]>
Commit | Line | Data |
---|---|---|
b411b363 PR |
1 | /* |
2 | drbd_receiver.c | |
3 | ||
4 | This file is part of DRBD by Philipp Reisner and Lars Ellenberg. | |
5 | ||
6 | Copyright (C) 2001-2008, LINBIT Information Technologies GmbH. | |
7 | Copyright (C) 1999-2008, Philipp Reisner <philipp.reisner@linbit.com>. | |
8 | Copyright (C) 2002-2008, Lars Ellenberg <lars.ellenberg@linbit.com>. | |
9 | ||
10 | drbd is free software; you can redistribute it and/or modify | |
11 | it under the terms of the GNU General Public License as published by | |
12 | the Free Software Foundation; either version 2, or (at your option) | |
13 | any later version. | |
14 | ||
15 | drbd is distributed in the hope that it will be useful, | |
16 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
17 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
18 | GNU General Public License for more details. | |
19 | ||
20 | You should have received a copy of the GNU General Public License | |
21 | along with drbd; see the file COPYING. If not, write to | |
22 | the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. | |
23 | */ | |
24 | ||
25 | ||
b411b363 PR |
26 | #include <linux/module.h> |
27 | ||
28 | #include <asm/uaccess.h> | |
29 | #include <net/sock.h> | |
30 | ||
b411b363 PR |
31 | #include <linux/drbd.h> |
32 | #include <linux/fs.h> | |
33 | #include <linux/file.h> | |
34 | #include <linux/in.h> | |
35 | #include <linux/mm.h> | |
36 | #include <linux/memcontrol.h> | |
37 | #include <linux/mm_inline.h> | |
38 | #include <linux/slab.h> | |
b411b363 PR |
39 | #include <linux/pkt_sched.h> |
40 | #define __KERNEL_SYSCALLS__ | |
41 | #include <linux/unistd.h> | |
42 | #include <linux/vmalloc.h> | |
43 | #include <linux/random.h> | |
b411b363 PR |
44 | #include <linux/string.h> |
45 | #include <linux/scatterlist.h> | |
46 | #include "drbd_int.h" | |
a3603a6e | 47 | #include "drbd_protocol.h" |
b411b363 | 48 | #include "drbd_req.h" |
b411b363 PR |
49 | #include "drbd_vli.h" |
50 | ||
20c68fde LE |
51 | #define PRO_FEATURES (FF_TRIM) |
52 | ||
77351055 PR |
53 | struct packet_info { |
54 | enum drbd_packet cmd; | |
e2857216 AG |
55 | unsigned int size; |
56 | unsigned int vnr; | |
e658983a | 57 | void *data; |
77351055 PR |
58 | }; |
59 | ||
b411b363 PR |
60 | enum finish_epoch { |
61 | FE_STILL_LIVE, | |
62 | FE_DESTROYED, | |
63 | FE_RECYCLED, | |
64 | }; | |
65 | ||
bde89a9e AG |
66 | static int drbd_do_features(struct drbd_connection *connection); |
67 | static int drbd_do_auth(struct drbd_connection *connection); | |
69a22773 | 68 | static int drbd_disconnected(struct drbd_peer_device *); |
a0fb3c47 | 69 | static void conn_wait_active_ee_empty(struct drbd_connection *connection); |
bde89a9e | 70 | static enum finish_epoch drbd_may_finish_epoch(struct drbd_connection *, struct drbd_epoch *, enum epoch_event); |
99920dc5 | 71 | static int e_end_block(struct drbd_work *, int); |
b411b363 | 72 | |
b411b363 PR |
73 | |
74 | #define GFP_TRY (__GFP_HIGHMEM | __GFP_NOWARN) | |
75 | ||
45bb912b LE |
76 | /* |
77 | * some helper functions to deal with single linked page lists, | |
78 | * page->private being our "next" pointer. | |
79 | */ | |
80 | ||
81 | /* If at least n pages are linked at head, get n pages off. | |
82 | * Otherwise, don't modify head, and return NULL. | |
83 | * Locking is the responsibility of the caller. | |
84 | */ | |
85 | static struct page *page_chain_del(struct page **head, int n) | |
86 | { | |
87 | struct page *page; | |
88 | struct page *tmp; | |
89 | ||
90 | BUG_ON(!n); | |
91 | BUG_ON(!head); | |
92 | ||
93 | page = *head; | |
23ce4227 PR |
94 | |
95 | if (!page) | |
96 | return NULL; | |
97 | ||
45bb912b LE |
98 | while (page) { |
99 | tmp = page_chain_next(page); | |
100 | if (--n == 0) | |
101 | break; /* found sufficient pages */ | |
102 | if (tmp == NULL) | |
103 | /* insufficient pages, don't use any of them. */ | |
104 | return NULL; | |
105 | page = tmp; | |
106 | } | |
107 | ||
108 | /* add end of list marker for the returned list */ | |
109 | set_page_private(page, 0); | |
110 | /* actual return value, and adjustment of head */ | |
111 | page = *head; | |
112 | *head = tmp; | |
113 | return page; | |
114 | } | |
115 | ||
116 | /* may be used outside of locks to find the tail of a (usually short) | |
117 | * "private" page chain, before adding it back to a global chain head | |
118 | * with page_chain_add() under a spinlock. */ | |
119 | static struct page *page_chain_tail(struct page *page, int *len) | |
120 | { | |
121 | struct page *tmp; | |
122 | int i = 1; | |
123 | while ((tmp = page_chain_next(page))) | |
124 | ++i, page = tmp; | |
125 | if (len) | |
126 | *len = i; | |
127 | return page; | |
128 | } | |
129 | ||
130 | static int page_chain_free(struct page *page) | |
131 | { | |
132 | struct page *tmp; | |
133 | int i = 0; | |
134 | page_chain_for_each_safe(page, tmp) { | |
135 | put_page(page); | |
136 | ++i; | |
137 | } | |
138 | return i; | |
139 | } | |
140 | ||
141 | static void page_chain_add(struct page **head, | |
142 | struct page *chain_first, struct page *chain_last) | |
143 | { | |
144 | #if 1 | |
145 | struct page *tmp; | |
146 | tmp = page_chain_tail(chain_first, NULL); | |
147 | BUG_ON(tmp != chain_last); | |
148 | #endif | |
149 | ||
150 | /* add chain to head */ | |
151 | set_page_private(chain_last, (unsigned long)*head); | |
152 | *head = chain_first; | |
153 | } | |
154 | ||
b30ab791 | 155 | static struct page *__drbd_alloc_pages(struct drbd_device *device, |
18c2d522 | 156 | unsigned int number) |
b411b363 PR |
157 | { |
158 | struct page *page = NULL; | |
45bb912b | 159 | struct page *tmp = NULL; |
18c2d522 | 160 | unsigned int i = 0; |
b411b363 PR |
161 | |
162 | /* Yes, testing drbd_pp_vacant outside the lock is racy. | |
163 | * So what. It saves a spin_lock. */ | |
45bb912b | 164 | if (drbd_pp_vacant >= number) { |
b411b363 | 165 | spin_lock(&drbd_pp_lock); |
45bb912b LE |
166 | page = page_chain_del(&drbd_pp_pool, number); |
167 | if (page) | |
168 | drbd_pp_vacant -= number; | |
b411b363 | 169 | spin_unlock(&drbd_pp_lock); |
45bb912b LE |
170 | if (page) |
171 | return page; | |
b411b363 | 172 | } |
45bb912b | 173 | |
b411b363 PR |
174 | /* GFP_TRY, because we must not cause arbitrary write-out: in a DRBD |
175 | * "criss-cross" setup, that might cause write-out on some other DRBD, | |
176 | * which in turn might block on the other node at this very place. */ | |
45bb912b LE |
177 | for (i = 0; i < number; i++) { |
178 | tmp = alloc_page(GFP_TRY); | |
179 | if (!tmp) | |
180 | break; | |
181 | set_page_private(tmp, (unsigned long)page); | |
182 | page = tmp; | |
183 | } | |
184 | ||
185 | if (i == number) | |
186 | return page; | |
187 | ||
188 | /* Not enough pages immediately available this time. | |
c37c8ecf | 189 | * No need to jump around here, drbd_alloc_pages will retry this |
45bb912b LE |
190 | * function "soon". */ |
191 | if (page) { | |
192 | tmp = page_chain_tail(page, NULL); | |
193 | spin_lock(&drbd_pp_lock); | |
194 | page_chain_add(&drbd_pp_pool, page, tmp); | |
195 | drbd_pp_vacant += i; | |
196 | spin_unlock(&drbd_pp_lock); | |
197 | } | |
198 | return NULL; | |
b411b363 PR |
199 | } |
200 | ||
b30ab791 | 201 | static void reclaim_finished_net_peer_reqs(struct drbd_device *device, |
a990be46 | 202 | struct list_head *to_be_freed) |
b411b363 | 203 | { |
a8cd15ba | 204 | struct drbd_peer_request *peer_req, *tmp; |
b411b363 PR |
205 | |
206 | /* The EEs are always appended to the end of the list. Since | |
207 | they are sent in order over the wire, they have to finish | |
208 | in order. As soon as we see the first not finished we can | |
209 | stop to examine the list... */ | |
210 | ||
a8cd15ba | 211 | list_for_each_entry_safe(peer_req, tmp, &device->net_ee, w.list) { |
045417f7 | 212 | if (drbd_peer_req_has_active_page(peer_req)) |
b411b363 | 213 | break; |
a8cd15ba | 214 | list_move(&peer_req->w.list, to_be_freed); |
b411b363 PR |
215 | } |
216 | } | |
217 | ||
668700b4 | 218 | static void drbd_reclaim_net_peer_reqs(struct drbd_device *device) |
b411b363 PR |
219 | { |
220 | LIST_HEAD(reclaimed); | |
db830c46 | 221 | struct drbd_peer_request *peer_req, *t; |
b411b363 | 222 | |
0500813f | 223 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 224 | reclaim_finished_net_peer_reqs(device, &reclaimed); |
0500813f | 225 | spin_unlock_irq(&device->resource->req_lock); |
a8cd15ba | 226 | list_for_each_entry_safe(peer_req, t, &reclaimed, w.list) |
b30ab791 | 227 | drbd_free_net_peer_req(device, peer_req); |
b411b363 PR |
228 | } |
229 | ||
668700b4 PR |
230 | static void conn_reclaim_net_peer_reqs(struct drbd_connection *connection) |
231 | { | |
232 | struct drbd_peer_device *peer_device; | |
233 | int vnr; | |
234 | ||
235 | rcu_read_lock(); | |
236 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { | |
237 | struct drbd_device *device = peer_device->device; | |
238 | if (!atomic_read(&device->pp_in_use_by_net)) | |
239 | continue; | |
240 | ||
241 | kref_get(&device->kref); | |
242 | rcu_read_unlock(); | |
243 | drbd_reclaim_net_peer_reqs(device); | |
244 | kref_put(&device->kref, drbd_destroy_device); | |
245 | rcu_read_lock(); | |
246 | } | |
247 | rcu_read_unlock(); | |
248 | } | |
249 | ||
b411b363 | 250 | /** |
c37c8ecf | 251 | * drbd_alloc_pages() - Returns @number pages, retries forever (or until signalled) |
b30ab791 | 252 | * @device: DRBD device. |
45bb912b LE |
253 | * @number: number of pages requested |
254 | * @retry: whether to retry, if not enough pages are available right now | |
255 | * | |
256 | * Tries to allocate number pages, first from our own page pool, then from | |
0e49d7b0 | 257 | * the kernel. |
45bb912b | 258 | * Possibly retry until DRBD frees sufficient pages somewhere else. |
b411b363 | 259 | * |
0e49d7b0 LE |
260 | * If this allocation would exceed the max_buffers setting, we throttle |
261 | * allocation (schedule_timeout) to give the system some room to breathe. | |
262 | * | |
263 | * We do not use max-buffers as hard limit, because it could lead to | |
264 | * congestion and further to a distributed deadlock during online-verify or | |
265 | * (checksum based) resync, if the max-buffers, socket buffer sizes and | |
266 | * resync-rate settings are mis-configured. | |
267 | * | |
45bb912b | 268 | * Returns a page chain linked via page->private. |
b411b363 | 269 | */ |
69a22773 | 270 | struct page *drbd_alloc_pages(struct drbd_peer_device *peer_device, unsigned int number, |
c37c8ecf | 271 | bool retry) |
b411b363 | 272 | { |
69a22773 | 273 | struct drbd_device *device = peer_device->device; |
b411b363 | 274 | struct page *page = NULL; |
44ed167d | 275 | struct net_conf *nc; |
b411b363 | 276 | DEFINE_WAIT(wait); |
0e49d7b0 | 277 | unsigned int mxb; |
b411b363 | 278 | |
44ed167d | 279 | rcu_read_lock(); |
69a22773 | 280 | nc = rcu_dereference(peer_device->connection->net_conf); |
44ed167d PR |
281 | mxb = nc ? nc->max_buffers : 1000000; |
282 | rcu_read_unlock(); | |
283 | ||
b30ab791 AG |
284 | if (atomic_read(&device->pp_in_use) < mxb) |
285 | page = __drbd_alloc_pages(device, number); | |
b411b363 | 286 | |
668700b4 PR |
287 | /* Try to keep the fast path fast, but occasionally we need |
288 | * to reclaim the pages we lended to the network stack. */ | |
289 | if (page && atomic_read(&device->pp_in_use_by_net) > 512) | |
290 | drbd_reclaim_net_peer_reqs(device); | |
291 | ||
45bb912b | 292 | while (page == NULL) { |
b411b363 PR |
293 | prepare_to_wait(&drbd_pp_wait, &wait, TASK_INTERRUPTIBLE); |
294 | ||
668700b4 | 295 | drbd_reclaim_net_peer_reqs(device); |
b411b363 | 296 | |
b30ab791 AG |
297 | if (atomic_read(&device->pp_in_use) < mxb) { |
298 | page = __drbd_alloc_pages(device, number); | |
b411b363 PR |
299 | if (page) |
300 | break; | |
301 | } | |
302 | ||
303 | if (!retry) | |
304 | break; | |
305 | ||
306 | if (signal_pending(current)) { | |
d0180171 | 307 | drbd_warn(device, "drbd_alloc_pages interrupted!\n"); |
b411b363 PR |
308 | break; |
309 | } | |
310 | ||
0e49d7b0 LE |
311 | if (schedule_timeout(HZ/10) == 0) |
312 | mxb = UINT_MAX; | |
b411b363 PR |
313 | } |
314 | finish_wait(&drbd_pp_wait, &wait); | |
315 | ||
45bb912b | 316 | if (page) |
b30ab791 | 317 | atomic_add(number, &device->pp_in_use); |
b411b363 PR |
318 | return page; |
319 | } | |
320 | ||
c37c8ecf | 321 | /* Must not be used from irq, as that may deadlock: see drbd_alloc_pages. |
0500813f | 322 | * Is also used from inside an other spin_lock_irq(&resource->req_lock); |
45bb912b LE |
323 | * Either links the page chain back to the global pool, |
324 | * or returns all pages to the system. */ | |
b30ab791 | 325 | static void drbd_free_pages(struct drbd_device *device, struct page *page, int is_net) |
b411b363 | 326 | { |
b30ab791 | 327 | atomic_t *a = is_net ? &device->pp_in_use_by_net : &device->pp_in_use; |
b411b363 | 328 | int i; |
435f0740 | 329 | |
a73ff323 LE |
330 | if (page == NULL) |
331 | return; | |
332 | ||
81a5d60e | 333 | if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE) * minor_count) |
45bb912b LE |
334 | i = page_chain_free(page); |
335 | else { | |
336 | struct page *tmp; | |
337 | tmp = page_chain_tail(page, &i); | |
338 | spin_lock(&drbd_pp_lock); | |
339 | page_chain_add(&drbd_pp_pool, page, tmp); | |
340 | drbd_pp_vacant += i; | |
341 | spin_unlock(&drbd_pp_lock); | |
b411b363 | 342 | } |
435f0740 | 343 | i = atomic_sub_return(i, a); |
45bb912b | 344 | if (i < 0) |
d0180171 | 345 | drbd_warn(device, "ASSERTION FAILED: %s: %d < 0\n", |
435f0740 | 346 | is_net ? "pp_in_use_by_net" : "pp_in_use", i); |
b411b363 PR |
347 | wake_up(&drbd_pp_wait); |
348 | } | |
349 | ||
350 | /* | |
351 | You need to hold the req_lock: | |
352 | _drbd_wait_ee_list_empty() | |
353 | ||
354 | You must not have the req_lock: | |
3967deb1 | 355 | drbd_free_peer_req() |
0db55363 | 356 | drbd_alloc_peer_req() |
7721f567 | 357 | drbd_free_peer_reqs() |
b411b363 | 358 | drbd_ee_fix_bhs() |
a990be46 | 359 | drbd_finish_peer_reqs() |
b411b363 PR |
360 | drbd_clear_done_ee() |
361 | drbd_wait_ee_list_empty() | |
362 | */ | |
363 | ||
f6ffca9f | 364 | struct drbd_peer_request * |
69a22773 | 365 | drbd_alloc_peer_req(struct drbd_peer_device *peer_device, u64 id, sector_t sector, |
a0fb3c47 | 366 | unsigned int data_size, bool has_payload, gfp_t gfp_mask) __must_hold(local) |
b411b363 | 367 | { |
69a22773 | 368 | struct drbd_device *device = peer_device->device; |
db830c46 | 369 | struct drbd_peer_request *peer_req; |
a73ff323 | 370 | struct page *page = NULL; |
45bb912b | 371 | unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT; |
b411b363 | 372 | |
b30ab791 | 373 | if (drbd_insert_fault(device, DRBD_FAULT_AL_EE)) |
b411b363 PR |
374 | return NULL; |
375 | ||
db830c46 AG |
376 | peer_req = mempool_alloc(drbd_ee_mempool, gfp_mask & ~__GFP_HIGHMEM); |
377 | if (!peer_req) { | |
b411b363 | 378 | if (!(gfp_mask & __GFP_NOWARN)) |
d0180171 | 379 | drbd_err(device, "%s: allocation failed\n", __func__); |
b411b363 PR |
380 | return NULL; |
381 | } | |
382 | ||
a0fb3c47 | 383 | if (has_payload && data_size) { |
d0164adc MG |
384 | page = drbd_alloc_pages(peer_device, nr_pages, |
385 | gfpflags_allow_blocking(gfp_mask)); | |
a73ff323 LE |
386 | if (!page) |
387 | goto fail; | |
388 | } | |
b411b363 | 389 | |
c5a2c150 LE |
390 | memset(peer_req, 0, sizeof(*peer_req)); |
391 | INIT_LIST_HEAD(&peer_req->w.list); | |
db830c46 AG |
392 | drbd_clear_interval(&peer_req->i); |
393 | peer_req->i.size = data_size; | |
394 | peer_req->i.sector = sector; | |
c5a2c150 | 395 | peer_req->submit_jif = jiffies; |
a8cd15ba | 396 | peer_req->peer_device = peer_device; |
db830c46 | 397 | peer_req->pages = page; |
9a8e7753 AG |
398 | /* |
399 | * The block_id is opaque to the receiver. It is not endianness | |
400 | * converted, and sent back to the sender unchanged. | |
401 | */ | |
db830c46 | 402 | peer_req->block_id = id; |
b411b363 | 403 | |
db830c46 | 404 | return peer_req; |
b411b363 | 405 | |
45bb912b | 406 | fail: |
db830c46 | 407 | mempool_free(peer_req, drbd_ee_mempool); |
b411b363 PR |
408 | return NULL; |
409 | } | |
410 | ||
b30ab791 | 411 | void __drbd_free_peer_req(struct drbd_device *device, struct drbd_peer_request *peer_req, |
f6ffca9f | 412 | int is_net) |
b411b363 | 413 | { |
21ae5d7f | 414 | might_sleep(); |
db830c46 AG |
415 | if (peer_req->flags & EE_HAS_DIGEST) |
416 | kfree(peer_req->digest); | |
b30ab791 | 417 | drbd_free_pages(device, peer_req->pages, is_net); |
0b0ba1ef AG |
418 | D_ASSERT(device, atomic_read(&peer_req->pending_bios) == 0); |
419 | D_ASSERT(device, drbd_interval_empty(&peer_req->i)); | |
21ae5d7f LE |
420 | if (!expect(!(peer_req->flags & EE_CALL_AL_COMPLETE_IO))) { |
421 | peer_req->flags &= ~EE_CALL_AL_COMPLETE_IO; | |
422 | drbd_al_complete_io(device, &peer_req->i); | |
423 | } | |
db830c46 | 424 | mempool_free(peer_req, drbd_ee_mempool); |
b411b363 PR |
425 | } |
426 | ||
b30ab791 | 427 | int drbd_free_peer_reqs(struct drbd_device *device, struct list_head *list) |
b411b363 PR |
428 | { |
429 | LIST_HEAD(work_list); | |
db830c46 | 430 | struct drbd_peer_request *peer_req, *t; |
b411b363 | 431 | int count = 0; |
b30ab791 | 432 | int is_net = list == &device->net_ee; |
b411b363 | 433 | |
0500813f | 434 | spin_lock_irq(&device->resource->req_lock); |
b411b363 | 435 | list_splice_init(list, &work_list); |
0500813f | 436 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 | 437 | |
a8cd15ba | 438 | list_for_each_entry_safe(peer_req, t, &work_list, w.list) { |
b30ab791 | 439 | __drbd_free_peer_req(device, peer_req, is_net); |
b411b363 PR |
440 | count++; |
441 | } | |
442 | return count; | |
443 | } | |
444 | ||
b411b363 | 445 | /* |
a990be46 | 446 | * See also comments in _req_mod(,BARRIER_ACKED) and receive_Barrier. |
b411b363 | 447 | */ |
b30ab791 | 448 | static int drbd_finish_peer_reqs(struct drbd_device *device) |
b411b363 PR |
449 | { |
450 | LIST_HEAD(work_list); | |
451 | LIST_HEAD(reclaimed); | |
db830c46 | 452 | struct drbd_peer_request *peer_req, *t; |
e2b3032b | 453 | int err = 0; |
b411b363 | 454 | |
0500813f | 455 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 AG |
456 | reclaim_finished_net_peer_reqs(device, &reclaimed); |
457 | list_splice_init(&device->done_ee, &work_list); | |
0500813f | 458 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 | 459 | |
a8cd15ba | 460 | list_for_each_entry_safe(peer_req, t, &reclaimed, w.list) |
b30ab791 | 461 | drbd_free_net_peer_req(device, peer_req); |
b411b363 PR |
462 | |
463 | /* possible callbacks here: | |
d4dabbe2 | 464 | * e_end_block, and e_end_resync_block, e_send_superseded. |
b411b363 PR |
465 | * all ignore the last argument. |
466 | */ | |
a8cd15ba | 467 | list_for_each_entry_safe(peer_req, t, &work_list, w.list) { |
e2b3032b AG |
468 | int err2; |
469 | ||
b411b363 | 470 | /* list_del not necessary, next/prev members not touched */ |
a8cd15ba | 471 | err2 = peer_req->w.cb(&peer_req->w, !!err); |
e2b3032b AG |
472 | if (!err) |
473 | err = err2; | |
b30ab791 | 474 | drbd_free_peer_req(device, peer_req); |
b411b363 | 475 | } |
b30ab791 | 476 | wake_up(&device->ee_wait); |
b411b363 | 477 | |
e2b3032b | 478 | return err; |
b411b363 PR |
479 | } |
480 | ||
b30ab791 | 481 | static void _drbd_wait_ee_list_empty(struct drbd_device *device, |
d4da1537 | 482 | struct list_head *head) |
b411b363 PR |
483 | { |
484 | DEFINE_WAIT(wait); | |
485 | ||
486 | /* avoids spin_lock/unlock | |
487 | * and calling prepare_to_wait in the fast path */ | |
488 | while (!list_empty(head)) { | |
b30ab791 | 489 | prepare_to_wait(&device->ee_wait, &wait, TASK_UNINTERRUPTIBLE); |
0500813f | 490 | spin_unlock_irq(&device->resource->req_lock); |
7eaceacc | 491 | io_schedule(); |
b30ab791 | 492 | finish_wait(&device->ee_wait, &wait); |
0500813f | 493 | spin_lock_irq(&device->resource->req_lock); |
b411b363 PR |
494 | } |
495 | } | |
496 | ||
b30ab791 | 497 | static void drbd_wait_ee_list_empty(struct drbd_device *device, |
d4da1537 | 498 | struct list_head *head) |
b411b363 | 499 | { |
0500813f | 500 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 501 | _drbd_wait_ee_list_empty(device, head); |
0500813f | 502 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 PR |
503 | } |
504 | ||
dbd9eea0 | 505 | static int drbd_recv_short(struct socket *sock, void *buf, size_t size, int flags) |
b411b363 | 506 | { |
b411b363 PR |
507 | struct kvec iov = { |
508 | .iov_base = buf, | |
509 | .iov_len = size, | |
510 | }; | |
511 | struct msghdr msg = { | |
b411b363 PR |
512 | .msg_flags = (flags ? flags : MSG_WAITALL | MSG_NOSIGNAL) |
513 | }; | |
f730c848 | 514 | return kernel_recvmsg(sock, &msg, &iov, 1, size, msg.msg_flags); |
b411b363 PR |
515 | } |
516 | ||
bde89a9e | 517 | static int drbd_recv(struct drbd_connection *connection, void *buf, size_t size) |
b411b363 | 518 | { |
b411b363 PR |
519 | int rv; |
520 | ||
bde89a9e | 521 | rv = drbd_recv_short(connection->data.socket, buf, size, 0); |
b411b363 | 522 | |
dbd0820c PR |
523 | if (rv < 0) { |
524 | if (rv == -ECONNRESET) | |
1ec861eb | 525 | drbd_info(connection, "sock was reset by peer\n"); |
dbd0820c | 526 | else if (rv != -ERESTARTSYS) |
1ec861eb | 527 | drbd_err(connection, "sock_recvmsg returned %d\n", rv); |
dbd0820c | 528 | } else if (rv == 0) { |
bde89a9e | 529 | if (test_bit(DISCONNECT_SENT, &connection->flags)) { |
b66623e3 PR |
530 | long t; |
531 | rcu_read_lock(); | |
bde89a9e | 532 | t = rcu_dereference(connection->net_conf)->ping_timeo * HZ/10; |
b66623e3 PR |
533 | rcu_read_unlock(); |
534 | ||
bde89a9e | 535 | t = wait_event_timeout(connection->ping_wait, connection->cstate < C_WF_REPORT_PARAMS, t); |
b66623e3 | 536 | |
599377ac PR |
537 | if (t) |
538 | goto out; | |
539 | } | |
1ec861eb | 540 | drbd_info(connection, "sock was shut down by peer\n"); |
599377ac PR |
541 | } |
542 | ||
b411b363 | 543 | if (rv != size) |
bde89a9e | 544 | conn_request_state(connection, NS(conn, C_BROKEN_PIPE), CS_HARD); |
b411b363 | 545 | |
599377ac | 546 | out: |
b411b363 PR |
547 | return rv; |
548 | } | |
549 | ||
bde89a9e | 550 | static int drbd_recv_all(struct drbd_connection *connection, void *buf, size_t size) |
c6967746 AG |
551 | { |
552 | int err; | |
553 | ||
bde89a9e | 554 | err = drbd_recv(connection, buf, size); |
c6967746 AG |
555 | if (err != size) { |
556 | if (err >= 0) | |
557 | err = -EIO; | |
558 | } else | |
559 | err = 0; | |
560 | return err; | |
561 | } | |
562 | ||
bde89a9e | 563 | static int drbd_recv_all_warn(struct drbd_connection *connection, void *buf, size_t size) |
a5c31904 AG |
564 | { |
565 | int err; | |
566 | ||
bde89a9e | 567 | err = drbd_recv_all(connection, buf, size); |
a5c31904 | 568 | if (err && !signal_pending(current)) |
1ec861eb | 569 | drbd_warn(connection, "short read (expected size %d)\n", (int)size); |
a5c31904 AG |
570 | return err; |
571 | } | |
572 | ||
5dbf1673 LE |
573 | /* quoting tcp(7): |
574 | * On individual connections, the socket buffer size must be set prior to the | |
575 | * listen(2) or connect(2) calls in order to have it take effect. | |
576 | * This is our wrapper to do so. | |
577 | */ | |
578 | static void drbd_setbufsize(struct socket *sock, unsigned int snd, | |
579 | unsigned int rcv) | |
580 | { | |
581 | /* open coded SO_SNDBUF, SO_RCVBUF */ | |
