terms of the GNU General Public License as published by
11 the Free Software Foundation.
12
13 This program is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
17
18 You should have received a copy of the GNU General Public License
19 along with this program; see the file COPYING. If not, write to
20 the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
21
22
23
Yehuda Sadehdfc56062010-11-19 14:51:04 -080024 For usage instructions, please refer to:
Yehuda Sadeh602adf42010-08-12 16:11:25 -070025
Yehuda Sadehdfc56062010-11-19 14:51:04 -080026 Documentation/ABI/testing/sysfs-bus-rbd
Yehuda Sadeh602adf42010-08-12 16:11:25 -070027
28 */
29
30#include <linux/ceph/libceph.h>
31#include <linux/ceph/osd_client.h>
32#include <linux/ceph/mon_client.h>
33#include <linux/ceph/decode.h>
Yehuda Sadeh59c2be12011-03-21 15:10:11 -070034#include <linux/parser.h>
Yehuda Sadeh602adf42010-08-12 16:11:25 -070035
36#include <linux/kernel.h>
37#include <linux/device.h>
38#include <linux/module.h>
39#include <linux/fs.h>
40#include <linux/blkdev.h>
41
42#include "rbd_types.h"
43
Alex Elder593a9e72012-02-07 12:03:37 -060044/*
45 * The basic unit of block I/O is a sector. It is interpreted in a
46 * number of contexts in Linux (blk, bio, genhd), but the default is
47 * universally 512 bytes. These symbols are just slightly more
48 * meaningful than the bare numbers they represent.
49 */
50#define SECTOR_SHIFT 9
51#define SECTOR_SIZE (1ULL << SECTOR_SHIFT)
52
Alex Elderf0f8cef2012-01-29 13:57:44 -060053#define RBD_DRV_NAME "rbd"
54#define RBD_DRV_NAME_LONG "rbd (rados block device)"
Yehuda Sadeh602adf42010-08-12 16:11:25 -070055
56#define RBD_MINORS_PER_MAJOR 256 /* max minors per blkdev */
57
Yehuda Sadeh602adf42010-08-12 16:11:25 -070058#define RBD_MAX_SNAP_NAME_LEN 32
59#define RBD_MAX_OPT_LEN 1024
60
61#define RBD_SNAP_HEAD_NAME "-"
62
Alex Elder81a89792012-02-02 08:13:30 -060063/*
64 * An RBD device name will be "rbd#", where the "rbd" comes from
65 * RBD_DRV_NAME above, and # is a unique integer identifier.
66 * MAX_INT_FORMAT_WIDTH is used in ensuring DEV_NAME_LEN is big
67 * enough to hold all possible device names.
68 */
Yehuda Sadeh602adf42010-08-12 16:11:25 -070069#define DEV_NAME_LEN 32
Alex Elder81a89792012-02-02 08:13:30 -060070#define MAX_INT_FORMAT_WIDTH ((5 * sizeof (int)) / 2 + 1)
Yehuda Sadeh602adf42010-08-12 16:11:25 -070071
Yehuda Sadeh59c2be12011-03-21 15:10:11 -070072#define RBD_NOTIFY_TIMEOUT_DEFAULT 10
73
Yehuda Sadeh602adf42010-08-12 16:11:25 -070074/*
75 * block device image metadata (in-memory version)
76 */
77struct rbd_image_header {
78 u64 image_size;
Alex Elder849b4262012-07-09 21:04:24 -050079 char *object_prefix;
Yehuda Sadeh602adf42010-08-12 16:11:25 -070080 __u8 obj_order;
81 __u8 crypt_type;
82 __u8 comp_type;
Yehuda Sadeh602adf42010-08-12 16:11:25 -070083 struct ceph_snap_context *snapc;
84 size_t snap_names_len;
85 u64 snap_seq;
86 u32 total_snaps;
87
88 char *snap_names;
89 u64 *snap_sizes;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -070090
91 u64 obj_version;
92};
93
94struct rbd_options {
95 int notify_timeout;
Yehuda Sadeh602adf42010-08-12 16:11:25 -070096};
97
98/*
Alex Elderf0f8cef2012-01-29 13:57:44 -060099 * an instance of the client. multiple devices may share an rbd client.
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700100 */
101struct rbd_client {
102 struct ceph_client *client;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700103 struct rbd_options *rbd_opts;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700104 struct kref kref;
105 struct list_head node;
106};
107
108/*
Alex Elderf0f8cef2012-01-29 13:57:44 -0600109 * a request completion status
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700110 */
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700111struct rbd_req_status {
112 int done;
113 int rc;
114 u64 bytes;
115};
116
117/*
118 * a collection of requests
119 */
120struct rbd_req_coll {
121 int total;
122 int num_done;
123 struct kref kref;
124 struct rbd_req_status status[0];
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700125};
126
Alex Elderf0f8cef2012-01-29 13:57:44 -0600127/*
128 * a single io request
129 */
130struct rbd_request {
131 struct request *rq; /* blk layer request */
132 struct bio *bio; /* cloned bio */
133 struct page **pages; /* list of used pages */
134 u64 len;
135 int coll_index;
136 struct rbd_req_coll *coll;
137};
138
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800139struct rbd_snap {
140 struct device dev;
141 const char *name;
Josh Durgin35915382011-12-05 18:25:13 -0800142 u64 size;
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800143 struct list_head node;
144 u64 id;
145};
146
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700147/*
148 * a single device
149 */
150struct rbd_device {
151 int id; /* blkdev unique id */
152
153 int major; /* blkdev assigned major */
154 struct gendisk *disk; /* blkdev's gendisk and rq */
155 struct request_queue *q;
156
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700157 struct rbd_client *rbd_client;
158
159 char name[DEV_NAME_LEN]; /* blkdev name, e.g. rbd3 */
160
161 spinlock_t lock; /* queue lock */
162
163 struct rbd_image_header header;
Alex Elderbf3e5ae2012-07-09 21:04:23 -0500164 char *obj; /* rbd image name */
165 size_t obj_len;
Alex Eldercb8627c2012-07-09 21:04:23 -0500166 char *obj_md_name; /* hdr nm. */
Alex Elderd22f76e2012-07-12 10:46:35 -0500167 char *pool_name;
Alex Elder9bb2f332012-07-12 10:46:35 -0500168 int pool_id;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700169
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700170 struct ceph_osd_event *watch_event;
171 struct ceph_osd_request *watch_request;
172
Josh Durginc6666012011-11-21 17:11:12 -0800173 /* protects updating the header */
174 struct rw_semaphore header_rwsem;
Alex Elder820a5f32012-07-09 21:04:24 -0500175 char *snap_name;
Josh Durgin77dfe992011-11-21 13:04:42 -0800176 u64 snap_id; /* current snapshot id */
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700177 int read_only;
178
179 struct list_head node;
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800180
181 /* list of snapshots */
182 struct list_head snaps;
183
184 /* sysfs related */
185 struct device dev;
186};
187
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700188static DEFINE_MUTEX(ctl_mutex); /* Serialize open/close/setup/teardown */
Alex Eldere124a82f2012-01-29 13:57:44 -0600189
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700190static LIST_HEAD(rbd_dev_list); /* devices */
Alex Eldere124a82f2012-01-29 13:57:44 -0600191static DEFINE_SPINLOCK(rbd_dev_list_lock);
192
Alex Elder432b8582012-01-29 13:57:44 -0600193static LIST_HEAD(rbd_client_list); /* clients */
194static DEFINE_SPINLOCK(rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700195
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800196static int __rbd_init_snaps_header(struct rbd_device *rbd_dev);
197static void rbd_dev_release(struct device *dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800198static ssize_t rbd_snap_add(struct device *dev,
199 struct device_attribute *attr,
200 const char *buf,
201 size_t count);
202static void __rbd_remove_snap_dev(struct rbd_device *rbd_dev,
Justin P. Mattock69932482011-07-26 23:06:29 -0700203 struct rbd_snap *snap);
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800204
Alex Elderf0f8cef2012-01-29 13:57:44 -0600205static ssize_t rbd_add(struct bus_type *bus, const char *buf,
206 size_t count);
207static ssize_t rbd_remove(struct bus_type *bus, const char *buf,
208 size_t count);
209
210static struct bus_attribute rbd_bus_attrs[] = {
211 __ATTR(add, S_IWUSR, NULL, rbd_add),
212 __ATTR(remove, S_IWUSR, NULL, rbd_remove),
213 __ATTR_NULL
214};
215
216static struct bus_type rbd_bus_type = {
217 .name = "rbd",
218 .bus_attrs = rbd_bus_attrs,
219};
220
221static void rbd_root_dev_release(struct device *dev)
222{
223}
224
225static struct device rbd_root_dev = {
226 .init_name = "rbd",
227 .release = rbd_root_dev_release,
228};
229
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800230
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800231static struct device *rbd_get_dev(struct rbd_device *rbd_dev)
232{
233 return get_device(&rbd_dev->dev);
234}
235
236static void rbd_put_dev(struct rbd_device *rbd_dev)
237{
238 put_device(&rbd_dev->dev);
239}
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700240
Josh Durgin263c6ca2011-12-05 10:43:42 -0800241static int __rbd_refresh_header(struct rbd_device *rbd_dev);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700242
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700243static int rbd_open(struct block_device *bdev, fmode_t mode)
244{
Alex Elderf0f8cef2012-01-29 13:57:44 -0600245 struct rbd_device *rbd_dev = bdev->bd_disk->private_data;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700246
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800247 rbd_get_dev(rbd_dev);
248
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700249 set_device_ro(bdev, rbd_dev->read_only);
250
251 if ((mode & FMODE_WRITE) && rbd_dev->read_only)
252 return -EROFS;
253
254 return 0;
255}
256
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800257static int rbd_release(struct gendisk *disk, fmode_t mode)
258{
259 struct rbd_device *rbd_dev = disk->private_data;
260
261 rbd_put_dev(rbd_dev);
262
263 return 0;
264}
265
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700266static const struct block_device_operations rbd_bd_ops = {
267 .owner = THIS_MODULE,
268 .open = rbd_open,
Yehuda Sadehdfc56062010-11-19 14:51:04 -0800269 .release = rbd_release,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700270};
271
272/*
273 * Initialize an rbd client instance.
274 * We own *opt.
275 */
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700276static struct rbd_client *rbd_client_create(struct ceph_options *opt,
277 struct rbd_options *rbd_opts)
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700278{
279 struct rbd_client *rbdc;
280 int ret = -ENOMEM;
281
282 dout("rbd_client_create\n");
283 rbdc = kmalloc(sizeof(struct rbd_client), GFP_KERNEL);
284 if (!rbdc)
285 goto out_opt;
286
287 kref_init(&rbdc->kref);
288 INIT_LIST_HEAD(&rbdc->node);
289
Alex Elderbc534d82012-01-29 13:57:44 -0600290 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
291
Sage Weil6ab00d42011-08-09 09:41:59 -0700292 rbdc->client = ceph_create_client(opt, rbdc, 0, 0);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700293 if (IS_ERR(rbdc->client))
Alex Elderbc534d82012-01-29 13:57:44 -0600294 goto out_mutex;
Vasiliy Kulikov28f259b2010-09-26 12:59:37 +0400295 opt = NULL; /* Now rbdc->client is responsible for opt */
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700296
297 ret = ceph_open_session(rbdc->client);
298 if (ret < 0)
299 goto out_err;
300
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700301 rbdc->rbd_opts = rbd_opts;
302
Alex Elder432b8582012-01-29 13:57:44 -0600303 spin_lock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700304 list_add_tail(&rbdc->node, &rbd_client_list);
Alex Elder432b8582012-01-29 13:57:44 -0600305 spin_unlock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700306
Alex Elderbc534d82012-01-29 13:57:44 -0600307 mutex_unlock(&ctl_mutex);
308
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700309 dout("rbd_client_create created %p\n", rbdc);
310 return rbdc;
311
312out_err:
313 ceph_destroy_client(rbdc->client);
Alex Elderbc534d82012-01-29 13:57:44 -0600314out_mutex:
315 mutex_unlock(&ctl_mutex);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700316 kfree(rbdc);
317out_opt:
Vasiliy Kulikov28f259b2010-09-26 12:59:37 +0400318 if (opt)
319 ceph_destroy_options(opt);
320 return ERR_PTR(ret);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700321}
322
323/*
324 * Find a ceph client with specific addr and configuration.
325 */
326static struct rbd_client *__rbd_client_find(struct ceph_options *opt)
327{
328 struct rbd_client *client_node;
329
330 if (opt->flags & CEPH_OPT_NOSHARE)
331 return NULL;
332
333 list_for_each_entry(client_node, &rbd_client_list, node)
334 if (ceph_compare_options(opt, client_node->client) == 0)
335 return client_node;
336 return NULL;
337}
338
339/*
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700340 * mount options
341 */
342enum {
343 Opt_notify_timeout,
344 Opt_last_int,
345 /* int args above */
346 Opt_last_string,
347 /* string args above */
348};
349
350static match_table_t rbdopt_tokens = {
351 {Opt_notify_timeout, "notify_timeout=%d"},
352 /* int args above */
353 /* string args above */
354 {-1, NULL}
355};
356
357static int parse_rbd_opts_token(char *c, void *private)
358{
359 struct rbd_options *rbdopt = private;
360 substring_t argstr[MAX_OPT_ARGS];
361 int token, intval, ret;
362
Alex Elder21079782012-01-24 10:08:36 -0600363 token = match_token(c, rbdopt_tokens, argstr);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700364 if (token < 0)
365 return -EINVAL;
366
367 if (token < Opt_last_int) {
368 ret = match_int(&argstr[0], &intval);
369 if (ret < 0) {
370 pr_err("bad mount option arg (not int) "
371 "at '%s'\n", c);
372 return ret;
373 }
374 dout("got int token %d val %d\n", token, intval);
375 } else if (token > Opt_last_int && token < Opt_last_string) {
376 dout("got string token %d val %s\n", token,
377 argstr[0].from);
378 } else {
379 dout("got token %d\n", token);
380 }
381
382 switch (token) {
383 case Opt_notify_timeout:
384 rbdopt->notify_timeout = intval;
385 break;
386 default:
387 BUG_ON(token);
388 }
389 return 0;
390}
391
392/*
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700393 * Get a ceph client with specific addr and configuration, if one does
394 * not exist create it.
395 */
Alex Elder5214ecc2012-02-02 08:13:30 -0600396static struct rbd_client *rbd_get_client(const char *mon_addr,
397 size_t mon_addr_len,
398 char *options)
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700399{
400 struct rbd_client *rbdc;
401 struct ceph_options *opt;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700402 struct rbd_options *rbd_opts;
403
404 rbd_opts = kzalloc(sizeof(*rbd_opts), GFP_KERNEL);
405 if (!rbd_opts)
Alex Elderd720bcb2012-02-02 08:13:30 -0600406 return ERR_PTR(-ENOMEM);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700407
408 rbd_opts->notify_timeout = RBD_NOTIFY_TIMEOUT_DEFAULT;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700409
Alex Elderee577412012-01-24 10:08:36 -0600410 opt = ceph_parse_options(options, mon_addr,
Alex Elder5214ecc2012-02-02 08:13:30 -0600411 mon_addr + mon_addr_len,
Alex Elder21079782012-01-24 10:08:36 -0600412 parse_rbd_opts_token, rbd_opts);
Alex Elderee577412012-01-24 10:08:36 -0600413 if (IS_ERR(opt)) {
Alex Elderd720bcb2012-02-02 08:13:30 -0600414 kfree(rbd_opts);
415 return ERR_CAST(opt);
Alex Elderee577412012-01-24 10:08:36 -0600416 }
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700417
Alex Elder432b8582012-01-29 13:57:44 -0600418 spin_lock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700419 rbdc = __rbd_client_find(opt);
420 if (rbdc) {
Alex Eldere6994d3d2012-01-29 13:57:44 -0600421 /* using an existing client */
422 kref_get(&rbdc->kref);
Alex Elder432b8582012-01-29 13:57:44 -0600423 spin_unlock(&rbd_client_list_lock);
Alex Eldere6994d3d2012-01-29 13:57:44 -0600424
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700425 ceph_destroy_options(opt);
Alex Elder97bb59a2012-01-24 10:08:36 -0600426 kfree(rbd_opts);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700427
Alex Elderd720bcb2012-02-02 08:13:30 -0600428 return rbdc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700429 }
Alex Elder432b8582012-01-29 13:57:44 -0600430 spin_unlock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700431
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700432 rbdc = rbd_client_create(opt, rbd_opts);
Alex Elderd97081b2012-01-29 13:57:44 -0600433
Alex Elderd720bcb2012-02-02 08:13:30 -0600434 if (IS_ERR(rbdc))
435 kfree(rbd_opts);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700436
Alex Elderd720bcb2012-02-02 08:13:30 -0600437 return rbdc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700438}
439
440/*
441 * Destroy ceph client
Alex Elderd23a4b32012-01-29 13:57:43 -0600442 *
Alex Elder432b8582012-01-29 13:57:44 -0600443 * Caller must hold rbd_client_list_lock.
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700444 */
445static void rbd_client_release(struct kref *kref)
446{
447 struct rbd_client *rbdc = container_of(kref, struct rbd_client, kref);
448
449 dout("rbd_release_client %p\n", rbdc);
Alex Eldercd9d9f52012-04-04 13:35:44 -0500450 spin_lock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700451 list_del(&rbdc->node);
Alex Eldercd9d9f52012-04-04 13:35:44 -0500452 spin_unlock(&rbd_client_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700453
454 ceph_destroy_client(rbdc->client);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700455 kfree(rbdc->rbd_opts);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700456 kfree(rbdc);
457}
458
459/*
460 * Drop reference to ceph client node. If it's not referenced anymore, release
461 * it.
462 */
463static void rbd_put_client(struct rbd_device *rbd_dev)
464{
465 kref_put(&rbd_dev->rbd_client->kref, rbd_client_release);
466 rbd_dev->rbd_client = NULL;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700467}
468
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700469/*
470 * Destroy requests collection
471 */
472static void rbd_coll_release(struct kref *kref)
473{
474 struct rbd_req_coll *coll =
475 container_of(kref, struct rbd_req_coll, kref);
476
477 dout("rbd_coll_release %p\n", coll);
478 kfree(coll);
479}
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700480
481/*
482 * Create a new header structure, translate header format from the on-disk
483 * header.
484 */
485static int rbd_header_from_disk(struct rbd_image_header *header,
486 struct rbd_image_header_ondisk *ondisk,
Xi Wang50f7c4c2012-04-20 15:49:44 -0500487 u32 allocated_snaps,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700488 gfp_t gfp_flags)
489{
Xi Wang50f7c4c2012-04-20 15:49:44 -0500490 u32 i, snap_count;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700491
Alex Elder21079782012-01-24 10:08:36 -0600492 if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT)))
Josh Durgin81e759f2011-11-15 14:49:53 -0800493 return -ENXIO;
Josh Durgin81e759f2011-11-15 14:49:53 -0800494
Alex Elder00f1f362012-02-07 12:03:36 -0600495 snap_count = le32_to_cpu(ondisk->snap_count);
Xi Wang50f7c4c2012-04-20 15:49:44 -0500496 if (snap_count > (UINT_MAX - sizeof(struct ceph_snap_context))
497 / sizeof (*ondisk))
498 return -EINVAL;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700499 header->snapc = kmalloc(sizeof(struct ceph_snap_context) +
Yan, Zhengf9f9a192012-06-06 09:15:33 -0500500 snap_count * sizeof(u64),
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700501 gfp_flags);
502 if (!header->snapc)
503 return -ENOMEM;
Alex Elder00f1f362012-02-07 12:03:36 -0600504
Alex Elder00f1f362012-02-07 12:03:36 -0600505 header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700506 if (snap_count) {
507 header->snap_names = kmalloc(header->snap_names_len,
Dan Carpenterf8ad4952012-04-20 15:49:44 -0500508 gfp_flags);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700509 if (!header->snap_names)
510 goto err_snapc;
511 header->snap_sizes = kmalloc(snap_count * sizeof(u64),
Dan Carpenterf8ad4952012-04-20 15:49:44 -0500512 gfp_flags);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700513 if (!header->snap_sizes)
514 goto err_names;
515 } else {
516 header->snap_names = NULL;
517 header->snap_sizes = NULL;
518 }
Alex Elder849b4262012-07-09 21:04:24 -0500519
520 header->object_prefix = kmalloc(sizeof (ondisk->block_name) + 1,
521 gfp_flags);
522 if (!header->object_prefix)
523 goto err_sizes;
524
Alex Elderca1e49a2012-07-10 20:30:09 -0500525 memcpy(header->object_prefix, ondisk->block_name,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700526 sizeof(ondisk->block_name));
Alex Elder849b4262012-07-09 21:04:24 -0500527 header->object_prefix[sizeof (ondisk->block_name)] = '\0';
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700528
529 header->image_size = le64_to_cpu(ondisk->image_size);
530 header->obj_order = ondisk->options.order;
531 header->crypt_type = ondisk->options.crypt_type;
532 header->comp_type = ondisk->options.comp_type;
533
534 atomic_set(&header->snapc->nref, 1);
535 header->snap_seq = le64_to_cpu(ondisk->snap_seq);
536 header->snapc->num_snaps = snap_count;
537 header->total_snaps = snap_count;
538
Alex Elder21079782012-01-24 10:08:36 -0600539 if (snap_count && allocated_snaps == snap_count) {
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700540 for (i = 0; i < snap_count; i++) {
541 header->snapc->snaps[i] =
542 le64_to_cpu(ondisk->snaps[i].id);
543 header->snap_sizes[i] =
544 le64_to_cpu(ondisk->snaps[i].image_size);
545 }
546
547 /* copy snapshot names */
548 memcpy(header->snap_names, &ondisk->snaps[i],
549 header->snap_names_len);
550 }
551
552 return 0;
553
Alex Elder849b4262012-07-09 21:04:24 -0500554err_sizes:
555 kfree(header->snap_sizes);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700556err_names:
557 kfree(header->snap_names);
558err_snapc:
559 kfree(header->snapc);
Alex Elder00f1f362012-02-07 12:03:36 -0600560 return -ENOMEM;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700561}
562
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700563static int snap_by_name(struct rbd_image_header *header, const char *snap_name,
564 u64 *seq, u64 *size)
565{
566 int i;
567 char *p = header->snap_names;
568
Alex Elder00f1f362012-02-07 12:03:36 -0600569 for (i = 0; i < header->total_snaps; i++) {
570 if (!strcmp(snap_name, p)) {
571
572 /* Found it. Pass back its id and/or size */
573
574 if (seq)
575 *seq = header->snapc->snaps[i];
576 if (size)
577 *size = header->snap_sizes[i];
578 return i;
579 }
580 p += strlen(p) + 1; /* Skip ahead to the next name */
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700581 }
Alex Elder00f1f362012-02-07 12:03:36 -0600582 return -ENOENT;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700583}
584
Alex Elder0ce1a792012-07-03 16:01:18 -0500585static int rbd_header_set_snap(struct rbd_device *rbd_dev, u64 *size)
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700586{
Alex Elder0ce1a792012-07-03 16:01:18 -0500587 struct rbd_image_header *header = &rbd_dev->header;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700588 struct ceph_snap_context *snapc = header->snapc;
589 int ret = -ENOENT;
590
Alex Elder0ce1a792012-07-03 16:01:18 -0500591 down_write(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700592
Alex Elder0ce1a792012-07-03 16:01:18 -0500593 if (!memcmp(rbd_dev->snap_name, RBD_SNAP_HEAD_NAME,
Josh Durgincc9d7342011-11-21 18:19:13 -0800594 sizeof (RBD_SNAP_HEAD_NAME))) {
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700595 if (header->total_snaps)
596 snapc->seq = header->snap_seq;
597 else
598 snapc->seq = 0;
Alex Elder0ce1a792012-07-03 16:01:18 -0500599 rbd_dev->snap_id = CEPH_NOSNAP;
600 rbd_dev->read_only = 0;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700601 if (size)
602 *size = header->image_size;
603 } else {
Alex Elder0ce1a792012-07-03 16:01:18 -0500604 ret = snap_by_name(header, rbd_dev->snap_name,
605 &snapc->seq, size);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700606 if (ret < 0)
607 goto done;
Alex Elder0ce1a792012-07-03 16:01:18 -0500608 rbd_dev->snap_id = snapc->seq;
609 rbd_dev->read_only = 1;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700610 }
611
612 ret = 0;
613done:
Alex Elder0ce1a792012-07-03 16:01:18 -0500614 up_write(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700615 return ret;
616}
617
618static void rbd_header_free(struct rbd_image_header *header)
619{
Alex Elder849b4262012-07-09 21:04:24 -0500620 kfree(header->object_prefix);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700621 kfree(header->snap_sizes);
Alex Elder849b4262012-07-09 21:04:24 -0500622 kfree(header->snap_names);
623 kfree(header->snapc);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700624}
625
626/*
627 * get the actual striped segment name, offset and length
628 */
629static u64 rbd_get_segment(struct rbd_image_header *header,
Alex Elderca1e49a2012-07-10 20:30:09 -0500630 const char *object_prefix,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700631 u64 ofs, u64 len,
632 char *seg_name, u64 *segofs)
633{
634 u64 seg = ofs >> header->obj_order;
635
636 if (seg_name)
637 snprintf(seg_name, RBD_MAX_SEG_NAME_LEN,
Alex Elderca1e49a2012-07-10 20:30:09 -0500638 "%s.%012llx", object_prefix, seg);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700639
640 ofs = ofs & ((1 << header->obj_order) - 1);
641 len = min_t(u64, len, (1 << header->obj_order) - ofs);
642
643 if (segofs)
644 *segofs = ofs;
645
646 return len;
647}
648
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700649static int rbd_get_num_segments(struct rbd_image_header *header,
650 u64 ofs, u64 len)
651{
652 u64 start_seg = ofs >> header->obj_order;
653 u64 end_seg = (ofs + len - 1) >> header->obj_order;
654 return end_seg - start_seg + 1;
655}
656
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700657/*
Josh Durgin029bcbd2011-07-22 11:35:23 -0700658 * returns the size of an object in the image
659 */
660static u64 rbd_obj_bytes(struct rbd_image_header *header)
661{
662 return 1 << header->obj_order;
663}
664
665/*
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700666 * bio helpers
667 */
668
669static void bio_chain_put(struct bio *chain)
670{
671 struct bio *tmp;
672
673 while (chain) {
674 tmp = chain;
675 chain = chain->bi_next;
676 bio_put(tmp);
677 }
678}
679
680/*
681 * zeros a bio chain, starting at specific offset
682 */
683static void zero_bio_chain(struct bio *chain, int start_ofs)
684{
685 struct bio_vec *bv;
686 unsigned long flags;
687 void *buf;
688 int i;
689 int pos = 0;
690
691 while (chain) {
692 bio_for_each_segment(bv, chain, i) {
693 if (pos + bv->bv_len > start_ofs) {
694 int remainder = max(start_ofs - pos, 0);
695 buf = bvec_kmap_irq(bv, &flags);
696 memset(buf + remainder, 0,
697 bv->bv_len - remainder);
Dan Carpenter85b5aaa2010-10-11 21:15:11 +0200698 bvec_kunmap_irq(buf, &flags);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700699 }
700 pos += bv->bv_len;
701 }
702
703 chain = chain->bi_next;
704 }
705}
706
707/*
708 * bio_chain_clone - clone a chain of bios up to a certain length.