582 | if (snd) { | |
583 | sock->sk->sk_sndbuf = snd; | |
584 | sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK; | |
585 | } | |
586 | if (rcv) { | |
587 | sock->sk->sk_rcvbuf = rcv; | |
588 | sock->sk->sk_userlocks |= SOCK_RCVBUF_LOCK; | |
589 | } | |
590 | } | |
591 | ||
bde89a9e | 592 | static struct socket *drbd_try_connect(struct drbd_connection *connection) |
b411b363 PR |
593 | { |
594 | const char *what; | |
595 | struct socket *sock; | |
596 | struct sockaddr_in6 src_in6; | |
44ed167d PR |
597 | struct sockaddr_in6 peer_in6; |
598 | struct net_conf *nc; | |
599 | int err, peer_addr_len, my_addr_len; | |
69ef82de | 600 | int sndbuf_size, rcvbuf_size, connect_int; |
b411b363 PR |
601 | int disconnect_on_error = 1; |
602 | ||
44ed167d | 603 | rcu_read_lock(); |
bde89a9e | 604 | nc = rcu_dereference(connection->net_conf); |
44ed167d PR |
605 | if (!nc) { |
606 | rcu_read_unlock(); | |
b411b363 | 607 | return NULL; |
44ed167d | 608 | } |
44ed167d PR |
609 | sndbuf_size = nc->sndbuf_size; |
610 | rcvbuf_size = nc->rcvbuf_size; | |
69ef82de | 611 | connect_int = nc->connect_int; |
089c075d | 612 | rcu_read_unlock(); |
44ed167d | 613 | |
bde89a9e AG |
614 | my_addr_len = min_t(int, connection->my_addr_len, sizeof(src_in6)); |
615 | memcpy(&src_in6, &connection->my_addr, my_addr_len); | |
44ed167d | 616 | |
bde89a9e | 617 | if (((struct sockaddr *)&connection->my_addr)->sa_family == AF_INET6) |
44ed167d PR |
618 | src_in6.sin6_port = 0; |
619 | else | |
620 | ((struct sockaddr_in *)&src_in6)->sin_port = 0; /* AF_INET & AF_SCI */ | |
621 | ||
bde89a9e AG |
622 | peer_addr_len = min_t(int, connection->peer_addr_len, sizeof(src_in6)); |
623 | memcpy(&peer_in6, &connection->peer_addr, peer_addr_len); | |
b411b363 PR |
624 | |
625 | what = "sock_create_kern"; | |
eeb1bd5c | 626 | err = sock_create_kern(&init_net, ((struct sockaddr *)&src_in6)->sa_family, |
44ed167d | 627 | SOCK_STREAM, IPPROTO_TCP, &sock); |
b411b363 PR |
628 | if (err < 0) { |
629 | sock = NULL; | |
630 | goto out; | |
631 | } | |
632 | ||
633 | sock->sk->sk_rcvtimeo = | |
69ef82de | 634 | sock->sk->sk_sndtimeo = connect_int * HZ; |
44ed167d | 635 | drbd_setbufsize(sock, sndbuf_size, rcvbuf_size); |
b411b363 PR |
636 | |
637 | /* explicitly bind to the configured IP as source IP | |
638 | * for the outgoing connections. | |
639 | * This is needed for multihomed hosts and to be | |
640 | * able to use lo: interfaces for drbd. | |
641 | * Make sure to use 0 as port number, so linux selects | |
642 | * a free one dynamically. | |
643 | */ | |
b411b363 | 644 | what = "bind before connect"; |
44ed167d | 645 | err = sock->ops->bind(sock, (struct sockaddr *) &src_in6, my_addr_len); |
b411b363 PR |
646 | if (err < 0) |
647 | goto out; | |
648 | ||
649 | /* connect may fail, peer not yet available. | |
650 | * stay C_WF_CONNECTION, don't go Disconnecting! */ | |
651 | disconnect_on_error = 0; | |
652 | what = "connect"; | |
44ed167d | 653 | err = sock->ops->connect(sock, (struct sockaddr *) &peer_in6, peer_addr_len, 0); |
b411b363 PR |
654 | |
655 | out: | |
656 | if (err < 0) { | |
657 | if (sock) { | |
658 | sock_release(sock); | |
659 | sock = NULL; | |
660 | } | |
661 | switch (-err) { | |
662 | /* timeout, busy, signal pending */ | |
663 | case ETIMEDOUT: case EAGAIN: case EINPROGRESS: | |
664 | case EINTR: case ERESTARTSYS: | |
665 | /* peer not (yet) available, network problem */ | |
666 | case ECONNREFUSED: case ENETUNREACH: | |
667 | case EHOSTDOWN: case EHOSTUNREACH: | |
668 | disconnect_on_error = 0; | |
669 | break; | |
670 | default: | |
1ec861eb | 671 | drbd_err(connection, "%s failed, err = %d\n", what, err); |
b411b363 PR |
672 | } |
673 | if (disconnect_on_error) | |
bde89a9e | 674 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b411b363 | 675 | } |
44ed167d | 676 | |
b411b363 PR |
677 | return sock; |
678 | } | |
679 | ||
7a426fd8 | 680 | struct accept_wait_data { |
bde89a9e | 681 | struct drbd_connection *connection; |
7a426fd8 PR |
682 | struct socket *s_listen; |
683 | struct completion door_bell; | |
684 | void (*original_sk_state_change)(struct sock *sk); | |
685 | ||
686 | }; | |
687 | ||
715306f6 | 688 | static void drbd_incoming_connection(struct sock *sk) |
7a426fd8 PR |
689 | { |
690 | struct accept_wait_data *ad = sk->sk_user_data; | |
715306f6 | 691 | void (*state_change)(struct sock *sk); |
7a426fd8 | 692 | |
715306f6 AG |
693 | state_change = ad->original_sk_state_change; |
694 | if (sk->sk_state == TCP_ESTABLISHED) | |
695 | complete(&ad->door_bell); | |
696 | state_change(sk); | |
7a426fd8 PR |
697 | } |
698 | ||
bde89a9e | 699 | static int prepare_listen_socket(struct drbd_connection *connection, struct accept_wait_data *ad) |
b411b363 | 700 | { |
1f3e509b | 701 | int err, sndbuf_size, rcvbuf_size, my_addr_len; |
44ed167d | 702 | struct sockaddr_in6 my_addr; |
1f3e509b | 703 | struct socket *s_listen; |
44ed167d | 704 | struct net_conf *nc; |
b411b363 PR |
705 | const char *what; |
706 | ||
44ed167d | 707 | rcu_read_lock(); |
bde89a9e | 708 | nc = rcu_dereference(connection->net_conf); |
44ed167d PR |
709 | if (!nc) { |
710 | rcu_read_unlock(); | |
7a426fd8 | 711 | return -EIO; |
44ed167d | 712 | } |
44ed167d PR |
713 | sndbuf_size = nc->sndbuf_size; |
714 | rcvbuf_size = nc->rcvbuf_size; | |
44ed167d | 715 | rcu_read_unlock(); |
b411b363 | 716 | |
bde89a9e AG |
717 | my_addr_len = min_t(int, connection->my_addr_len, sizeof(struct sockaddr_in6)); |
718 | memcpy(&my_addr, &connection->my_addr, my_addr_len); | |
b411b363 PR |
719 | |
720 | what = "sock_create_kern"; | |
eeb1bd5c | 721 | err = sock_create_kern(&init_net, ((struct sockaddr *)&my_addr)->sa_family, |
1f3e509b | 722 | SOCK_STREAM, IPPROTO_TCP, &s_listen); |
b411b363 PR |
723 | if (err) { |
724 | s_listen = NULL; | |
725 | goto out; | |
726 | } | |
727 | ||
98683650 | 728 | s_listen->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ |
44ed167d | 729 | drbd_setbufsize(s_listen, sndbuf_size, rcvbuf_size); |
b411b363 PR |
730 | |
731 | what = "bind before listen"; | |
44ed167d | 732 | err = s_listen->ops->bind(s_listen, (struct sockaddr *)&my_addr, my_addr_len); |
b411b363 PR |
733 | if (err < 0) |
734 | goto out; | |
735 | ||
7a426fd8 PR |
736 | ad->s_listen = s_listen; |
737 | write_lock_bh(&s_listen->sk->sk_callback_lock); | |
738 | ad->original_sk_state_change = s_listen->sk->sk_state_change; | |
715306f6 | 739 | s_listen->sk->sk_state_change = drbd_incoming_connection; |
7a426fd8 PR |
740 | s_listen->sk->sk_user_data = ad; |
741 | write_unlock_bh(&s_listen->sk->sk_callback_lock); | |
b411b363 | 742 | |
2820fd39 PR |
743 | what = "listen"; |
744 | err = s_listen->ops->listen(s_listen, 5); | |
745 | if (err < 0) | |
746 | goto out; | |
747 | ||
7a426fd8 | 748 | return 0; |
b411b363 PR |
749 | out: |
750 | if (s_listen) | |
751 | sock_release(s_listen); | |
752 | if (err < 0) { | |
753 | if (err != -EAGAIN && err != -EINTR && err != -ERESTARTSYS) { | |
1ec861eb | 754 | drbd_err(connection, "%s failed, err = %d\n", what, err); |
bde89a9e | 755 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b411b363 PR |
756 | } |
757 | } | |
b411b363 | 758 | |
7a426fd8 | 759 | return -EIO; |
b411b363 PR |
760 | } |
761 | ||
715306f6 | 762 | static void unregister_state_change(struct sock *sk, struct accept_wait_data *ad) |
b411b363 | 763 | { |
715306f6 AG |
764 | write_lock_bh(&sk->sk_callback_lock); |
765 | sk->sk_state_change = ad->original_sk_state_change; | |
766 | sk->sk_user_data = NULL; | |
767 | write_unlock_bh(&sk->sk_callback_lock); | |
b411b363 PR |
768 | } |
769 | ||
bde89a9e | 770 | static struct socket *drbd_wait_for_connect(struct drbd_connection *connection, struct accept_wait_data *ad) |
b411b363 | 771 | { |
1f3e509b PR |
772 | int timeo, connect_int, err = 0; |
773 | struct socket *s_estab = NULL; | |
1f3e509b PR |
774 | struct net_conf *nc; |
775 | ||
776 | rcu_read_lock(); | |
bde89a9e | 777 | nc = rcu_dereference(connection->net_conf); |
1f3e509b PR |
778 | if (!nc) { |
779 | rcu_read_unlock(); | |
780 | return NULL; | |
781 | } | |
782 | connect_int = nc->connect_int; | |
783 | rcu_read_unlock(); | |
784 | ||
785 | timeo = connect_int * HZ; | |
38b682b2 AM |
786 | /* 28.5% random jitter */ |
787 | timeo += (prandom_u32() & 1) ? timeo / 7 : -timeo / 7; | |
1f3e509b | 788 | |
7a426fd8 PR |
789 | err = wait_for_completion_interruptible_timeout(&ad->door_bell, timeo); |
790 | if (err <= 0) | |
791 | return NULL; | |
b411b363 | 792 | |
7a426fd8 | 793 | err = kernel_accept(ad->s_listen, &s_estab, 0); |
b411b363 PR |
794 | if (err < 0) { |
795 | if (err != -EAGAIN && err != -EINTR && err != -ERESTARTSYS) { | |
1ec861eb | 796 | drbd_err(connection, "accept failed, err = %d\n", err); |
bde89a9e | 797 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b411b363 PR |
798 | } |
799 | } | |
b411b363 | 800 | |
715306f6 AG |
801 | if (s_estab) |
802 | unregister_state_change(s_estab->sk, ad); | |
b411b363 | 803 | |
b411b363 PR |
804 | return s_estab; |
805 | } | |
b411b363 | 806 | |
bde89a9e | 807 | static int decode_header(struct drbd_connection *, void *, struct packet_info *); |
b411b363 | 808 | |
bde89a9e | 809 | static int send_first_packet(struct drbd_connection *connection, struct drbd_socket *sock, |
9f5bdc33 AG |
810 | enum drbd_packet cmd) |
811 | { | |
bde89a9e | 812 | if (!conn_prepare_command(connection, sock)) |
9f5bdc33 | 813 | return -EIO; |
bde89a9e | 814 | return conn_send_command(connection, sock, cmd, 0, NULL, 0); |
b411b363 PR |
815 | } |
816 | ||
bde89a9e | 817 | static int receive_first_packet(struct drbd_connection *connection, struct socket *sock) |
b411b363 | 818 | { |
bde89a9e | 819 | unsigned int header_size = drbd_header_size(connection); |
9f5bdc33 | 820 | struct packet_info pi; |
4920e37a | 821 | struct net_conf *nc; |
9f5bdc33 | 822 | int err; |
b411b363 | 823 | |
4920e37a PR |
824 | rcu_read_lock(); |
825 | nc = rcu_dereference(connection->net_conf); | |
826 | if (!nc) { | |
827 | rcu_read_unlock(); | |
828 | return -EIO; | |
829 | } | |
830 | sock->sk->sk_rcvtimeo = nc->ping_timeo * 4 * HZ / 10; | |
831 | rcu_read_unlock(); | |
832 | ||
bde89a9e | 833 | err = drbd_recv_short(sock, connection->data.rbuf, header_size, 0); |
9f5bdc33 AG |
834 | if (err != header_size) { |
835 | if (err >= 0) | |
836 | err = -EIO; | |
837 | return err; | |
838 | } | |
bde89a9e | 839 | err = decode_header(connection, connection->data.rbuf, &pi); |
9f5bdc33 AG |
840 | if (err) |
841 | return err; | |
842 | return pi.cmd; | |
b411b363 PR |
843 | } |
844 | ||
845 | /** | |
846 | * drbd_socket_okay() - Free the socket if its connection is not okay | |
b411b363 PR |
847 | * @sock: pointer to the pointer to the socket. |
848 | */ | |
5d0b17f1 | 849 | static bool drbd_socket_okay(struct socket **sock) |
b411b363 PR |
850 | { |
851 | int rr; | |
852 | char tb[4]; | |
853 | ||
854 | if (!*sock) | |
81e84650 | 855 | return false; |
b411b363 | 856 | |
dbd9eea0 | 857 | rr = drbd_recv_short(*sock, tb, 4, MSG_DONTWAIT | MSG_PEEK); |
b411b363 PR |
858 | |
859 | if (rr > 0 || rr == -EAGAIN) { | |
81e84650 | 860 | return true; |
b411b363 PR |
861 | } else { |
862 | sock_release(*sock); | |
863 | *sock = NULL; | |
81e84650 | 864 | return false; |
b411b363 PR |
865 | } |
866 | } | |
5d0b17f1 PR |
867 | |
868 | static bool connection_established(struct drbd_connection *connection, | |
869 | struct socket **sock1, | |
870 | struct socket **sock2) | |
871 | { | |
872 | struct net_conf *nc; | |
873 | int timeout; | |
874 | bool ok; | |
875 | ||
876 | if (!*sock1 || !*sock2) | |
877 | return false; | |
878 | ||
879 | rcu_read_lock(); | |
880 | nc = rcu_dereference(connection->net_conf); | |
881 | timeout = (nc->sock_check_timeo ?: nc->ping_timeo) * HZ / 10; | |
882 | rcu_read_unlock(); | |
883 | schedule_timeout_interruptible(timeout); | |
884 | ||
885 | ok = drbd_socket_okay(sock1); | |
886 | ok = drbd_socket_okay(sock2) && ok; | |
887 | ||
888 | return ok; | |
889 | } | |
890 | ||
2325eb66 PR |
891 | /* Gets called if a connection is established, or if a new minor gets created |
892 | in a connection */ | |
69a22773 | 893 | int drbd_connected(struct drbd_peer_device *peer_device) |
907599e0 | 894 | { |
69a22773 | 895 | struct drbd_device *device = peer_device->device; |
0829f5ed | 896 | int err; |
907599e0 | 897 | |
b30ab791 AG |
898 | atomic_set(&device->packet_seq, 0); |
899 | device->peer_seq = 0; | |
907599e0 | 900 | |
69a22773 AG |
901 | device->state_mutex = peer_device->connection->agreed_pro_version < 100 ? |
902 | &peer_device->connection->cstate_mutex : | |
b30ab791 | 903 | &device->own_state_mutex; |
8410da8f | 904 | |
69a22773 | 905 | err = drbd_send_sync_param(peer_device); |
0829f5ed | 906 | if (!err) |
69a22773 | 907 | err = drbd_send_sizes(peer_device, 0, 0); |
0829f5ed | 908 | if (!err) |
69a22773 | 909 | err = drbd_send_uuids(peer_device); |
0829f5ed | 910 | if (!err) |
69a22773 | 911 | err = drbd_send_current_state(peer_device); |
b30ab791 AG |
912 | clear_bit(USE_DEGR_WFC_T, &device->flags); |
913 | clear_bit(RESIZE_PENDING, &device->flags); | |
914 | atomic_set(&device->ap_in_flight, 0); | |
915 | mod_timer(&device->request_timer, jiffies + HZ); /* just start it here. */ | |
0829f5ed | 916 | return err; |
907599e0 | 917 | } |
b411b363 PR |
918 | |
919 | /* | |
920 | * return values: | |
921 | * 1 yes, we have a valid connection | |
922 | * 0 oops, did not work out, please try again | |
923 | * -1 peer talks different language, | |
924 | * no point in trying again, please go standalone. | |
925 | * -2 We do not have a network config... | |
926 | */ | |
bde89a9e | 927 | static int conn_connect(struct drbd_connection *connection) |
b411b363 | 928 | { |
7da35862 | 929 | struct drbd_socket sock, msock; |
c06ece6b | 930 | struct drbd_peer_device *peer_device; |
44ed167d | 931 | struct net_conf *nc; |
5d0b17f1 PR |
932 | int vnr, timeout, h; |
933 | bool discard_my_data, ok; | |
197296ff | 934 | enum drbd_state_rv rv; |
7a426fd8 | 935 | struct accept_wait_data ad = { |
bde89a9e | 936 | .connection = connection, |
7a426fd8 PR |
937 | .door_bell = COMPLETION_INITIALIZER_ONSTACK(ad.door_bell), |
938 | }; | |
b411b363 | 939 | |
bde89a9e AG |
940 | clear_bit(DISCONNECT_SENT, &connection->flags); |
941 | if (conn_request_state(connection, NS(conn, C_WF_CONNECTION), CS_VERBOSE) < SS_SUCCESS) | |
b411b363 PR |
942 | return -2; |
943 | ||
7da35862 | 944 | mutex_init(&sock.mutex); |
bde89a9e AG |
945 | sock.sbuf = connection->data.sbuf; |
946 | sock.rbuf = connection->data.rbuf; | |
7da35862 PR |
947 | sock.socket = NULL; |
948 | mutex_init(&msock.mutex); | |
bde89a9e AG |
949 | msock.sbuf = connection->meta.sbuf; |
950 | msock.rbuf = connection->meta.rbuf; | |
7da35862 PR |
951 | msock.socket = NULL; |
952 | ||
0916e0e3 | 953 | /* Assume that the peer only understands protocol 80 until we know better. */ |
bde89a9e | 954 | connection->agreed_pro_version = 80; |
b411b363 | 955 | |
bde89a9e | 956 | if (prepare_listen_socket(connection, &ad)) |
7a426fd8 | 957 | return 0; |
b411b363 PR |
958 | |
959 | do { | |
2bf89621 | 960 | struct socket *s; |
b411b363 | 961 | |
bde89a9e | 962 | s = drbd_try_connect(connection); |
b411b363 | 963 | if (s) { |
7da35862 PR |
964 | if (!sock.socket) { |
965 | sock.socket = s; | |
bde89a9e | 966 | send_first_packet(connection, &sock, P_INITIAL_DATA); |
7da35862 | 967 | } else if (!msock.socket) { |
bde89a9e | 968 | clear_bit(RESOLVE_CONFLICTS, &connection->flags); |
7da35862 | 969 | msock.socket = s; |
bde89a9e | 970 | send_first_packet(connection, &msock, P_INITIAL_META); |
b411b363 | 971 | } else { |
1ec861eb | 972 | drbd_err(connection, "Logic error in conn_connect()\n"); |
b411b363 PR |
973 | goto out_release_sockets; |
974 | } | |
975 | } | |
976 | ||
5d0b17f1 PR |
977 | if (connection_established(connection, &sock.socket, &msock.socket)) |
978 | break; | |
b411b363 PR |
979 | |
980 | retry: | |
bde89a9e | 981 | s = drbd_wait_for_connect(connection, &ad); |
b411b363 | 982 | if (s) { |
bde89a9e | 983 | int fp = receive_first_packet(connection, s); |
7da35862 PR |
984 | drbd_socket_okay(&sock.socket); |
985 | drbd_socket_okay(&msock.socket); | |
92f14951 | 986 | switch (fp) { |
e5d6f33a | 987 | case P_INITIAL_DATA: |
7da35862 | 988 | if (sock.socket) { |
1ec861eb | 989 | drbd_warn(connection, "initial packet S crossed\n"); |
7da35862 | 990 | sock_release(sock.socket); |
80c6eed4 PR |
991 | sock.socket = s; |
992 | goto randomize; | |
b411b363 | 993 | } |
7da35862 | 994 | sock.socket = s; |
b411b363 | 995 | break; |
e5d6f33a | 996 | case P_INITIAL_META: |
bde89a9e | 997 | set_bit(RESOLVE_CONFLICTS, &connection->flags); |
7da35862 | 998 | if (msock.socket) { |
1ec861eb | 999 | drbd_warn(connection, "initial packet M crossed\n"); |
7da35862 | 1000 | sock_release(msock.socket); |
80c6eed4 PR |
1001 | msock.socket = s; |
1002 | goto randomize; | |
b411b363 | 1003 | } |
7da35862 | 1004 | msock.socket = s; |
b411b363 PR |
1005 | break; |
1006 | default: | |
1ec861eb | 1007 | drbd_warn(connection, "Error receiving initial packet\n"); |
b411b363 | 1008 | sock_release(s); |
80c6eed4 | 1009 | randomize: |
38b682b2 | 1010 | if (prandom_u32() & 1) |
b411b363 PR |
1011 | goto retry; |
1012 | } | |
1013 | } | |
1014 | ||
bde89a9e | 1015 | if (connection->cstate <= C_DISCONNECTING) |
b411b363 PR |
1016 | goto out_release_sockets; |
1017 | if (signal_pending(current)) { | |
1018 | flush_signals(current); | |
1019 | smp_rmb(); | |
bde89a9e | 1020 | if (get_t_state(&connection->receiver) == EXITING) |
b411b363 PR |
1021 | goto out_release_sockets; |
1022 | } | |
1023 | ||
5d0b17f1 | 1024 | ok = connection_established(connection, &sock.socket, &msock.socket); |
b666dbf8 | 1025 | } while (!ok); |
b411b363 | 1026 | |
7a426fd8 PR |
1027 | if (ad.s_listen) |
1028 | sock_release(ad.s_listen); | |
b411b363 | 1029 | |
98683650 PR |
1030 | sock.socket->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ |
1031 | msock.socket->sk->sk_reuse = SK_CAN_REUSE; /* SO_REUSEADDR */ | |
b411b363 | 1032 | |
7da35862 PR |
1033 | sock.socket->sk->sk_allocation = GFP_NOIO; |
1034 | msock.socket->sk->sk_allocation = GFP_NOIO; | |
b411b363 | 1035 | |
7da35862 PR |
1036 | sock.socket->sk->sk_priority = TC_PRIO_INTERACTIVE_BULK; |
1037 | msock.socket->sk->sk_priority = TC_PRIO_INTERACTIVE; | |
b411b363 | 1038 | |
b411b363 | 1039 | /* NOT YET ... |
bde89a9e | 1040 | * sock.socket->sk->sk_sndtimeo = connection->net_conf->timeout*HZ/10; |
7da35862 | 1041 | * sock.socket->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; |
6038178e | 1042 | * first set it to the P_CONNECTION_FEATURES timeout, |
b411b363 | 1043 | * which we set to 4x the configured ping_timeout. */ |
44ed167d | 1044 | rcu_read_lock(); |
bde89a9e | 1045 | nc = rcu_dereference(connection->net_conf); |
44ed167d | 1046 | |
7da35862 PR |
1047 | sock.socket->sk->sk_sndtimeo = |
1048 | sock.socket->sk->sk_rcvtimeo = nc->ping_timeo*4*HZ/10; | |
b411b363 | 1049 | |
7da35862 | 1050 | msock.socket->sk->sk_rcvtimeo = nc->ping_int*HZ; |
44ed167d | 1051 | timeout = nc->timeout * HZ / 10; |
08b165ba | 1052 | discard_my_data = nc->discard_my_data; |
44ed167d | 1053 | rcu_read_unlock(); |
b411b363 | 1054 | |
7da35862 | 1055 | msock.socket->sk->sk_sndtimeo = timeout; |
b411b363 PR |
1056 | |
1057 | /* we don't want delays. | |
25985edc | 1058 | * we use TCP_CORK where appropriate, though */ |
7da35862 PR |
1059 | drbd_tcp_nodelay(sock.socket); |
1060 | drbd_tcp_nodelay(msock.socket); | |
b411b363 | 1061 | |
bde89a9e AG |
1062 | connection->data.socket = sock.socket; |
1063 | connection->meta.socket = msock.socket; | |
1064 | connection->last_received = jiffies; | |
b411b363 | 1065 | |
bde89a9e | 1066 | h = drbd_do_features(connection); |
b411b363 PR |
1067 | if (h <= 0) |
1068 | return h; | |
1069 | ||
bde89a9e | 1070 | if (connection->cram_hmac_tfm) { |
b30ab791 | 1071 | /* drbd_request_state(device, NS(conn, WFAuth)); */ |
bde89a9e | 1072 | switch (drbd_do_auth(connection)) { |
b10d96cb | 1073 | case -1: |
1ec861eb | 1074 | drbd_err(connection, "Authentication of peer failed\n"); |
b411b363 | 1075 | return -1; |
b10d96cb | 1076 | case 0: |
1ec861eb | 1077 | drbd_err(connection, "Authentication of peer failed, trying again.\n"); |
b10d96cb | 1078 | return 0; |
b411b363 PR |
1079 | } |
1080 | } | |
1081 | ||
bde89a9e AG |
1082 | connection->data.socket->sk->sk_sndtimeo = timeout; |
1083 | connection->data.socket->sk->sk_rcvtimeo = MAX_SCHEDULE_TIMEOUT; | |
b411b363 | 1084 | |
bde89a9e | 1085 | if (drbd_send_protocol(connection) == -EOPNOTSUPP) |
7e2455c1 | 1086 | return -1; |
b411b363 | 1087 | |
31007745 PR |
1088 | /* Prevent a race between resync-handshake and |
1089 | * being promoted to Primary. | |
1090 | * | |
1091 | * Grab and release the state mutex, so we know that any current | |
1092 | * drbd_set_role() is finished, and any incoming drbd_set_role | |
1093 | * will see the STATE_SENT flag, and wait for it to be cleared. | |
1094 | */ | |
1095 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) | |
1096 | mutex_lock(peer_device->device->state_mutex); | |
1097 | ||
bde89a9e | 1098 | set_bit(STATE_SENT, &connection->flags); |
a1096a6e | 1099 | |
31007745 PR |
1100 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) |
1101 | mutex_unlock(peer_device->device->state_mutex); | |
1102 | ||
c141ebda | 1103 | rcu_read_lock(); |
c06ece6b AG |
1104 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { |
1105 | struct drbd_device *device = peer_device->device; | |
b30ab791 | 1106 | kref_get(&device->kref); |
26ea8f92 AG |
1107 | rcu_read_unlock(); |
1108 | ||
08b165ba | 1109 | if (discard_my_data) |
b30ab791 | 1110 | set_bit(DISCARD_MY_DATA, &device->flags); |
08b165ba | 1111 | else |
b30ab791 | 1112 | clear_bit(DISCARD_MY_DATA, &device->flags); |
08b165ba | 1113 | |
69a22773 | 1114 | drbd_connected(peer_device); |
05a10ec7 | 1115 | kref_put(&device->kref, drbd_destroy_device); |
c141ebda PR |
1116 | rcu_read_lock(); |
1117 | } | |
1118 | rcu_read_unlock(); | |
1119 | ||
bde89a9e AG |
1120 | rv = conn_request_state(connection, NS(conn, C_WF_REPORT_PARAMS), CS_VERBOSE); |
1121 | if (rv < SS_SUCCESS || connection->cstate != C_WF_REPORT_PARAMS) { | |
1122 | clear_bit(STATE_SENT, &connection->flags); | |
1e86ac48 | 1123 | return 0; |
a1096a6e | 1124 | } |
1e86ac48 | 1125 | |
1c03e520 | 1126 | drbd_thread_start(&connection->ack_receiver); |
39e91a60 LE |
1127 | /* opencoded create_singlethread_workqueue(), |
1128 | * to be able to use format string arguments */ | |
1129 | connection->ack_sender = | |
1130 | alloc_ordered_workqueue("drbd_as_%s", WQ_MEM_RECLAIM, connection->resource->name); | |
668700b4 PR |
1131 | if (!connection->ack_sender) { |
1132 | drbd_err(connection, "Failed to create workqueue ack_sender\n"); | |
1133 | return 0; | |
1134 | } | |
b411b363 | 1135 | |
0500813f | 1136 | mutex_lock(&connection->resource->conf_update); |
08b165ba PR |
1137 | /* The discard_my_data flag is a single-shot modifier to the next |
1138 | * connection attempt, the handshake of which is now well underway. | |
1139 | * No need for rcu style copying of the whole struct | |
1140 | * just to clear a single value. */ | |
bde89a9e | 1141 | connection->net_conf->discard_my_data = 0; |
0500813f | 1142 | mutex_unlock(&connection->resource->conf_update); |