709 * might return a bio_pair that will need to be released.
710 */
711static struct bio *bio_chain_clone(struct bio **old, struct bio **next,
712 struct bio_pair **bp,
713 int len, gfp_t gfpmask)
714{
715 struct bio *tmp, *old_chain = *old, *new_chain = NULL, *tail = NULL;
716 int total = 0;
717
718 if (*bp) {
719 bio_pair_release(*bp);
720 *bp = NULL;
721 }
722
723 while (old_chain && (total < len)) {
724 tmp = bio_kmalloc(gfpmask, old_chain->bi_max_vecs);
725 if (!tmp)
726 goto err_out;
727
728 if (total + old_chain->bi_size > len) {
729 struct bio_pair *bp;
730
731 /*
732 * this split can only happen with a single paged bio,
733 * split_bio will BUG_ON if this is not the case
734 */
735 dout("bio_chain_clone split! total=%d remaining=%d"
736 "bi_size=%d\n",
737 (int)total, (int)len-total,
738 (int)old_chain->bi_size);
739
740 /* split the bio. We'll release it either in the next
741 call, or it will have to be released outside */
Alex Elder593a9e72012-02-07 12:03:37 -0600742 bp = bio_split(old_chain, (len - total) / SECTOR_SIZE);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700743 if (!bp)
744 goto err_out;
745
746 __bio_clone(tmp, &bp->bio1);
747
748 *next = &bp->bio2;
749 } else {
750 __bio_clone(tmp, old_chain);
751 *next = old_chain->bi_next;
752 }
753
754 tmp->bi_bdev = NULL;
755 gfpmask &= ~__GFP_WAIT;
756 tmp->bi_next = NULL;
757
758 if (!new_chain) {
759 new_chain = tail = tmp;
760 } else {
761 tail->bi_next = tmp;
762 tail = tmp;
763 }
764 old_chain = old_chain->bi_next;
765
766 total += tmp->bi_size;
767 }
768
769 BUG_ON(total < len);
770
771 if (tail)
772 tail->bi_next = NULL;
773
774 *old = old_chain;
775
776 return new_chain;
777
778err_out:
779 dout("bio_chain_clone with err\n");
780 bio_chain_put(new_chain);
781 return NULL;
782}
783
784/*
785 * helpers for osd request op vectors.
786 */
787static int rbd_create_rw_ops(struct ceph_osd_req_op **ops,
788 int num_ops,
789 int opcode,
790 u32 payload_len)
791{
792 *ops = kzalloc(sizeof(struct ceph_osd_req_op) * (num_ops + 1),
793 GFP_NOIO);
794 if (!*ops)
795 return -ENOMEM;
796 (*ops)[0].op = opcode;
797 /*
798 * op extent offset and length will be set later on
799 * in calc_raw_layout()
800 */
801 (*ops)[0].payload_len = payload_len;
802 return 0;
803}
804
805static void rbd_destroy_ops(struct ceph_osd_req_op *ops)
806{
807 kfree(ops);
808}
809
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700810static void rbd_coll_end_req_index(struct request *rq,
811 struct rbd_req_coll *coll,
812 int index,
813 int ret, u64 len)
814{
815 struct request_queue *q;
816 int min, max, i;
817
818 dout("rbd_coll_end_req_index %p index %d ret %d len %lld\n",
819 coll, index, ret, len);
820
821 if (!rq)
822 return;
823
824 if (!coll) {
825 blk_end_request(rq, ret, len);
826 return;
827 }
828
829 q = rq->q;
830
831 spin_lock_irq(q->queue_lock);
832 coll->status[index].done = 1;
833 coll->status[index].rc = ret;
834 coll->status[index].bytes = len;
835 max = min = coll->num_done;
836 while (max < coll->total && coll->status[max].done)
837 max++;
838
839 for (i = min; i<max; i++) {
840 __blk_end_request(rq, coll->status[i].rc,
841 coll->status[i].bytes);
842 coll->num_done++;
843 kref_put(&coll->kref, rbd_coll_release);
844 }
845 spin_unlock_irq(q->queue_lock);
846}
847
848static void rbd_coll_end_req(struct rbd_request *req,
849 int ret, u64 len)
850{
851 rbd_coll_end_req_index(req->rq, req->coll, req->coll_index, ret, len);
852}
853
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700854/*
855 * Send ceph osd request
856 */
857static int rbd_do_request(struct request *rq,
Alex Elder0ce1a792012-07-03 16:01:18 -0500858 struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700859 struct ceph_snap_context *snapc,
860 u64 snapid,
861 const char *obj, u64 ofs, u64 len,
862 struct bio *bio,
863 struct page **pages,
864 int num_pages,
865 int flags,
866 struct ceph_osd_req_op *ops,
867 int num_reply,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700868 struct rbd_req_coll *coll,
869 int coll_index,
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700870 void (*rbd_cb)(struct ceph_osd_request *req,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700871 struct ceph_msg *msg),
872 struct ceph_osd_request **linger_req,
873 u64 *ver)
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700874{
875 struct ceph_osd_request *req;
876 struct ceph_file_layout *layout;
877 int ret;
878 u64 bno;
879 struct timespec mtime = CURRENT_TIME;
880 struct rbd_request *req_data;
881 struct ceph_osd_request_head *reqhead;
Alex Elder1dbb4392012-01-24 10:08:37 -0600882 struct ceph_osd_client *osdc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700883
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700884 req_data = kzalloc(sizeof(*req_data), GFP_NOIO);
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700885 if (!req_data) {
886 if (coll)
887 rbd_coll_end_req_index(rq, coll, coll_index,
888 -ENOMEM, len);
889 return -ENOMEM;
890 }
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700891
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700892 if (coll) {
893 req_data->coll = coll;
894 req_data->coll_index = coll_index;
895 }
896
897 dout("rbd_do_request obj=%s ofs=%lld len=%lld\n", obj, len, ofs);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700898
Alex Elder0ce1a792012-07-03 16:01:18 -0500899 down_read(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700900
Alex Elder0ce1a792012-07-03 16:01:18 -0500901 osdc = &rbd_dev->rbd_client->client->osdc;
Alex Elder1dbb4392012-01-24 10:08:37 -0600902 req = ceph_osdc_alloc_request(osdc, flags, snapc, ops,
903 false, GFP_NOIO, pages, bio);
Sage Weil4ad12622011-05-03 09:23:36 -0700904 if (!req) {
Alex Elder0ce1a792012-07-03 16:01:18 -0500905 up_read(&rbd_dev->header_rwsem);
Sage Weil4ad12622011-05-03 09:23:36 -0700906 ret = -ENOMEM;
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700907 goto done_pages;
908 }
909
910 req->r_callback = rbd_cb;
911
912 req_data->rq = rq;
913 req_data->bio = bio;
914 req_data->pages = pages;
915 req_data->len = len;
916
917 req->r_priv = req_data;
918
919 reqhead = req->r_request->front.iov_base;
920 reqhead->snapid = cpu_to_le64(CEPH_NOSNAP);
921
922 strncpy(req->r_oid, obj, sizeof(req->r_oid));
923 req->r_oid_len = strlen(req->r_oid);
924
925 layout = &req->r_file_layout;
926 memset(layout, 0, sizeof(*layout));
927 layout->fl_stripe_unit = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
928 layout->fl_stripe_count = cpu_to_le32(1);
929 layout->fl_object_size = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
Alex Elder0ce1a792012-07-03 16:01:18 -0500930 layout->fl_pg_pool = cpu_to_le32(rbd_dev->pool_id);
Alex Elder1dbb4392012-01-24 10:08:37 -0600931 ceph_calc_raw_layout(osdc, layout, snapid, ofs, &len, &bno,
932 req, ops);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700933
934 ceph_osdc_build_request(req, ofs, &len,
935 ops,
936 snapc,
937 &mtime,
938 req->r_oid, req->r_oid_len);
Alex Elder0ce1a792012-07-03 16:01:18 -0500939 up_read(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700940
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700941 if (linger_req) {
Alex Elder1dbb4392012-01-24 10:08:37 -0600942 ceph_osdc_set_request_linger(osdc, req);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700943 *linger_req = req;
944 }
945
Alex Elder1dbb4392012-01-24 10:08:37 -0600946 ret = ceph_osdc_start_request(osdc, req, false);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700947 if (ret < 0)
948 goto done_err;
949
950 if (!rbd_cb) {
Alex Elder1dbb4392012-01-24 10:08:37 -0600951 ret = ceph_osdc_wait_request(osdc, req);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -0700952 if (ver)
953 *ver = le64_to_cpu(req->r_reassert_version.version);
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700954 dout("reassert_ver=%lld\n",
955 le64_to_cpu(req->r_reassert_version.version));
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700956 ceph_osdc_put_request(req);
957 }
958 return ret;
959
960done_err:
961 bio_chain_put(req_data->bio);
962 ceph_osdc_put_request(req);
963done_pages:
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700964 rbd_coll_end_req(req_data, ret, len);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700965 kfree(req_data);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700966 return ret;
967}
968
969/*
970 * Ceph osd op callback
971 */
972static void rbd_req_cb(struct ceph_osd_request *req, struct ceph_msg *msg)
973{
974 struct rbd_request *req_data = req->r_priv;
975 struct ceph_osd_reply_head *replyhead;
976 struct ceph_osd_op *op;
977 __s32 rc;
978 u64 bytes;
979 int read_op;
980
981 /* parse reply */
982 replyhead = msg->front.iov_base;
983 WARN_ON(le32_to_cpu(replyhead->num_ops) == 0);
984 op = (void *)(replyhead + 1);
985 rc = le32_to_cpu(replyhead->result);
986 bytes = le64_to_cpu(op->extent.length);
Dan Carpenter895cfcc2012-06-06 09:15:33 -0500987 read_op = (le16_to_cpu(op->op) == CEPH_OSD_OP_READ);
Yehuda Sadeh602adf42010-08-12 16:11:25 -0700988
989 dout("rbd_req_cb bytes=%lld readop=%d rc=%d\n", bytes, read_op, rc);
990
991 if (rc == -ENOENT && read_op) {
992 zero_bio_chain(req_data->bio, 0);
993 rc = 0;
994 } else if (rc == 0 && read_op && bytes < req_data->len) {
995 zero_bio_chain(req_data->bio, bytes);
996 bytes = req_data->len;
997 }
998
Yehuda Sadeh1fec7092011-05-13 13:52:56 -0700999 rbd_coll_end_req(req_data, rc, bytes);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001000
1001 if (req_data->bio)
1002 bio_chain_put(req_data->bio);
1003
1004 ceph_osdc_put_request(req);
1005 kfree(req_data);
1006}
1007
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001008static void rbd_simple_req_cb(struct ceph_osd_request *req, struct ceph_msg *msg)
1009{
1010 ceph_osdc_put_request(req);
1011}
1012
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001013/*
1014 * Do a synchronous ceph osd operation
1015 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001016static int rbd_req_sync_op(struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001017 struct ceph_snap_context *snapc,
1018 u64 snapid,
1019 int opcode,
1020 int flags,
1021 struct ceph_osd_req_op *orig_ops,
1022 int num_reply,
1023 const char *obj,
1024 u64 ofs, u64 len,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001025 char *buf,
1026 struct ceph_osd_request **linger_req,
1027 u64 *ver)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001028{
1029 int ret;
1030 struct page **pages;
1031 int num_pages;
1032 struct ceph_osd_req_op *ops = orig_ops;
1033 u32 payload_len;
1034
1035 num_pages = calc_pages_for(ofs , len);
1036 pages = ceph_alloc_page_vector(num_pages, GFP_KERNEL);
Dan Carpenterb8d06382010-10-11 21:14:23 +02001037 if (IS_ERR(pages))
1038 return PTR_ERR(pages);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001039
1040 if (!orig_ops) {
1041 payload_len = (flags & CEPH_OSD_FLAG_WRITE ? len : 0);
1042 ret = rbd_create_rw_ops(&ops, 1, opcode, payload_len);
1043 if (ret < 0)
1044 goto done;
1045
1046 if ((flags & CEPH_OSD_FLAG_WRITE) && buf) {
1047 ret = ceph_copy_to_page_vector(pages, buf, ofs, len);
1048 if (ret < 0)
1049 goto done_ops;
1050 }
1051 }
1052
Alex Elder0ce1a792012-07-03 16:01:18 -05001053 ret = rbd_do_request(NULL, rbd_dev, snapc, snapid,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001054 obj, ofs, len, NULL,
1055 pages, num_pages,
1056 flags,
1057 ops,
1058 2,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001059 NULL, 0,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001060 NULL,
1061 linger_req, ver);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001062 if (ret < 0)
1063 goto done_ops;
1064
1065 if ((flags & CEPH_OSD_FLAG_READ) && buf)
1066 ret = ceph_copy_from_page_vector(pages, buf, ofs, ret);
1067
1068done_ops:
1069 if (!orig_ops)
1070 rbd_destroy_ops(ops);
1071done:
1072 ceph_release_page_vector(pages, num_pages);
1073 return ret;
1074}
1075
1076/*
1077 * Do an asynchronous ceph osd operation
1078 */
1079static int rbd_do_op(struct request *rq,
Alex Elder0ce1a792012-07-03 16:01:18 -05001080 struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001081 struct ceph_snap_context *snapc,
1082 u64 snapid,
1083 int opcode, int flags, int num_reply,
1084 u64 ofs, u64 len,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001085 struct bio *bio,
1086 struct rbd_req_coll *coll,
1087 int coll_index)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001088{
1089 char *seg_name;
1090 u64 seg_ofs;
1091 u64 seg_len;
1092 int ret;
1093 struct ceph_osd_req_op *ops;
1094 u32 payload_len;
1095
1096 seg_name = kmalloc(RBD_MAX_SEG_NAME_LEN + 1, GFP_NOIO);
1097 if (!seg_name)
1098 return -ENOMEM;
1099
1100 seg_len = rbd_get_segment(&rbd_dev->header,
Alex Elderca1e49a2012-07-10 20:30:09 -05001101 rbd_dev->header.object_prefix,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001102 ofs, len,
1103 seg_name, &seg_ofs);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001104
1105 payload_len = (flags & CEPH_OSD_FLAG_WRITE ? seg_len : 0);
1106
1107 ret = rbd_create_rw_ops(&ops, 1, opcode, payload_len);
1108 if (ret < 0)
1109 goto done;
1110
1111 /* we've taken care of segment sizes earlier when we
1112 cloned the bios. We should never have a segment
1113 truncated at this point */
1114 BUG_ON(seg_len < len);
1115
1116 ret = rbd_do_request(rq, rbd_dev, snapc, snapid,
1117 seg_name, seg_ofs, seg_len,
1118 bio,
1119 NULL, 0,
1120 flags,
1121 ops,
1122 num_reply,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001123 coll, coll_index,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001124 rbd_req_cb, 0, NULL);
Sage Weil11f77002011-05-12 16:13:54 -07001125
1126 rbd_destroy_ops(ops);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001127done:
1128 kfree(seg_name);
1129 return ret;