08b165ba | 1143 | |
d3fcb490 | 1144 | return h; |
b411b363 PR |
1145 | |
1146 | out_release_sockets: | |
7a426fd8 PR |
1147 | if (ad.s_listen) |
1148 | sock_release(ad.s_listen); | |
7da35862 PR |
1149 | if (sock.socket) |
1150 | sock_release(sock.socket); | |
1151 | if (msock.socket) | |
1152 | sock_release(msock.socket); | |
b411b363 PR |
1153 | return -1; |
1154 | } | |
1155 | ||
bde89a9e | 1156 | static int decode_header(struct drbd_connection *connection, void *header, struct packet_info *pi) |
b411b363 | 1157 | { |
bde89a9e | 1158 | unsigned int header_size = drbd_header_size(connection); |
e658983a | 1159 | |
0c8e36d9 AG |
1160 | if (header_size == sizeof(struct p_header100) && |
1161 | *(__be32 *)header == cpu_to_be32(DRBD_MAGIC_100)) { | |
1162 | struct p_header100 *h = header; | |
1163 | if (h->pad != 0) { | |
1ec861eb | 1164 | drbd_err(connection, "Header padding is not zero\n"); |
0c8e36d9 AG |
1165 | return -EINVAL; |
1166 | } | |
1167 | pi->vnr = be16_to_cpu(h->volume); | |
1168 | pi->cmd = be16_to_cpu(h->command); | |
1169 | pi->size = be32_to_cpu(h->length); | |
1170 | } else if (header_size == sizeof(struct p_header95) && | |
1171 | *(__be16 *)header == cpu_to_be16(DRBD_MAGIC_BIG)) { | |
e658983a | 1172 | struct p_header95 *h = header; |
e658983a | 1173 | pi->cmd = be16_to_cpu(h->command); |
b55d84ba AG |
1174 | pi->size = be32_to_cpu(h->length); |
1175 | pi->vnr = 0; | |
e658983a AG |
1176 | } else if (header_size == sizeof(struct p_header80) && |
1177 | *(__be32 *)header == cpu_to_be32(DRBD_MAGIC)) { | |
1178 | struct p_header80 *h = header; | |
1179 | pi->cmd = be16_to_cpu(h->command); | |
1180 | pi->size = be16_to_cpu(h->length); | |
77351055 | 1181 | pi->vnr = 0; |
02918be2 | 1182 | } else { |
1ec861eb | 1183 | drbd_err(connection, "Wrong magic value 0x%08x in protocol version %d\n", |
e658983a | 1184 | be32_to_cpu(*(__be32 *)header), |
bde89a9e | 1185 | connection->agreed_pro_version); |
8172f3e9 | 1186 | return -EINVAL; |
b411b363 | 1187 | } |
e658983a | 1188 | pi->data = header + header_size; |
8172f3e9 | 1189 | return 0; |
257d0af6 | 1190 | } |
b411b363 | 1191 | |
bde89a9e | 1192 | static int drbd_recv_header(struct drbd_connection *connection, struct packet_info *pi) |
257d0af6 | 1193 | { |
bde89a9e | 1194 | void *buffer = connection->data.rbuf; |
69bc7bc3 | 1195 | int err; |
257d0af6 | 1196 | |
bde89a9e | 1197 | err = drbd_recv_all_warn(connection, buffer, drbd_header_size(connection)); |
a5c31904 | 1198 | if (err) |
69bc7bc3 | 1199 | return err; |
257d0af6 | 1200 | |
bde89a9e AG |
1201 | err = decode_header(connection, buffer, pi); |
1202 | connection->last_received = jiffies; | |
b411b363 | 1203 | |
69bc7bc3 | 1204 | return err; |
b411b363 PR |
1205 | } |
1206 | ||
bde89a9e | 1207 | static void drbd_flush(struct drbd_connection *connection) |
b411b363 PR |
1208 | { |
1209 | int rv; | |
c06ece6b | 1210 | struct drbd_peer_device *peer_device; |
4b0007c0 PR |
1211 | int vnr; |
1212 | ||
f6ba8636 | 1213 | if (connection->resource->write_ordering >= WO_BDEV_FLUSH) { |
615e087f | 1214 | rcu_read_lock(); |
c06ece6b AG |
1215 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { |
1216 | struct drbd_device *device = peer_device->device; | |
1217 | ||
b30ab791 | 1218 | if (!get_ldev(device)) |
615e087f | 1219 | continue; |
b30ab791 | 1220 | kref_get(&device->kref); |
615e087f LE |
1221 | rcu_read_unlock(); |
1222 | ||
f418815f LE |
1223 | /* Right now, we have only this one synchronous code path |
1224 | * for flushes between request epochs. | |
1225 | * We may want to make those asynchronous, | |
1226 | * or at least parallelize the flushes to the volume devices. | |
1227 | */ | |
1228 | device->flush_jif = jiffies; | |
1229 | set_bit(FLUSH_PENDING, &device->flags); | |
b30ab791 | 1230 | rv = blkdev_issue_flush(device->ldev->backing_bdev, |
615e087f | 1231 | GFP_NOIO, NULL); |
f418815f | 1232 | clear_bit(FLUSH_PENDING, &device->flags); |
615e087f | 1233 | if (rv) { |
d0180171 | 1234 | drbd_info(device, "local disk flush failed with status %d\n", rv); |
615e087f LE |
1235 | /* would rather check on EOPNOTSUPP, but that is not reliable. |
1236 | * don't try again for ANY return value != 0 | |
1237 | * if (rv == -EOPNOTSUPP) */ | |
f6ba8636 | 1238 | drbd_bump_write_ordering(connection->resource, NULL, WO_DRAIN_IO); |
4b0007c0 | 1239 | } |
b30ab791 | 1240 | put_ldev(device); |
05a10ec7 | 1241 | kref_put(&device->kref, drbd_destroy_device); |
b411b363 | 1242 | |
615e087f LE |
1243 | rcu_read_lock(); |
1244 | if (rv) | |
1245 | break; | |
b411b363 | 1246 | } |
615e087f | 1247 | rcu_read_unlock(); |
b411b363 | 1248 | } |
b411b363 PR |
1249 | } |
1250 | ||
1251 | /** | |
1252 | * drbd_may_finish_epoch() - Applies an epoch_event to the epoch's state, eventually finishes it. | |
b30ab791 | 1253 | * @device: DRBD device. |
b411b363 PR |
1254 | * @epoch: Epoch object. |
1255 | * @ev: Epoch event. | |
1256 | */ | |
bde89a9e | 1257 | static enum finish_epoch drbd_may_finish_epoch(struct drbd_connection *connection, |
b411b363 PR |
1258 | struct drbd_epoch *epoch, |
1259 | enum epoch_event ev) | |
1260 | { | |
2451fc3b | 1261 | int epoch_size; |
b411b363 | 1262 | struct drbd_epoch *next_epoch; |
b411b363 PR |
1263 | enum finish_epoch rv = FE_STILL_LIVE; |
1264 | ||
bde89a9e | 1265 | spin_lock(&connection->epoch_lock); |
b411b363 PR |
1266 | do { |
1267 | next_epoch = NULL; | |
b411b363 PR |
1268 | |
1269 | epoch_size = atomic_read(&epoch->epoch_size); | |
1270 | ||
1271 | switch (ev & ~EV_CLEANUP) { | |
1272 | case EV_PUT: | |
1273 | atomic_dec(&epoch->active); | |
1274 | break; | |
1275 | case EV_GOT_BARRIER_NR: | |
1276 | set_bit(DE_HAVE_BARRIER_NUMBER, &epoch->flags); | |
b411b363 PR |
1277 | break; |
1278 | case EV_BECAME_LAST: | |
1279 | /* nothing to do*/ | |
1280 | break; | |
1281 | } | |
1282 | ||
b411b363 PR |
1283 | if (epoch_size != 0 && |
1284 | atomic_read(&epoch->active) == 0 && | |
80f9fd55 | 1285 | (test_bit(DE_HAVE_BARRIER_NUMBER, &epoch->flags) || ev & EV_CLEANUP)) { |
b411b363 | 1286 | if (!(ev & EV_CLEANUP)) { |
bde89a9e AG |
1287 | spin_unlock(&connection->epoch_lock); |
1288 | drbd_send_b_ack(epoch->connection, epoch->barrier_nr, epoch_size); | |
1289 | spin_lock(&connection->epoch_lock); | |
b411b363 | 1290 | } |
9ed57dcb LE |
1291 | #if 0 |
1292 | /* FIXME: dec unacked on connection, once we have | |
1293 | * something to count pending connection packets in. */ | |
80f9fd55 | 1294 | if (test_bit(DE_HAVE_BARRIER_NUMBER, &epoch->flags)) |
bde89a9e | 1295 | dec_unacked(epoch->connection); |
9ed57dcb | 1296 | #endif |
b411b363 | 1297 | |
bde89a9e | 1298 | if (connection->current_epoch != epoch) { |
b411b363 PR |
1299 | next_epoch = list_entry(epoch->list.next, struct drbd_epoch, list); |
1300 | list_del(&epoch->list); | |
1301 | ev = EV_BECAME_LAST | (ev & EV_CLEANUP); | |
bde89a9e | 1302 | connection->epochs--; |
b411b363 PR |
1303 | kfree(epoch); |
1304 | ||
1305 | if (rv == FE_STILL_LIVE) | |
1306 | rv = FE_DESTROYED; | |
1307 | } else { | |
1308 | epoch->flags = 0; | |
1309 | atomic_set(&epoch->epoch_size, 0); | |
698f9315 | 1310 | /* atomic_set(&epoch->active, 0); is already zero */ |
b411b363 PR |
1311 | if (rv == FE_STILL_LIVE) |
1312 | rv = FE_RECYCLED; | |
1313 | } | |
1314 | } | |
1315 | ||
1316 | if (!next_epoch) | |
1317 | break; | |
1318 | ||
1319 | epoch = next_epoch; | |
1320 | } while (1); | |
1321 | ||
bde89a9e | 1322 | spin_unlock(&connection->epoch_lock); |
b411b363 | 1323 | |
b411b363 PR |
1324 | return rv; |
1325 | } | |
1326 | ||
8fe39aac PR |
1327 | static enum write_ordering_e |
1328 | max_allowed_wo(struct drbd_backing_dev *bdev, enum write_ordering_e wo) | |
1329 | { | |
1330 | struct disk_conf *dc; | |
1331 | ||
1332 | dc = rcu_dereference(bdev->disk_conf); | |
1333 | ||
f6ba8636 AG |
1334 | if (wo == WO_BDEV_FLUSH && !dc->disk_flushes) |
1335 | wo = WO_DRAIN_IO; | |
1336 | if (wo == WO_DRAIN_IO && !dc->disk_drain) | |
1337 | wo = WO_NONE; | |
8fe39aac PR |
1338 | |
1339 | return wo; | |
1340 | } | |
1341 | ||
b411b363 PR |
1342 | /** |
1343 | * drbd_bump_write_ordering() - Fall back to an other write ordering method | |
bde89a9e | 1344 | * @connection: DRBD connection. |
b411b363 PR |
1345 | * @wo: Write ordering method to try. |
1346 | */ | |
8fe39aac PR |
1347 | void drbd_bump_write_ordering(struct drbd_resource *resource, struct drbd_backing_dev *bdev, |
1348 | enum write_ordering_e wo) | |
b411b363 | 1349 | { |
e9526580 | 1350 | struct drbd_device *device; |
b411b363 | 1351 | enum write_ordering_e pwo; |
4b0007c0 | 1352 | int vnr; |
b411b363 | 1353 | static char *write_ordering_str[] = { |
f6ba8636 AG |
1354 | [WO_NONE] = "none", |
1355 | [WO_DRAIN_IO] = "drain", | |
1356 | [WO_BDEV_FLUSH] = "flush", | |
b411b363 PR |
1357 | }; |
1358 | ||
e9526580 | 1359 | pwo = resource->write_ordering; |
f6ba8636 | 1360 | if (wo != WO_BDEV_FLUSH) |
70df7092 | 1361 | wo = min(pwo, wo); |
daeda1cc | 1362 | rcu_read_lock(); |
e9526580 | 1363 | idr_for_each_entry(&resource->devices, device, vnr) { |
8fe39aac PR |
1364 | if (get_ldev(device)) { |
1365 | wo = max_allowed_wo(device->ldev, wo); | |
1366 | if (device->ldev == bdev) | |
1367 | bdev = NULL; | |
1368 | put_ldev(device); | |
1369 | } | |
4b0007c0 | 1370 | } |
8fe39aac PR |
1371 | |
1372 | if (bdev) | |
1373 | wo = max_allowed_wo(bdev, wo); | |
1374 | ||
70df7092 LE |
1375 | rcu_read_unlock(); |
1376 | ||
e9526580 | 1377 | resource->write_ordering = wo; |
f6ba8636 | 1378 | if (pwo != resource->write_ordering || wo == WO_BDEV_FLUSH) |
e9526580 | 1379 | drbd_info(resource, "Method to ensure write ordering: %s\n", write_ordering_str[resource->write_ordering]); |
b411b363 PR |
1380 | } |
1381 | ||
45bb912b | 1382 | /** |
fbe29dec | 1383 | * drbd_submit_peer_request() |
b30ab791 | 1384 | * @device: DRBD device. |
db830c46 | 1385 | * @peer_req: peer request |
45bb912b | 1386 | * @rw: flag field, see bio->bi_rw |
10f6d992 LE |
1387 | * |
1388 | * May spread the pages to multiple bios, | |
1389 | * depending on bio_add_page restrictions. | |
1390 | * | |
1391 | * Returns 0 if all bios have been submitted, | |
1392 | * -ENOMEM if we could not allocate enough bios, | |
1393 | * -ENOSPC (any better suggestion?) if we have not been able to bio_add_page a | |
1394 | * single page to an empty bio (which should never happen and likely indicates | |
1395 | * that the lower level IO stack is in some way broken). This has been observed | |
1396 | * on certain Xen deployments. | |
45bb912b LE |
1397 | */ |
1398 | /* TODO allocate from our own bio_set. */ | |
b30ab791 | 1399 | int drbd_submit_peer_request(struct drbd_device *device, |
fbe29dec AG |
1400 | struct drbd_peer_request *peer_req, |
1401 | const unsigned rw, const int fault_type) | |
45bb912b LE |
1402 | { |
1403 | struct bio *bios = NULL; | |
1404 | struct bio *bio; | |
db830c46 AG |
1405 | struct page *page = peer_req->pages; |
1406 | sector_t sector = peer_req->i.sector; | |
11f8b2b6 | 1407 | unsigned data_size = peer_req->i.size; |
45bb912b | 1408 | unsigned n_bios = 0; |
11f8b2b6 | 1409 | unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT; |
10f6d992 | 1410 | int err = -ENOMEM; |
45bb912b | 1411 | |
a0fb3c47 LE |
1412 | if (peer_req->flags & EE_IS_TRIM_USE_ZEROOUT) { |
1413 | /* wait for all pending IO completions, before we start | |
1414 | * zeroing things out. */ | |
5dd2ca19 | 1415 | conn_wait_active_ee_empty(peer_req->peer_device->connection); |
45d2933c LE |
1416 | /* add it to the active list now, |
1417 | * so we can find it to present it in debugfs */ | |
21ae5d7f LE |
1418 | peer_req->submit_jif = jiffies; |
1419 | peer_req->flags |= EE_SUBMITTED; | |
45d2933c LE |
1420 | spin_lock_irq(&device->resource->req_lock); |
1421 | list_add_tail(&peer_req->w.list, &device->active_ee); | |
1422 | spin_unlock_irq(&device->resource->req_lock); | |
a0fb3c47 | 1423 | if (blkdev_issue_zeroout(device->ldev->backing_bdev, |
d93ba7a5 | 1424 | sector, data_size >> 9, GFP_NOIO, false)) |
a0fb3c47 LE |
1425 | peer_req->flags |= EE_WAS_ERROR; |
1426 | drbd_endio_write_sec_final(peer_req); | |
1427 | return 0; | |
1428 | } | |
1429 | ||
54ed4ed8 LE |
1430 | /* Discards don't have any payload. |
1431 | * But the scsi layer still expects a bio_vec it can use internally, | |
1432 | * see sd_setup_discard_cmnd() and blk_add_request_payload(). */ | |
a0fb3c47 | 1433 | if (peer_req->flags & EE_IS_TRIM) |
54ed4ed8 | 1434 | nr_pages = 1; |
a0fb3c47 | 1435 | |
45bb912b LE |
1436 | /* In most cases, we will only need one bio. But in case the lower |
1437 | * level restrictions happen to be different at this offset on this | |
1438 | * side than those of the sending peer, we may need to submit the | |
9476f39d LE |
1439 | * request in more than one bio. |
1440 | * | |
1441 | * Plain bio_alloc is good enough here, this is no DRBD internally | |
1442 | * generated bio, but a bio allocated on behalf of the peer. | |
1443 | */ | |
45bb912b LE |
1444 | next_bio: |
1445 | bio = bio_alloc(GFP_NOIO, nr_pages); | |
1446 | if (!bio) { | |
a0fb3c47 | 1447 | drbd_err(device, "submit_ee: Allocation of a bio failed (nr_pages=%u)\n", nr_pages); |
45bb912b LE |
1448 | goto fail; |
1449 | } | |
db830c46 | 1450 | /* > peer_req->i.sector, unless this is the first bio */ |
4f024f37 | 1451 | bio->bi_iter.bi_sector = sector; |
b30ab791 | 1452 | bio->bi_bdev = device->ldev->backing_bdev; |
45bb912b | 1453 | bio->bi_rw = rw; |
db830c46 | 1454 | bio->bi_private = peer_req; |
fcefa62e | 1455 | bio->bi_end_io = drbd_peer_request_endio; |
45bb912b LE |
1456 | |
1457 | bio->bi_next = bios; | |
1458 | bios = bio; | |
1459 | ++n_bios; | |
1460 | ||
a0fb3c47 | 1461 | if (rw & REQ_DISCARD) { |
11f8b2b6 | 1462 | bio->bi_iter.bi_size = data_size; |
a0fb3c47 LE |
1463 | goto submit; |
1464 | } | |
1465 | ||
45bb912b | 1466 | page_chain_for_each(page) { |
11f8b2b6 | 1467 | unsigned len = min_t(unsigned, data_size, PAGE_SIZE); |
45bb912b | 1468 | if (!bio_add_page(bio, page, len, 0)) { |
10f6d992 LE |
1469 | /* A single page must always be possible! |
1470 | * But in case it fails anyways, | |
1471 | * we deal with it, and complain (below). */ | |
1472 | if (bio->bi_vcnt == 0) { | |
d0180171 | 1473 | drbd_err(device, |
10f6d992 LE |
1474 | "bio_add_page failed for len=%u, " |
1475 | "bi_vcnt=0 (bi_sector=%llu)\n", | |
4f024f37 | 1476 | len, (uint64_t)bio->bi_iter.bi_sector); |
10f6d992 LE |
1477 | err = -ENOSPC; |
1478 | goto fail; | |
1479 | } | |
45bb912b LE |
1480 | goto next_bio; |
1481 | } | |
11f8b2b6 | 1482 | data_size -= len; |
45bb912b LE |
1483 | sector += len >> 9; |
1484 | --nr_pages; | |
1485 | } | |
11f8b2b6 | 1486 | D_ASSERT(device, data_size == 0); |
a0fb3c47 LE |
1487 | submit: |
1488 | D_ASSERT(device, page == NULL); | |
45bb912b | 1489 | |
db830c46 | 1490 | atomic_set(&peer_req->pending_bios, n_bios); |
21ae5d7f LE |
1491 | /* for debugfs: update timestamp, mark as submitted */ |
1492 | peer_req->submit_jif = jiffies; | |
1493 | peer_req->flags |= EE_SUBMITTED; | |
45bb912b LE |
1494 | do { |
1495 | bio = bios; | |
1496 | bios = bios->bi_next; | |
1497 | bio->bi_next = NULL; | |
1498 | ||
b30ab791 | 1499 | drbd_generic_make_request(device, fault_type, bio); |
45bb912b | 1500 | } while (bios); |
45bb912b LE |
1501 | return 0; |
1502 | ||
1503 | fail: | |
1504 | while (bios) { | |
1505 | bio = bios; | |
1506 | bios = bios->bi_next; | |
1507 | bio_put(bio); | |
1508 | } | |
10f6d992 | 1509 | return err; |
45bb912b LE |
1510 | } |
1511 | ||
b30ab791 | 1512 | static void drbd_remove_epoch_entry_interval(struct drbd_device *device, |
db830c46 | 1513 | struct drbd_peer_request *peer_req) |
53840641 | 1514 | { |
db830c46 | 1515 | struct drbd_interval *i = &peer_req->i; |
53840641 | 1516 | |
b30ab791 | 1517 | drbd_remove_interval(&device->write_requests, i); |
53840641 AG |
1518 | drbd_clear_interval(i); |
1519 | ||
6c852bec | 1520 | /* Wake up any processes waiting for this peer request to complete. */ |
53840641 | 1521 | if (i->waiting) |
b30ab791 | 1522 | wake_up(&device->misc_wait); |
53840641 AG |
1523 | } |
1524 | ||
bde89a9e | 1525 | static void conn_wait_active_ee_empty(struct drbd_connection *connection) |
77fede51 | 1526 | { |
c06ece6b | 1527 | struct drbd_peer_device *peer_device; |
77fede51 PR |
1528 | int vnr; |
1529 | ||
1530 | rcu_read_lock(); | |
c06ece6b AG |
1531 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { |
1532 | struct drbd_device *device = peer_device->device; | |
1533 | ||
b30ab791 | 1534 | kref_get(&device->kref); |
77fede51 | 1535 | rcu_read_unlock(); |
b30ab791 | 1536 | drbd_wait_ee_list_empty(device, &device->active_ee); |
05a10ec7 | 1537 | kref_put(&device->kref, drbd_destroy_device); |
77fede51 PR |
1538 | rcu_read_lock(); |
1539 | } | |
1540 | rcu_read_unlock(); | |
1541 | } | |
1542 | ||
bde89a9e | 1543 | static int receive_Barrier(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 1544 | { |
2451fc3b | 1545 | int rv; |
e658983a | 1546 | struct p_barrier *p = pi->data; |
b411b363 PR |
1547 | struct drbd_epoch *epoch; |
1548 | ||
9ed57dcb LE |
1549 | /* FIXME these are unacked on connection, |
1550 | * not a specific (peer)device. | |
1551 | */ | |
bde89a9e AG |
1552 | connection->current_epoch->barrier_nr = p->barrier; |
1553 | connection->current_epoch->connection = connection; | |
1554 | rv = drbd_may_finish_epoch(connection, connection->current_epoch, EV_GOT_BARRIER_NR); | |
b411b363 PR |
1555 | |
1556 | /* P_BARRIER_ACK may imply that the corresponding extent is dropped from | |
1557 | * the activity log, which means it would not be resynced in case the | |
1558 | * R_PRIMARY crashes now. | |
1559 | * Therefore we must send the barrier_ack after the barrier request was | |
1560 | * completed. */ | |
e9526580 | 1561 | switch (connection->resource->write_ordering) { |
f6ba8636 | 1562 | case WO_NONE: |
b411b363 | 1563 | if (rv == FE_RECYCLED) |
82bc0194 | 1564 | return 0; |
2451fc3b PR |
1565 | |
1566 | /* receiver context, in the writeout path of the other node. | |
1567 | * avoid potential distributed deadlock */ | |
1568 | epoch = kmalloc(sizeof(struct drbd_epoch), GFP_NOIO); | |
1569 | if (epoch) | |
1570 | break; | |
1571 | else | |
1ec861eb | 1572 | drbd_warn(connection, "Allocation of an epoch failed, slowing down\n"); |
2451fc3b | 1573 | /* Fall through */ |
b411b363 | 1574 | |
f6ba8636 AG |
1575 | case WO_BDEV_FLUSH: |
1576 | case WO_DRAIN_IO: | |
bde89a9e AG |
1577 | conn_wait_active_ee_empty(connection); |
1578 | drbd_flush(connection); | |
2451fc3b | 1579 | |
bde89a9e | 1580 | if (atomic_read(&connection->current_epoch->epoch_size)) { |
2451fc3b PR |
1581 | epoch = kmalloc(sizeof(struct drbd_epoch), GFP_NOIO); |
1582 | if (epoch) | |
1583 | break; | |
b411b363 PR |
1584 | } |
1585 | ||
82bc0194 | 1586 | return 0; |
2451fc3b | 1587 | default: |
e9526580 PR |
1588 | drbd_err(connection, "Strangeness in connection->write_ordering %d\n", |
1589 | connection->resource->write_ordering); | |
82bc0194 | 1590 | return -EIO; |
b411b363 PR |
1591 | } |
1592 | ||
1593 | epoch->flags = 0; | |
1594 | atomic_set(&epoch->epoch_size, 0); | |
1595 | atomic_set(&epoch->active, 0); | |
1596 | ||
bde89a9e AG |
1597 | spin_lock(&connection->epoch_lock); |
1598 | if (atomic_read(&connection->current_epoch->epoch_size)) { | |
1599 | list_add(&epoch->list, &connection->current_epoch->list); | |
1600 | connection->current_epoch = epoch; | |
1601 | connection->epochs++; | |
b411b363 PR |
1602 | } else { |
1603 | /* The current_epoch got recycled while we allocated this one... */ | |
1604 | kfree(epoch); | |
1605 | } | |
bde89a9e | 1606 | spin_unlock(&connection->epoch_lock); |
b411b363 | 1607 | |
82bc0194 | 1608 | return 0; |
b411b363 PR |
1609 | } |
1610 | ||
1611 | /* used from receive_RSDataReply (recv_resync_read) | |
1612 | * and from receive_Data */ | |
f6ffca9f | 1613 | static struct drbd_peer_request * |
69a22773 | 1614 | read_in_block(struct drbd_peer_device *peer_device, u64 id, sector_t sector, |
a0fb3c47 | 1615 | struct packet_info *pi) __must_hold(local) |
b411b363 | 1616 | { |
69a22773 | 1617 | struct drbd_device *device = peer_device->device; |
b30ab791 | 1618 | const sector_t capacity = drbd_get_capacity(device->this_bdev); |
db830c46 | 1619 | struct drbd_peer_request *peer_req; |
b411b363 | 1620 | struct page *page; |
11f8b2b6 AG |
1621 | int digest_size, err; |
1622 | unsigned int data_size = pi->size, ds; | |
69a22773 AG |
1623 | void *dig_in = peer_device->connection->int_dig_in; |
1624 | void *dig_vv = peer_device->connection->int_dig_vv; | |
6b4388ac | 1625 | unsigned long *data; |
a0fb3c47 | 1626 | struct p_trim *trim = (pi->cmd == P_TRIM) ? pi->data : NULL; |
b411b363 | 1627 | |
11f8b2b6 | 1628 | digest_size = 0; |
a0fb3c47 | 1629 | if (!trim && peer_device->connection->peer_integrity_tfm) { |
11f8b2b6 | 1630 | digest_size = crypto_hash_digestsize(peer_device->connection->peer_integrity_tfm); |
9f5bdc33 AG |
1631 | /* |
1632 | * FIXME: Receive the incoming digest into the receive buffer | |
1633 | * here, together with its struct p_data? | |
1634 | */ | |
11f8b2b6 | 1635 | err = drbd_recv_all_warn(peer_device->connection, dig_in, digest_size); |
a5c31904 | 1636 | if (err) |
b411b363 | 1637 | return NULL; |
11f8b2b6 | 1638 | data_size -= digest_size; |
b411b363 PR |
1639 | } |
1640 | ||
a0fb3c47 LE |
1641 | if (trim) { |
1642 | D_ASSERT(peer_device, data_size == 0); | |
1643 | data_size = be32_to_cpu(trim->size); | |
1644 | } | |
1645 | ||
841ce241 AG |
1646 | if (!expect(IS_ALIGNED(data_size, 512))) |
1647 | return NULL; | |
a0fb3c47 LE |
1648 | /* prepare for larger trim requests. */ |
1649 | if (!trim && !expect(data_size <= DRBD_MAX_BIO_SIZE)) | |
841ce241 | 1650 | return NULL; |
b411b363 | 1651 | |
6666032a LE |
1652 | /* even though we trust out peer, |
1653 | * we sometimes have to double check. */ | |
1654 | if (sector + (data_size>>9) > capacity) { | |
d0180171 | 1655 | drbd_err(device, "request from peer beyond end of local disk: " |
fdda6544 | 1656 | "capacity: %llus < sector: %llus + size: %u\n", |
6666032a LE |
1657 | (unsigned long long)capacity, |
1658 | (unsigned long long)sector, data_size); | |
1659 | return NULL; | |
1660 | } | |
1661 | ||
b411b363 PR |
1662 | /* GFP_NOIO, because we must not cause arbitrary write-out: in a DRBD |
1663 | * "criss-cross" setup, that might cause write-out on some other DRBD, | |
1664 | * which in turn might block on the other node at this very place. */ | |
a0fb3c47 | 1665 | peer_req = drbd_alloc_peer_req(peer_device, id, sector, data_size, trim == NULL, GFP_NOIO); |
db830c46 | 1666 | if (!peer_req) |
b411b363 | 1667 | return NULL; |
45bb912b | 1668 | |
21ae5d7f | 1669 | peer_req->flags |= EE_WRITE; |
a0fb3c47 | 1670 | if (trim) |
81a3537a | 1671 | return peer_req; |
a73ff323 | 1672 | |
b411b363 | 1673 | ds = data_size; |
db830c46 | 1674 | page = peer_req->pages; |
45bb912b LE |
1675 | page_chain_for_each(page) { |
1676 | unsigned len = min_t(int, ds, PAGE_SIZE); | |
6b4388ac | 1677 | data = kmap(page); |
69a22773 | 1678 | err = drbd_recv_all_warn(peer_device->connection, data, len); |
b30ab791 | 1679 | if (drbd_insert_fault(device, DRBD_FAULT_RECEIVE)) { |
d0180171 | 1680 | drbd_err(device, "Fault injection: Corrupting data on receive\n"); |
6b4388ac PR |