1130}
1131
1132/*
1133 * Request async osd write
1134 */
1135static int rbd_req_write(struct request *rq,
1136 struct rbd_device *rbd_dev,
1137 struct ceph_snap_context *snapc,
1138 u64 ofs, u64 len,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001139 struct bio *bio,
1140 struct rbd_req_coll *coll,
1141 int coll_index)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001142{
1143 return rbd_do_op(rq, rbd_dev, snapc, CEPH_NOSNAP,
1144 CEPH_OSD_OP_WRITE,
1145 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
1146 2,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001147 ofs, len, bio, coll, coll_index);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001148}
1149
1150/*
1151 * Request async osd read
1152 */
1153static int rbd_req_read(struct request *rq,
1154 struct rbd_device *rbd_dev,
1155 u64 snapid,
1156 u64 ofs, u64 len,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001157 struct bio *bio,
1158 struct rbd_req_coll *coll,
1159 int coll_index)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001160{
1161 return rbd_do_op(rq, rbd_dev, NULL,
Josh Durginb06e6a62011-11-21 18:16:52 -08001162 snapid,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001163 CEPH_OSD_OP_READ,
1164 CEPH_OSD_FLAG_READ,
1165 2,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001166 ofs, len, bio, coll, coll_index);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001167}
1168
1169/*
1170 * Request sync osd read
1171 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001172static int rbd_req_sync_read(struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001173 struct ceph_snap_context *snapc,
1174 u64 snapid,
1175 const char *obj,
1176 u64 ofs, u64 len,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001177 char *buf,
1178 u64 *ver)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001179{
Alex Elder0ce1a792012-07-03 16:01:18 -05001180 return rbd_req_sync_op(rbd_dev, NULL,
Josh Durginb06e6a62011-11-21 18:16:52 -08001181 snapid,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001182 CEPH_OSD_OP_READ,
1183 CEPH_OSD_FLAG_READ,
1184 NULL,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001185 1, obj, ofs, len, buf, NULL, ver);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001186}
1187
1188/*
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001189 * Request sync osd watch
1190 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001191static int rbd_req_sync_notify_ack(struct rbd_device *rbd_dev,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001192 u64 ver,
1193 u64 notify_id,
1194 const char *obj)
1195{
1196 struct ceph_osd_req_op *ops;
Sage Weil11f77002011-05-12 16:13:54 -07001197 int ret;
1198
1199 ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_NOTIFY_ACK, 0);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001200 if (ret < 0)
1201 return ret;
1202
Alex Elder0ce1a792012-07-03 16:01:18 -05001203 ops[0].watch.ver = cpu_to_le64(rbd_dev->header.obj_version);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001204 ops[0].watch.cookie = notify_id;
1205 ops[0].watch.flag = 0;
1206
Alex Elder0ce1a792012-07-03 16:01:18 -05001207 ret = rbd_do_request(NULL, rbd_dev, NULL, CEPH_NOSNAP,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001208 obj, 0, 0, NULL,
Alex Elderad4f2322012-07-03 16:01:19 -05001209 NULL, 0,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001210 CEPH_OSD_FLAG_READ,
1211 ops,
1212 1,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001213 NULL, 0,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001214 rbd_simple_req_cb, 0, NULL);
1215
1216 rbd_destroy_ops(ops);
1217 return ret;
1218}
1219
1220static void rbd_watch_cb(u64 ver, u64 notify_id, u8 opcode, void *data)
1221{
Alex Elder0ce1a792012-07-03 16:01:18 -05001222 struct rbd_device *rbd_dev = (struct rbd_device *)data;
Sage Weil13143d22011-05-12 16:08:30 -07001223 int rc;
1224
Alex Elder0ce1a792012-07-03 16:01:18 -05001225 if (!rbd_dev)
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001226 return;
1227
Alex Elder0ce1a792012-07-03 16:01:18 -05001228 dout("rbd_watch_cb %s notify_id=%lld opcode=%d\n", rbd_dev->obj_md_name,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001229 notify_id, (int)opcode);
1230 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
Alex Elder0ce1a792012-07-03 16:01:18 -05001231 rc = __rbd_refresh_header(rbd_dev);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001232 mutex_unlock(&ctl_mutex);
Sage Weil13143d22011-05-12 16:08:30 -07001233 if (rc)
Alex Elderf0f8cef2012-01-29 13:57:44 -06001234 pr_warning(RBD_DRV_NAME "%d got notification but failed to "
Alex Elder0ce1a792012-07-03 16:01:18 -05001235 " update snaps: %d\n", rbd_dev->major, rc);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001236
Alex Elder0ce1a792012-07-03 16:01:18 -05001237 rbd_req_sync_notify_ack(rbd_dev, ver, notify_id, rbd_dev->obj_md_name);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001238}
1239
1240/*
1241 * Request sync osd watch
1242 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001243static int rbd_req_sync_watch(struct rbd_device *rbd_dev,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001244 const char *obj,
1245 u64 ver)
1246{
1247 struct ceph_osd_req_op *ops;
Alex Elder0ce1a792012-07-03 16:01:18 -05001248 struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001249
1250 int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_WATCH, 0);
1251 if (ret < 0)
1252 return ret;
1253
1254 ret = ceph_osdc_create_event(osdc, rbd_watch_cb, 0,
Alex Elder0ce1a792012-07-03 16:01:18 -05001255 (void *)rbd_dev, &rbd_dev->watch_event);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001256 if (ret < 0)
1257 goto fail;
1258
1259 ops[0].watch.ver = cpu_to_le64(ver);
Alex Elder0ce1a792012-07-03 16:01:18 -05001260 ops[0].watch.cookie = cpu_to_le64(rbd_dev->watch_event->cookie);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001261 ops[0].watch.flag = 1;
1262
Alex Elder0ce1a792012-07-03 16:01:18 -05001263 ret = rbd_req_sync_op(rbd_dev, NULL,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001264 CEPH_NOSNAP,
1265 0,
1266 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
1267 ops,
1268 1, obj, 0, 0, NULL,
Alex Elder0ce1a792012-07-03 16:01:18 -05001269 &rbd_dev->watch_request, NULL);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001270
1271 if (ret < 0)
1272 goto fail_event;
1273
1274 rbd_destroy_ops(ops);
1275 return 0;
1276
1277fail_event:
Alex Elder0ce1a792012-07-03 16:01:18 -05001278 ceph_osdc_cancel_event(rbd_dev->watch_event);
1279 rbd_dev->watch_event = NULL;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001280fail:
1281 rbd_destroy_ops(ops);
1282 return ret;
1283}
1284
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07001285/*
1286 * Request sync osd unwatch
1287 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001288static int rbd_req_sync_unwatch(struct rbd_device *rbd_dev,
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07001289 const char *obj)
1290{
1291 struct ceph_osd_req_op *ops;
1292
1293 int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_WATCH, 0);
1294 if (ret < 0)
1295 return ret;
1296
1297 ops[0].watch.ver = 0;
Alex Elder0ce1a792012-07-03 16:01:18 -05001298 ops[0].watch.cookie = cpu_to_le64(rbd_dev->watch_event->cookie);
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07001299 ops[0].watch.flag = 0;
1300
Alex Elder0ce1a792012-07-03 16:01:18 -05001301 ret = rbd_req_sync_op(rbd_dev, NULL,
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07001302 CEPH_NOSNAP,
1303 0,
1304 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
1305 ops,
1306 1, obj, 0, 0, NULL, NULL, NULL);
1307
1308 rbd_destroy_ops(ops);
Alex Elder0ce1a792012-07-03 16:01:18 -05001309 ceph_osdc_cancel_event(rbd_dev->watch_event);
1310 rbd_dev->watch_event = NULL;
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07001311 return ret;
1312}
1313
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001314struct rbd_notify_info {
Alex Elder0ce1a792012-07-03 16:01:18 -05001315 struct rbd_device *rbd_dev;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001316};
1317
1318static void rbd_notify_cb(u64 ver, u64 notify_id, u8 opcode, void *data)
1319{
Alex Elder0ce1a792012-07-03 16:01:18 -05001320 struct rbd_device *rbd_dev = (struct rbd_device *)data;
1321 if (!rbd_dev)
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001322 return;
1323
Alex Elder0ce1a792012-07-03 16:01:18 -05001324 dout("rbd_notify_cb %s notify_id=%lld opcode=%d\n",
1325 rbd_dev->obj_md_name,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001326 notify_id, (int)opcode);
1327}
1328
1329/*
1330 * Request sync osd notify
1331 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001332static int rbd_req_sync_notify(struct rbd_device *rbd_dev,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001333 const char *obj)
1334{
1335 struct ceph_osd_req_op *ops;
Alex Elder0ce1a792012-07-03 16:01:18 -05001336 struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001337 struct ceph_osd_event *event;
1338 struct rbd_notify_info info;
1339 int payload_len = sizeof(u32) + sizeof(u32);
1340 int ret;
1341
1342 ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_NOTIFY, payload_len);
1343 if (ret < 0)
1344 return ret;
1345
Alex Elder0ce1a792012-07-03 16:01:18 -05001346 info.rbd_dev = rbd_dev;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001347
1348 ret = ceph_osdc_create_event(osdc, rbd_notify_cb, 1,
1349 (void *)&info, &event);
1350 if (ret < 0)
1351 goto fail;
1352
1353 ops[0].watch.ver = 1;
1354 ops[0].watch.flag = 1;
1355 ops[0].watch.cookie = event->cookie;
1356 ops[0].watch.prot_ver = RADOS_NOTIFY_VER;
1357 ops[0].watch.timeout = 12;
1358
Alex Elder0ce1a792012-07-03 16:01:18 -05001359 ret = rbd_req_sync_op(rbd_dev, NULL,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001360 CEPH_NOSNAP,
1361 0,
1362 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
1363 ops,
1364 1, obj, 0, 0, NULL, NULL, NULL);
1365 if (ret < 0)
1366 goto fail_event;
1367
1368 ret = ceph_osdc_wait_event(event, CEPH_OSD_TIMEOUT_DEFAULT);
1369 dout("ceph_osdc_wait_event returned %d\n", ret);
1370 rbd_destroy_ops(ops);
1371 return 0;
1372
1373fail_event:
1374 ceph_osdc_cancel_event(event);
1375fail:
1376 rbd_destroy_ops(ops);
1377 return ret;
1378}
1379
1380/*
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001381 * Request sync osd read
1382 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001383static int rbd_req_sync_exec(struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001384 const char *obj,
1385 const char *cls,
1386 const char *method,
1387 const char *data,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001388 int len,
1389 u64 *ver)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001390{
1391 struct ceph_osd_req_op *ops;
1392 int cls_len = strlen(cls);
1393 int method_len = strlen(method);
1394 int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_CALL,
1395 cls_len + method_len + len);
1396 if (ret < 0)
1397 return ret;
1398
1399 ops[0].cls.class_name = cls;
1400 ops[0].cls.class_len = (__u8)cls_len;
1401 ops[0].cls.method_name = method;
1402 ops[0].cls.method_len = (__u8)method_len;
1403 ops[0].cls.argc = 0;
1404 ops[0].cls.indata = data;
1405 ops[0].cls.indata_len = len;
1406
Alex Elder0ce1a792012-07-03 16:01:18 -05001407 ret = rbd_req_sync_op(rbd_dev, NULL,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001408 CEPH_NOSNAP,
1409 0,
1410 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
1411 ops,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001412 1, obj, 0, 0, NULL, NULL, ver);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001413
1414 rbd_destroy_ops(ops);
1415
1416 dout("cls_exec returned %d\n", ret);
1417 return ret;
1418}
1419
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001420static struct rbd_req_coll *rbd_alloc_coll(int num_reqs)
1421{
1422 struct rbd_req_coll *coll =
1423 kzalloc(sizeof(struct rbd_req_coll) +
1424 sizeof(struct rbd_req_status) * num_reqs,
1425 GFP_ATOMIC);
1426
1427 if (!coll)
1428 return NULL;
1429 coll->total = num_reqs;
1430 kref_init(&coll->kref);
1431 return coll;
1432}
1433
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001434/*
1435 * block device queue callback
1436 */
1437static void rbd_rq_fn(struct request_queue *q)
1438{
1439 struct rbd_device *rbd_dev = q->queuedata;
1440 struct request *rq;
1441 struct bio_pair *bp = NULL;
1442
Alex Elder00f1f362012-02-07 12:03:36 -06001443 while ((rq = blk_fetch_request(q))) {
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001444 struct bio *bio;
1445 struct bio *rq_bio, *next_bio = NULL;
1446 bool do_write;
1447 int size, op_size = 0;
1448 u64 ofs;
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001449 int num_segs, cur_seg = 0;
1450 struct rbd_req_coll *coll;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001451
1452 /* peek at request from block layer */
1453 if (!rq)
1454 break;
1455
1456 dout("fetched request\n");
1457
1458 /* filter out block requests we don't understand */
1459 if ((rq->cmd_type != REQ_TYPE_FS)) {
1460 __blk_end_request_all(rq, 0);
Alex Elder00f1f362012-02-07 12:03:36 -06001461 continue;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001462 }
1463
1464 /* deduce our operation (read, write) */
1465 do_write = (rq_data_dir(rq) == WRITE);
1466
1467 size = blk_rq_bytes(rq);
Alex Elder593a9e72012-02-07 12:03:37 -06001468 ofs = blk_rq_pos(rq) * SECTOR_SIZE;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001469 rq_bio = rq->bio;
1470 if (do_write && rbd_dev->read_only) {
1471 __blk_end_request_all(rq, -EROFS);
Alex Elder00f1f362012-02-07 12:03:36 -06001472 continue;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001473 }
1474
1475 spin_unlock_irq(q->queue_lock);
1476
1477 dout("%s 0x%x bytes at 0x%llx\n",