1681 | data[0] = data[0] ^ (unsigned long)-1; |
1682 | } | |
b411b363 | 1683 | kunmap(page); |
a5c31904 | 1684 | if (err) { |
b30ab791 | 1685 | drbd_free_peer_req(device, peer_req); |
b411b363 PR |
1686 | return NULL; |
1687 | } | |
a5c31904 | 1688 | ds -= len; |
b411b363 PR |
1689 | } |
1690 | ||
11f8b2b6 | 1691 | if (digest_size) { |
69a22773 | 1692 | drbd_csum_ee(peer_device->connection->peer_integrity_tfm, peer_req, dig_vv); |
11f8b2b6 | 1693 | if (memcmp(dig_in, dig_vv, digest_size)) { |
d0180171 | 1694 | drbd_err(device, "Digest integrity check FAILED: %llus +%u\n", |
470be44a | 1695 | (unsigned long long)sector, data_size); |
b30ab791 | 1696 | drbd_free_peer_req(device, peer_req); |
b411b363 PR |
1697 | return NULL; |
1698 | } | |
1699 | } | |
11f8b2b6 | 1700 | device->recv_cnt += data_size >> 9; |
db830c46 | 1701 | return peer_req; |
b411b363 PR |
1702 | } |
1703 | ||
1704 | /* drbd_drain_block() just takes a data block | |
1705 | * out of the socket input buffer, and discards it. | |
1706 | */ | |
69a22773 | 1707 | static int drbd_drain_block(struct drbd_peer_device *peer_device, int data_size) |
b411b363 PR |
1708 | { |
1709 | struct page *page; | |
a5c31904 | 1710 | int err = 0; |
b411b363 PR |
1711 | void *data; |
1712 | ||
c3470cde | 1713 | if (!data_size) |
fc5be839 | 1714 | return 0; |
c3470cde | 1715 | |
69a22773 | 1716 | page = drbd_alloc_pages(peer_device, 1, 1); |
b411b363 PR |
1717 | |
1718 | data = kmap(page); | |
1719 | while (data_size) { | |
fc5be839 AG |
1720 | unsigned int len = min_t(int, data_size, PAGE_SIZE); |
1721 | ||
69a22773 | 1722 | err = drbd_recv_all_warn(peer_device->connection, data, len); |
a5c31904 | 1723 | if (err) |
b411b363 | 1724 | break; |
a5c31904 | 1725 | data_size -= len; |
b411b363 PR |
1726 | } |
1727 | kunmap(page); | |
69a22773 | 1728 | drbd_free_pages(peer_device->device, page, 0); |
fc5be839 | 1729 | return err; |
b411b363 PR |
1730 | } |
1731 | ||
69a22773 | 1732 | static int recv_dless_read(struct drbd_peer_device *peer_device, struct drbd_request *req, |
b411b363 PR |
1733 | sector_t sector, int data_size) |
1734 | { | |
7988613b KO |
1735 | struct bio_vec bvec; |
1736 | struct bvec_iter iter; | |
b411b363 | 1737 | struct bio *bio; |
11f8b2b6 | 1738 | int digest_size, err, expect; |
69a22773 AG |
1739 | void *dig_in = peer_device->connection->int_dig_in; |
1740 | void *dig_vv = peer_device->connection->int_dig_vv; | |
b411b363 | 1741 | |
11f8b2b6 | 1742 | digest_size = 0; |
69a22773 | 1743 | if (peer_device->connection->peer_integrity_tfm) { |
11f8b2b6 AG |
1744 | digest_size = crypto_hash_digestsize(peer_device->connection->peer_integrity_tfm); |
1745 | err = drbd_recv_all_warn(peer_device->connection, dig_in, digest_size); | |
a5c31904 AG |
1746 | if (err) |
1747 | return err; | |
11f8b2b6 | 1748 | data_size -= digest_size; |
b411b363 PR |
1749 | } |
1750 | ||
b411b363 PR |
1751 | /* optimistically update recv_cnt. if receiving fails below, |
1752 | * we disconnect anyways, and counters will be reset. */ | |
69a22773 | 1753 | peer_device->device->recv_cnt += data_size>>9; |
b411b363 PR |
1754 | |
1755 | bio = req->master_bio; | |
69a22773 | 1756 | D_ASSERT(peer_device->device, sector == bio->bi_iter.bi_sector); |
b411b363 | 1757 | |
7988613b KO |
1758 | bio_for_each_segment(bvec, bio, iter) { |
1759 | void *mapped = kmap(bvec.bv_page) + bvec.bv_offset; | |
1760 | expect = min_t(int, data_size, bvec.bv_len); | |
69a22773 | 1761 | err = drbd_recv_all_warn(peer_device->connection, mapped, expect); |
7988613b | 1762 | kunmap(bvec.bv_page); |
a5c31904 AG |
1763 | if (err) |
1764 | return err; | |
1765 | data_size -= expect; | |
b411b363 PR |
1766 | } |
1767 | ||
11f8b2b6 | 1768 | if (digest_size) { |
69a22773 | 1769 | drbd_csum_bio(peer_device->connection->peer_integrity_tfm, bio, dig_vv); |
11f8b2b6 | 1770 | if (memcmp(dig_in, dig_vv, digest_size)) { |
69a22773 | 1771 | drbd_err(peer_device, "Digest integrity check FAILED. Broken NICs?\n"); |
28284cef | 1772 | return -EINVAL; |
b411b363 PR |
1773 | } |
1774 | } | |
1775 | ||
69a22773 | 1776 | D_ASSERT(peer_device->device, data_size == 0); |
28284cef | 1777 | return 0; |
b411b363 PR |
1778 | } |
1779 | ||
a990be46 | 1780 | /* |
668700b4 | 1781 | * e_end_resync_block() is called in ack_sender context via |
a990be46 AG |
1782 | * drbd_finish_peer_reqs(). |
1783 | */ | |
99920dc5 | 1784 | static int e_end_resync_block(struct drbd_work *w, int unused) |
b411b363 | 1785 | { |
8050e6d0 | 1786 | struct drbd_peer_request *peer_req = |
a8cd15ba AG |
1787 | container_of(w, struct drbd_peer_request, w); |
1788 | struct drbd_peer_device *peer_device = peer_req->peer_device; | |
1789 | struct drbd_device *device = peer_device->device; | |
db830c46 | 1790 | sector_t sector = peer_req->i.sector; |
99920dc5 | 1791 | int err; |
b411b363 | 1792 | |
0b0ba1ef | 1793 | D_ASSERT(device, drbd_interval_empty(&peer_req->i)); |
b411b363 | 1794 | |
db830c46 | 1795 | if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { |
b30ab791 | 1796 | drbd_set_in_sync(device, sector, peer_req->i.size); |
a8cd15ba | 1797 | err = drbd_send_ack(peer_device, P_RS_WRITE_ACK, peer_req); |
b411b363 PR |
1798 | } else { |
1799 | /* Record failure to sync */ | |
b30ab791 | 1800 | drbd_rs_failed_io(device, sector, peer_req->i.size); |
b411b363 | 1801 | |
a8cd15ba | 1802 | err = drbd_send_ack(peer_device, P_NEG_ACK, peer_req); |
b411b363 | 1803 | } |
b30ab791 | 1804 | dec_unacked(device); |
b411b363 | 1805 | |
99920dc5 | 1806 | return err; |
b411b363 PR |
1807 | } |
1808 | ||
69a22773 | 1809 | static int recv_resync_read(struct drbd_peer_device *peer_device, sector_t sector, |
a0fb3c47 | 1810 | struct packet_info *pi) __releases(local) |
b411b363 | 1811 | { |
69a22773 | 1812 | struct drbd_device *device = peer_device->device; |
db830c46 | 1813 | struct drbd_peer_request *peer_req; |
b411b363 | 1814 | |
a0fb3c47 | 1815 | peer_req = read_in_block(peer_device, ID_SYNCER, sector, pi); |
db830c46 | 1816 | if (!peer_req) |
45bb912b | 1817 | goto fail; |
b411b363 | 1818 | |
b30ab791 | 1819 | dec_rs_pending(device); |
b411b363 | 1820 | |
b30ab791 | 1821 | inc_unacked(device); |
b411b363 PR |
1822 | /* corresponding dec_unacked() in e_end_resync_block() |
1823 | * respective _drbd_clear_done_ee */ | |
1824 | ||
a8cd15ba | 1825 | peer_req->w.cb = e_end_resync_block; |
21ae5d7f | 1826 | peer_req->submit_jif = jiffies; |
45bb912b | 1827 | |
0500813f | 1828 | spin_lock_irq(&device->resource->req_lock); |
b9ed7080 | 1829 | list_add_tail(&peer_req->w.list, &device->sync_ee); |
0500813f | 1830 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 | 1831 | |
a0fb3c47 | 1832 | atomic_add(pi->size >> 9, &device->rs_sect_ev); |
b30ab791 | 1833 | if (drbd_submit_peer_request(device, peer_req, WRITE, DRBD_FAULT_RS_WR) == 0) |
e1c1b0fc | 1834 | return 0; |
b411b363 | 1835 | |
10f6d992 | 1836 | /* don't care for the reason here */ |
d0180171 | 1837 | drbd_err(device, "submit failed, triggering re-connect\n"); |
0500813f | 1838 | spin_lock_irq(&device->resource->req_lock); |
a8cd15ba | 1839 | list_del(&peer_req->w.list); |
0500813f | 1840 | spin_unlock_irq(&device->resource->req_lock); |
22cc37a9 | 1841 | |
b30ab791 | 1842 | drbd_free_peer_req(device, peer_req); |
45bb912b | 1843 | fail: |
b30ab791 | 1844 | put_ldev(device); |
e1c1b0fc | 1845 | return -EIO; |
b411b363 PR |
1846 | } |
1847 | ||
668eebc6 | 1848 | static struct drbd_request * |
b30ab791 | 1849 | find_request(struct drbd_device *device, struct rb_root *root, u64 id, |
bc9c5c41 | 1850 | sector_t sector, bool missing_ok, const char *func) |
51624585 | 1851 | { |
51624585 AG |
1852 | struct drbd_request *req; |
1853 | ||
bc9c5c41 AG |
1854 | /* Request object according to our peer */ |
1855 | req = (struct drbd_request *)(unsigned long)id; | |
5e472264 | 1856 | if (drbd_contains_interval(root, sector, &req->i) && req->i.local) |
668eebc6 | 1857 | return req; |
c3afd8f5 | 1858 | if (!missing_ok) { |
d0180171 | 1859 | drbd_err(device, "%s: failed to find request 0x%lx, sector %llus\n", func, |
c3afd8f5 AG |
1860 | (unsigned long)id, (unsigned long long)sector); |
1861 | } | |
51624585 | 1862 | return NULL; |
b411b363 PR |
1863 | } |
1864 | ||
bde89a9e | 1865 | static int receive_DataReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 1866 | { |
9f4fe9ad | 1867 | struct drbd_peer_device *peer_device; |
b30ab791 | 1868 | struct drbd_device *device; |
b411b363 PR |
1869 | struct drbd_request *req; |
1870 | sector_t sector; | |
82bc0194 | 1871 | int err; |
e658983a | 1872 | struct p_data *p = pi->data; |
4a76b161 | 1873 | |
9f4fe9ad AG |
1874 | peer_device = conn_peer_device(connection, pi->vnr); |
1875 | if (!peer_device) | |
4a76b161 | 1876 | return -EIO; |
9f4fe9ad | 1877 | device = peer_device->device; |
b411b363 PR |
1878 | |
1879 | sector = be64_to_cpu(p->sector); | |
1880 | ||
0500813f | 1881 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 1882 | req = find_request(device, &device->read_requests, p->block_id, sector, false, __func__); |
0500813f | 1883 | spin_unlock_irq(&device->resource->req_lock); |
c3afd8f5 | 1884 | if (unlikely(!req)) |
82bc0194 | 1885 | return -EIO; |
b411b363 | 1886 | |
24c4830c | 1887 | /* hlist_del(&req->collision) is done in _req_may_be_done, to avoid |
b411b363 PR |
1888 | * special casing it there for the various failure cases. |
1889 | * still no race with drbd_fail_pending_reads */ | |
69a22773 | 1890 | err = recv_dless_read(peer_device, req, sector, pi->size); |
82bc0194 | 1891 | if (!err) |
8554df1c | 1892 | req_mod(req, DATA_RECEIVED); |
b411b363 PR |
1893 | /* else: nothing. handled from drbd_disconnect... |
1894 | * I don't think we may complete this just yet | |
1895 | * in case we are "on-disconnect: freeze" */ | |
1896 | ||
82bc0194 | 1897 | return err; |
b411b363 PR |
1898 | } |
1899 | ||
bde89a9e | 1900 | static int receive_RSDataReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 1901 | { |
9f4fe9ad | 1902 | struct drbd_peer_device *peer_device; |
b30ab791 | 1903 | struct drbd_device *device; |
b411b363 | 1904 | sector_t sector; |
82bc0194 | 1905 | int err; |
e658983a | 1906 | struct p_data *p = pi->data; |
4a76b161 | 1907 | |
9f4fe9ad AG |
1908 | peer_device = conn_peer_device(connection, pi->vnr); |
1909 | if (!peer_device) | |
4a76b161 | 1910 | return -EIO; |
9f4fe9ad | 1911 | device = peer_device->device; |
b411b363 PR |
1912 | |
1913 | sector = be64_to_cpu(p->sector); | |
0b0ba1ef | 1914 | D_ASSERT(device, p->block_id == ID_SYNCER); |
b411b363 | 1915 | |
b30ab791 | 1916 | if (get_ldev(device)) { |
b411b363 PR |
1917 | /* data is submitted to disk within recv_resync_read. |
1918 | * corresponding put_ldev done below on error, | |
fcefa62e | 1919 | * or in drbd_peer_request_endio. */ |
a0fb3c47 | 1920 | err = recv_resync_read(peer_device, sector, pi); |
b411b363 PR |
1921 | } else { |
1922 | if (__ratelimit(&drbd_ratelimit_state)) | |
d0180171 | 1923 | drbd_err(device, "Can not write resync data to local disk.\n"); |
b411b363 | 1924 | |
69a22773 | 1925 | err = drbd_drain_block(peer_device, pi->size); |
b411b363 | 1926 | |
69a22773 | 1927 | drbd_send_ack_dp(peer_device, P_NEG_ACK, p, pi->size); |
b411b363 PR |
1928 | } |
1929 | ||
b30ab791 | 1930 | atomic_add(pi->size >> 9, &device->rs_sect_in); |
778f271d | 1931 | |
82bc0194 | 1932 | return err; |
b411b363 PR |
1933 | } |
1934 | ||
b30ab791 | 1935 | static void restart_conflicting_writes(struct drbd_device *device, |
7be8da07 | 1936 | sector_t sector, int size) |
b411b363 | 1937 | { |
7be8da07 AG |
1938 | struct drbd_interval *i; |
1939 | struct drbd_request *req; | |
1940 | ||
b30ab791 | 1941 | drbd_for_each_overlap(i, &device->write_requests, sector, size) { |
7be8da07 AG |
1942 | if (!i->local) |
1943 | continue; | |
1944 | req = container_of(i, struct drbd_request, i); | |
1945 | if (req->rq_state & RQ_LOCAL_PENDING || | |
1946 | !(req->rq_state & RQ_POSTPONED)) | |
1947 | continue; | |
2312f0b3 LE |
1948 | /* as it is RQ_POSTPONED, this will cause it to |
1949 | * be queued on the retry workqueue. */ | |
d4dabbe2 | 1950 | __req_mod(req, CONFLICT_RESOLVED, NULL); |
7be8da07 AG |
1951 | } |
1952 | } | |
b411b363 | 1953 | |
a990be46 | 1954 | /* |
668700b4 | 1955 | * e_end_block() is called in ack_sender context via drbd_finish_peer_reqs(). |
b411b363 | 1956 | */ |
99920dc5 | 1957 | static int e_end_block(struct drbd_work *w, int cancel) |
b411b363 | 1958 | { |
8050e6d0 | 1959 | struct drbd_peer_request *peer_req = |
a8cd15ba AG |
1960 | container_of(w, struct drbd_peer_request, w); |
1961 | struct drbd_peer_device *peer_device = peer_req->peer_device; | |
1962 | struct drbd_device *device = peer_device->device; | |
db830c46 | 1963 | sector_t sector = peer_req->i.sector; |
99920dc5 | 1964 | int err = 0, pcmd; |
b411b363 | 1965 | |
303d1448 | 1966 | if (peer_req->flags & EE_SEND_WRITE_ACK) { |
db830c46 | 1967 | if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) { |
b30ab791 AG |
1968 | pcmd = (device->state.conn >= C_SYNC_SOURCE && |
1969 | device->state.conn <= C_PAUSED_SYNC_T && | |
db830c46 | 1970 | peer_req->flags & EE_MAY_SET_IN_SYNC) ? |
b411b363 | 1971 | P_RS_WRITE_ACK : P_WRITE_ACK; |
a8cd15ba | 1972 | err = drbd_send_ack(peer_device, pcmd, peer_req); |
b411b363 | 1973 | if (pcmd == P_RS_WRITE_ACK) |
b30ab791 | 1974 | drbd_set_in_sync(device, sector, peer_req->i.size); |
b411b363 | 1975 | } else { |
a8cd15ba | 1976 | err = drbd_send_ack(peer_device, P_NEG_ACK, peer_req); |
b411b363 PR |
1977 | /* we expect it to be marked out of sync anyways... |
1978 | * maybe assert this? */ | |
1979 | } | |
b30ab791 | 1980 | dec_unacked(device); |
b411b363 | 1981 | } |
08d0dabf | 1982 | |
b411b363 PR |
1983 | /* we delete from the conflict detection hash _after_ we sent out the |
1984 | * P_WRITE_ACK / P_NEG_ACK, to get the sequence number right. */ | |
302bdeae | 1985 | if (peer_req->flags & EE_IN_INTERVAL_TREE) { |
0500813f | 1986 | spin_lock_irq(&device->resource->req_lock); |
0b0ba1ef | 1987 | D_ASSERT(device, !drbd_interval_empty(&peer_req->i)); |
b30ab791 | 1988 | drbd_remove_epoch_entry_interval(device, peer_req); |
7be8da07 | 1989 | if (peer_req->flags & EE_RESTART_REQUESTS) |
b30ab791 | 1990 | restart_conflicting_writes(device, sector, peer_req->i.size); |
0500813f | 1991 | spin_unlock_irq(&device->resource->req_lock); |
bb3bfe96 | 1992 | } else |
0b0ba1ef | 1993 | D_ASSERT(device, drbd_interval_empty(&peer_req->i)); |
b411b363 | 1994 | |
5dd2ca19 | 1995 | drbd_may_finish_epoch(peer_device->connection, peer_req->epoch, EV_PUT + (cancel ? EV_CLEANUP : 0)); |
b411b363 | 1996 | |
99920dc5 | 1997 | return err; |
b411b363 PR |
1998 | } |
1999 | ||
a8cd15ba | 2000 | static int e_send_ack(struct drbd_work *w, enum drbd_packet ack) |
b411b363 | 2001 | { |
8050e6d0 | 2002 | struct drbd_peer_request *peer_req = |
a8cd15ba AG |
2003 | container_of(w, struct drbd_peer_request, w); |
2004 | struct drbd_peer_device *peer_device = peer_req->peer_device; | |
99920dc5 | 2005 | int err; |
b411b363 | 2006 | |
a8cd15ba AG |
2007 | err = drbd_send_ack(peer_device, ack, peer_req); |
2008 | dec_unacked(peer_device->device); | |
b411b363 | 2009 | |
99920dc5 | 2010 | return err; |
b411b363 PR |
2011 | } |
2012 | ||
d4dabbe2 | 2013 | static int e_send_superseded(struct drbd_work *w, int unused) |
7be8da07 | 2014 | { |
a8cd15ba | 2015 | return e_send_ack(w, P_SUPERSEDED); |
7be8da07 AG |
2016 | } |
2017 | ||
99920dc5 | 2018 | static int e_send_retry_write(struct drbd_work *w, int unused) |
7be8da07 | 2019 | { |
a8cd15ba AG |
2020 | struct drbd_peer_request *peer_req = |
2021 | container_of(w, struct drbd_peer_request, w); | |
2022 | struct drbd_connection *connection = peer_req->peer_device->connection; | |
7be8da07 | 2023 | |
a8cd15ba | 2024 | return e_send_ack(w, connection->agreed_pro_version >= 100 ? |
d4dabbe2 | 2025 | P_RETRY_WRITE : P_SUPERSEDED); |
7be8da07 | 2026 | } |
b411b363 | 2027 | |
3e394da1 AG |
2028 | static bool seq_greater(u32 a, u32 b) |
2029 | { | |
2030 | /* | |
2031 | * We assume 32-bit wrap-around here. | |
2032 | * For 24-bit wrap-around, we would have to shift: | |
2033 | * a <<= 8; b <<= 8; | |
2034 | */ | |
2035 | return (s32)a - (s32)b > 0; | |
2036 | } | |
b411b363 | 2037 | |
3e394da1 AG |
2038 | static u32 seq_max(u32 a, u32 b) |
2039 | { | |
2040 | return seq_greater(a, b) ? a : b; | |
b411b363 PR |
2041 | } |
2042 | ||
69a22773 | 2043 | static void update_peer_seq(struct drbd_peer_device *peer_device, unsigned int peer_seq) |
3e394da1 | 2044 | { |
69a22773 | 2045 | struct drbd_device *device = peer_device->device; |
3c13b680 | 2046 | unsigned int newest_peer_seq; |
3e394da1 | 2047 | |
69a22773 | 2048 | if (test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags)) { |
b30ab791 AG |
2049 | spin_lock(&device->peer_seq_lock); |
2050 | newest_peer_seq = seq_max(device->peer_seq, peer_seq); | |
2051 | device->peer_seq = newest_peer_seq; | |
2052 | spin_unlock(&device->peer_seq_lock); | |
2053 | /* wake up only if we actually changed device->peer_seq */ | |
3c13b680 | 2054 | if (peer_seq == newest_peer_seq) |
b30ab791 | 2055 | wake_up(&device->seq_wait); |
7be8da07 | 2056 | } |
b411b363 PR |
2057 | } |
2058 | ||
d93f6302 | 2059 | static inline int overlaps(sector_t s1, int l1, sector_t s2, int l2) |
b6a370ba | 2060 | { |
d93f6302 LE |
2061 | return !((s1 + (l1>>9) <= s2) || (s1 >= s2 + (l2>>9))); |
2062 | } | |
b6a370ba | 2063 | |
d93f6302 | 2064 | /* maybe change sync_ee into interval trees as well? */ |
b30ab791 | 2065 | static bool overlapping_resync_write(struct drbd_device *device, struct drbd_peer_request *peer_req) |
d93f6302 LE |
2066 | { |
2067 | struct drbd_peer_request *rs_req; | |
b6a370ba PR |
2068 | bool rv = 0; |
2069 | ||
0500813f | 2070 | spin_lock_irq(&device->resource->req_lock); |
a8cd15ba | 2071 | list_for_each_entry(rs_req, &device->sync_ee, w.list) { |
d93f6302 LE |
2072 | if (overlaps(peer_req->i.sector, peer_req->i.size, |
2073 | rs_req->i.sector, rs_req->i.size)) { | |
b6a370ba PR |
2074 | rv = 1; |
2075 | break; | |
2076 | } | |
2077 | } | |
0500813f | 2078 | spin_unlock_irq(&device->resource->req_lock); |
b6a370ba PR |
2079 | |
2080 | return rv; | |
2081 | } | |
2082 | ||
b411b363 PR |
2083 | /* Called from receive_Data. |
2084 | * Synchronize packets on sock with packets on msock. | |
2085 | * | |
2086 | * This is here so even when a P_DATA packet traveling via sock overtook an Ack | |
2087 | * packet traveling on msock, they are still processed in the order they have | |
2088 | * been sent. | |
2089 | * | |
2090 | * Note: we don't care for Ack packets overtaking P_DATA packets. | |
2091 | * | |
b30ab791 | 2092 | * In case packet_seq is larger than device->peer_seq number, there are |
b411b363 | 2093 | * outstanding packets on the msock. We wait for them to arrive. |
b30ab791 | 2094 | * In case we are the logically next packet, we update device->peer_seq |
b411b363 PR |
2095 | * ourselves. Correctly handles 32bit wrap around. |
2096 | * | |
2097 | * Assume we have a 10 GBit connection, that is about 1<<30 byte per second, | |
2098 | * about 1<<21 sectors per second. So "worst" case, we have 1<<3 == 8 seconds | |
2099 | * for the 24bit wrap (historical atomic_t guarantee on some archs), and we have | |
2100 | * 1<<9 == 512 seconds aka ages for the 32bit wrap around... | |
2101 | * | |
2102 | * returns 0 if we may process the packet, | |
2103 | * -ERESTARTSYS if we were interrupted (by disconnect signal). */ | |
69a22773 | 2104 | static int wait_for_and_update_peer_seq(struct drbd_peer_device *peer_device, const u32 peer_seq) |
b411b363 | 2105 | { |
69a22773 | 2106 | struct drbd_device *device = peer_device->device; |
b411b363 | 2107 | DEFINE_WAIT(wait); |
b411b363 | 2108 | long timeout; |
b874d231 | 2109 | int ret = 0, tp; |
7be8da07 | 2110 | |
69a22773 | 2111 | if (!test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags)) |
7be8da07 AG |
2112 | return 0; |
2113 | ||
b30ab791 | 2114 | spin_lock(&device->peer_seq_lock); |
b411b363 | 2115 | for (;;) { |
b30ab791 AG |
2116 | if (!seq_greater(peer_seq - 1, device->peer_seq)) { |
2117 | device->peer_seq = seq_max(device->peer_seq, peer_seq); | |
b411b363 | 2118 | break; |
7be8da07 | 2119 | } |
b874d231 | 2120 | |
b411b363 PR |
2121 | if (signal_pending(current)) { |
2122 | ret = -ERESTARTSYS; | |
2123 | break; | |
2124 | } | |
b874d231 PR |
2125 | |
2126 | rcu_read_lock(); | |
5dd2ca19 | 2127 | tp = rcu_dereference(peer_device->connection->net_conf)->two_primaries; |
b874d231 PR |
2128 | rcu_read_unlock(); |
2129 | ||
2130 | if (!tp) | |
2131 | break; | |
2132 | ||
2133 | /* Only need to wait if two_primaries is enabled */ | |
b30ab791 AG |
2134 | prepare_to_wait(&device->seq_wait, &wait, TASK_INTERRUPTIBLE); |
2135 | spin_unlock(&device->peer_seq_lock); | |
44ed167d | 2136 | rcu_read_lock(); |
69a22773 | 2137 | timeout = rcu_dereference(peer_device->connection->net_conf)->ping_timeo*HZ/10; |
44ed167d | 2138 | rcu_read_unlock(); |
71b1c1eb | 2139 | timeout = schedule_timeout(timeout); |
b30ab791 | 2140 | spin_lock(&device->peer_seq_lock); |
7be8da07 | 2141 | if (!timeout) { |
b411b363 | 2142 | ret = -ETIMEDOUT; |
d0180171 | 2143 | drbd_err(device, "Timed out waiting for missing ack packets; disconnecting\n"); |
b411b363 PR |
2144 | break; |
2145 | } | |
2146 | } | |
b30ab791 AG |
2147 | spin_unlock(&device->peer_seq_lock); |
2148 | finish_wait(&device->seq_wait, &wait); | |
b411b363 PR |
2149 | return ret; |
2150 | } | |
2151 | ||
688593c5 LE |
2152 | /* see also bio_flags_to_wire() |
2153 | * DRBD_REQ_*, because we need to semantically map the flags to data packet | |
2154 | * flags and back. We may replicate to other kernel versions. */ | |
81f0ffd2 | 2155 | static unsigned long wire_flags_to_bio(u32 dpf) |
76d2e7ec | 2156 | { |
688593c5 LE |
2157 | return (dpf & DP_RW_SYNC ? REQ_SYNC : 0) | |
2158 | (dpf & DP_FUA ? REQ_FUA : 0) | | |
2159 | (dpf & DP_FLUSH ? REQ_FLUSH : 0) | | |
2160 | (dpf & DP_DISCARD ? REQ_DISCARD : 0); | |
76d2e7ec PR |
2161 | } |
2162 | ||
b30ab791 | 2163 | static void fail_postponed_requests(struct drbd_device *device, sector_t sector, |
7be8da07 AG |
2164 | unsigned int size) |
2165 | { | |
2166 | struct drbd_interval *i; | |
2167 | ||
2168 | repeat: | |
b30ab791 | 2169 | drbd_for_each_overlap(i, &device->write_requests, sector, size) { |
7be8da07 AG |
2170 | struct drbd_request *req; |
2171 | struct bio_and_error m; | |
2172 | ||
2173 | if (!i->local) | |
2174 | continue; | |
2175 | req = container_of(i, struct drbd_request, i); | |
2176 | if (!(req->rq_state & RQ_POSTPONED)) | |
2177 | continue; | |
2178 | req->rq_state &= ~RQ_POSTPONED; | |
2179 | __req_mod(req, NEG_ACKED, &m); | |
0500813f | 2180 | spin_unlock_irq(&device->resource->req_lock); |
7be8da07 | 2181 | if (m.bio) |
b30ab791 | 2182 | complete_master_bio(device, &m); |
0500813f | 2183 | spin_lock_irq(&device->resource->req_lock); |
7be8da07 AG |
2184 | goto repeat; |
2185 | } | |
2186 | } | |
2187 | ||
b30ab791 | 2188 | static int handle_write_conflicts(struct drbd_device *device, |
7be8da07 AG |
2189 | struct drbd_peer_request *peer_req) |
2190 | { | |
e33b32de | 2191 | struct drbd_connection *connection = peer_req->peer_device->connection; |
bde89a9e | 2192 | bool resolve_conflicts = test_bit(RESOLVE_CONFLICTS, &connection->flags); |
7be8da07 AG |