1478 do_write ? "write" : "read",
Alex Elder593a9e72012-02-07 12:03:37 -06001479 size, blk_rq_pos(rq) * SECTOR_SIZE);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001480
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001481 num_segs = rbd_get_num_segments(&rbd_dev->header, ofs, size);
1482 coll = rbd_alloc_coll(num_segs);
1483 if (!coll) {
1484 spin_lock_irq(q->queue_lock);
1485 __blk_end_request_all(rq, -ENOMEM);
Alex Elder00f1f362012-02-07 12:03:36 -06001486 continue;
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001487 }
1488
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001489 do {
1490 /* a bio clone to be passed down to OSD req */
1491 dout("rq->bio->bi_vcnt=%d\n", rq->bio->bi_vcnt);
1492 op_size = rbd_get_segment(&rbd_dev->header,
Alex Elderca1e49a2012-07-10 20:30:09 -05001493 rbd_dev->header.object_prefix,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001494 ofs, size,
1495 NULL, NULL);
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001496 kref_get(&coll->kref);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001497 bio = bio_chain_clone(&rq_bio, &next_bio, &bp,
1498 op_size, GFP_ATOMIC);
1499 if (!bio) {
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001500 rbd_coll_end_req_index(rq, coll, cur_seg,
1501 -ENOMEM, op_size);
1502 goto next_seg;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001503 }
1504
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001505
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001506 /* init OSD command: write or read */
1507 if (do_write)
1508 rbd_req_write(rq, rbd_dev,
1509 rbd_dev->header.snapc,
1510 ofs,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001511 op_size, bio,
1512 coll, cur_seg);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001513 else
1514 rbd_req_read(rq, rbd_dev,
Josh Durgin77dfe992011-11-21 13:04:42 -08001515 rbd_dev->snap_id,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001516 ofs,
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001517 op_size, bio,
1518 coll, cur_seg);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001519
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001520next_seg:
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001521 size -= op_size;
1522 ofs += op_size;
1523
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001524 cur_seg++;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001525 rq_bio = next_bio;
1526 } while (size > 0);
Yehuda Sadeh1fec7092011-05-13 13:52:56 -07001527 kref_put(&coll->kref, rbd_coll_release);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001528
1529 if (bp)
1530 bio_pair_release(bp);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001531 spin_lock_irq(q->queue_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001532 }
1533}
1534
1535/*
1536 * a queue callback. Makes sure that we don't create a bio that spans across
1537 * multiple osd objects. One exception would be with a single page bios,
1538 * which we handle later at bio_chain_clone
1539 */
1540static int rbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bmd,
1541 struct bio_vec *bvec)
1542{
1543 struct rbd_device *rbd_dev = q->queuedata;
Alex Elder593a9e72012-02-07 12:03:37 -06001544 unsigned int chunk_sectors;
1545 sector_t sector;
1546 unsigned int bio_sectors;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001547 int max;
1548
Alex Elder593a9e72012-02-07 12:03:37 -06001549 chunk_sectors = 1 << (rbd_dev->header.obj_order - SECTOR_SHIFT);
1550 sector = bmd->bi_sector + get_start_sect(bmd->bi_bdev);
1551 bio_sectors = bmd->bi_size >> SECTOR_SHIFT;
1552
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001553 max = (chunk_sectors - ((sector & (chunk_sectors - 1))
Alex Elder593a9e72012-02-07 12:03:37 -06001554 + bio_sectors)) << SECTOR_SHIFT;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001555 if (max < 0)
1556 max = 0; /* bio_add cannot handle a negative return */
1557 if (max <= bvec->bv_len && bio_sectors == 0)
1558 return bvec->bv_len;
1559 return max;
1560}
1561
1562static void rbd_free_disk(struct rbd_device *rbd_dev)
1563{
1564 struct gendisk *disk = rbd_dev->disk;
1565
1566 if (!disk)
1567 return;
1568
1569 rbd_header_free(&rbd_dev->header);
1570
1571 if (disk->flags & GENHD_FL_UP)
1572 del_gendisk(disk);
1573 if (disk->queue)
1574 blk_cleanup_queue(disk->queue);
1575 put_disk(disk);
1576}
1577
1578/*
1579 * reload the ondisk the header
1580 */
1581static int rbd_read_header(struct rbd_device *rbd_dev,
1582 struct rbd_image_header *header)
1583{
1584 ssize_t rc;
1585 struct rbd_image_header_ondisk *dh;
Xi Wang50f7c4c2012-04-20 15:49:44 -05001586 u32 snap_count = 0;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001587 u64 ver;
Alex Elder00f1f362012-02-07 12:03:36 -06001588 size_t len;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001589
Alex Elder00f1f362012-02-07 12:03:36 -06001590 /*
1591 * First reads the fixed-size header to determine the number
1592 * of snapshots, then re-reads it, along with all snapshot
1593 * records as well as their stored names.
1594 */
1595 len = sizeof (*dh);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001596 while (1) {
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001597 dh = kmalloc(len, GFP_KERNEL);
1598 if (!dh)
1599 return -ENOMEM;
1600
1601 rc = rbd_req_sync_read(rbd_dev,
1602 NULL, CEPH_NOSNAP,
1603 rbd_dev->obj_md_name,
1604 0, len,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001605 (char *)dh, &ver);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001606 if (rc < 0)
1607 goto out_dh;
1608
1609 rc = rbd_header_from_disk(header, dh, snap_count, GFP_KERNEL);
Josh Durgin81e759f2011-11-15 14:49:53 -08001610 if (rc < 0) {
Alex Elder00f1f362012-02-07 12:03:36 -06001611 if (rc == -ENXIO)
Josh Durgin81e759f2011-11-15 14:49:53 -08001612 pr_warning("unrecognized header format"
1613 " for image %s", rbd_dev->obj);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001614 goto out_dh;
Josh Durgin81e759f2011-11-15 14:49:53 -08001615 }
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001616
Alex Elder00f1f362012-02-07 12:03:36 -06001617 if (snap_count == header->total_snaps)
1618 break;
1619
1620 snap_count = header->total_snaps;
1621 len = sizeof (*dh) +
1622 snap_count * sizeof(struct rbd_image_snap_ondisk) +
1623 header->snap_names_len;
1624
1625 rbd_header_free(header);
1626 kfree(dh);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001627 }
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001628 header->obj_version = ver;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001629
1630out_dh:
1631 kfree(dh);
1632 return rc;
1633}
1634
1635/*
1636 * create a snapshot
1637 */
Alex Elder0ce1a792012-07-03 16:01:18 -05001638static int rbd_header_add_snap(struct rbd_device *rbd_dev,
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001639 const char *snap_name,
1640 gfp_t gfp_flags)
1641{
1642 int name_len = strlen(snap_name);
1643 u64 new_snapid;
1644 int ret;
Sage Weil916d4d62011-05-12 16:10:50 -07001645 void *data, *p, *e;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001646 u64 ver;
Alex Elder1dbb4392012-01-24 10:08:37 -06001647 struct ceph_mon_client *monc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001648
1649 /* we should create a snapshot only if we're pointing at the head */
Alex Elder0ce1a792012-07-03 16:01:18 -05001650 if (rbd_dev->snap_id != CEPH_NOSNAP)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001651 return -EINVAL;
1652
Alex Elder0ce1a792012-07-03 16:01:18 -05001653 monc = &rbd_dev->rbd_client->client->monc;
1654 ret = ceph_monc_create_snapid(monc, rbd_dev->pool_id, &new_snapid);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001655 dout("created snapid=%lld\n", new_snapid);
1656 if (ret < 0)
1657 return ret;
1658
1659 data = kmalloc(name_len + 16, gfp_flags);
1660 if (!data)
1661 return -ENOMEM;
1662
Sage Weil916d4d62011-05-12 16:10:50 -07001663 p = data;
1664 e = data + name_len + 16;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001665
Sage Weil916d4d62011-05-12 16:10:50 -07001666 ceph_encode_string_safe(&p, e, snap_name, name_len, bad);
1667 ceph_encode_64_safe(&p, e, new_snapid, bad);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001668
Alex Elder0ce1a792012-07-03 16:01:18 -05001669 ret = rbd_req_sync_exec(rbd_dev, rbd_dev->obj_md_name,
1670 "rbd", "snap_add",
Sage Weil916d4d62011-05-12 16:10:50 -07001671 data, p - data, &ver);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001672
Sage Weil916d4d62011-05-12 16:10:50 -07001673 kfree(data);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001674
1675 if (ret < 0)
1676 return ret;
1677
Alex Elder0ce1a792012-07-03 16:01:18 -05001678 down_write(&rbd_dev->header_rwsem);
1679 rbd_dev->header.snapc->seq = new_snapid;
1680 up_write(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001681
1682 return 0;
1683bad:
1684 return -ERANGE;
1685}
1686
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001687static void __rbd_remove_all_snaps(struct rbd_device *rbd_dev)
1688{
1689 struct rbd_snap *snap;
1690
1691 while (!list_empty(&rbd_dev->snaps)) {
1692 snap = list_first_entry(&rbd_dev->snaps, struct rbd_snap, node);
1693 __rbd_remove_snap_dev(rbd_dev, snap);
1694 }
1695}
1696
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001697/*
1698 * only read the first part of the ondisk header, without the snaps info
1699 */
Josh Durgin263c6ca2011-12-05 10:43:42 -08001700static int __rbd_refresh_header(struct rbd_device *rbd_dev)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001701{
1702 int ret;
1703 struct rbd_image_header h;
1704 u64 snap_seq;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001705 int follow_seq = 0;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001706
1707 ret = rbd_read_header(rbd_dev, &h);
1708 if (ret < 0)
1709 return ret;
1710
Sage Weil9db4b3e2011-04-19 22:49:06 -07001711 /* resized? */
Alex Elder593a9e72012-02-07 12:03:37 -06001712 set_capacity(rbd_dev->disk, h.image_size / SECTOR_SIZE);
Sage Weil9db4b3e2011-04-19 22:49:06 -07001713
Josh Durginc6666012011-11-21 17:11:12 -08001714 down_write(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001715
1716 snap_seq = rbd_dev->header.snapc->seq;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001717 if (rbd_dev->header.total_snaps &&
1718 rbd_dev->header.snapc->snaps[0] == snap_seq)
1719 /* pointing at the head, will need to follow that
1720 if head moves */
1721 follow_seq = 1;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001722
Alex Elder849b4262012-07-09 21:04:24 -05001723 /* rbd_dev->header.object_prefix shouldn't change */
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001724 kfree(rbd_dev->header.snap_sizes);
Alex Elder849b4262012-07-09 21:04:24 -05001725 kfree(rbd_dev->header.snap_names);
1726 kfree(rbd_dev->header.snapc);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001727
1728 rbd_dev->header.total_snaps = h.total_snaps;
1729 rbd_dev->header.snapc = h.snapc;
1730 rbd_dev->header.snap_names = h.snap_names;
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001731 rbd_dev->header.snap_names_len = h.snap_names_len;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001732 rbd_dev->header.snap_sizes = h.snap_sizes;
Alex Elder849b4262012-07-09 21:04:24 -05001733 /* Free the extra copy of the object prefix */
1734 WARN_ON(strcmp(rbd_dev->header.object_prefix, h.object_prefix));
1735 kfree(h.object_prefix);
1736
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07001737 if (follow_seq)
1738 rbd_dev->header.snapc->seq = rbd_dev->header.snapc->snaps[0];
1739 else
1740 rbd_dev->header.snapc->seq = snap_seq;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001741
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001742 ret = __rbd_init_snaps_header(rbd_dev);
1743
Josh Durginc6666012011-11-21 17:11:12 -08001744 up_write(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001745
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001746 return ret;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001747}
1748
1749static int rbd_init_disk(struct rbd_device *rbd_dev)
1750{
1751 struct gendisk *disk;
1752 struct request_queue *q;
1753 int rc;
Alex Elder593a9e72012-02-07 12:03:37 -06001754 u64 segment_size;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001755 u64 total_size = 0;
1756
1757 /* contact OSD, request size info about the object being mapped */
1758 rc = rbd_read_header(rbd_dev, &rbd_dev->header);
1759 if (rc)
1760 return rc;
1761
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001762 /* no need to lock here, as rbd_dev is not registered yet */
1763 rc = __rbd_init_snaps_header(rbd_dev);
1764 if (rc)
1765 return rc;
1766
Josh Durgincc9d7342011-11-21 18:19:13 -08001767 rc = rbd_header_set_snap(rbd_dev, &total_size);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001768 if (rc)
1769 return rc;
1770
1771 /* create gendisk info */
1772 rc = -ENOMEM;
1773 disk = alloc_disk(RBD_MINORS_PER_MAJOR);
1774 if (!disk)
1775 goto out;
1776
Alex Elderf0f8cef2012-01-29 13:57:44 -06001777 snprintf(disk->disk_name, sizeof(disk->disk_name), RBD_DRV_NAME "%d",
Sage Weilaedfec52011-05-12 20:57:03 -07001778 rbd_dev->id);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001779 disk->major = rbd_dev->major;
1780 disk->first_minor = 0;
1781 disk->fops = &rbd_bd_ops;
1782 disk->private_data = rbd_dev;
1783
1784 /* init rq */
1785 rc = -ENOMEM;
1786 q = blk_init_queue(rbd_rq_fn, &rbd_dev->lock);
1787 if (!q)
1788 goto out_disk;
Josh Durgin029bcbd2011-07-22 11:35:23 -07001789
Alex Elder593a9e72012-02-07 12:03:37 -06001790 /* We use the default size, but let's be explicit about it. */
1791 blk_queue_physical_block_size(q, SECTOR_SIZE);
1792
Josh Durgin029bcbd2011-07-22 11:35:23 -07001793 /* set io sizes to object size */