2193 | sector_t sector = peer_req->i.sector; |
2194 | const unsigned int size = peer_req->i.size; | |
2195 | struct drbd_interval *i; | |
2196 | bool equal; | |
2197 | int err; | |
2198 | ||
2199 | /* | |
2200 | * Inserting the peer request into the write_requests tree will prevent | |
2201 | * new conflicting local requests from being added. | |
2202 | */ | |
b30ab791 | 2203 | drbd_insert_interval(&device->write_requests, &peer_req->i); |
7be8da07 AG |
2204 | |
2205 | repeat: | |
b30ab791 | 2206 | drbd_for_each_overlap(i, &device->write_requests, sector, size) { |
7be8da07 AG |
2207 | if (i == &peer_req->i) |
2208 | continue; | |
08d0dabf LE |
2209 | if (i->completed) |
2210 | continue; | |
7be8da07 AG |
2211 | |
2212 | if (!i->local) { | |
2213 | /* | |
2214 | * Our peer has sent a conflicting remote request; this | |
2215 | * should not happen in a two-node setup. Wait for the | |
2216 | * earlier peer request to complete. | |
2217 | */ | |
b30ab791 | 2218 | err = drbd_wait_misc(device, i); |
7be8da07 AG |
2219 | if (err) |
2220 | goto out; | |
2221 | goto repeat; | |
2222 | } | |
2223 | ||
2224 | equal = i->sector == sector && i->size == size; | |
2225 | if (resolve_conflicts) { | |
2226 | /* | |
2227 | * If the peer request is fully contained within the | |
d4dabbe2 LE |
2228 | * overlapping request, it can be considered overwritten |
2229 | * and thus superseded; otherwise, it will be retried | |
2230 | * once all overlapping requests have completed. | |
7be8da07 | 2231 | */ |
d4dabbe2 | 2232 | bool superseded = i->sector <= sector && i->sector + |
7be8da07 AG |
2233 | (i->size >> 9) >= sector + (size >> 9); |
2234 | ||
2235 | if (!equal) | |
d0180171 | 2236 | drbd_alert(device, "Concurrent writes detected: " |
7be8da07 AG |
2237 | "local=%llus +%u, remote=%llus +%u, " |
2238 | "assuming %s came first\n", | |
2239 | (unsigned long long)i->sector, i->size, | |
2240 | (unsigned long long)sector, size, | |
d4dabbe2 | 2241 | superseded ? "local" : "remote"); |
7be8da07 | 2242 | |
a8cd15ba | 2243 | peer_req->w.cb = superseded ? e_send_superseded : |
7be8da07 | 2244 | e_send_retry_write; |
a8cd15ba | 2245 | list_add_tail(&peer_req->w.list, &device->done_ee); |
668700b4 | 2246 | queue_work(connection->ack_sender, &peer_req->peer_device->send_acks_work); |
7be8da07 AG |
2247 | |
2248 | err = -ENOENT; | |
2249 | goto out; | |
2250 | } else { | |
2251 | struct drbd_request *req = | |
2252 | container_of(i, struct drbd_request, i); | |
2253 | ||
2254 | if (!equal) | |
d0180171 | 2255 | drbd_alert(device, "Concurrent writes detected: " |
7be8da07 AG |
2256 | "local=%llus +%u, remote=%llus +%u\n", |
2257 | (unsigned long long)i->sector, i->size, | |
2258 | (unsigned long long)sector, size); | |
2259 | ||
2260 | if (req->rq_state & RQ_LOCAL_PENDING || | |
2261 | !(req->rq_state & RQ_POSTPONED)) { | |
2262 | /* | |
2263 | * Wait for the node with the discard flag to | |
d4dabbe2 LE |
2264 | * decide if this request has been superseded |
2265 | * or needs to be retried. | |
2266 | * Requests that have been superseded will | |
7be8da07 AG |
2267 | * disappear from the write_requests tree. |
2268 | * | |
2269 | * In addition, wait for the conflicting | |
2270 | * request to finish locally before submitting | |
2271 | * the conflicting peer request. | |
2272 | */ | |
b30ab791 | 2273 | err = drbd_wait_misc(device, &req->i); |
7be8da07 | 2274 | if (err) { |
e33b32de | 2275 | _conn_request_state(connection, NS(conn, C_TIMEOUT), CS_HARD); |
b30ab791 | 2276 | fail_postponed_requests(device, sector, size); |
7be8da07 AG |
2277 | goto out; |
2278 | } | |
2279 | goto repeat; | |
2280 | } | |
2281 | /* | |
2282 | * Remember to restart the conflicting requests after | |
2283 | * the new peer request has completed. | |
2284 | */ | |
2285 | peer_req->flags |= EE_RESTART_REQUESTS; | |
2286 | } | |
2287 | } | |
2288 | err = 0; | |
2289 | ||
2290 | out: | |
2291 | if (err) | |
b30ab791 | 2292 | drbd_remove_epoch_entry_interval(device, peer_req); |
7be8da07 AG |
2293 | return err; |
2294 | } | |
2295 | ||
b411b363 | 2296 | /* mirrored write */ |
bde89a9e | 2297 | static int receive_Data(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 2298 | { |
9f4fe9ad | 2299 | struct drbd_peer_device *peer_device; |
b30ab791 | 2300 | struct drbd_device *device; |
21ae5d7f | 2301 | struct net_conf *nc; |
b411b363 | 2302 | sector_t sector; |
db830c46 | 2303 | struct drbd_peer_request *peer_req; |
e658983a | 2304 | struct p_data *p = pi->data; |
7be8da07 | 2305 | u32 peer_seq = be32_to_cpu(p->seq_num); |
b411b363 PR |
2306 | int rw = WRITE; |
2307 | u32 dp_flags; | |
302bdeae | 2308 | int err, tp; |
b411b363 | 2309 | |
9f4fe9ad AG |
2310 | peer_device = conn_peer_device(connection, pi->vnr); |
2311 | if (!peer_device) | |
4a76b161 | 2312 | return -EIO; |
9f4fe9ad | 2313 | device = peer_device->device; |
b411b363 | 2314 | |
b30ab791 | 2315 | if (!get_ldev(device)) { |
82bc0194 AG |
2316 | int err2; |
2317 | ||
69a22773 AG |
2318 | err = wait_for_and_update_peer_seq(peer_device, peer_seq); |
2319 | drbd_send_ack_dp(peer_device, P_NEG_ACK, p, pi->size); | |
bde89a9e | 2320 | atomic_inc(&connection->current_epoch->epoch_size); |
69a22773 | 2321 | err2 = drbd_drain_block(peer_device, pi->size); |
82bc0194 AG |
2322 | if (!err) |
2323 | err = err2; | |
2324 | return err; | |
b411b363 PR |
2325 | } |
2326 | ||
fcefa62e AG |
2327 | /* |
2328 | * Corresponding put_ldev done either below (on various errors), or in | |
2329 | * drbd_peer_request_endio, if we successfully submit the data at the | |
2330 | * end of this function. | |
2331 | */ | |
b411b363 PR |
2332 | |
2333 | sector = be64_to_cpu(p->sector); | |
a0fb3c47 | 2334 | peer_req = read_in_block(peer_device, p->block_id, sector, pi); |
db830c46 | 2335 | if (!peer_req) { |
b30ab791 | 2336 | put_ldev(device); |
82bc0194 | 2337 | return -EIO; |
b411b363 PR |
2338 | } |
2339 | ||
a8cd15ba | 2340 | peer_req->w.cb = e_end_block; |
21ae5d7f LE |
2341 | peer_req->submit_jif = jiffies; |
2342 | peer_req->flags |= EE_APPLICATION; | |
b411b363 | 2343 | |
688593c5 | 2344 | dp_flags = be32_to_cpu(p->dp_flags); |
81f0ffd2 | 2345 | rw |= wire_flags_to_bio(dp_flags); |
a0fb3c47 LE |
2346 | if (pi->cmd == P_TRIM) { |
2347 | struct request_queue *q = bdev_get_queue(device->ldev->backing_bdev); | |
2348 | peer_req->flags |= EE_IS_TRIM; | |
2349 | if (!blk_queue_discard(q)) | |
2350 | peer_req->flags |= EE_IS_TRIM_USE_ZEROOUT; | |
2351 | D_ASSERT(peer_device, peer_req->i.size > 0); | |
2352 | D_ASSERT(peer_device, rw & REQ_DISCARD); | |
2353 | D_ASSERT(peer_device, peer_req->pages == NULL); | |
2354 | } else if (peer_req->pages == NULL) { | |
0b0ba1ef AG |
2355 | D_ASSERT(device, peer_req->i.size == 0); |
2356 | D_ASSERT(device, dp_flags & DP_FLUSH); | |
a73ff323 | 2357 | } |
688593c5 LE |
2358 | |
2359 | if (dp_flags & DP_MAY_SET_IN_SYNC) | |
db830c46 | 2360 | peer_req->flags |= EE_MAY_SET_IN_SYNC; |
688593c5 | 2361 | |
bde89a9e AG |
2362 | spin_lock(&connection->epoch_lock); |
2363 | peer_req->epoch = connection->current_epoch; | |
db830c46 AG |
2364 | atomic_inc(&peer_req->epoch->epoch_size); |
2365 | atomic_inc(&peer_req->epoch->active); | |
bde89a9e | 2366 | spin_unlock(&connection->epoch_lock); |
b411b363 | 2367 | |
302bdeae | 2368 | rcu_read_lock(); |
21ae5d7f LE |
2369 | nc = rcu_dereference(peer_device->connection->net_conf); |
2370 | tp = nc->two_primaries; | |
2371 | if (peer_device->connection->agreed_pro_version < 100) { | |
2372 | switch (nc->wire_protocol) { | |
2373 | case DRBD_PROT_C: | |
2374 | dp_flags |= DP_SEND_WRITE_ACK; | |
2375 | break; | |
2376 | case DRBD_PROT_B: | |
2377 | dp_flags |= DP_SEND_RECEIVE_ACK; | |
2378 | break; | |
2379 | } | |
2380 | } | |
302bdeae | 2381 | rcu_read_unlock(); |
21ae5d7f LE |
2382 | |
2383 | if (dp_flags & DP_SEND_WRITE_ACK) { | |
2384 | peer_req->flags |= EE_SEND_WRITE_ACK; | |
2385 | inc_unacked(device); | |
2386 | /* corresponding dec_unacked() in e_end_block() | |
2387 | * respective _drbd_clear_done_ee */ | |
2388 | } | |
2389 | ||
2390 | if (dp_flags & DP_SEND_RECEIVE_ACK) { | |
2391 | /* I really don't like it that the receiver thread | |
2392 | * sends on the msock, but anyways */ | |
5dd2ca19 | 2393 | drbd_send_ack(peer_device, P_RECV_ACK, peer_req); |
21ae5d7f LE |
2394 | } |
2395 | ||
302bdeae | 2396 | if (tp) { |
21ae5d7f LE |
2397 | /* two primaries implies protocol C */ |
2398 | D_ASSERT(device, dp_flags & DP_SEND_WRITE_ACK); | |
302bdeae | 2399 | peer_req->flags |= EE_IN_INTERVAL_TREE; |
69a22773 | 2400 | err = wait_for_and_update_peer_seq(peer_device, peer_seq); |
7be8da07 | 2401 | if (err) |
b411b363 | 2402 | goto out_interrupted; |
0500813f | 2403 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 2404 | err = handle_write_conflicts(device, peer_req); |
7be8da07 | 2405 | if (err) { |
0500813f | 2406 | spin_unlock_irq(&device->resource->req_lock); |
7be8da07 | 2407 | if (err == -ENOENT) { |
b30ab791 | 2408 | put_ldev(device); |
82bc0194 | 2409 | return 0; |
b411b363 | 2410 | } |
7be8da07 | 2411 | goto out_interrupted; |
b411b363 | 2412 | } |
b874d231 | 2413 | } else { |
69a22773 | 2414 | update_peer_seq(peer_device, peer_seq); |
0500813f | 2415 | spin_lock_irq(&device->resource->req_lock); |
b874d231 | 2416 | } |
a0fb3c47 LE |
2417 | /* if we use the zeroout fallback code, we process synchronously |
2418 | * and we wait for all pending requests, respectively wait for | |
2419 | * active_ee to become empty in drbd_submit_peer_request(); | |
2420 | * better not add ourselves here. */ | |
2421 | if ((peer_req->flags & EE_IS_TRIM_USE_ZEROOUT) == 0) | |
b9ed7080 | 2422 | list_add_tail(&peer_req->w.list, &device->active_ee); |
0500813f | 2423 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 | 2424 | |
b30ab791 AG |
2425 | if (device->state.conn == C_SYNC_TARGET) |
2426 | wait_event(device->ee_wait, !overlapping_resync_write(device, peer_req)); | |
b411b363 | 2427 | |
b30ab791 | 2428 | if (device->state.pdsk < D_INCONSISTENT) { |
b411b363 | 2429 | /* In case we have the only disk of the cluster, */ |
b30ab791 | 2430 | drbd_set_out_of_sync(device, peer_req->i.sector, peer_req->i.size); |
db830c46 | 2431 | peer_req->flags &= ~EE_MAY_SET_IN_SYNC; |
4dd726f0 | 2432 | drbd_al_begin_io(device, &peer_req->i); |
21ae5d7f | 2433 | peer_req->flags |= EE_CALL_AL_COMPLETE_IO; |
b411b363 PR |
2434 | } |
2435 | ||
b30ab791 | 2436 | err = drbd_submit_peer_request(device, peer_req, rw, DRBD_FAULT_DT_WR); |
82bc0194 AG |
2437 | if (!err) |
2438 | return 0; | |
b411b363 | 2439 | |
10f6d992 | 2440 | /* don't care for the reason here */ |
d0180171 | 2441 | drbd_err(device, "submit failed, triggering re-connect\n"); |
0500813f | 2442 | spin_lock_irq(&device->resource->req_lock); |
a8cd15ba | 2443 | list_del(&peer_req->w.list); |
b30ab791 | 2444 | drbd_remove_epoch_entry_interval(device, peer_req); |
0500813f | 2445 | spin_unlock_irq(&device->resource->req_lock); |
21ae5d7f LE |
2446 | if (peer_req->flags & EE_CALL_AL_COMPLETE_IO) { |
2447 | peer_req->flags &= ~EE_CALL_AL_COMPLETE_IO; | |
b30ab791 | 2448 | drbd_al_complete_io(device, &peer_req->i); |
21ae5d7f | 2449 | } |
22cc37a9 | 2450 | |
b411b363 | 2451 | out_interrupted: |
bde89a9e | 2452 | drbd_may_finish_epoch(connection, peer_req->epoch, EV_PUT + EV_CLEANUP); |
b30ab791 AG |
2453 | put_ldev(device); |
2454 | drbd_free_peer_req(device, peer_req); | |
82bc0194 | 2455 | return err; |
b411b363 PR |
2456 | } |
2457 | ||
0f0601f4 LE |
2458 | /* We may throttle resync, if the lower device seems to be busy, |
2459 | * and current sync rate is above c_min_rate. | |
2460 | * | |
2461 | * To decide whether or not the lower device is busy, we use a scheme similar | |
2462 | * to MD RAID is_mddev_idle(): if the partition stats reveal "significant" | |
2463 | * (more than 64 sectors) of activity we cannot account for with our own resync | |
2464 | * activity, it obviously is "busy". | |
2465 | * | |
2466 | * The current sync rate used here uses only the most recent two step marks, | |
2467 | * to have a short time average so we can react faster. | |
2468 | */ | |
ad3fee79 LE |
2469 | bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector, |
2470 | bool throttle_if_app_is_waiting) | |
0f0601f4 | 2471 | { |
e3555d85 | 2472 | struct lc_element *tmp; |
ad3fee79 | 2473 | bool throttle = drbd_rs_c_min_rate_throttle(device); |
daeda1cc | 2474 | |
ad3fee79 LE |
2475 | if (!throttle || throttle_if_app_is_waiting) |
2476 | return throttle; | |
0f0601f4 | 2477 | |
b30ab791 AG |
2478 | spin_lock_irq(&device->al_lock); |
2479 | tmp = lc_find(device->resync, BM_SECT_TO_EXT(sector)); | |
e3555d85 PR |
2480 | if (tmp) { |
2481 | struct bm_extent *bm_ext = lc_entry(tmp, struct bm_extent, lce); | |
e8299874 LE |
2482 | if (test_bit(BME_PRIORITY, &bm_ext->flags)) |
2483 | throttle = false; | |
ad3fee79 LE |
2484 | /* Do not slow down if app IO is already waiting for this extent, |
2485 | * and our progress is necessary for application IO to complete. */ | |
e3555d85 | 2486 | } |
b30ab791 | 2487 | spin_unlock_irq(&device->al_lock); |
e3555d85 | 2488 | |
e8299874 LE |
2489 | return throttle; |
2490 | } | |
2491 | ||
2492 | bool drbd_rs_c_min_rate_throttle(struct drbd_device *device) | |
2493 | { | |
2494 | struct gendisk *disk = device->ldev->backing_bdev->bd_contains->bd_disk; | |
2495 | unsigned long db, dt, dbdt; | |
2496 | unsigned int c_min_rate; | |
2497 | int curr_events; | |
2498 | ||
2499 | rcu_read_lock(); | |
2500 | c_min_rate = rcu_dereference(device->ldev->disk_conf)->c_min_rate; | |
2501 | rcu_read_unlock(); | |
2502 | ||
2503 | /* feature disabled? */ | |
2504 | if (c_min_rate == 0) | |
2505 | return false; | |
2506 | ||
0f0601f4 LE |
2507 | curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + |
2508 | (int)part_stat_read(&disk->part0, sectors[1]) - | |
b30ab791 | 2509 | atomic_read(&device->rs_sect_ev); |
ad3fee79 LE |
2510 | |
2511 | if (atomic_read(&device->ap_actlog_cnt) | |
ff8bd88b | 2512 | || curr_events - device->rs_last_events > 64) { |
0f0601f4 LE |
2513 | unsigned long rs_left; |
2514 | int i; | |
2515 | ||
b30ab791 | 2516 | device->rs_last_events = curr_events; |
0f0601f4 LE |
2517 | |
2518 | /* sync speed average over the last 2*DRBD_SYNC_MARK_STEP, | |
2519 | * approx. */ | |
b30ab791 | 2520 | i = (device->rs_last_mark + DRBD_SYNC_MARKS-1) % DRBD_SYNC_MARKS; |
2649f080 | 2521 | |
b30ab791 AG |
2522 | if (device->state.conn == C_VERIFY_S || device->state.conn == C_VERIFY_T) |
2523 | rs_left = device->ov_left; | |
2649f080 | 2524 | else |
b30ab791 | 2525 | rs_left = drbd_bm_total_weight(device) - device->rs_failed; |
0f0601f4 | 2526 | |
b30ab791 | 2527 | dt = ((long)jiffies - (long)device->rs_mark_time[i]) / HZ; |
0f0601f4 LE |
2528 | if (!dt) |
2529 | dt++; | |
b30ab791 | 2530 | db = device->rs_mark_left[i] - rs_left; |
0f0601f4 LE |
2531 | dbdt = Bit2KB(db/dt); |
2532 | ||
daeda1cc | 2533 | if (dbdt > c_min_rate) |
e8299874 | 2534 | return true; |
0f0601f4 | 2535 | } |
e8299874 | 2536 | return false; |
0f0601f4 LE |
2537 | } |
2538 | ||
bde89a9e | 2539 | static int receive_DataRequest(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 2540 | { |
9f4fe9ad | 2541 | struct drbd_peer_device *peer_device; |
b30ab791 | 2542 | struct drbd_device *device; |
b411b363 | 2543 | sector_t sector; |
4a76b161 | 2544 | sector_t capacity; |
db830c46 | 2545 | struct drbd_peer_request *peer_req; |
b411b363 | 2546 | struct digest_info *di = NULL; |
b18b37be | 2547 | int size, verb; |
b411b363 | 2548 | unsigned int fault_type; |
e658983a | 2549 | struct p_block_req *p = pi->data; |
4a76b161 | 2550 | |
9f4fe9ad AG |
2551 | peer_device = conn_peer_device(connection, pi->vnr); |
2552 | if (!peer_device) | |
4a76b161 | 2553 | return -EIO; |
9f4fe9ad | 2554 | device = peer_device->device; |
b30ab791 | 2555 | capacity = drbd_get_capacity(device->this_bdev); |
b411b363 PR |
2556 | |
2557 | sector = be64_to_cpu(p->sector); | |
2558 | size = be32_to_cpu(p->blksize); | |
2559 | ||
c670a398 | 2560 | if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) { |
d0180171 | 2561 | drbd_err(device, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__, |
b411b363 | 2562 | (unsigned long long)sector, size); |
82bc0194 | 2563 | return -EINVAL; |
b411b363 PR |
2564 | } |
2565 | if (sector + (size>>9) > capacity) { | |
d0180171 | 2566 | drbd_err(device, "%s:%d: sector: %llus, size: %u\n", __FILE__, __LINE__, |
b411b363 | 2567 | (unsigned long long)sector, size); |
82bc0194 | 2568 | return -EINVAL; |
b411b363 PR |
2569 | } |
2570 | ||
b30ab791 | 2571 | if (!get_ldev_if_state(device, D_UP_TO_DATE)) { |
b18b37be | 2572 | verb = 1; |
e2857216 | 2573 | switch (pi->cmd) { |
b18b37be | 2574 | case P_DATA_REQUEST: |
69a22773 | 2575 | drbd_send_ack_rp(peer_device, P_NEG_DREPLY, p); |
b18b37be PR |
2576 | break; |
2577 | case P_RS_DATA_REQUEST: | |
2578 | case P_CSUM_RS_REQUEST: | |
2579 | case P_OV_REQUEST: | |
69a22773 | 2580 | drbd_send_ack_rp(peer_device, P_NEG_RS_DREPLY , p); |
b18b37be PR |
2581 | break; |
2582 | case P_OV_REPLY: | |
2583 | verb = 0; | |
b30ab791 | 2584 | dec_rs_pending(device); |
69a22773 | 2585 | drbd_send_ack_ex(peer_device, P_OV_RESULT, sector, size, ID_IN_SYNC); |
b18b37be PR |
2586 | break; |
2587 | default: | |
49ba9b1b | 2588 | BUG(); |
b18b37be PR |
2589 | } |
2590 | if (verb && __ratelimit(&drbd_ratelimit_state)) | |
d0180171 | 2591 | drbd_err(device, "Can not satisfy peer's read request, " |
b411b363 | 2592 | "no local data.\n"); |
b18b37be | 2593 | |
a821cc4a | 2594 | /* drain possibly payload */ |
69a22773 | 2595 | return drbd_drain_block(peer_device, pi->size); |
b411b363 PR |
2596 | } |
2597 | ||
2598 | /* GFP_NOIO, because we must not cause arbitrary write-out: in a DRBD | |
2599 | * "criss-cross" setup, that might cause write-out on some other DRBD, | |
2600 | * which in turn might block on the other node at this very place. */ | |
a0fb3c47 LE |
2601 | peer_req = drbd_alloc_peer_req(peer_device, p->block_id, sector, size, |
2602 | true /* has real payload */, GFP_NOIO); | |
db830c46 | 2603 | if (!peer_req) { |
b30ab791 | 2604 | put_ldev(device); |
82bc0194 | 2605 | return -ENOMEM; |
b411b363 PR |
2606 | } |
2607 | ||
e2857216 | 2608 | switch (pi->cmd) { |
b411b363 | 2609 | case P_DATA_REQUEST: |
a8cd15ba | 2610 | peer_req->w.cb = w_e_end_data_req; |
b411b363 | 2611 | fault_type = DRBD_FAULT_DT_RD; |
80a40e43 | 2612 | /* application IO, don't drbd_rs_begin_io */ |
21ae5d7f | 2613 | peer_req->flags |= EE_APPLICATION; |
80a40e43 LE |
2614 | goto submit; |
2615 | ||
b411b363 | 2616 | case P_RS_DATA_REQUEST: |
a8cd15ba | 2617 | peer_req->w.cb = w_e_end_rsdata_req; |
b411b363 | 2618 | fault_type = DRBD_FAULT_RS_RD; |
5f9915bb | 2619 | /* used in the sector offset progress display */ |
b30ab791 | 2620 | device->bm_resync_fo = BM_SECT_TO_BIT(sector); |
b411b363 PR |
2621 | break; |
2622 | ||
2623 | case P_OV_REPLY: | |
2624 | case P_CSUM_RS_REQUEST: | |
2625 | fault_type = DRBD_FAULT_RS_RD; | |
e2857216 | 2626 | di = kmalloc(sizeof(*di) + pi->size, GFP_NOIO); |
b411b363 PR |
2627 | if (!di) |
2628 | goto out_free_e; | |
2629 | ||
e2857216 | 2630 | di->digest_size = pi->size; |
b411b363 PR |
2631 | di->digest = (((char *)di)+sizeof(struct digest_info)); |
2632 | ||
db830c46 AG |
2633 | peer_req->digest = di; |
2634 | peer_req->flags |= EE_HAS_DIGEST; | |
c36c3ced | 2635 | |
9f4fe9ad | 2636 | if (drbd_recv_all(peer_device->connection, di->digest, pi->size)) |
b411b363 PR |
2637 | goto out_free_e; |
2638 | ||
e2857216 | 2639 | if (pi->cmd == P_CSUM_RS_REQUEST) { |
9f4fe9ad | 2640 | D_ASSERT(device, peer_device->connection->agreed_pro_version >= 89); |
a8cd15ba | 2641 | peer_req->w.cb = w_e_end_csum_rs_req; |
5f9915bb | 2642 | /* used in the sector offset progress display */ |
b30ab791 | 2643 | device->bm_resync_fo = BM_SECT_TO_BIT(sector); |
aaaba345 LE |
2644 | /* remember to report stats in drbd_resync_finished */ |
2645 | device->use_csums = true; | |
e2857216 | 2646 | } else if (pi->cmd == P_OV_REPLY) { |
2649f080 | 2647 | /* track progress, we may need to throttle */ |
b30ab791 | 2648 | atomic_add(size >> 9, &device->rs_sect_in); |
a8cd15ba | 2649 | peer_req->w.cb = w_e_end_ov_reply; |
b30ab791 | 2650 | dec_rs_pending(device); |
0f0601f4 LE |
2651 | /* drbd_rs_begin_io done when we sent this request, |
2652 | * but accounting still needs to be done. */ | |
2653 | goto submit_for_resync; | |
b411b363 PR |
2654 | } |
2655 | break; | |
2656 | ||
2657 | case P_OV_REQUEST: | |
b30ab791 | 2658 | if (device->ov_start_sector == ~(sector_t)0 && |
9f4fe9ad | 2659 | peer_device->connection->agreed_pro_version >= 90) { |
de228bba LE |
2660 | unsigned long now = jiffies; |
2661 | int i; | |
b30ab791 AG |
2662 | device->ov_start_sector = sector; |
2663 | device->ov_position = sector; | |
2664 | device->ov_left = drbd_bm_bits(device) - BM_SECT_TO_BIT(sector); | |
2665 | device->rs_total = device->ov_left; | |
de228bba | 2666 | for (i = 0; i < DRBD_SYNC_MARKS; i++) { |
b30ab791 AG |
2667 | device->rs_mark_left[i] = device->ov_left; |
2668 | device->rs_mark_time[i] = now; | |
de228bba | 2669 | } |
d0180171 | 2670 | drbd_info(device, "Online Verify start sector: %llu\n", |
b411b363 PR |
2671 | (unsigned long long)sector); |
2672 | } | |
a8cd15ba | 2673 | peer_req->w.cb = w_e_end_ov_req; |
b411b363 | 2674 | fault_type = DRBD_FAULT_RS_RD; |
b411b363 PR |
2675 | break; |
2676 | ||
b411b363 | 2677 | default: |
49ba9b1b | 2678 | BUG(); |
b411b363 PR |
2679 | } |
2680 | ||
0f0601f4 LE |
2681 | /* Throttle, drbd_rs_begin_io and submit should become asynchronous |
2682 | * wrt the receiver, but it is not as straightforward as it may seem. | |
2683 | * Various places in the resync start and stop logic assume resync | |
2684 | * requests are processed in order, requeuing this on the worker thread | |
2685 | * introduces a bunch of new code for synchronization between threads. | |
2686 | * | |
2687 | * Unlimited throttling before drbd_rs_begin_io may stall the resync | |
2688 | * "forever", throttling after drbd_rs_begin_io will lock that extent | |
2689 | * for application writes for the same time. For now, just throttle | |
2690 | * here, where the rest of the code expects the receiver to sleep for | |
2691 | * a while, anyways. | |
2692 | */ | |
2693 | ||
2694 | /* Throttle before drbd_rs_begin_io, as that locks out application IO; | |
2695 | * this defers syncer requests for some time, before letting at least | |
2696 | * on request through. The resync controller on the receiving side | |
2697 | * will adapt to the incoming rate accordingly. | |
2698 | * | |
2699 | * We cannot throttle here if remote is Primary/SyncTarget: | |
2700 | * we would also throttle its application reads. | |
2701 | * In that case, throttling is done on the SyncTarget only. | |
2702 | */ | |
c5a2c150 LE |
2703 | |
2704 | /* Even though this may be a resync request, we do add to "read_ee"; | |
2705 | * "sync_ee" is only used for resync WRITEs. | |
2706 | * Add to list early, so debugfs can find this request | |
2707 | * even if we have to sleep below. */ | |
2708 | spin_lock_irq(&device->resource->req_lock); | |
2709 | list_add_tail(&peer_req->w.list, &device->read_ee); | |
2710 | spin_unlock_irq(&device->resource->req_lock); | |
2711 | ||