Alex Elder593a9e72012-02-07 12:03:37 -06001794 segment_size = rbd_obj_bytes(&rbd_dev->header);
1795 blk_queue_max_hw_sectors(q, segment_size / SECTOR_SIZE);
1796 blk_queue_max_segment_size(q, segment_size);
1797 blk_queue_io_min(q, segment_size);
1798 blk_queue_io_opt(q, segment_size);
Josh Durgin029bcbd2011-07-22 11:35:23 -07001799
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001800 blk_queue_merge_bvec(q, rbd_merge_bvec);
1801 disk->queue = q;
1802
1803 q->queuedata = rbd_dev;
1804
1805 rbd_dev->disk = disk;
1806 rbd_dev->q = q;
1807
1808 /* finally, announce the disk to the world */
Alex Elder593a9e72012-02-07 12:03:37 -06001809 set_capacity(disk, total_size / SECTOR_SIZE);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001810 add_disk(disk);
1811
1812 pr_info("%s: added with size 0x%llx\n",
1813 disk->disk_name, (unsigned long long)total_size);
1814 return 0;
1815
1816out_disk:
1817 put_disk(disk);
1818out:
1819 return rc;
1820}
1821
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001822/*
1823 sysfs
1824*/
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001825
Alex Elder593a9e72012-02-07 12:03:37 -06001826static struct rbd_device *dev_to_rbd_dev(struct device *dev)
1827{
1828 return container_of(dev, struct rbd_device, dev);
1829}
1830
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001831static ssize_t rbd_size_show(struct device *dev,
1832 struct device_attribute *attr, char *buf)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001833{
Alex Elder593a9e72012-02-07 12:03:37 -06001834 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001835
1836 return sprintf(buf, "%llu\n", (unsigned long long)rbd_dev->header.image_size);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001837}
1838
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001839static ssize_t rbd_major_show(struct device *dev,
1840 struct device_attribute *attr, char *buf)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001841{
Alex Elder593a9e72012-02-07 12:03:37 -06001842 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001843
1844 return sprintf(buf, "%d\n", rbd_dev->major);
1845}
1846
1847static ssize_t rbd_client_id_show(struct device *dev,
1848 struct device_attribute *attr, char *buf)
1849{
Alex Elder593a9e72012-02-07 12:03:37 -06001850 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001851
Alex Elder1dbb4392012-01-24 10:08:37 -06001852 return sprintf(buf, "client%lld\n",
1853 ceph_client_id(rbd_dev->rbd_client->client));
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001854}
1855
1856static ssize_t rbd_pool_show(struct device *dev,
1857 struct device_attribute *attr, char *buf)
1858{
Alex Elder593a9e72012-02-07 12:03:37 -06001859 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001860
1861 return sprintf(buf, "%s\n", rbd_dev->pool_name);
1862}
1863
Alex Elder9bb2f332012-07-12 10:46:35 -05001864static ssize_t rbd_pool_id_show(struct device *dev,
1865 struct device_attribute *attr, char *buf)
1866{
1867 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
1868
1869 return sprintf(buf, "%d\n", rbd_dev->pool_id);
1870}
1871
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001872static ssize_t rbd_name_show(struct device *dev,
1873 struct device_attribute *attr, char *buf)
1874{
Alex Elder593a9e72012-02-07 12:03:37 -06001875 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001876
1877 return sprintf(buf, "%s\n", rbd_dev->obj);
1878}
1879
1880static ssize_t rbd_snap_show(struct device *dev,
1881 struct device_attribute *attr,
1882 char *buf)
1883{
Alex Elder593a9e72012-02-07 12:03:37 -06001884 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001885
1886 return sprintf(buf, "%s\n", rbd_dev->snap_name);
1887}
1888
1889static ssize_t rbd_image_refresh(struct device *dev,
1890 struct device_attribute *attr,
1891 const char *buf,
1892 size_t size)
1893{
Alex Elder593a9e72012-02-07 12:03:37 -06001894 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001895 int rc;
1896 int ret = size;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001897
1898 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
1899
Josh Durgin263c6ca2011-12-05 10:43:42 -08001900 rc = __rbd_refresh_header(rbd_dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001901 if (rc < 0)
1902 ret = rc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001903
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001904 mutex_unlock(&ctl_mutex);
1905 return ret;
1906}
Yehuda Sadeh602adf42010-08-12 16:11:25 -07001907
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001908static DEVICE_ATTR(size, S_IRUGO, rbd_size_show, NULL);
1909static DEVICE_ATTR(major, S_IRUGO, rbd_major_show, NULL);
1910static DEVICE_ATTR(client_id, S_IRUGO, rbd_client_id_show, NULL);
1911static DEVICE_ATTR(pool, S_IRUGO, rbd_pool_show, NULL);
Alex Elder9bb2f332012-07-12 10:46:35 -05001912static DEVICE_ATTR(pool_id, S_IRUGO, rbd_pool_id_show, NULL);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001913static DEVICE_ATTR(name, S_IRUGO, rbd_name_show, NULL);
1914static DEVICE_ATTR(refresh, S_IWUSR, NULL, rbd_image_refresh);
1915static DEVICE_ATTR(current_snap, S_IRUGO, rbd_snap_show, NULL);
1916static DEVICE_ATTR(create_snap, S_IWUSR, NULL, rbd_snap_add);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001917
1918static struct attribute *rbd_attrs[] = {
1919 &dev_attr_size.attr,
1920 &dev_attr_major.attr,
1921 &dev_attr_client_id.attr,
1922 &dev_attr_pool.attr,
Alex Elder9bb2f332012-07-12 10:46:35 -05001923 &dev_attr_pool_id.attr,
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001924 &dev_attr_name.attr,
1925 &dev_attr_current_snap.attr,
1926 &dev_attr_refresh.attr,
1927 &dev_attr_create_snap.attr,
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001928 NULL
1929};
1930
1931static struct attribute_group rbd_attr_group = {
1932 .attrs = rbd_attrs,
1933};
1934
1935static const struct attribute_group *rbd_attr_groups[] = {
1936 &rbd_attr_group,
1937 NULL
1938};
1939
1940static void rbd_sysfs_dev_release(struct device *dev)
1941{
1942}
1943
1944static struct device_type rbd_device_type = {
1945 .name = "rbd",
1946 .groups = rbd_attr_groups,
1947 .release = rbd_sysfs_dev_release,
1948};
1949
1950
1951/*
1952 sysfs - snapshots
1953*/
1954
1955static ssize_t rbd_snap_size_show(struct device *dev,
1956 struct device_attribute *attr,
1957 char *buf)
1958{
1959 struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
1960
Josh Durgin35915382011-12-05 18:25:13 -08001961 return sprintf(buf, "%llu\n", (unsigned long long)snap->size);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001962}
1963
1964static ssize_t rbd_snap_id_show(struct device *dev,
1965 struct device_attribute *attr,
1966 char *buf)
1967{
1968 struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
1969
Josh Durgin35915382011-12-05 18:25:13 -08001970 return sprintf(buf, "%llu\n", (unsigned long long)snap->id);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08001971}
1972
1973static DEVICE_ATTR(snap_size, S_IRUGO, rbd_snap_size_show, NULL);
1974static DEVICE_ATTR(snap_id, S_IRUGO, rbd_snap_id_show, NULL);
1975
1976static struct attribute *rbd_snap_attrs[] = {
1977 &dev_attr_snap_size.attr,
1978 &dev_attr_snap_id.attr,
1979 NULL,
1980};
1981
1982static struct attribute_group rbd_snap_attr_group = {
1983 .attrs = rbd_snap_attrs,
1984};
1985
1986static void rbd_snap_dev_release(struct device *dev)
1987{
1988 struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
1989 kfree(snap->name);
1990 kfree(snap);
1991}
1992
1993static const struct attribute_group *rbd_snap_attr_groups[] = {
1994 &rbd_snap_attr_group,
1995 NULL
1996};
1997
1998static struct device_type rbd_snap_device_type = {
1999 .groups = rbd_snap_attr_groups,
2000 .release = rbd_snap_dev_release,
2001};
2002
2003static void __rbd_remove_snap_dev(struct rbd_device *rbd_dev,
2004 struct rbd_snap *snap)
2005{
2006 list_del(&snap->node);
2007 device_unregister(&snap->dev);
2008}
2009
2010static int rbd_register_snap_dev(struct rbd_device *rbd_dev,
2011 struct rbd_snap *snap,
2012 struct device *parent)
2013{
2014 struct device *dev = &snap->dev;
2015 int ret;
2016
2017 dev->type = &rbd_snap_device_type;
2018 dev->parent = parent;
2019 dev->release = rbd_snap_dev_release;
2020 dev_set_name(dev, "snap_%s", snap->name);
2021 ret = device_register(dev);
2022
2023 return ret;
2024}
2025
2026static int __rbd_add_snap_dev(struct rbd_device *rbd_dev,
2027 int i, const char *name,
2028 struct rbd_snap **snapp)
2029{
2030 int ret;
2031 struct rbd_snap *snap = kzalloc(sizeof(*snap), GFP_KERNEL);
2032 if (!snap)
2033 return -ENOMEM;
2034 snap->name = kstrdup(name, GFP_KERNEL);
2035 snap->size = rbd_dev->header.snap_sizes[i];
2036 snap->id = rbd_dev->header.snapc->snaps[i];
2037 if (device_is_registered(&rbd_dev->dev)) {
2038 ret = rbd_register_snap_dev(rbd_dev, snap,
2039 &rbd_dev->dev);
2040 if (ret < 0)
2041 goto err;
2042 }
2043 *snapp = snap;
2044 return 0;
2045err:
2046 kfree(snap->name);
2047 kfree(snap);
2048 return ret;
2049}
2050
2051/*
2052 * search for the previous snap in a null delimited string list
2053 */
2054const char *rbd_prev_snap_name(const char *name, const char *start)
2055{
2056 if (name < start + 2)
2057 return NULL;
2058
2059 name -= 2;
2060 while (*name) {
2061 if (name == start)
2062 return start;
2063 name--;
2064 }
2065 return name + 1;
2066}
2067
2068/*
2069 * compare the old list of snapshots that we have to what's in the header
2070 * and update it accordingly. Note that the header holds the snapshots
2071 * in a reverse order (from newest to oldest) and we need to go from
2072 * older to new so that we don't get a duplicate snap name when
2073 * doing the process (e.g., removed snapshot and recreated a new
2074 * one with the same name.
2075 */
2076static int __rbd_init_snaps_header(struct rbd_device *rbd_dev)
2077{
2078 const char *name, *first_name;
2079 int i = rbd_dev->header.total_snaps;
2080 struct rbd_snap *snap, *old_snap = NULL;
2081 int ret;
2082 struct list_head *p, *n;
2083
2084 first_name = rbd_dev->header.snap_names;
2085 name = first_name + rbd_dev->header.snap_names_len;
2086
2087 list_for_each_prev_safe(p, n, &rbd_dev->snaps) {
2088 u64 cur_id;
2089
2090 old_snap = list_entry(p, struct rbd_snap, node);
2091
2092 if (i)
2093 cur_id = rbd_dev->header.snapc->snaps[i - 1];
2094
2095 if (!i || old_snap->id < cur_id) {
2096 /* old_snap->id was skipped, thus was removed */
2097 __rbd_remove_snap_dev(rbd_dev, old_snap);
2098 continue;
2099 }
2100 if (old_snap->id == cur_id) {
2101 /* we have this snapshot already */
2102 i--;
2103 name = rbd_prev_snap_name(name, first_name);
2104 continue;
2105 }
2106 for (; i > 0;
2107 i--, name = rbd_prev_snap_name(name, first_name)) {
2108 if (!name) {
2109 WARN_ON(1);
2110 return -EINVAL;
2111 }
2112 cur_id = rbd_dev->header.snapc->snaps[i];
2113 /* snapshot removal? handle it above */
2114 if (cur_id >= old_snap->id)
2115 break;
2116 /* a new snapshot */
2117 ret = __rbd_add_snap_dev(rbd_dev, i - 1, name, &snap);
2118 if (ret < 0)
2119 return ret;
2120
2121 /* note that we add it backward so using n and not p */
2122 list_add(&snap->node, n);
2123 p = &snap->node;
2124 }
2125 }
2126 /* we're done going over the old snap list, just add what's left */
2127 for (; i > 0; i--) {
2128 name = rbd_prev_snap_name(name, first_name);
2129 if (!name) {
2130 WARN_ON(1);
2131 return -EINVAL;
2132 }
2133 ret = __rbd_add_snap_dev(rbd_dev, i - 1, name, &snap);
2134 if (ret < 0)
2135 return ret;
2136 list_add(&snap->node, &rbd_dev->snaps);
2137 }
2138
2139 return 0;
2140}
2141
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002142static int rbd_bus_add_dev(struct rbd_device *rbd_dev)
2143{
Alex Elderf0f8cef2012-01-29 13:57:44 -06002144 int ret;
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002145 struct device *dev;
2146 struct rbd_snap *snap;
2147
2148 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
2149 dev = &rbd_dev->dev;
2150
2151 dev->bus = &rbd_bus_type;
2152 dev->type = &rbd_device_type;
2153 dev->parent = &rbd_root_dev;
2154 dev->release = rbd_dev_release;
2155 dev_set_name(dev, "%d", rbd_dev->id);
2156 ret = device_register(dev);
2157 if (ret < 0)
Alex Elderf0f8cef2012-01-29 13:57:44 -06002158 goto out;
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002159
2160 list_for_each_entry(snap, &rbd_dev->snaps, node) {
2161 ret = rbd_register_snap_dev(rbd_dev, snap,
2162 &rbd_dev->dev);
2163 if (ret < 0)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002164 break;
2165 }
Alex Elderf0f8cef2012-01-29 13:57:44 -06002166out:
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002167 mutex_unlock(&ctl_mutex);
2168 return ret;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002169}
2170
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002171static void rbd_bus_del_dev(struct rbd_device *rbd_dev)
2172{
2173 device_unregister(&rbd_dev->dev);
2174}
2175
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002176static int rbd_init_watch_dev(struct rbd_device *rbd_dev)
2177{
2178 int ret, rc;
2179
2180 do {
2181 ret = rbd_req_sync_watch(rbd_dev, rbd_dev->obj_md_name,
2182 rbd_dev->header.obj_version);
2183 if (ret == -ERANGE) {
2184 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
Josh Durgin263c6ca2011-12-05 10:43:42 -08002185 rc = __rbd_refresh_header(rbd_dev);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002186 mutex_unlock(&ctl_mutex);
2187 if (rc < 0)
2188 return rc;
2189 }
2190 } while (ret == -ERANGE);
2191
2192 return ret;
2193}
2194
Alex Elder1ddbe942012-01-29 13:57:44 -06002195static atomic64_t rbd_id_max = ATOMIC64_INIT(0);
2196
2197/*
Alex Elder499afd52012-02-02 08:13:29 -06002198 * Get a unique rbd identifier for the given new rbd_dev, and add
2199 * the rbd_dev to the global list. The minimum rbd id is 1.