944410e9 | 2712 | update_receiver_timing_details(connection, drbd_rs_should_slow_down); |
ad3fee79 LE |
2713 | if (device->state.peer != R_PRIMARY |
2714 | && drbd_rs_should_slow_down(device, sector, false)) | |
e3555d85 | 2715 | schedule_timeout_uninterruptible(HZ/10); |
944410e9 | 2716 | update_receiver_timing_details(connection, drbd_rs_begin_io); |
b30ab791 | 2717 | if (drbd_rs_begin_io(device, sector)) |
80a40e43 | 2718 | goto out_free_e; |
b411b363 | 2719 | |
0f0601f4 | 2720 | submit_for_resync: |
b30ab791 | 2721 | atomic_add(size >> 9, &device->rs_sect_ev); |
0f0601f4 | 2722 | |
80a40e43 | 2723 | submit: |
944410e9 | 2724 | update_receiver_timing_details(connection, drbd_submit_peer_request); |
b30ab791 | 2725 | inc_unacked(device); |
b30ab791 | 2726 | if (drbd_submit_peer_request(device, peer_req, READ, fault_type) == 0) |
82bc0194 | 2727 | return 0; |
b411b363 | 2728 | |
10f6d992 | 2729 | /* don't care for the reason here */ |
d0180171 | 2730 | drbd_err(device, "submit failed, triggering re-connect\n"); |
c5a2c150 LE |
2731 | |
2732 | out_free_e: | |
0500813f | 2733 | spin_lock_irq(&device->resource->req_lock); |
a8cd15ba | 2734 | list_del(&peer_req->w.list); |
0500813f | 2735 | spin_unlock_irq(&device->resource->req_lock); |
22cc37a9 LE |
2736 | /* no drbd_rs_complete_io(), we are dropping the connection anyways */ |
2737 | ||
b30ab791 AG |
2738 | put_ldev(device); |
2739 | drbd_free_peer_req(device, peer_req); | |
82bc0194 | 2740 | return -EIO; |
b411b363 PR |
2741 | } |
2742 | ||
69a22773 AG |
2743 | /** |
2744 | * drbd_asb_recover_0p - Recover after split-brain with no remaining primaries | |
2745 | */ | |
2746 | static int drbd_asb_recover_0p(struct drbd_peer_device *peer_device) __must_hold(local) | |
b411b363 | 2747 | { |
69a22773 | 2748 | struct drbd_device *device = peer_device->device; |
b411b363 PR |
2749 | int self, peer, rv = -100; |
2750 | unsigned long ch_self, ch_peer; | |
44ed167d | 2751 | enum drbd_after_sb_p after_sb_0p; |
b411b363 | 2752 | |
b30ab791 AG |
2753 | self = device->ldev->md.uuid[UI_BITMAP] & 1; |
2754 | peer = device->p_uuid[UI_BITMAP] & 1; | |
b411b363 | 2755 | |
b30ab791 AG |
2756 | ch_peer = device->p_uuid[UI_SIZE]; |
2757 | ch_self = device->comm_bm_set; | |
b411b363 | 2758 | |
44ed167d | 2759 | rcu_read_lock(); |
69a22773 | 2760 | after_sb_0p = rcu_dereference(peer_device->connection->net_conf)->after_sb_0p; |
44ed167d PR |
2761 | rcu_read_unlock(); |
2762 | switch (after_sb_0p) { | |
b411b363 PR |
2763 | case ASB_CONSENSUS: |
2764 | case ASB_DISCARD_SECONDARY: | |
2765 | case ASB_CALL_HELPER: | |
44ed167d | 2766 | case ASB_VIOLENTLY: |
d0180171 | 2767 | drbd_err(device, "Configuration error.\n"); |
b411b363 PR |
2768 | break; |
2769 | case ASB_DISCONNECT: | |
2770 | break; | |
2771 | case ASB_DISCARD_YOUNGER_PRI: | |
2772 | if (self == 0 && peer == 1) { | |
2773 | rv = -1; | |
2774 | break; | |
2775 | } | |
2776 | if (self == 1 && peer == 0) { | |
2777 | rv = 1; | |
2778 | break; | |
2779 | } | |
2780 | /* Else fall through to one of the other strategies... */ | |
2781 | case ASB_DISCARD_OLDER_PRI: | |
2782 | if (self == 0 && peer == 1) { | |
2783 | rv = 1; | |
2784 | break; | |
2785 | } | |
2786 | if (self == 1 && peer == 0) { | |
2787 | rv = -1; | |
2788 | break; | |
2789 | } | |
2790 | /* Else fall through to one of the other strategies... */ | |
d0180171 | 2791 | drbd_warn(device, "Discard younger/older primary did not find a decision\n" |
b411b363 PR |
2792 | "Using discard-least-changes instead\n"); |
2793 | case ASB_DISCARD_ZERO_CHG: | |
2794 | if (ch_peer == 0 && ch_self == 0) { | |
69a22773 | 2795 | rv = test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags) |
b411b363 PR |
2796 | ? -1 : 1; |
2797 | break; | |
2798 | } else { | |
2799 | if (ch_peer == 0) { rv = 1; break; } | |
2800 | if (ch_self == 0) { rv = -1; break; } | |
2801 | } | |
44ed167d | 2802 | if (after_sb_0p == ASB_DISCARD_ZERO_CHG) |
b411b363 PR |
2803 | break; |
2804 | case ASB_DISCARD_LEAST_CHG: | |
2805 | if (ch_self < ch_peer) | |
2806 | rv = -1; | |
2807 | else if (ch_self > ch_peer) | |
2808 | rv = 1; | |
2809 | else /* ( ch_self == ch_peer ) */ | |
2810 | /* Well, then use something else. */ | |
69a22773 | 2811 | rv = test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags) |
b411b363 PR |
2812 | ? -1 : 1; |
2813 | break; | |
2814 | case ASB_DISCARD_LOCAL: | |
2815 | rv = -1; | |
2816 | break; | |
2817 | case ASB_DISCARD_REMOTE: | |
2818 | rv = 1; | |
2819 | } | |
2820 | ||
2821 | return rv; | |
2822 | } | |
2823 | ||
69a22773 AG |
2824 | /** |
2825 | * drbd_asb_recover_1p - Recover after split-brain with one remaining primary | |
2826 | */ | |
2827 | static int drbd_asb_recover_1p(struct drbd_peer_device *peer_device) __must_hold(local) | |
b411b363 | 2828 | { |
69a22773 | 2829 | struct drbd_device *device = peer_device->device; |
6184ea21 | 2830 | int hg, rv = -100; |
44ed167d | 2831 | enum drbd_after_sb_p after_sb_1p; |
b411b363 | 2832 | |
44ed167d | 2833 | rcu_read_lock(); |
69a22773 | 2834 | after_sb_1p = rcu_dereference(peer_device->connection->net_conf)->after_sb_1p; |
44ed167d PR |
2835 | rcu_read_unlock(); |
2836 | switch (after_sb_1p) { | |
b411b363 PR |
2837 | case ASB_DISCARD_YOUNGER_PRI: |
2838 | case ASB_DISCARD_OLDER_PRI: | |
2839 | case ASB_DISCARD_LEAST_CHG: | |
2840 | case ASB_DISCARD_LOCAL: | |
2841 | case ASB_DISCARD_REMOTE: | |
44ed167d | 2842 | case ASB_DISCARD_ZERO_CHG: |
d0180171 | 2843 | drbd_err(device, "Configuration error.\n"); |
b411b363 PR |
2844 | break; |
2845 | case ASB_DISCONNECT: | |
2846 | break; | |
2847 | case ASB_CONSENSUS: | |
69a22773 | 2848 | hg = drbd_asb_recover_0p(peer_device); |
b30ab791 | 2849 | if (hg == -1 && device->state.role == R_SECONDARY) |
b411b363 | 2850 | rv = hg; |
b30ab791 | 2851 | if (hg == 1 && device->state.role == R_PRIMARY) |
b411b363 PR |
2852 | rv = hg; |
2853 | break; | |
2854 | case ASB_VIOLENTLY: | |
69a22773 | 2855 | rv = drbd_asb_recover_0p(peer_device); |
b411b363 PR |
2856 | break; |
2857 | case ASB_DISCARD_SECONDARY: | |
b30ab791 | 2858 | return device->state.role == R_PRIMARY ? 1 : -1; |
b411b363 | 2859 | case ASB_CALL_HELPER: |
69a22773 | 2860 | hg = drbd_asb_recover_0p(peer_device); |
b30ab791 | 2861 | if (hg == -1 && device->state.role == R_PRIMARY) { |
bb437946 AG |
2862 | enum drbd_state_rv rv2; |
2863 | ||
b411b363 PR |
2864 | /* drbd_change_state() does not sleep while in SS_IN_TRANSIENT_STATE, |
2865 | * we might be here in C_WF_REPORT_PARAMS which is transient. | |
2866 | * we do not need to wait for the after state change work either. */ | |
b30ab791 | 2867 | rv2 = drbd_change_state(device, CS_VERBOSE, NS(role, R_SECONDARY)); |
bb437946 | 2868 | if (rv2 != SS_SUCCESS) { |
b30ab791 | 2869 | drbd_khelper(device, "pri-lost-after-sb"); |
b411b363 | 2870 | } else { |
d0180171 | 2871 | drbd_warn(device, "Successfully gave up primary role.\n"); |
b411b363 PR |
2872 | rv = hg; |
2873 | } | |
2874 | } else | |
2875 | rv = hg; | |
2876 | } | |
2877 | ||
2878 | return rv; | |
2879 | } | |
2880 | ||
69a22773 AG |
2881 | /** |
2882 | * drbd_asb_recover_2p - Recover after split-brain with two remaining primaries | |
2883 | */ | |
2884 | static int drbd_asb_recover_2p(struct drbd_peer_device *peer_device) __must_hold(local) | |
b411b363 | 2885 | { |
69a22773 | 2886 | struct drbd_device *device = peer_device->device; |
6184ea21 | 2887 | int hg, rv = -100; |
44ed167d | 2888 | enum drbd_after_sb_p after_sb_2p; |
b411b363 | 2889 | |
44ed167d | 2890 | rcu_read_lock(); |
69a22773 | 2891 | after_sb_2p = rcu_dereference(peer_device->connection->net_conf)->after_sb_2p; |
44ed167d PR |
2892 | rcu_read_unlock(); |
2893 | switch (after_sb_2p) { | |
b411b363 PR |
2894 | case ASB_DISCARD_YOUNGER_PRI: |
2895 | case ASB_DISCARD_OLDER_PRI: | |
2896 | case ASB_DISCARD_LEAST_CHG: | |
2897 | case ASB_DISCARD_LOCAL: | |
2898 | case ASB_DISCARD_REMOTE: | |
2899 | case ASB_CONSENSUS: | |
2900 | case ASB_DISCARD_SECONDARY: | |
44ed167d | 2901 | case ASB_DISCARD_ZERO_CHG: |
d0180171 | 2902 | drbd_err(device, "Configuration error.\n"); |
b411b363 PR |
2903 | break; |
2904 | case ASB_VIOLENTLY: | |
69a22773 | 2905 | rv = drbd_asb_recover_0p(peer_device); |
b411b363 PR |
2906 | break; |
2907 | case ASB_DISCONNECT: | |
2908 | break; | |
2909 | case ASB_CALL_HELPER: | |
69a22773 | 2910 | hg = drbd_asb_recover_0p(peer_device); |
b411b363 | 2911 | if (hg == -1) { |
bb437946 AG |
2912 | enum drbd_state_rv rv2; |
2913 | ||
b411b363 PR |
2914 | /* drbd_change_state() does not sleep while in SS_IN_TRANSIENT_STATE, |
2915 | * we might be here in C_WF_REPORT_PARAMS which is transient. | |
2916 | * we do not need to wait for the after state change work either. */ | |
b30ab791 | 2917 | rv2 = drbd_change_state(device, CS_VERBOSE, NS(role, R_SECONDARY)); |
bb437946 | 2918 | if (rv2 != SS_SUCCESS) { |
b30ab791 | 2919 | drbd_khelper(device, "pri-lost-after-sb"); |
b411b363 | 2920 | } else { |
d0180171 | 2921 | drbd_warn(device, "Successfully gave up primary role.\n"); |
b411b363 PR |
2922 | rv = hg; |
2923 | } | |
2924 | } else | |
2925 | rv = hg; | |
2926 | } | |
2927 | ||
2928 | return rv; | |
2929 | } | |
2930 | ||
b30ab791 | 2931 | static void drbd_uuid_dump(struct drbd_device *device, char *text, u64 *uuid, |
b411b363 PR |
2932 | u64 bits, u64 flags) |
2933 | { | |
2934 | if (!uuid) { | |
d0180171 | 2935 | drbd_info(device, "%s uuid info vanished while I was looking!\n", text); |
b411b363 PR |
2936 | return; |
2937 | } | |
d0180171 | 2938 | drbd_info(device, "%s %016llX:%016llX:%016llX:%016llX bits:%llu flags:%llX\n", |
b411b363 PR |
2939 | text, |
2940 | (unsigned long long)uuid[UI_CURRENT], | |
2941 | (unsigned long long)uuid[UI_BITMAP], | |
2942 | (unsigned long long)uuid[UI_HISTORY_START], | |
2943 | (unsigned long long)uuid[UI_HISTORY_END], | |
2944 | (unsigned long long)bits, | |
2945 | (unsigned long long)flags); | |
2946 | } | |
2947 | ||
2948 | /* | |
2949 | 100 after split brain try auto recover | |
2950 | 2 C_SYNC_SOURCE set BitMap | |
2951 | 1 C_SYNC_SOURCE use BitMap | |
2952 | 0 no Sync | |
2953 | -1 C_SYNC_TARGET use BitMap | |
2954 | -2 C_SYNC_TARGET set BitMap | |
2955 | -100 after split brain, disconnect | |
2956 | -1000 unrelated data | |
4a23f264 PR |
2957 | -1091 requires proto 91 |
2958 | -1096 requires proto 96 | |
b411b363 | 2959 | */ |
44a4d551 | 2960 | static int drbd_uuid_compare(struct drbd_device *const device, int *rule_nr) __must_hold(local) |
b411b363 | 2961 | { |
44a4d551 LE |
2962 | struct drbd_peer_device *const peer_device = first_peer_device(device); |
2963 | struct drbd_connection *const connection = peer_device ? peer_device->connection : NULL; | |
b411b363 PR |
2964 | u64 self, peer; |
2965 | int i, j; | |
2966 | ||
b30ab791 AG |
2967 | self = device->ldev->md.uuid[UI_CURRENT] & ~((u64)1); |
2968 | peer = device->p_uuid[UI_CURRENT] & ~((u64)1); | |
b411b363 PR |
2969 | |
2970 | *rule_nr = 10; | |
2971 | if (self == UUID_JUST_CREATED && peer == UUID_JUST_CREATED) | |
2972 | return 0; | |
2973 | ||
2974 | *rule_nr = 20; | |
2975 | if ((self == UUID_JUST_CREATED || self == (u64)0) && | |
2976 | peer != UUID_JUST_CREATED) | |
2977 | return -2; | |
2978 | ||
2979 | *rule_nr = 30; | |
2980 | if (self != UUID_JUST_CREATED && | |
2981 | (peer == UUID_JUST_CREATED || peer == (u64)0)) | |
2982 | return 2; | |
2983 | ||
2984 | if (self == peer) { | |
2985 | int rct, dc; /* roles at crash time */ | |
2986 | ||
b30ab791 | 2987 | if (device->p_uuid[UI_BITMAP] == (u64)0 && device->ldev->md.uuid[UI_BITMAP] != (u64)0) { |
b411b363 | 2988 | |
44a4d551 | 2989 | if (connection->agreed_pro_version < 91) |
4a23f264 | 2990 | return -1091; |
b411b363 | 2991 | |
b30ab791 AG |
2992 | if ((device->ldev->md.uuid[UI_BITMAP] & ~((u64)1)) == (device->p_uuid[UI_HISTORY_START] & ~((u64)1)) && |
2993 | (device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (device->p_uuid[UI_HISTORY_START + 1] & ~((u64)1))) { | |
d0180171 | 2994 | drbd_info(device, "was SyncSource, missed the resync finished event, corrected myself:\n"); |
b30ab791 AG |
2995 | drbd_uuid_move_history(device); |
2996 | device->ldev->md.uuid[UI_HISTORY_START] = device->ldev->md.uuid[UI_BITMAP]; | |
2997 | device->ldev->md.uuid[UI_BITMAP] = 0; | |
b411b363 | 2998 | |
b30ab791 AG |
2999 | drbd_uuid_dump(device, "self", device->ldev->md.uuid, |
3000 | device->state.disk >= D_NEGOTIATING ? drbd_bm_total_weight(device) : 0, 0); | |
b411b363 PR |
3001 | *rule_nr = 34; |
3002 | } else { | |
d0180171 | 3003 | drbd_info(device, "was SyncSource (peer failed to write sync_uuid)\n"); |
b411b363 PR |
3004 | *rule_nr = 36; |
3005 | } | |
3006 | ||
3007 | return 1; | |
3008 | } | |
3009 | ||
b30ab791 | 3010 | if (device->ldev->md.uuid[UI_BITMAP] == (u64)0 && device->p_uuid[UI_BITMAP] != (u64)0) { |
b411b363 | 3011 | |
44a4d551 | 3012 | if (connection->agreed_pro_version < 91) |
4a23f264 | 3013 | return -1091; |
b411b363 | 3014 | |
b30ab791 AG |
3015 | if ((device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (device->p_uuid[UI_BITMAP] & ~((u64)1)) && |
3016 | (device->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) == (device->p_uuid[UI_HISTORY_START] & ~((u64)1))) { | |
d0180171 | 3017 | drbd_info(device, "was SyncTarget, peer missed the resync finished event, corrected peer:\n"); |
b411b363 | 3018 | |
b30ab791 AG |
3019 | device->p_uuid[UI_HISTORY_START + 1] = device->p_uuid[UI_HISTORY_START]; |
3020 | device->p_uuid[UI_HISTORY_START] = device->p_uuid[UI_BITMAP]; | |
3021 | device->p_uuid[UI_BITMAP] = 0UL; | |
b411b363 | 3022 | |
b30ab791 | 3023 | drbd_uuid_dump(device, "peer", device->p_uuid, device->p_uuid[UI_SIZE], device->p_uuid[UI_FLAGS]); |
b411b363 PR |
3024 | *rule_nr = 35; |
3025 | } else { | |
d0180171 | 3026 | drbd_info(device, "was SyncTarget (failed to write sync_uuid)\n"); |
b411b363 PR |
3027 | *rule_nr = 37; |
3028 | } | |
3029 | ||
3030 | return -1; | |
3031 | } | |
3032 | ||
3033 | /* Common power [off|failure] */ | |
b30ab791 AG |
3034 | rct = (test_bit(CRASHED_PRIMARY, &device->flags) ? 1 : 0) + |
3035 | (device->p_uuid[UI_FLAGS] & 2); | |
b411b363 PR |
3036 | /* lowest bit is set when we were primary, |
3037 | * next bit (weight 2) is set when peer was primary */ | |
3038 | *rule_nr = 40; | |
3039 | ||
3040 | switch (rct) { | |
3041 | case 0: /* !self_pri && !peer_pri */ return 0; | |
3042 | case 1: /* self_pri && !peer_pri */ return 1; | |
3043 | case 2: /* !self_pri && peer_pri */ return -1; | |
3044 | case 3: /* self_pri && peer_pri */ | |
44a4d551 | 3045 | dc = test_bit(RESOLVE_CONFLICTS, &connection->flags); |
b411b363 PR |
3046 | return dc ? -1 : 1; |
3047 | } | |
3048 | } | |
3049 | ||
3050 | *rule_nr = 50; | |
b30ab791 | 3051 | peer = device->p_uuid[UI_BITMAP] & ~((u64)1); |
b411b363 PR |
3052 | if (self == peer) |
3053 | return -1; | |
3054 | ||
3055 | *rule_nr = 51; | |
b30ab791 | 3056 | peer = device->p_uuid[UI_HISTORY_START] & ~((u64)1); |
b411b363 | 3057 | if (self == peer) { |
44a4d551 | 3058 | if (connection->agreed_pro_version < 96 ? |
b30ab791 AG |
3059 | (device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == |
3060 | (device->p_uuid[UI_HISTORY_START + 1] & ~((u64)1)) : | |
3061 | peer + UUID_NEW_BM_OFFSET == (device->p_uuid[UI_BITMAP] & ~((u64)1))) { | |
b411b363 PR |
3062 | /* The last P_SYNC_UUID did not get though. Undo the last start of |
3063 | resync as sync source modifications of the peer's UUIDs. */ | |
3064 | ||
44a4d551 | 3065 | if (connection->agreed_pro_version < 91) |
4a23f264 | 3066 | return -1091; |
b411b363 | 3067 | |
b30ab791 AG |
3068 | device->p_uuid[UI_BITMAP] = device->p_uuid[UI_HISTORY_START]; |
3069 | device->p_uuid[UI_HISTORY_START] = device->p_uuid[UI_HISTORY_START + 1]; | |
4a23f264 | 3070 | |
d0180171 | 3071 | drbd_info(device, "Lost last syncUUID packet, corrected:\n"); |
b30ab791 | 3072 | drbd_uuid_dump(device, "peer", device->p_uuid, device->p_uuid[UI_SIZE], device->p_uuid[UI_FLAGS]); |
4a23f264 | 3073 | |
b411b363 PR |
3074 | return -1; |
3075 | } | |
3076 | } | |
3077 | ||
3078 | *rule_nr = 60; | |
b30ab791 | 3079 | self = device->ldev->md.uuid[UI_CURRENT] & ~((u64)1); |
b411b363 | 3080 | for (i = UI_HISTORY_START; i <= UI_HISTORY_END; i++) { |
b30ab791 | 3081 | peer = device->p_uuid[i] & ~((u64)1); |
b411b363 PR |
3082 | if (self == peer) |
3083 | return -2; | |
3084 | } | |
3085 | ||
3086 | *rule_nr = 70; | |
b30ab791 AG |
3087 | self = device->ldev->md.uuid[UI_BITMAP] & ~((u64)1); |
3088 | peer = device->p_uuid[UI_CURRENT] & ~((u64)1); | |
b411b363 PR |
3089 | if (self == peer) |
3090 | return 1; | |
3091 | ||
3092 | *rule_nr = 71; | |
b30ab791 | 3093 | self = device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1); |
b411b363 | 3094 | if (self == peer) { |
44a4d551 | 3095 | if (connection->agreed_pro_version < 96 ? |
b30ab791 AG |
3096 | (device->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) == |
3097 | (device->p_uuid[UI_HISTORY_START] & ~((u64)1)) : | |
3098 | self + UUID_NEW_BM_OFFSET == (device->ldev->md.uuid[UI_BITMAP] & ~((u64)1))) { | |
b411b363 PR |
3099 | /* The last P_SYNC_UUID did not get though. Undo the last start of |
3100 | resync as sync source modifications of our UUIDs. */ | |
3101 | ||
44a4d551 | 3102 | if (connection->agreed_pro_version < 91) |
4a23f264 | 3103 | return -1091; |
b411b363 | 3104 | |
b30ab791 AG |
3105 | __drbd_uuid_set(device, UI_BITMAP, device->ldev->md.uuid[UI_HISTORY_START]); |
3106 | __drbd_uuid_set(device, UI_HISTORY_START, device->ldev->md.uuid[UI_HISTORY_START + 1]); | |
b411b363 | 3107 | |
d0180171 | 3108 | drbd_info(device, "Last syncUUID did not get through, corrected:\n"); |
b30ab791 AG |
3109 | drbd_uuid_dump(device, "self", device->ldev->md.uuid, |
3110 | device->state.disk >= D_NEGOTIATING ? drbd_bm_total_weight(device) : 0, 0); | |
b411b363 PR |
3111 | |
3112 | return 1; | |
3113 | } | |
3114 | } | |
3115 | ||
3116 | ||
3117 | *rule_nr = 80; | |
b30ab791 | 3118 | peer = device->p_uuid[UI_CURRENT] & ~((u64)1); |
b411b363 | 3119 | for (i = UI_HISTORY_START; i <= UI_HISTORY_END; i++) { |
b30ab791 | 3120 | self = device->ldev->md.uuid[i] & ~((u64)1); |
b411b363 PR |
3121 | if (self == peer) |
3122 | return 2; | |
3123 | } | |
3124 | ||
3125 | *rule_nr = 90; | |
b30ab791 AG |
3126 | self = device->ldev->md.uuid[UI_BITMAP] & ~((u64)1); |
3127 | peer = device->p_uuid[UI_BITMAP] & ~((u64)1); | |
b411b363 PR |
3128 | if (self == peer && self != ((u64)0)) |
3129 | return 100; | |
3130 | ||
3131 | *rule_nr = 100; | |
3132 | for (i = UI_HISTORY_START; i <= UI_HISTORY_END; i++) { | |
b30ab791 | 3133 | self = device->ldev->md.uuid[i] & ~((u64)1); |
b411b363 | 3134 | for (j = UI_HISTORY_START; j <= UI_HISTORY_END; j++) { |
b30ab791 | 3135 | peer = device->p_uuid[j] & ~((u64)1); |
b411b363 PR |
3136 | if (self == peer) |
3137 | return -100; | |
3138 | } | |
3139 | } | |
3140 | ||
3141 | return -1000; | |
3142 | } | |
3143 | ||
3144 | /* drbd_sync_handshake() returns the new conn state on success, or | |
3145 | CONN_MASK (-1) on failure. | |
3146 | */ | |
69a22773 AG |
3147 | static enum drbd_conns drbd_sync_handshake(struct drbd_peer_device *peer_device, |
3148 | enum drbd_role peer_role, | |
b411b363 PR |
3149 | enum drbd_disk_state peer_disk) __must_hold(local) |
3150 | { | |
69a22773 | 3151 | struct drbd_device *device = peer_device->device; |
b411b363 PR |
3152 | enum drbd_conns rv = C_MASK; |
3153 | enum drbd_disk_state mydisk; | |
44ed167d | 3154 | struct net_conf *nc; |
6dff2902 | 3155 | int hg, rule_nr, rr_conflict, tentative; |
b411b363 | 3156 | |
b30ab791 | 3157 | mydisk = device->state.disk; |
b411b363 | 3158 | if (mydisk == D_NEGOTIATING) |
b30ab791 | 3159 | mydisk = device->new_state_tmp.disk; |
b411b363 | 3160 | |
d0180171 | 3161 | drbd_info(device, "drbd_sync_handshake:\n"); |
9f2247bb | 3162 | |
b30ab791 AG |
3163 | spin_lock_irq(&device->ldev->md.uuid_lock); |
3164 | drbd_uuid_dump(device, "self", device->ldev->md.uuid, device->comm_bm_set, 0); | |
3165 | drbd_uuid_dump(device, "peer", device->p_uuid, | |
3166 | device->p_uuid[UI_SIZE], device->p_uuid[UI_FLAGS]); | |
b411b363 | 3167 | |
b30ab791 AG |
3168 | hg = drbd_uuid_compare(device, &rule_nr); |
3169 | spin_unlock_irq(&device->ldev->md.uuid_lock); | |
b411b363 | 3170 | |
d0180171 | 3171 | drbd_info(device, "uuid_compare()=%d by rule %d\n", hg, rule_nr); |
b411b363 PR |
3172 | |
3173 | if (hg == -1000) { | |
d0180171 | 3174 | drbd_alert(device, "Unrelated data, aborting!\n"); |
b411b363 PR |
3175 | return C_MASK; |
3176 | } | |
4a23f264 | 3177 | if (hg < -1000) { |
d0180171 | 3178 | drbd_alert(device, "To resolve this both sides have to support at least protocol %d\n", -hg - 1000); |
b411b363 PR |
3179 | return C_MASK; |
3180 | } | |
3181 | ||
3182 | if ((mydisk == D_INCONSISTENT && peer_disk > D_INCONSISTENT) || | |
3183 | (peer_disk == D_INCONSISTENT && mydisk > D_INCONSISTENT)) { | |
3184 | int f = (hg == -100) || abs(hg) == 2; | |
3185 | hg = mydisk > D_INCONSISTENT ? 1 : -1; | |
3186 | if (f) | |
3187 | hg = hg*2; | |
d0180171 | 3188 | drbd_info(device, "Becoming sync %s due to disk states.\n", |
b411b363 PR |
3189 | hg > 0 ? "source" : "target"); |
3190 | } | |
3191 | ||
3a11a487 | 3192 | if (abs(hg) == 100) |
b30ab791 | 3193 | drbd_khelper(device, "initial-split-brain"); |
3a11a487 | 3194 | |
44ed167d | 3195 | rcu_read_lock(); |
69a22773 | 3196 | nc = rcu_dereference(peer_device->connection->net_conf); |
44ed167d PR |
3197 | |
3198 | if (hg == 100 || (hg == -100 && nc->always_asbp)) { | |
b30ab791 | 3199 | int pcount = (device->state.role == R_PRIMARY) |
b411b363 PR |
3200 | + (peer_role == R_PRIMARY); |
3201 | int forced = (hg == -100); | |
3202 | ||
3203 | switch (pcount) { | |
3204 | case 0: | |
69a22773 | 3205 | hg = drbd_asb_recover_0p(peer_device); |
b411b363 PR |
3206 | break; |
3207 | case 1: | |
69a22773 | 3208 | hg = drbd_asb_recover_1p(peer_device); |
b411b363 PR |
3209 | break; |
3210 | case 2: | |
69a22773 | 3211 | hg = drbd_asb_recover_2p(peer_device); |
b411b363 PR |
3212 | break; |
3213 | } | |
3214 | if (abs(hg) < 100) { | |
d0180171 | 3215 | drbd_warn(device, "Split-Brain detected, %d primaries, " |
b411b363 PR |
3216 | "automatically solved. Sync from %s node\n", |
3217 | pcount, (hg < 0) ? "peer" : "this"); | |
3218 | if (forced) { | |
d0180171 | 3219 | drbd_warn(device, "Doing a full sync, since" |
b411b363 PR |
3220 | " UUIDs where ambiguous.\n"); |
3221 | hg = hg*2; | |
3222 | } | |
3223 | } | |
3224 | } | |
3225 | ||
3226 | if (hg == -100) { | |
b30ab791 | 3227 | if (test_bit(DISCARD_MY_DATA, &device->flags) && !(device->p_uuid[UI_FLAGS]&1)) |
b411b363 | 3228 | hg = -1; |
b30ab791 | 3229 | if (!test_bit(DISCARD_MY_DATA, &device->flags) && (device->p_uuid[UI_FLAGS]&1)) |
b411b363 PR |
3230 | hg = 1; |
3231 | ||
3232 | if (abs(hg) < 100) | |
d0180171 | 3233 | drbd_warn(device, "Split-Brain detected, manually solved. " |
b411b363 PR |
3234 | "Sync from %s node\n", |
3235 | (hg < 0) ? "peer" : "this"); | |
3236 | } | |
44ed167d | 3237 | rr_conflict = nc->rr_conflict; |
6dff2902 | 3238 | tentative = nc->tentative; |
44ed167d | 3239 | rcu_read_unlock(); |
b411b363 PR |
3240 | |
3241 | if (hg == -100) { | |
580b9767 LE |
3242 | /* FIXME this log message is not correct if we end up here |
3243 | * after an attempted attach on a diskless node. | |