Alex Elder1ddbe942012-01-29 13:57:44 -06002200 */
Alex Elder499afd52012-02-02 08:13:29 -06002201static void rbd_id_get(struct rbd_device *rbd_dev)
Alex Elderb7f23c32012-01-29 13:57:43 -06002202{
Alex Elder499afd52012-02-02 08:13:29 -06002203 rbd_dev->id = atomic64_inc_return(&rbd_id_max);
2204
2205 spin_lock(&rbd_dev_list_lock);
2206 list_add_tail(&rbd_dev->node, &rbd_dev_list);
2207 spin_unlock(&rbd_dev_list_lock);
Alex Elder1ddbe942012-01-29 13:57:44 -06002208}
Alex Elderb7f23c32012-01-29 13:57:43 -06002209
Alex Elder1ddbe942012-01-29 13:57:44 -06002210/*
Alex Elder499afd52012-02-02 08:13:29 -06002211 * Remove an rbd_dev from the global list, and record that its
2212 * identifier is no longer in use.
Alex Elder1ddbe942012-01-29 13:57:44 -06002213 */
Alex Elder499afd52012-02-02 08:13:29 -06002214static void rbd_id_put(struct rbd_device *rbd_dev)
Alex Elder1ddbe942012-01-29 13:57:44 -06002215{
Alex Elderd184f6b2012-01-29 13:57:44 -06002216 struct list_head *tmp;
2217 int rbd_id = rbd_dev->id;
2218 int max_id;
2219
2220 BUG_ON(rbd_id < 1);
Alex Elder499afd52012-02-02 08:13:29 -06002221
2222 spin_lock(&rbd_dev_list_lock);
2223 list_del_init(&rbd_dev->node);
Alex Elderd184f6b2012-01-29 13:57:44 -06002224
2225 /*
2226 * If the id being "put" is not the current maximum, there
2227 * is nothing special we need to do.
2228 */
2229 if (rbd_id != atomic64_read(&rbd_id_max)) {
2230 spin_unlock(&rbd_dev_list_lock);
2231 return;
2232 }
2233
2234 /*
2235 * We need to update the current maximum id. Search the
2236 * list to find out what it is. We're more likely to find
2237 * the maximum at the end, so search the list backward.
2238 */
2239 max_id = 0;
2240 list_for_each_prev(tmp, &rbd_dev_list) {
2241 struct rbd_device *rbd_dev;
2242
2243 rbd_dev = list_entry(tmp, struct rbd_device, node);
2244 if (rbd_id > max_id)
2245 max_id = rbd_id;
2246 }
Alex Elder499afd52012-02-02 08:13:29 -06002247 spin_unlock(&rbd_dev_list_lock);
Alex Elderb7f23c32012-01-29 13:57:43 -06002248
Alex Elder1ddbe942012-01-29 13:57:44 -06002249 /*
Alex Elderd184f6b2012-01-29 13:57:44 -06002250 * The max id could have been updated by rbd_id_get(), in
2251 * which case it now accurately reflects the new maximum.
2252 * Be careful not to overwrite the maximum value in that
2253 * case.
Alex Elder1ddbe942012-01-29 13:57:44 -06002254 */
Alex Elderd184f6b2012-01-29 13:57:44 -06002255 atomic64_cmpxchg(&rbd_id_max, rbd_id, max_id);
Alex Elderb7f23c32012-01-29 13:57:43 -06002256}
2257
Alex Eldera725f65e2012-02-02 08:13:30 -06002258/*
Alex Eldere28fff262012-02-02 08:13:30 -06002259 * Skips over white space at *buf, and updates *buf to point to the
2260 * first found non-space character (if any). Returns the length of
Alex Elder593a9e72012-02-07 12:03:37 -06002261 * the token (string of non-white space characters) found. Note
2262 * that *buf must be terminated with '\0'.
Alex Eldere28fff262012-02-02 08:13:30 -06002263 */
2264static inline size_t next_token(const char **buf)
2265{
2266 /*
2267 * These are the characters that produce nonzero for
2268 * isspace() in the "C" and "POSIX" locales.
2269 */
2270 const char *spaces = " \f\n\r\t\v";
2271
2272 *buf += strspn(*buf, spaces); /* Find start of token */
2273
2274 return strcspn(*buf, spaces); /* Return token length */
2275}
2276
2277/*
2278 * Finds the next token in *buf, and if the provided token buffer is
2279 * big enough, copies the found token into it. The result, if
Alex Elder593a9e72012-02-07 12:03:37 -06002280 * copied, is guaranteed to be terminated with '\0'. Note that *buf
2281 * must be terminated with '\0' on entry.
Alex Eldere28fff262012-02-02 08:13:30 -06002282 *
2283 * Returns the length of the token found (not including the '\0').
2284 * Return value will be 0 if no token is found, and it will be >=
2285 * token_size if the token would not fit.
2286 *
Alex Elder593a9e72012-02-07 12:03:37 -06002287 * The *buf pointer will be updated to point beyond the end of the
Alex Eldere28fff262012-02-02 08:13:30 -06002288 * found token. Note that this occurs even if the token buffer is
2289 * too small to hold it.
2290 */
2291static inline size_t copy_token(const char **buf,
2292 char *token,
2293 size_t token_size)
2294{
2295 size_t len;
2296
2297 len = next_token(buf);
2298 if (len < token_size) {
2299 memcpy(token, *buf, len);
2300 *(token + len) = '\0';
2301 }
2302 *buf += len;
2303
2304 return len;
2305}
2306
2307/*
Alex Elderea3352f2012-07-09 21:04:23 -05002308 * Finds the next token in *buf, dynamically allocates a buffer big
2309 * enough to hold a copy of it, and copies the token into the new
2310 * buffer. The copy is guaranteed to be terminated with '\0'. Note
2311 * that a duplicate buffer is created even for a zero-length token.
2312 *
2313 * Returns a pointer to the newly-allocated duplicate, or a null
2314 * pointer if memory for the duplicate was not available. If
2315 * the lenp argument is a non-null pointer, the length of the token
2316 * (not including the '\0') is returned in *lenp.
2317 *
2318 * If successful, the *buf pointer will be updated to point beyond
2319 * the end of the found token.
2320 *
2321 * Note: uses GFP_KERNEL for allocation.
2322 */
2323static inline char *dup_token(const char **buf, size_t *lenp)
2324{
2325 char *dup;
2326 size_t len;
2327
2328 len = next_token(buf);
2329 dup = kmalloc(len + 1, GFP_KERNEL);
2330 if (!dup)
2331 return NULL;
2332
2333 memcpy(dup, *buf, len);
2334 *(dup + len) = '\0';
2335 *buf += len;
2336
2337 if (lenp)
2338 *lenp = len;
2339
2340 return dup;
2341}
2342
2343/*
Alex Eldera725f65e2012-02-02 08:13:30 -06002344 * This fills in the pool_name, obj, obj_len, snap_name, obj_len,
2345 * rbd_dev, rbd_md_name, and name fields of the given rbd_dev, based
2346 * on the list of monitor addresses and other options provided via
2347 * /sys/bus/rbd/add.
Alex Elderd22f76e2012-07-12 10:46:35 -05002348 *
2349 * Note: rbd_dev is assumed to have been initially zero-filled.
Alex Eldera725f65e2012-02-02 08:13:30 -06002350 */
2351static int rbd_add_parse_args(struct rbd_device *rbd_dev,
2352 const char *buf,
Alex Elder7ef32142012-02-02 08:13:30 -06002353 const char **mon_addrs,
Alex Elder5214ecc2012-02-02 08:13:30 -06002354 size_t *mon_addrs_size,
Alex Eldere28fff262012-02-02 08:13:30 -06002355 char *options,
2356 size_t options_size)
Alex Eldera725f65e2012-02-02 08:13:30 -06002357{
Alex Elderd22f76e2012-07-12 10:46:35 -05002358 size_t len;
2359 int ret;
Alex Eldere28fff262012-02-02 08:13:30 -06002360
2361 /* The first four tokens are required */
2362
Alex Elder7ef32142012-02-02 08:13:30 -06002363 len = next_token(&buf);
2364 if (!len)
Alex Eldera725f65e2012-02-02 08:13:30 -06002365 return -EINVAL;
Alex Elder5214ecc2012-02-02 08:13:30 -06002366 *mon_addrs_size = len + 1;
Alex Elder7ef32142012-02-02 08:13:30 -06002367 *mon_addrs = buf;
2368
2369 buf += len;
Alex Eldera725f65e2012-02-02 08:13:30 -06002370
Alex Eldere28fff262012-02-02 08:13:30 -06002371 len = copy_token(&buf, options, options_size);
2372 if (!len || len >= options_size)
2373 return -EINVAL;
Alex Eldera725f65e2012-02-02 08:13:30 -06002374
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002375 ret = -ENOMEM;
Alex Elderd22f76e2012-07-12 10:46:35 -05002376 rbd_dev->pool_name = dup_token(&buf, NULL);
2377 if (!rbd_dev->pool_name)
Alex Elderd22f76e2012-07-12 10:46:35 -05002378 goto out_err;
Alex Eldere28fff262012-02-02 08:13:30 -06002379
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002380 rbd_dev->obj = dup_token(&buf, &rbd_dev->obj_len);
2381 if (!rbd_dev->obj)
2382 goto out_err;
Alex Eldere28fff262012-02-02 08:13:30 -06002383
Alex Eldercb8627c2012-07-09 21:04:23 -05002384 /* Create the name of the header object */
2385
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002386 rbd_dev->obj_md_name = kmalloc(rbd_dev->obj_len
2387 + sizeof (RBD_SUFFIX),
2388 GFP_KERNEL);
2389 if (!rbd_dev->obj_md_name)
Alex Eldercb8627c2012-07-09 21:04:23 -05002390 goto out_err;
Alex Elder81a89792012-02-02 08:13:30 -06002391 sprintf(rbd_dev->obj_md_name, "%s%s", rbd_dev->obj, RBD_SUFFIX);
Alex Eldera725f65e2012-02-02 08:13:30 -06002392
Alex Eldere28fff262012-02-02 08:13:30 -06002393 /*
Alex Elder820a5f32012-07-09 21:04:24 -05002394 * The snapshot name is optional. If none is is supplied,
2395 * we use the default value.