3244 | * We just refuse to attach -- well, we drop the "connection" | |
3245 | * to that disk, in a way... */ | |
d0180171 | 3246 | drbd_alert(device, "Split-Brain detected but unresolved, dropping connection!\n"); |
b30ab791 | 3247 | drbd_khelper(device, "split-brain"); |
b411b363 PR |
3248 | return C_MASK; |
3249 | } | |
3250 | ||
3251 | if (hg > 0 && mydisk <= D_INCONSISTENT) { | |
d0180171 | 3252 | drbd_err(device, "I shall become SyncSource, but I am inconsistent!\n"); |
b411b363 PR |
3253 | return C_MASK; |
3254 | } | |
3255 | ||
3256 | if (hg < 0 && /* by intention we do not use mydisk here. */ | |
b30ab791 | 3257 | device->state.role == R_PRIMARY && device->state.disk >= D_CONSISTENT) { |
44ed167d | 3258 | switch (rr_conflict) { |
b411b363 | 3259 | case ASB_CALL_HELPER: |
b30ab791 | 3260 | drbd_khelper(device, "pri-lost"); |
b411b363 PR |
3261 | /* fall through */ |
3262 | case ASB_DISCONNECT: | |
d0180171 | 3263 | drbd_err(device, "I shall become SyncTarget, but I am primary!\n"); |
b411b363 PR |
3264 | return C_MASK; |
3265 | case ASB_VIOLENTLY: | |
d0180171 | 3266 | drbd_warn(device, "Becoming SyncTarget, violating the stable-data" |
b411b363 PR |
3267 | "assumption\n"); |
3268 | } | |
3269 | } | |
3270 | ||
69a22773 | 3271 | if (tentative || test_bit(CONN_DRY_RUN, &peer_device->connection->flags)) { |
cf14c2e9 | 3272 | if (hg == 0) |
d0180171 | 3273 | drbd_info(device, "dry-run connect: No resync, would become Connected immediately.\n"); |
cf14c2e9 | 3274 | else |
d0180171 | 3275 | drbd_info(device, "dry-run connect: Would become %s, doing a %s resync.", |
cf14c2e9 PR |
3276 | drbd_conn_str(hg > 0 ? C_SYNC_SOURCE : C_SYNC_TARGET), |
3277 | abs(hg) >= 2 ? "full" : "bit-map based"); | |
3278 | return C_MASK; | |
3279 | } | |
3280 | ||
b411b363 | 3281 | if (abs(hg) >= 2) { |
d0180171 | 3282 | drbd_info(device, "Writing the whole bitmap, full sync required after drbd_sync_handshake.\n"); |
b30ab791 | 3283 | if (drbd_bitmap_io(device, &drbd_bmio_set_n_write, "set_n_write from sync_handshake", |
20ceb2b2 | 3284 | BM_LOCKED_SET_ALLOWED)) |
b411b363 PR |
3285 | return C_MASK; |
3286 | } | |
3287 | ||
3288 | if (hg > 0) { /* become sync source. */ | |
3289 | rv = C_WF_BITMAP_S; | |
3290 | } else if (hg < 0) { /* become sync target */ | |
3291 | rv = C_WF_BITMAP_T; | |
3292 | } else { | |
3293 | rv = C_CONNECTED; | |
b30ab791 | 3294 | if (drbd_bm_total_weight(device)) { |
d0180171 | 3295 | drbd_info(device, "No resync, but %lu bits in bitmap!\n", |
b30ab791 | 3296 | drbd_bm_total_weight(device)); |
b411b363 PR |
3297 | } |
3298 | } | |
3299 | ||
3300 | return rv; | |
3301 | } | |
3302 | ||
f179d76d | 3303 | static enum drbd_after_sb_p convert_after_sb(enum drbd_after_sb_p peer) |
b411b363 PR |
3304 | { |
3305 | /* ASB_DISCARD_REMOTE - ASB_DISCARD_LOCAL is valid */ | |
f179d76d PR |
3306 | if (peer == ASB_DISCARD_REMOTE) |
3307 | return ASB_DISCARD_LOCAL; | |
b411b363 PR |
3308 | |
3309 | /* any other things with ASB_DISCARD_REMOTE or ASB_DISCARD_LOCAL are invalid */ | |
f179d76d PR |
3310 | if (peer == ASB_DISCARD_LOCAL) |
3311 | return ASB_DISCARD_REMOTE; | |
b411b363 PR |
3312 | |
3313 | /* everything else is valid if they are equal on both sides. */ | |
f179d76d | 3314 | return peer; |
b411b363 PR |
3315 | } |
3316 | ||
bde89a9e | 3317 | static int receive_protocol(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 3318 | { |
e658983a | 3319 | struct p_protocol *p = pi->data; |
036b17ea PR |
3320 | enum drbd_after_sb_p p_after_sb_0p, p_after_sb_1p, p_after_sb_2p; |
3321 | int p_proto, p_discard_my_data, p_two_primaries, cf; | |
3322 | struct net_conf *nc, *old_net_conf, *new_net_conf = NULL; | |
3323 | char integrity_alg[SHARED_SECRET_MAX] = ""; | |
accdbcc5 | 3324 | struct crypto_hash *peer_integrity_tfm = NULL; |
7aca6c75 | 3325 | void *int_dig_in = NULL, *int_dig_vv = NULL; |
b411b363 | 3326 | |
b411b363 PR |
3327 | p_proto = be32_to_cpu(p->protocol); |
3328 | p_after_sb_0p = be32_to_cpu(p->after_sb_0p); | |
3329 | p_after_sb_1p = be32_to_cpu(p->after_sb_1p); | |
3330 | p_after_sb_2p = be32_to_cpu(p->after_sb_2p); | |
b411b363 | 3331 | p_two_primaries = be32_to_cpu(p->two_primaries); |
cf14c2e9 | 3332 | cf = be32_to_cpu(p->conn_flags); |
6139f60d | 3333 | p_discard_my_data = cf & CF_DISCARD_MY_DATA; |
cf14c2e9 | 3334 | |
bde89a9e | 3335 | if (connection->agreed_pro_version >= 87) { |
86db0618 | 3336 | int err; |
cf14c2e9 | 3337 | |
88104ca4 | 3338 | if (pi->size > sizeof(integrity_alg)) |
86db0618 | 3339 | return -EIO; |
bde89a9e | 3340 | err = drbd_recv_all(connection, integrity_alg, pi->size); |
86db0618 AG |
3341 | if (err) |
3342 | return err; | |
036b17ea | 3343 | integrity_alg[SHARED_SECRET_MAX - 1] = 0; |
b411b363 PR |
3344 | } |
3345 | ||
7d4c782c | 3346 | if (pi->cmd != P_PROTOCOL_UPDATE) { |
bde89a9e | 3347 | clear_bit(CONN_DRY_RUN, &connection->flags); |
b411b363 | 3348 | |
fbc12f45 | 3349 | if (cf & CF_DRY_RUN) |
bde89a9e | 3350 | set_bit(CONN_DRY_RUN, &connection->flags); |
b411b363 | 3351 | |
fbc12f45 | 3352 | rcu_read_lock(); |
bde89a9e | 3353 | nc = rcu_dereference(connection->net_conf); |
b411b363 | 3354 | |
fbc12f45 | 3355 | if (p_proto != nc->wire_protocol) { |
1ec861eb | 3356 | drbd_err(connection, "incompatible %s settings\n", "protocol"); |
fbc12f45 AG |
3357 | goto disconnect_rcu_unlock; |
3358 | } | |
b411b363 | 3359 | |
fbc12f45 | 3360 | if (convert_after_sb(p_after_sb_0p) != nc->after_sb_0p) { |
1ec861eb | 3361 | drbd_err(connection, "incompatible %s settings\n", "after-sb-0pri"); |
fbc12f45 AG |
3362 | goto disconnect_rcu_unlock; |
3363 | } | |
b411b363 | 3364 | |
fbc12f45 | 3365 | if (convert_after_sb(p_after_sb_1p) != nc->after_sb_1p) { |
1ec861eb | 3366 | drbd_err(connection, "incompatible %s settings\n", "after-sb-1pri"); |
fbc12f45 AG |
3367 | goto disconnect_rcu_unlock; |
3368 | } | |
b411b363 | 3369 | |
fbc12f45 | 3370 | if (convert_after_sb(p_after_sb_2p) != nc->after_sb_2p) { |
1ec861eb | 3371 | drbd_err(connection, "incompatible %s settings\n", "after-sb-2pri"); |
fbc12f45 AG |
3372 | goto disconnect_rcu_unlock; |
3373 | } | |
b411b363 | 3374 | |
fbc12f45 | 3375 | if (p_discard_my_data && nc->discard_my_data) { |
1ec861eb | 3376 | drbd_err(connection, "incompatible %s settings\n", "discard-my-data"); |
fbc12f45 AG |
3377 | goto disconnect_rcu_unlock; |
3378 | } | |
b411b363 | 3379 | |
fbc12f45 | 3380 | if (p_two_primaries != nc->two_primaries) { |
1ec861eb | 3381 | drbd_err(connection, "incompatible %s settings\n", "allow-two-primaries"); |
fbc12f45 AG |
3382 | goto disconnect_rcu_unlock; |
3383 | } | |
b411b363 | 3384 | |
fbc12f45 | 3385 | if (strcmp(integrity_alg, nc->integrity_alg)) { |
1ec861eb | 3386 | drbd_err(connection, "incompatible %s settings\n", "data-integrity-alg"); |
fbc12f45 AG |
3387 | goto disconnect_rcu_unlock; |
3388 | } | |
b411b363 | 3389 | |
fbc12f45 | 3390 | rcu_read_unlock(); |
b411b363 PR |
3391 | } |
3392 | ||
7d4c782c AG |
3393 | if (integrity_alg[0]) { |
3394 | int hash_size; | |
3395 | ||
3396 | /* | |
3397 | * We can only change the peer data integrity algorithm | |
3398 | * here. Changing our own data integrity algorithm | |
3399 | * requires that we send a P_PROTOCOL_UPDATE packet at | |
3400 | * the same time; otherwise, the peer has no way to | |
3401 | * tell between which packets the algorithm should | |
3402 | * change. | |
3403 | */ | |
b411b363 | 3404 | |
7d4c782c AG |
3405 | peer_integrity_tfm = crypto_alloc_hash(integrity_alg, 0, CRYPTO_ALG_ASYNC); |
3406 | if (!peer_integrity_tfm) { | |
1ec861eb | 3407 | drbd_err(connection, "peer data-integrity-alg %s not supported\n", |
7d4c782c AG |
3408 | integrity_alg); |
3409 | goto disconnect; | |
3410 | } | |
b411b363 | 3411 | |
7d4c782c AG |
3412 | hash_size = crypto_hash_digestsize(peer_integrity_tfm); |
3413 | int_dig_in = kmalloc(hash_size, GFP_KERNEL); | |
3414 | int_dig_vv = kmalloc(hash_size, GFP_KERNEL); | |
3415 | if (!(int_dig_in && int_dig_vv)) { | |
1ec861eb | 3416 | drbd_err(connection, "Allocation of buffers for data integrity checking failed\n"); |
b411b363 PR |
3417 | goto disconnect; |
3418 | } | |
b411b363 PR |
3419 | } |
3420 | ||
7d4c782c AG |
3421 | new_net_conf = kmalloc(sizeof(struct net_conf), GFP_KERNEL); |
3422 | if (!new_net_conf) { | |
1ec861eb | 3423 | drbd_err(connection, "Allocation of new net_conf failed\n"); |
7d4c782c AG |
3424 | goto disconnect; |
3425 | } | |
3426 | ||
bde89a9e | 3427 | mutex_lock(&connection->data.mutex); |
0500813f | 3428 | mutex_lock(&connection->resource->conf_update); |
bde89a9e | 3429 | old_net_conf = connection->net_conf; |
7d4c782c AG |
3430 | *new_net_conf = *old_net_conf; |
3431 | ||
3432 | new_net_conf->wire_protocol = p_proto; | |
3433 | new_net_conf->after_sb_0p = convert_after_sb(p_after_sb_0p); | |
3434 | new_net_conf->after_sb_1p = convert_after_sb(p_after_sb_1p); | |
3435 | new_net_conf->after_sb_2p = convert_after_sb(p_after_sb_2p); | |
3436 | new_net_conf->two_primaries = p_two_primaries; | |
3437 | ||
bde89a9e | 3438 | rcu_assign_pointer(connection->net_conf, new_net_conf); |
0500813f | 3439 | mutex_unlock(&connection->resource->conf_update); |
bde89a9e | 3440 | mutex_unlock(&connection->data.mutex); |
7d4c782c | 3441 | |
bde89a9e AG |
3442 | crypto_free_hash(connection->peer_integrity_tfm); |
3443 | kfree(connection->int_dig_in); | |
3444 | kfree(connection->int_dig_vv); | |
3445 | connection->peer_integrity_tfm = peer_integrity_tfm; | |
3446 | connection->int_dig_in = int_dig_in; | |
3447 | connection->int_dig_vv = int_dig_vv; | |
7d4c782c AG |
3448 | |
3449 | if (strcmp(old_net_conf->integrity_alg, integrity_alg)) | |
1ec861eb | 3450 | drbd_info(connection, "peer data-integrity-alg: %s\n", |
7d4c782c AG |
3451 | integrity_alg[0] ? integrity_alg : "(none)"); |
3452 | ||
3453 | synchronize_rcu(); | |
3454 | kfree(old_net_conf); | |
82bc0194 | 3455 | return 0; |
b411b363 | 3456 | |
44ed167d PR |
3457 | disconnect_rcu_unlock: |
3458 | rcu_read_unlock(); | |
b411b363 | 3459 | disconnect: |
b792c35c | 3460 | crypto_free_hash(peer_integrity_tfm); |
036b17ea PR |
3461 | kfree(int_dig_in); |
3462 | kfree(int_dig_vv); | |
bde89a9e | 3463 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
82bc0194 | 3464 | return -EIO; |
b411b363 PR |
3465 | } |
3466 | ||
3467 | /* helper function | |
3468 | * input: alg name, feature name | |
3469 | * return: NULL (alg name was "") | |
3470 | * ERR_PTR(error) if something goes wrong | |
3471 | * or the crypto hash ptr, if it worked out ok. */ | |
8ce953aa | 3472 | static struct crypto_hash *drbd_crypto_alloc_digest_safe(const struct drbd_device *device, |
b411b363 PR |
3473 | const char *alg, const char *name) |
3474 | { | |
3475 | struct crypto_hash *tfm; | |
3476 | ||
3477 | if (!alg[0]) | |
3478 | return NULL; | |
3479 | ||
3480 | tfm = crypto_alloc_hash(alg, 0, CRYPTO_ALG_ASYNC); | |
3481 | if (IS_ERR(tfm)) { | |
d0180171 | 3482 | drbd_err(device, "Can not allocate \"%s\" as %s (reason: %ld)\n", |
b411b363 PR |
3483 | alg, name, PTR_ERR(tfm)); |
3484 | return tfm; | |
3485 | } | |
b411b363 PR |
3486 | return tfm; |
3487 | } | |
3488 | ||
bde89a9e | 3489 | static int ignore_remaining_packet(struct drbd_connection *connection, struct packet_info *pi) |
4a76b161 | 3490 | { |
bde89a9e | 3491 | void *buffer = connection->data.rbuf; |
4a76b161 AG |
3492 | int size = pi->size; |
3493 | ||
3494 | while (size) { | |
3495 | int s = min_t(int, size, DRBD_SOCKET_BUFFER_SIZE); | |
bde89a9e | 3496 | s = drbd_recv(connection, buffer, s); |
4a76b161 AG |
3497 | if (s <= 0) { |
3498 | if (s < 0) | |
3499 | return s; | |
3500 | break; | |
3501 | } | |
3502 | size -= s; | |
3503 | } | |
3504 | if (size) | |
3505 | return -EIO; | |
3506 | return 0; | |
3507 | } | |
3508 | ||
3509 | /* | |
3510 | * config_unknown_volume - device configuration command for unknown volume | |
3511 | * | |
3512 | * When a device is added to an existing connection, the node on which the | |
3513 | * device is added first will send configuration commands to its peer but the | |
3514 | * peer will not know about the device yet. It will warn and ignore these | |
3515 | * commands. Once the device is added on the second node, the second node will | |
3516 | * send the same device configuration commands, but in the other direction. | |
3517 | * | |
3518 | * (We can also end up here if drbd is misconfigured.) | |
3519 | */ | |
bde89a9e | 3520 | static int config_unknown_volume(struct drbd_connection *connection, struct packet_info *pi) |
4a76b161 | 3521 | { |
1ec861eb | 3522 | drbd_warn(connection, "%s packet received for volume %u, which is not configured locally\n", |
2fcb8f30 | 3523 | cmdname(pi->cmd), pi->vnr); |
bde89a9e | 3524 | return ignore_remaining_packet(connection, pi); |
4a76b161 AG |
3525 | } |
3526 | ||
bde89a9e | 3527 | static int receive_SyncParam(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 3528 | { |
9f4fe9ad | 3529 | struct drbd_peer_device *peer_device; |
b30ab791 | 3530 | struct drbd_device *device; |
e658983a | 3531 | struct p_rs_param_95 *p; |
b411b363 PR |
3532 | unsigned int header_size, data_size, exp_max_sz; |
3533 | struct crypto_hash *verify_tfm = NULL; | |
3534 | struct crypto_hash *csums_tfm = NULL; | |
2ec91e0e | 3535 | struct net_conf *old_net_conf, *new_net_conf = NULL; |
813472ce | 3536 | struct disk_conf *old_disk_conf = NULL, *new_disk_conf = NULL; |
bde89a9e | 3537 | const int apv = connection->agreed_pro_version; |
813472ce | 3538 | struct fifo_buffer *old_plan = NULL, *new_plan = NULL; |
778f271d | 3539 | int fifo_size = 0; |
82bc0194 | 3540 | int err; |
b411b363 | 3541 | |
9f4fe9ad AG |
3542 | peer_device = conn_peer_device(connection, pi->vnr); |
3543 | if (!peer_device) | |
bde89a9e | 3544 | return config_unknown_volume(connection, pi); |
9f4fe9ad | 3545 | device = peer_device->device; |
b411b363 PR |
3546 | |
3547 | exp_max_sz = apv <= 87 ? sizeof(struct p_rs_param) | |
3548 | : apv == 88 ? sizeof(struct p_rs_param) | |
3549 | + SHARED_SECRET_MAX | |
8e26f9cc PR |
3550 | : apv <= 94 ? sizeof(struct p_rs_param_89) |
3551 | : /* apv >= 95 */ sizeof(struct p_rs_param_95); | |
b411b363 | 3552 | |
e2857216 | 3553 | if (pi->size > exp_max_sz) { |
d0180171 | 3554 | drbd_err(device, "SyncParam packet too long: received %u, expected <= %u bytes\n", |
e2857216 | 3555 | pi->size, exp_max_sz); |
82bc0194 | 3556 | return -EIO; |
b411b363 PR |
3557 | } |
3558 | ||
3559 | if (apv <= 88) { | |
e658983a | 3560 | header_size = sizeof(struct p_rs_param); |
e2857216 | 3561 | data_size = pi->size - header_size; |
8e26f9cc | 3562 | } else if (apv <= 94) { |
e658983a | 3563 | header_size = sizeof(struct p_rs_param_89); |
e2857216 | 3564 | data_size = pi->size - header_size; |
0b0ba1ef | 3565 | D_ASSERT(device, data_size == 0); |
8e26f9cc | 3566 | } else { |
e658983a | 3567 | header_size = sizeof(struct p_rs_param_95); |
e2857216 | 3568 | data_size = pi->size - header_size; |
0b0ba1ef | 3569 | D_ASSERT(device, data_size == 0); |
b411b363 PR |
3570 | } |
3571 | ||
3572 | /* initialize verify_alg and csums_alg */ | |
e658983a | 3573 | p = pi->data; |
b411b363 PR |
3574 | memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); |
3575 | ||
9f4fe9ad | 3576 | err = drbd_recv_all(peer_device->connection, p, header_size); |
82bc0194 AG |
3577 | if (err) |
3578 | return err; | |
b411b363 | 3579 | |
0500813f | 3580 | mutex_lock(&connection->resource->conf_update); |
9f4fe9ad | 3581 | old_net_conf = peer_device->connection->net_conf; |
b30ab791 | 3582 | if (get_ldev(device)) { |
813472ce PR |
3583 | new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL); |
3584 | if (!new_disk_conf) { | |
b30ab791 | 3585 | put_ldev(device); |
0500813f | 3586 | mutex_unlock(&connection->resource->conf_update); |
d0180171 | 3587 | drbd_err(device, "Allocation of new disk_conf failed\n"); |
813472ce PR |
3588 | return -ENOMEM; |
3589 | } | |
daeda1cc | 3590 | |
b30ab791 | 3591 | old_disk_conf = device->ldev->disk_conf; |
813472ce | 3592 | *new_disk_conf = *old_disk_conf; |
b411b363 | 3593 | |
6394b935 | 3594 | new_disk_conf->resync_rate = be32_to_cpu(p->resync_rate); |
813472ce | 3595 | } |
b411b363 PR |
3596 | |
3597 | if (apv >= 88) { | |
3598 | if (apv == 88) { | |
5de73827 | 3599 | if (data_size > SHARED_SECRET_MAX || data_size == 0) { |
d0180171 | 3600 | drbd_err(device, "verify-alg of wrong size, " |
5de73827 PR |
3601 | "peer wants %u, accepting only up to %u byte\n", |
3602 | data_size, SHARED_SECRET_MAX); | |
813472ce PR |
3603 | err = -EIO; |
3604 | goto reconnect; | |
b411b363 PR |
3605 | } |
3606 | ||
9f4fe9ad | 3607 | err = drbd_recv_all(peer_device->connection, p->verify_alg, data_size); |
813472ce PR |
3608 | if (err) |
3609 | goto reconnect; | |
b411b363 PR |
3610 | /* we expect NUL terminated string */ |
3611 | /* but just in case someone tries to be evil */ | |
0b0ba1ef | 3612 | D_ASSERT(device, p->verify_alg[data_size-1] == 0); |
b411b363 PR |
3613 | p->verify_alg[data_size-1] = 0; |
3614 | ||
3615 | } else /* apv >= 89 */ { | |
3616 | /* we still expect NUL terminated strings */ | |
3617 | /* but just in case someone tries to be evil */ | |
0b0ba1ef AG |
3618 | D_ASSERT(device, p->verify_alg[SHARED_SECRET_MAX-1] == 0); |
3619 | D_ASSERT(device, p->csums_alg[SHARED_SECRET_MAX-1] == 0); | |
b411b363 PR |
3620 | p->verify_alg[SHARED_SECRET_MAX-1] = 0; |
3621 | p->csums_alg[SHARED_SECRET_MAX-1] = 0; | |
3622 | } | |
3623 | ||
2ec91e0e | 3624 | if (strcmp(old_net_conf->verify_alg, p->verify_alg)) { |
b30ab791 | 3625 | if (device->state.conn == C_WF_REPORT_PARAMS) { |
d0180171 | 3626 | drbd_err(device, "Different verify-alg settings. me=\"%s\" peer=\"%s\"\n", |
2ec91e0e | 3627 | old_net_conf->verify_alg, p->verify_alg); |
b411b363 PR |
3628 | goto disconnect; |
3629 | } | |
b30ab791 | 3630 | verify_tfm = drbd_crypto_alloc_digest_safe(device, |
b411b363 PR |
3631 | p->verify_alg, "verify-alg"); |
3632 | if (IS_ERR(verify_tfm)) { | |
3633 | verify_tfm = NULL; | |
3634 | goto disconnect; | |
3635 | } | |
3636 | } | |
3637 | ||
2ec91e0e | 3638 | if (apv >= 89 && strcmp(old_net_conf->csums_alg, p->csums_alg)) { |
b30ab791 | 3639 | if (device->state.conn == C_WF_REPORT_PARAMS) { |
d0180171 | 3640 | drbd_err(device, "Different csums-alg settings. me=\"%s\" peer=\"%s\"\n", |
2ec91e0e | 3641 | old_net_conf->csums_alg, p->csums_alg); |
b411b363 PR |
3642 | goto disconnect; |
3643 | } | |
b30ab791 | 3644 | csums_tfm = drbd_crypto_alloc_digest_safe(device, |
b411b363 PR |
3645 | p->csums_alg, "csums-alg"); |
3646 | if (IS_ERR(csums_tfm)) { | |
3647 | csums_tfm = NULL; | |
3648 | goto disconnect; | |
3649 | } | |
3650 | } | |
3651 | ||
813472ce | 3652 | if (apv > 94 && new_disk_conf) { |
daeda1cc PR |
3653 | new_disk_conf->c_plan_ahead = be32_to_cpu(p->c_plan_ahead); |
3654 | new_disk_conf->c_delay_target = be32_to_cpu(p->c_delay_target); | |
3655 | new_disk_conf->c_fill_target = be32_to_cpu(p->c_fill_target); | |
3656 | new_disk_conf->c_max_rate = be32_to_cpu(p->c_max_rate); | |
778f271d | 3657 | |
daeda1cc | 3658 | fifo_size = (new_disk_conf->c_plan_ahead * 10 * SLEEP_TIME) / HZ; |
b30ab791 | 3659 | if (fifo_size != device->rs_plan_s->size) { |
813472ce PR |
3660 | new_plan = fifo_alloc(fifo_size); |
3661 | if (!new_plan) { | |
d0180171 | 3662 | drbd_err(device, "kmalloc of fifo_buffer failed"); |
b30ab791 | 3663 | put_ldev(device); |
778f271d PR |
3664 | goto disconnect; |
3665 | } | |
3666 | } | |
8e26f9cc | 3667 | } |
b411b363 | 3668 | |
91fd4dad | 3669 | if (verify_tfm || csums_tfm) { |
2ec91e0e PR |
3670 | new_net_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL); |
3671 | if (!new_net_conf) { | |
d0180171 | 3672 | drbd_err(device, "Allocation of new net_conf failed\n"); |
91fd4dad PR |
3673 | goto disconnect; |
3674 | } | |
3675 | ||
2ec91e0e | 3676 | *new_net_conf = *old_net_conf; |
91fd4dad PR |
3677 | |
3678 | if (verify_tfm) { | |
2ec91e0e PR |
3679 | strcpy(new_net_conf->verify_alg, p->verify_alg); |
3680 | new_net_conf->verify_alg_len = strlen(p->verify_alg) + 1; | |
9f4fe9ad AG |
3681 | crypto_free_hash(peer_device->connection->verify_tfm); |
3682 | peer_device->connection->verify_tfm = verify_tfm; | |
d0180171 | 3683 | drbd_info(device, "using verify-alg: \"%s\"\n", p->verify_alg); |
91fd4dad PR |
3684 | } |
3685 | if (csums_tfm) { | |
2ec91e0e PR |
3686 | strcpy(new_net_conf->csums_alg, p->csums_alg); |
3687 | new_net_conf->csums_alg_len = strlen(p->csums_alg) + 1; | |
9f4fe9ad AG |
3688 | crypto_free_hash(peer_device->connection->csums_tfm); |
3689 | peer_device->connection->csums_tfm = csums_tfm; | |
d0180171 | 3690 | drbd_info(device, "using csums-alg: \"%s\"\n", p->csums_alg); |
91fd4dad | 3691 | } |
bde89a9e | 3692 | rcu_assign_pointer(connection->net_conf, new_net_conf); |
778f271d | 3693 | } |
b411b363 PR |
3694 | } |
3695 | ||
813472ce | 3696 | if (new_disk_conf) { |
b30ab791 AG |
3697 | rcu_assign_pointer(device->ldev->disk_conf, new_disk_conf); |
3698 | put_ldev(device); | |
813472ce PR |
3699 | } |
3700 | ||
3701 | if (new_plan) { | |
b30ab791 AG |
3702 | old_plan = device->rs_plan_s; |
3703 | rcu_assign_pointer(device->rs_plan_s, new_plan); | |
b411b363 | 3704 | } |
daeda1cc | 3705 | |
0500813f | 3706 | mutex_unlock(&connection->resource->conf_update); |
daeda1cc PR |
3707 | synchronize_rcu(); |
3708 | if (new_net_conf) | |
3709 | kfree(old_net_conf); | |
3710 | kfree(old_disk_conf); | |
813472ce | 3711 | kfree(old_plan); |
daeda1cc | 3712 | |
82bc0194 | 3713 | return 0; |
b411b363 | 3714 | |
813472ce PR |
3715 | reconnect: |
3716 | if (new_disk_conf) { | |
b30ab791 | 3717 | put_ldev(device); |
813472ce PR |
3718 | kfree(new_disk_conf); |
3719 | } | |
0500813f | 3720 | mutex_unlock(&connection->resource->conf_update); |
813472ce PR |
3721 | return -EIO; |
3722 | ||
b411b363 | 3723 | disconnect: |
813472ce PR |
3724 | kfree(new_plan); |
3725 | if (new_disk_conf) { | |
b30ab791 | 3726 | put_ldev(device); |
813472ce PR |
3727 | kfree(new_disk_conf); |
3728 | } | |
0500813f | 3729 | mutex_unlock(&connection->resource->conf_update); |
b411b363 PR |
3730 | /* just for completeness: actually not needed, |
3731 | * as this is not reached if csums_tfm was ok. */ | |
3732 | crypto_free_hash(csums_tfm); | |
3733 | /* but free the verify_tfm again, if csums_tfm did not work out */ | |
3734 | crypto_free_hash(verify_tfm); | |
9f4fe9ad | 3735 | conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD); |
82bc0194 | 3736 | return -EIO; |
b411b363 PR |
3737 | } |
3738 | ||
b411b363 | 3739 | /* warn if the arguments differ by more than 12.5% */ |
b30ab791 | 3740 | static void warn_if_differ_considerably(struct drbd_device *device, |
b411b363 PR |
3741 | const char *s, sector_t a, sector_t b) |
3742 | { | |
3743 | sector_t d; | |
3744 | if (a == 0 || b == 0) | |
3745 | return; | |
3746 | d = (a > b) ? (a - b) : (b - a); | |
3747 | if (d > (a>>3) || d > (b>>3)) | |
d0180171 | 3748 | drbd_warn(device, "Considerable difference in %s: %llus vs. %llus\n", s, |
b411b363 PR |
3749 | (unsigned long long)a, (unsigned long long)b); |
3750 | } | |
3751 | ||
bde89a9e | 3752 | static int receive_sizes(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 3753 | { |
9f4fe9ad | 3754 | struct drbd_peer_device *peer_device; |
b30ab791 | 3755 | struct drbd_device *device; |
e658983a | 3756 | struct p_sizes *p = pi->data; |