Alex Eldere28fff262012-02-02 08:13:30 -06002396 */
Alex Elder820a5f32012-07-09 21:04:24 -05002397 rbd_dev->snap_name = dup_token(&buf, &len);
2398 if (!rbd_dev->snap_name)
2399 goto out_err;
2400 if (!len) {
2401 /* Replace the empty name with the default */
2402 kfree(rbd_dev->snap_name);
2403 rbd_dev->snap_name
2404 = kmalloc(sizeof (RBD_SNAP_HEAD_NAME), GFP_KERNEL);
2405 if (!rbd_dev->snap_name)
2406 goto out_err;
2407
Alex Eldere28fff262012-02-02 08:13:30 -06002408 memcpy(rbd_dev->snap_name, RBD_SNAP_HEAD_NAME,
2409 sizeof (RBD_SNAP_HEAD_NAME));
Alex Elder849b4262012-07-09 21:04:24 -05002410 }
Alex Eldere28fff262012-02-02 08:13:30 -06002411
Alex Eldera725f65e2012-02-02 08:13:30 -06002412 return 0;
Alex Elderd22f76e2012-07-12 10:46:35 -05002413
2414out_err:
Alex Eldercb8627c2012-07-09 21:04:23 -05002415 kfree(rbd_dev->obj_md_name);
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002416 kfree(rbd_dev->obj);
Alex Elderd22f76e2012-07-12 10:46:35 -05002417 kfree(rbd_dev->pool_name);
2418 rbd_dev->pool_name = NULL;
2419
2420 return ret;
Alex Eldera725f65e2012-02-02 08:13:30 -06002421}
2422
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002423static ssize_t rbd_add(struct bus_type *bus,
2424 const char *buf,
2425 size_t count)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002426{
Alex Eldercb8627c2012-07-09 21:04:23 -05002427 char *options;
2428 struct rbd_device *rbd_dev = NULL;
Alex Elder7ef32142012-02-02 08:13:30 -06002429 const char *mon_addrs = NULL;
2430 size_t mon_addrs_size = 0;
Alex Elder27cc2592012-02-02 08:13:30 -06002431 struct ceph_osd_client *osdc;
2432 int rc = -ENOMEM;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002433
2434 if (!try_module_get(THIS_MODULE))
2435 return -ENODEV;
2436
Alex Elder27cc2592012-02-02 08:13:30 -06002437 options = kmalloc(count, GFP_KERNEL);
2438 if (!options)
2439 goto err_nomem;
Alex Eldercb8627c2012-07-09 21:04:23 -05002440 rbd_dev = kzalloc(sizeof(*rbd_dev), GFP_KERNEL);
2441 if (!rbd_dev)
2442 goto err_nomem;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002443
2444 /* static rbd_device initialization */
2445 spin_lock_init(&rbd_dev->lock);
2446 INIT_LIST_HEAD(&rbd_dev->node);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002447 INIT_LIST_HEAD(&rbd_dev->snaps);
Josh Durginc6666012011-11-21 17:11:12 -08002448 init_rwsem(&rbd_dev->header_rwsem);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002449
Josh Durginc6666012011-11-21 17:11:12 -08002450 init_rwsem(&rbd_dev->header_rwsem);
Alex Elder0e805a12012-01-11 19:42:15 -08002451
Alex Elderd184f6b2012-01-29 13:57:44 -06002452 /* generate unique id: find highest unique id, add one */
Alex Elder499afd52012-02-02 08:13:29 -06002453 rbd_id_get(rbd_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002454
Alex Eldera725f65e2012-02-02 08:13:30 -06002455 /* Fill in the device name, now that we have its id. */
Alex Elder81a89792012-02-02 08:13:30 -06002456 BUILD_BUG_ON(DEV_NAME_LEN
2457 < sizeof (RBD_DRV_NAME) + MAX_INT_FORMAT_WIDTH);
2458 sprintf(rbd_dev->name, "%s%d", RBD_DRV_NAME, rbd_dev->id);
Alex Eldere124a82f2012-01-29 13:57:44 -06002459
Alex Eldera725f65e2012-02-02 08:13:30 -06002460 /* parse add command */
Alex Elder7ef32142012-02-02 08:13:30 -06002461 rc = rbd_add_parse_args(rbd_dev, buf, &mon_addrs, &mon_addrs_size,
Alex Eldere28fff262012-02-02 08:13:30 -06002462 options, count);
Alex Eldera725f65e2012-02-02 08:13:30 -06002463 if (rc)
2464 goto err_put_id;
2465
Alex Elder5214ecc2012-02-02 08:13:30 -06002466 rbd_dev->rbd_client = rbd_get_client(mon_addrs, mon_addrs_size - 1,
2467 options);
Alex Elderd720bcb2012-02-02 08:13:30 -06002468 if (IS_ERR(rbd_dev->rbd_client)) {
2469 rc = PTR_ERR(rbd_dev->rbd_client);
Alex Elderf0f8cef2012-01-29 13:57:44 -06002470 goto err_put_id;
Alex Elderd720bcb2012-02-02 08:13:30 -06002471 }
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002472
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002473 /* pick the pool */
Alex Elder1dbb4392012-01-24 10:08:37 -06002474 osdc = &rbd_dev->rbd_client->client->osdc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002475 rc = ceph_pg_poolid_by_name(osdc->osdmap, rbd_dev->pool_name);
2476 if (rc < 0)
2477 goto err_out_client;
Alex Elder9bb2f332012-07-12 10:46:35 -05002478 rbd_dev->pool_id = rc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002479
2480 /* register our block device */
Alex Elder27cc2592012-02-02 08:13:30 -06002481 rc = register_blkdev(0, rbd_dev->name);
2482 if (rc < 0)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002483 goto err_out_client;
Alex Elder27cc2592012-02-02 08:13:30 -06002484 rbd_dev->major = rc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002485
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002486 rc = rbd_bus_add_dev(rbd_dev);
2487 if (rc)
Yehuda Sadeh766fc432011-01-07 14:58:42 -08002488 goto err_out_blkdev;
2489
Alex Elder32eec682012-02-08 16:11:14 -06002490 /*
2491 * At this point cleanup in the event of an error is the job
2492 * of the sysfs code (initiated by rbd_bus_del_dev()).
2493 *
2494 * Set up and announce blkdev mapping.
2495 */
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002496 rc = rbd_init_disk(rbd_dev);
2497 if (rc)
Yehuda Sadeh766fc432011-01-07 14:58:42 -08002498 goto err_out_bus;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002499
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002500 rc = rbd_init_watch_dev(rbd_dev);
2501 if (rc)
2502 goto err_out_bus;
2503
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002504 return count;
2505
Yehuda Sadeh766fc432011-01-07 14:58:42 -08002506err_out_bus:
Yehuda Sadeh766fc432011-01-07 14:58:42 -08002507 /* this will also clean up rest of rbd_dev stuff */
2508
2509 rbd_bus_del_dev(rbd_dev);
2510 kfree(options);
Yehuda Sadeh766fc432011-01-07 14:58:42 -08002511 return rc;
2512
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002513err_out_blkdev:
2514 unregister_blkdev(rbd_dev->major, rbd_dev->name);
2515err_out_client:
2516 rbd_put_client(rbd_dev);
Alex Elderf0f8cef2012-01-29 13:57:44 -06002517err_put_id:
Alex Eldercb8627c2012-07-09 21:04:23 -05002518 if (rbd_dev->pool_name) {
Alex Elder820a5f32012-07-09 21:04:24 -05002519 kfree(rbd_dev->snap_name);
Alex Eldercb8627c2012-07-09 21:04:23 -05002520 kfree(rbd_dev->obj_md_name);
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002521 kfree(rbd_dev->obj);
Alex Eldercb8627c2012-07-09 21:04:23 -05002522 kfree(rbd_dev->pool_name);
2523 }
Alex Elder499afd52012-02-02 08:13:29 -06002524 rbd_id_put(rbd_dev);
Alex Elder27cc2592012-02-02 08:13:30 -06002525err_nomem:
Alex Elder27cc2592012-02-02 08:13:30 -06002526 kfree(rbd_dev);
Alex Eldercb8627c2012-07-09 21:04:23 -05002527 kfree(options);
Alex Elder27cc2592012-02-02 08:13:30 -06002528
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002529 dout("Error adding device %s\n", buf);
2530 module_put(THIS_MODULE);
Alex Elder27cc2592012-02-02 08:13:30 -06002531
2532 return (ssize_t) rc;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002533}
2534
2535static struct rbd_device *__rbd_get_dev(unsigned long id)
2536{
2537 struct list_head *tmp;
2538 struct rbd_device *rbd_dev;
2539
Alex Eldere124a82f2012-01-29 13:57:44 -06002540 spin_lock(&rbd_dev_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002541 list_for_each(tmp, &rbd_dev_list) {
2542 rbd_dev = list_entry(tmp, struct rbd_device, node);
Alex Eldere124a82f2012-01-29 13:57:44 -06002543 if (rbd_dev->id == id) {
2544 spin_unlock(&rbd_dev_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002545 return rbd_dev;
Alex Eldere124a82f2012-01-29 13:57:44 -06002546 }
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002547 }
Alex Eldere124a82f2012-01-29 13:57:44 -06002548 spin_unlock(&rbd_dev_list_lock);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002549 return NULL;
2550}
2551
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002552static void rbd_dev_release(struct device *dev)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002553{
Alex Elder593a9e72012-02-07 12:03:37 -06002554 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002555
Alex Elder1dbb4392012-01-24 10:08:37 -06002556 if (rbd_dev->watch_request) {
2557 struct ceph_client *client = rbd_dev->rbd_client->client;
2558
2559 ceph_osdc_unregister_linger_request(&client->osdc,
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002560 rbd_dev->watch_request);
Alex Elder1dbb4392012-01-24 10:08:37 -06002561 }
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002562 if (rbd_dev->watch_event)
Yehuda Sadeh79e3057c2011-07-12 16:56:57 -07002563 rbd_req_sync_unwatch(rbd_dev, rbd_dev->obj_md_name);
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002564
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002565 rbd_put_client(rbd_dev);
2566
2567 /* clean up and free blkdev */
2568 rbd_free_disk(rbd_dev);
2569 unregister_blkdev(rbd_dev->major, rbd_dev->name);
Alex Elder32eec682012-02-08 16:11:14 -06002570
2571 /* done with the id, and with the rbd_dev */
Alex Elder820a5f32012-07-09 21:04:24 -05002572 kfree(rbd_dev->snap_name);
Alex Eldercb8627c2012-07-09 21:04:23 -05002573 kfree(rbd_dev->obj_md_name);
Alex Elderd22f76e2012-07-12 10:46:35 -05002574 kfree(rbd_dev->pool_name);
Alex Elderbf3e5ae2012-07-09 21:04:23 -05002575 kfree(rbd_dev->obj);
Alex Elder32eec682012-02-08 16:11:14 -06002576 rbd_id_put(rbd_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002577 kfree(rbd_dev);
2578
2579 /* release module ref */
2580 module_put(THIS_MODULE);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002581}
2582
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002583static ssize_t rbd_remove(struct bus_type *bus,
2584 const char *buf,
2585 size_t count)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002586{
2587 struct rbd_device *rbd_dev = NULL;
2588 int target_id, rc;
2589 unsigned long ul;
2590 int ret = count;
2591
2592 rc = strict_strtoul(buf, 10, &ul);
2593 if (rc)
2594 return rc;
2595
2596 /* convert to int; abort if we lost anything in the conversion */
2597 target_id = (int) ul;
2598 if (target_id != ul)
2599 return -EINVAL;
2600
2601 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
2602
2603 rbd_dev = __rbd_get_dev(target_id);
2604 if (!rbd_dev) {
2605 ret = -ENOENT;
2606 goto done;
2607 }
2608
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002609 __rbd_remove_all_snaps(rbd_dev);
2610 rbd_bus_del_dev(rbd_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002611
2612done:
2613 mutex_unlock(&ctl_mutex);
2614 return ret;
2615}
2616
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002617static ssize_t rbd_snap_add(struct device *dev,
2618 struct device_attribute *attr,
2619 const char *buf,
2620 size_t count)
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002621{
Alex Elder593a9e72012-02-07 12:03:37 -06002622 struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002623 int ret;
2624 char *name = kmalloc(count + 1, GFP_KERNEL);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002625 if (!name)
2626 return -ENOMEM;
2627
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002628 snprintf(name, count, "%s", buf);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002629
2630 mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
2631
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002632 ret = rbd_header_add_snap(rbd_dev,
2633 name, GFP_KERNEL);
2634 if (ret < 0)
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002635 goto err_unlock;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002636
Josh Durgin263c6ca2011-12-05 10:43:42 -08002637 ret = __rbd_refresh_header(rbd_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002638 if (ret < 0)
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002639 goto err_unlock;
2640
2641 /* shouldn't hold ctl_mutex when notifying.. notify might
2642 trigger a watch callback that would need to get that mutex */
2643 mutex_unlock(&ctl_mutex);
2644
2645 /* make a best effort, don't error if failed */
2646 rbd_req_sync_notify(rbd_dev, rbd_dev->obj_md_name);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002647
2648 ret = count;
Yehuda Sadeh59c2be12011-03-21 15:10:11 -07002649 kfree(name);
2650 return ret;
2651
2652err_unlock:
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002653 mutex_unlock(&ctl_mutex);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002654 kfree(name);
2655 return ret;
2656}
2657
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002658/*
2659 * create control files in sysfs
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002660 * /sys/bus/rbd/...
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002661 */
2662static int rbd_sysfs_init(void)
2663{
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002664 int ret;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002665
Alex Elderfed4c142012-02-07 12:03:36 -06002666 ret = device_register(&rbd_root_dev);
Alex Elder21079782012-01-24 10:08:36 -06002667 if (ret < 0)
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002668 return ret;
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002669
Alex Elderfed4c142012-02-07 12:03:36 -06002670 ret = bus_register(&rbd_bus_type);
2671 if (ret < 0)
2672 device_unregister(&rbd_root_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002673
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002674 return ret;
2675}
2676
2677static void rbd_sysfs_cleanup(void)
2678{
Yehuda Sadehdfc56062010-11-19 14:51:04 -08002679 bus_unregister(&rbd_bus_type);
Alex Elderfed4c142012-02-07 12:03:36 -06002680 device_unregister(&rbd_root_dev);
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002681}
2682
2683int __init rbd_init(void)
2684{
2685 int rc;
2686
2687 rc = rbd_sysfs_init();
2688 if (rc)
2689 return rc;
Alex Elderf0f8cef2012-01-29 13:57:44 -06002690 pr_info("loaded " RBD_DRV_NAME_LONG "\n");
Yehuda Sadeh602adf42010-08-12 16:11:25 -07002691 return 0;
2692}
2693
2694void __exit rbd_exit(void)
2695{
2696 rbd_sysfs_cleanup();
2697}
2698
2699module_init(rbd_init);
2700module_exit(rbd_exit);
2701
2702MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
2703MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
2704MODULE_DESCRIPTION("rados block device");
2705
2706/* following authorship retained from original osdblk.c */
2707MODULE_AUTHOR("Jeff Garzik <jeff@garzik.org>");
2708
2709MODULE_LICENSE("GPL");