e96c9633 | 3757 | enum determine_dev_size dd = DS_UNCHANGED; |
6a8d68b1 | 3758 | sector_t p_size, p_usize, p_csize, my_usize; |
b411b363 | 3759 | int ldsc = 0; /* local disk size changed */ |
e89b591c | 3760 | enum dds_flags ddsf; |
b411b363 | 3761 | |
9f4fe9ad AG |
3762 | peer_device = conn_peer_device(connection, pi->vnr); |
3763 | if (!peer_device) | |
bde89a9e | 3764 | return config_unknown_volume(connection, pi); |
9f4fe9ad | 3765 | device = peer_device->device; |
4a76b161 | 3766 | |
b411b363 PR |
3767 | p_size = be64_to_cpu(p->d_size); |
3768 | p_usize = be64_to_cpu(p->u_size); | |
6a8d68b1 | 3769 | p_csize = be64_to_cpu(p->c_size); |
b411b363 | 3770 | |
b411b363 PR |
3771 | /* just store the peer's disk size for now. |
3772 | * we still need to figure out whether we accept that. */ | |
b30ab791 | 3773 | device->p_size = p_size; |
b411b363 | 3774 | |
b30ab791 | 3775 | if (get_ldev(device)) { |
daeda1cc | 3776 | rcu_read_lock(); |
b30ab791 | 3777 | my_usize = rcu_dereference(device->ldev->disk_conf)->disk_size; |
daeda1cc PR |
3778 | rcu_read_unlock(); |
3779 | ||
b30ab791 AG |
3780 | warn_if_differ_considerably(device, "lower level device sizes", |
3781 | p_size, drbd_get_max_capacity(device->ldev)); | |
3782 | warn_if_differ_considerably(device, "user requested size", | |
daeda1cc | 3783 | p_usize, my_usize); |
b411b363 PR |
3784 | |
3785 | /* if this is the first connect, or an otherwise expected | |
3786 | * param exchange, choose the minimum */ | |
b30ab791 | 3787 | if (device->state.conn == C_WF_REPORT_PARAMS) |
daeda1cc | 3788 | p_usize = min_not_zero(my_usize, p_usize); |
b411b363 PR |
3789 | |
3790 | /* Never shrink a device with usable data during connect. | |
3791 | But allow online shrinking if we are connected. */ | |
b30ab791 AG |
3792 | if (drbd_new_dev_size(device, device->ldev, p_usize, 0) < |
3793 | drbd_get_capacity(device->this_bdev) && | |
3794 | device->state.disk >= D_OUTDATED && | |
3795 | device->state.conn < C_CONNECTED) { | |
d0180171 | 3796 | drbd_err(device, "The peer's disk size is too small!\n"); |
9f4fe9ad | 3797 | conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b30ab791 | 3798 | put_ldev(device); |
82bc0194 | 3799 | return -EIO; |
b411b363 | 3800 | } |
daeda1cc PR |
3801 | |
3802 | if (my_usize != p_usize) { | |
3803 | struct disk_conf *old_disk_conf, *new_disk_conf = NULL; | |
3804 | ||
3805 | new_disk_conf = kzalloc(sizeof(struct disk_conf), GFP_KERNEL); | |
3806 | if (!new_disk_conf) { | |
d0180171 | 3807 | drbd_err(device, "Allocation of new disk_conf failed\n"); |
b30ab791 | 3808 | put_ldev(device); |
daeda1cc PR |
3809 | return -ENOMEM; |
3810 | } | |
3811 | ||
0500813f | 3812 | mutex_lock(&connection->resource->conf_update); |
b30ab791 | 3813 | old_disk_conf = device->ldev->disk_conf; |
daeda1cc PR |
3814 | *new_disk_conf = *old_disk_conf; |
3815 | new_disk_conf->disk_size = p_usize; | |
3816 | ||
b30ab791 | 3817 | rcu_assign_pointer(device->ldev->disk_conf, new_disk_conf); |
0500813f | 3818 | mutex_unlock(&connection->resource->conf_update); |
daeda1cc PR |
3819 | synchronize_rcu(); |
3820 | kfree(old_disk_conf); | |
3821 | ||
d0180171 | 3822 | drbd_info(device, "Peer sets u_size to %lu sectors\n", |
daeda1cc | 3823 | (unsigned long)my_usize); |
b411b363 | 3824 | } |
daeda1cc | 3825 | |
b30ab791 | 3826 | put_ldev(device); |
b411b363 | 3827 | } |
b411b363 | 3828 | |
20c68fde | 3829 | device->peer_max_bio_size = be32_to_cpu(p->max_bio_size); |
20c68fde LE |
3830 | /* Leave drbd_reconsider_max_bio_size() before drbd_determine_dev_size(). |
3831 | In case we cleared the QUEUE_FLAG_DISCARD from our queue in | |
3832 | drbd_reconsider_max_bio_size(), we can be sure that after | |
3833 | drbd_determine_dev_size() no REQ_DISCARDs are in the queue. */ | |
3834 | ||
e89b591c | 3835 | ddsf = be16_to_cpu(p->dds_flags); |
b30ab791 | 3836 | if (get_ldev(device)) { |
8fe39aac | 3837 | drbd_reconsider_max_bio_size(device, device->ldev); |
b30ab791 AG |
3838 | dd = drbd_determine_dev_size(device, ddsf, NULL); |
3839 | put_ldev(device); | |
e96c9633 | 3840 | if (dd == DS_ERROR) |
82bc0194 | 3841 | return -EIO; |
b30ab791 | 3842 | drbd_md_sync(device); |
b411b363 | 3843 | } else { |
6a8d68b1 LE |
3844 | /* |
3845 | * I am diskless, need to accept the peer's *current* size. | |
3846 | * I must NOT accept the peers backing disk size, | |
3847 | * it may have been larger than mine all along... | |
3848 | * | |
3849 | * At this point, the peer knows more about my disk, or at | |
3850 | * least about what we last agreed upon, than myself. | |
3851 | * So if his c_size is less than his d_size, the most likely | |
3852 | * reason is that *my* d_size was smaller last time we checked. | |
3853 | * | |
3854 | * However, if he sends a zero current size, | |
3855 | * take his (user-capped or) backing disk size anyways. | |
3856 | */ | |
8fe39aac | 3857 | drbd_reconsider_max_bio_size(device, NULL); |
6a8d68b1 | 3858 | drbd_set_my_capacity(device, p_csize ?: p_usize ?: p_size); |
b411b363 PR |
3859 | } |
3860 | ||
b30ab791 AG |
3861 | if (get_ldev(device)) { |
3862 | if (device->ldev->known_size != drbd_get_capacity(device->ldev->backing_bdev)) { | |
3863 | device->ldev->known_size = drbd_get_capacity(device->ldev->backing_bdev); | |
b411b363 PR |
3864 | ldsc = 1; |
3865 | } | |
3866 | ||
b30ab791 | 3867 | put_ldev(device); |
b411b363 PR |
3868 | } |
3869 | ||
b30ab791 | 3870 | if (device->state.conn > C_WF_REPORT_PARAMS) { |
b411b363 | 3871 | if (be64_to_cpu(p->c_size) != |
b30ab791 | 3872 | drbd_get_capacity(device->this_bdev) || ldsc) { |
b411b363 PR |
3873 | /* we have different sizes, probably peer |
3874 | * needs to know my new size... */ | |
69a22773 | 3875 | drbd_send_sizes(peer_device, 0, ddsf); |
b411b363 | 3876 | } |
b30ab791 AG |
3877 | if (test_and_clear_bit(RESIZE_PENDING, &device->flags) || |
3878 | (dd == DS_GREW && device->state.conn == C_CONNECTED)) { | |
3879 | if (device->state.pdsk >= D_INCONSISTENT && | |
3880 | device->state.disk >= D_INCONSISTENT) { | |
e89b591c | 3881 | if (ddsf & DDSF_NO_RESYNC) |
d0180171 | 3882 | drbd_info(device, "Resync of new storage suppressed with --assume-clean\n"); |
e89b591c | 3883 | else |
b30ab791 | 3884 | resync_after_online_grow(device); |
e89b591c | 3885 | } else |
b30ab791 | 3886 | set_bit(RESYNC_AFTER_NEG, &device->flags); |
b411b363 PR |
3887 | } |
3888 | } | |
3889 | ||
82bc0194 | 3890 | return 0; |
b411b363 PR |
3891 | } |
3892 | ||
bde89a9e | 3893 | static int receive_uuids(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 3894 | { |
9f4fe9ad | 3895 | struct drbd_peer_device *peer_device; |
b30ab791 | 3896 | struct drbd_device *device; |
e658983a | 3897 | struct p_uuids *p = pi->data; |
b411b363 | 3898 | u64 *p_uuid; |
62b0da3a | 3899 | int i, updated_uuids = 0; |
b411b363 | 3900 | |
9f4fe9ad AG |
3901 | peer_device = conn_peer_device(connection, pi->vnr); |
3902 | if (!peer_device) | |
bde89a9e | 3903 | return config_unknown_volume(connection, pi); |
9f4fe9ad | 3904 | device = peer_device->device; |
4a76b161 | 3905 | |
b411b363 | 3906 | p_uuid = kmalloc(sizeof(u64)*UI_EXTENDED_SIZE, GFP_NOIO); |
063eacf8 | 3907 | if (!p_uuid) { |
d0180171 | 3908 | drbd_err(device, "kmalloc of p_uuid failed\n"); |
063eacf8 JW |
3909 | return false; |
3910 | } | |
b411b363 PR |
3911 | |
3912 | for (i = UI_CURRENT; i < UI_EXTENDED_SIZE; i++) | |
3913 | p_uuid[i] = be64_to_cpu(p->uuid[i]); | |
3914 | ||
b30ab791 AG |
3915 | kfree(device->p_uuid); |
3916 | device->p_uuid = p_uuid; | |
b411b363 | 3917 | |
b30ab791 AG |
3918 | if (device->state.conn < C_CONNECTED && |
3919 | device->state.disk < D_INCONSISTENT && | |
3920 | device->state.role == R_PRIMARY && | |
3921 | (device->ed_uuid & ~((u64)1)) != (p_uuid[UI_CURRENT] & ~((u64)1))) { | |
d0180171 | 3922 | drbd_err(device, "Can only connect to data with current UUID=%016llX\n", |
b30ab791 | 3923 | (unsigned long long)device->ed_uuid); |
9f4fe9ad | 3924 | conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD); |
82bc0194 | 3925 | return -EIO; |
b411b363 PR |
3926 | } |
3927 | ||
b30ab791 | 3928 | if (get_ldev(device)) { |
b411b363 | 3929 | int skip_initial_sync = |
b30ab791 | 3930 | device->state.conn == C_CONNECTED && |
9f4fe9ad | 3931 | peer_device->connection->agreed_pro_version >= 90 && |
b30ab791 | 3932 | device->ldev->md.uuid[UI_CURRENT] == UUID_JUST_CREATED && |
b411b363 PR |
3933 | (p_uuid[UI_FLAGS] & 8); |
3934 | if (skip_initial_sync) { | |
d0180171 | 3935 | drbd_info(device, "Accepted new current UUID, preparing to skip initial sync\n"); |
b30ab791 | 3936 | drbd_bitmap_io(device, &drbd_bmio_clear_n_write, |
20ceb2b2 LE |
3937 | "clear_n_write from receive_uuids", |
3938 | BM_LOCKED_TEST_ALLOWED); | |
b30ab791 AG |
3939 | _drbd_uuid_set(device, UI_CURRENT, p_uuid[UI_CURRENT]); |
3940 | _drbd_uuid_set(device, UI_BITMAP, 0); | |
3941 | _drbd_set_state(_NS2(device, disk, D_UP_TO_DATE, pdsk, D_UP_TO_DATE), | |
b411b363 | 3942 | CS_VERBOSE, NULL); |
b30ab791 | 3943 | drbd_md_sync(device); |
62b0da3a | 3944 | updated_uuids = 1; |
b411b363 | 3945 | } |
b30ab791 AG |
3946 | put_ldev(device); |
3947 | } else if (device->state.disk < D_INCONSISTENT && | |
3948 | device->state.role == R_PRIMARY) { | |
18a50fa2 PR |
3949 | /* I am a diskless primary, the peer just created a new current UUID |
3950 | for me. */ | |
b30ab791 | 3951 | updated_uuids = drbd_set_ed_uuid(device, p_uuid[UI_CURRENT]); |
b411b363 PR |
3952 | } |
3953 | ||
3954 | /* Before we test for the disk state, we should wait until an eventually | |
3955 | ongoing cluster wide state change is finished. That is important if | |
3956 | we are primary and are detaching from our disk. We need to see the | |
3957 | new disk state... */ | |
b30ab791 AG |
3958 | mutex_lock(device->state_mutex); |
3959 | mutex_unlock(device->state_mutex); | |
3960 | if (device->state.conn >= C_CONNECTED && device->state.disk < D_INCONSISTENT) | |
3961 | updated_uuids |= drbd_set_ed_uuid(device, p_uuid[UI_CURRENT]); | |
62b0da3a LE |
3962 | |
3963 | if (updated_uuids) | |
b30ab791 | 3964 | drbd_print_uuids(device, "receiver updated UUIDs to"); |
b411b363 | 3965 | |
82bc0194 | 3966 | return 0; |
b411b363 PR |
3967 | } |
3968 | ||
3969 | /** | |
3970 | * convert_state() - Converts the peer's view of the cluster state to our point of view | |
3971 | * @ps: The state as seen by the peer. | |
3972 | */ | |
3973 | static union drbd_state convert_state(union drbd_state ps) | |
3974 | { | |
3975 | union drbd_state ms; | |
3976 | ||
3977 | static enum drbd_conns c_tab[] = { | |
369bea63 | 3978 | [C_WF_REPORT_PARAMS] = C_WF_REPORT_PARAMS, |
b411b363 PR |
3979 | [C_CONNECTED] = C_CONNECTED, |
3980 | ||
3981 | [C_STARTING_SYNC_S] = C_STARTING_SYNC_T, | |
3982 | [C_STARTING_SYNC_T] = C_STARTING_SYNC_S, | |
3983 | [C_DISCONNECTING] = C_TEAR_DOWN, /* C_NETWORK_FAILURE, */ | |
3984 | [C_VERIFY_S] = C_VERIFY_T, | |
3985 | [C_MASK] = C_MASK, | |
3986 | }; | |
3987 | ||
3988 | ms.i = ps.i; | |
3989 | ||
3990 | ms.conn = c_tab[ps.conn]; | |
3991 | ms.peer = ps.role; | |
3992 | ms.role = ps.peer; | |
3993 | ms.pdsk = ps.disk; | |
3994 | ms.disk = ps.pdsk; | |
3995 | ms.peer_isp = (ps.aftr_isp | ps.user_isp); | |
3996 | ||
3997 | return ms; | |
3998 | } | |
3999 | ||
bde89a9e | 4000 | static int receive_req_state(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4001 | { |
9f4fe9ad | 4002 | struct drbd_peer_device *peer_device; |
b30ab791 | 4003 | struct drbd_device *device; |
e658983a | 4004 | struct p_req_state *p = pi->data; |
b411b363 | 4005 | union drbd_state mask, val; |
bf885f8a | 4006 | enum drbd_state_rv rv; |
b411b363 | 4007 | |
9f4fe9ad AG |
4008 | peer_device = conn_peer_device(connection, pi->vnr); |
4009 | if (!peer_device) | |
4a76b161 | 4010 | return -EIO; |
9f4fe9ad | 4011 | device = peer_device->device; |
4a76b161 | 4012 | |
b411b363 PR |
4013 | mask.i = be32_to_cpu(p->mask); |
4014 | val.i = be32_to_cpu(p->val); | |
4015 | ||
9f4fe9ad | 4016 | if (test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags) && |
b30ab791 | 4017 | mutex_is_locked(device->state_mutex)) { |
69a22773 | 4018 | drbd_send_sr_reply(peer_device, SS_CONCURRENT_ST_CHG); |
82bc0194 | 4019 | return 0; |
b411b363 PR |
4020 | } |
4021 | ||
4022 | mask = convert_state(mask); | |
4023 | val = convert_state(val); | |
4024 | ||
b30ab791 | 4025 | rv = drbd_change_state(device, CS_VERBOSE, mask, val); |
69a22773 | 4026 | drbd_send_sr_reply(peer_device, rv); |
b411b363 | 4027 | |
b30ab791 | 4028 | drbd_md_sync(device); |
b411b363 | 4029 | |
82bc0194 | 4030 | return 0; |
b411b363 PR |
4031 | } |
4032 | ||
bde89a9e | 4033 | static int receive_req_conn_state(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4034 | { |
e658983a | 4035 | struct p_req_state *p = pi->data; |
b411b363 | 4036 | union drbd_state mask, val; |
bf885f8a | 4037 | enum drbd_state_rv rv; |
b411b363 | 4038 | |
b411b363 PR |
4039 | mask.i = be32_to_cpu(p->mask); |
4040 | val.i = be32_to_cpu(p->val); | |
4041 | ||
bde89a9e AG |
4042 | if (test_bit(RESOLVE_CONFLICTS, &connection->flags) && |
4043 | mutex_is_locked(&connection->cstate_mutex)) { | |
4044 | conn_send_sr_reply(connection, SS_CONCURRENT_ST_CHG); | |
82bc0194 | 4045 | return 0; |
b411b363 PR |
4046 | } |
4047 | ||
4048 | mask = convert_state(mask); | |
4049 | val = convert_state(val); | |
4050 | ||
bde89a9e AG |
4051 | rv = conn_request_state(connection, mask, val, CS_VERBOSE | CS_LOCAL_ONLY | CS_IGN_OUTD_FAIL); |
4052 | conn_send_sr_reply(connection, rv); | |
b411b363 | 4053 | |
82bc0194 | 4054 | return 0; |
b411b363 PR |
4055 | } |
4056 | ||
bde89a9e | 4057 | static int receive_state(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4058 | { |
9f4fe9ad | 4059 | struct drbd_peer_device *peer_device; |
b30ab791 | 4060 | struct drbd_device *device; |
e658983a | 4061 | struct p_state *p = pi->data; |
4ac4aada | 4062 | union drbd_state os, ns, peer_state; |
b411b363 | 4063 | enum drbd_disk_state real_peer_disk; |
65d922c3 | 4064 | enum chg_state_flags cs_flags; |
b411b363 PR |
4065 | int rv; |
4066 | ||
9f4fe9ad AG |
4067 | peer_device = conn_peer_device(connection, pi->vnr); |
4068 | if (!peer_device) | |
bde89a9e | 4069 | return config_unknown_volume(connection, pi); |
9f4fe9ad | 4070 | device = peer_device->device; |
4a76b161 | 4071 | |
b411b363 PR |
4072 | peer_state.i = be32_to_cpu(p->state); |
4073 | ||
4074 | real_peer_disk = peer_state.disk; | |
4075 | if (peer_state.disk == D_NEGOTIATING) { | |
b30ab791 | 4076 | real_peer_disk = device->p_uuid[UI_FLAGS] & 4 ? D_INCONSISTENT : D_CONSISTENT; |
d0180171 | 4077 | drbd_info(device, "real peer disk state = %s\n", drbd_disk_str(real_peer_disk)); |
b411b363 PR |
4078 | } |
4079 | ||
0500813f | 4080 | spin_lock_irq(&device->resource->req_lock); |
b411b363 | 4081 | retry: |
b30ab791 | 4082 | os = ns = drbd_read_state(device); |
0500813f | 4083 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 | 4084 | |
668700b4 | 4085 | /* If some other part of the code (ack_receiver thread, timeout) |
545752d5 LE |
4086 | * already decided to close the connection again, |
4087 | * we must not "re-establish" it here. */ | |
4088 | if (os.conn <= C_TEAR_DOWN) | |
58ffa580 | 4089 | return -ECONNRESET; |
545752d5 | 4090 | |
40424e4a LE |
4091 | /* If this is the "end of sync" confirmation, usually the peer disk |
4092 | * transitions from D_INCONSISTENT to D_UP_TO_DATE. For empty (0 bits | |
4093 | * set) resync started in PausedSyncT, or if the timing of pause-/ | |
4094 | * unpause-sync events has been "just right", the peer disk may | |
4095 | * transition from D_CONSISTENT to D_UP_TO_DATE as well. | |
4096 | */ | |
4097 | if ((os.pdsk == D_INCONSISTENT || os.pdsk == D_CONSISTENT) && | |
4098 | real_peer_disk == D_UP_TO_DATE && | |
e9ef7bb6 LE |
4099 | os.conn > C_CONNECTED && os.disk == D_UP_TO_DATE) { |
4100 | /* If we are (becoming) SyncSource, but peer is still in sync | |
4101 | * preparation, ignore its uptodate-ness to avoid flapping, it | |
4102 | * will change to inconsistent once the peer reaches active | |
4103 | * syncing states. | |
4104 | * It may have changed syncer-paused flags, however, so we | |
4105 | * cannot ignore this completely. */ | |
4106 | if (peer_state.conn > C_CONNECTED && | |
4107 | peer_state.conn < C_SYNC_SOURCE) | |
4108 | real_peer_disk = D_INCONSISTENT; | |
4109 | ||
4110 | /* if peer_state changes to connected at the same time, | |
4111 | * it explicitly notifies us that it finished resync. | |
4112 | * Maybe we should finish it up, too? */ | |
4113 | else if (os.conn >= C_SYNC_SOURCE && | |
4114 | peer_state.conn == C_CONNECTED) { | |
b30ab791 AG |
4115 | if (drbd_bm_total_weight(device) <= device->rs_failed) |
4116 | drbd_resync_finished(device); | |
82bc0194 | 4117 | return 0; |
e9ef7bb6 LE |
4118 | } |
4119 | } | |
4120 | ||
02b91b55 LE |
4121 | /* explicit verify finished notification, stop sector reached. */ |
4122 | if (os.conn == C_VERIFY_T && os.disk == D_UP_TO_DATE && | |
4123 | peer_state.conn == C_CONNECTED && real_peer_disk == D_UP_TO_DATE) { | |
b30ab791 AG |
4124 | ov_out_of_sync_print(device); |
4125 | drbd_resync_finished(device); | |
58ffa580 | 4126 | return 0; |
02b91b55 LE |
4127 | } |
4128 | ||
e9ef7bb6 LE |
4129 | /* peer says his disk is inconsistent, while we think it is uptodate, |
4130 | * and this happens while the peer still thinks we have a sync going on, | |
4131 | * but we think we are already done with the sync. | |
4132 | * We ignore this to avoid flapping pdsk. | |
4133 | * This should not happen, if the peer is a recent version of drbd. */ | |
4134 | if (os.pdsk == D_UP_TO_DATE && real_peer_disk == D_INCONSISTENT && | |
4135 | os.conn == C_CONNECTED && peer_state.conn > C_SYNC_SOURCE) | |
4136 | real_peer_disk = D_UP_TO_DATE; | |
4137 | ||
4ac4aada LE |
4138 | if (ns.conn == C_WF_REPORT_PARAMS) |
4139 | ns.conn = C_CONNECTED; | |
b411b363 | 4140 | |
67531718 PR |
4141 | if (peer_state.conn == C_AHEAD) |
4142 | ns.conn = C_BEHIND; | |
4143 | ||
b30ab791 AG |
4144 | if (device->p_uuid && peer_state.disk >= D_NEGOTIATING && |
4145 | get_ldev_if_state(device, D_NEGOTIATING)) { | |
b411b363 PR |
4146 | int cr; /* consider resync */ |
4147 | ||
4148 | /* if we established a new connection */ | |
4ac4aada | 4149 | cr = (os.conn < C_CONNECTED); |
b411b363 PR |
4150 | /* if we had an established connection |
4151 | * and one of the nodes newly attaches a disk */ | |
4ac4aada | 4152 | cr |= (os.conn == C_CONNECTED && |
b411b363 | 4153 | (peer_state.disk == D_NEGOTIATING || |
4ac4aada | 4154 | os.disk == D_NEGOTIATING)); |
b411b363 PR |
4155 | /* if we have both been inconsistent, and the peer has been |
4156 | * forced to be UpToDate with --overwrite-data */ | |
b30ab791 | 4157 | cr |= test_bit(CONSIDER_RESYNC, &device->flags); |
b411b363 PR |
4158 | /* if we had been plain connected, and the admin requested to |
4159 | * start a sync by "invalidate" or "invalidate-remote" */ | |
4ac4aada | 4160 | cr |= (os.conn == C_CONNECTED && |
b411b363 PR |
4161 | (peer_state.conn >= C_STARTING_SYNC_S && |
4162 | peer_state.conn <= C_WF_BITMAP_T)); | |
4163 | ||
4164 | if (cr) | |
69a22773 | 4165 | ns.conn = drbd_sync_handshake(peer_device, peer_state.role, real_peer_disk); |
b411b363 | 4166 | |
b30ab791 | 4167 | put_ldev(device); |
4ac4aada LE |
4168 | if (ns.conn == C_MASK) { |
4169 | ns.conn = C_CONNECTED; | |
b30ab791 AG |
4170 | if (device->state.disk == D_NEGOTIATING) { |
4171 | drbd_force_state(device, NS(disk, D_FAILED)); | |
b411b363 | 4172 | } else if (peer_state.disk == D_NEGOTIATING) { |
d0180171 | 4173 | drbd_err(device, "Disk attach process on the peer node was aborted.\n"); |
b411b363 | 4174 | peer_state.disk = D_DISKLESS; |
580b9767 | 4175 | real_peer_disk = D_DISKLESS; |
b411b363 | 4176 | } else { |
9f4fe9ad | 4177 | if (test_and_clear_bit(CONN_DRY_RUN, &peer_device->connection->flags)) |
82bc0194 | 4178 | return -EIO; |
0b0ba1ef | 4179 | D_ASSERT(device, os.conn == C_WF_REPORT_PARAMS); |
9f4fe9ad | 4180 | conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD); |
82bc0194 | 4181 | return -EIO; |
b411b363 PR |
4182 | } |
4183 | } | |
4184 | } | |
4185 | ||
0500813f | 4186 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 4187 | if (os.i != drbd_read_state(device).i) |
b411b363 | 4188 | goto retry; |
b30ab791 | 4189 | clear_bit(CONSIDER_RESYNC, &device->flags); |
b411b363 PR |
4190 | ns.peer = peer_state.role; |
4191 | ns.pdsk = real_peer_disk; | |
4192 | ns.peer_isp = (peer_state.aftr_isp | peer_state.user_isp); | |
4ac4aada | 4193 | if ((ns.conn == C_CONNECTED || ns.conn == C_WF_BITMAP_S) && ns.disk == D_NEGOTIATING) |
b30ab791 | 4194 | ns.disk = device->new_state_tmp.disk; |
4ac4aada | 4195 | cs_flags = CS_VERBOSE + (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED ? 0 : CS_HARD); |
b30ab791 AG |
4196 | if (ns.pdsk == D_CONSISTENT && drbd_suspended(device) && ns.conn == C_CONNECTED && os.conn < C_CONNECTED && |
4197 | test_bit(NEW_CUR_UUID, &device->flags)) { | |
8554df1c | 4198 | /* Do not allow tl_restart(RESEND) for a rebooted peer. We can only allow this |
481c6f50 | 4199 | for temporal network outages! */ |
0500813f | 4200 | spin_unlock_irq(&device->resource->req_lock); |
d0180171 | 4201 | drbd_err(device, "Aborting Connect, can not thaw IO with an only Consistent peer\n"); |
9f4fe9ad | 4202 | tl_clear(peer_device->connection); |
b30ab791 AG |
4203 | drbd_uuid_new_current(device); |
4204 | clear_bit(NEW_CUR_UUID, &device->flags); | |
9f4fe9ad | 4205 | conn_request_state(peer_device->connection, NS2(conn, C_PROTOCOL_ERROR, susp, 0), CS_HARD); |
82bc0194 | 4206 | return -EIO; |
481c6f50 | 4207 | } |
b30ab791 AG |
4208 | rv = _drbd_set_state(device, ns, cs_flags, NULL); |
4209 | ns = drbd_read_state(device); | |
0500813f | 4210 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 PR |
4211 | |
4212 | if (rv < SS_SUCCESS) { | |
9f4fe9ad | 4213 | conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD); |
82bc0194 | 4214 | return -EIO; |
b411b363 PR |
4215 | } |
4216 | ||
4ac4aada LE |
4217 | if (os.conn > C_WF_REPORT_PARAMS) { |
4218 | if (ns.conn > C_CONNECTED && peer_state.conn <= C_CONNECTED && | |
b411b363 PR |
4219 | peer_state.disk != D_NEGOTIATING ) { |
4220 | /* we want resync, peer has not yet decided to sync... */ | |
4221 | /* Nowadays only used when forcing a node into primary role and | |
4222 | setting its disk to UpToDate with that */ | |
69a22773 AG |
4223 | drbd_send_uuids(peer_device); |
4224 | drbd_send_current_state(peer_device); | |
b411b363 PR |
4225 | } |
4226 | } | |
4227 | ||
b30ab791 | 4228 | clear_bit(DISCARD_MY_DATA, &device->flags); |
b411b363 | 4229 | |
b30ab791 | 4230 | drbd_md_sync(device); /* update connected indicator, la_size_sect, ... */ |
b411b363 | 4231 | |
82bc0194 | 4232 | return 0; |
b411b363 PR |
4233 | } |
4234 | ||
bde89a9e | 4235 | static int receive_sync_uuid(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4236 | { |
9f4fe9ad | 4237 | struct drbd_peer_device *peer_device; |
b30ab791 | 4238 | struct drbd_device *device; |
e658983a | 4239 | struct p_rs_uuid *p = pi->data; |
4a76b161 | 4240 | |
9f4fe9ad AG |
4241 | peer_device = conn_peer_device(connection, pi->vnr); |
4242 | if (!peer_device) | |
4a76b161 | 4243 | return -EIO; |
9f4fe9ad | 4244 | device = peer_device->device; |
b411b363 | 4245 | |
b30ab791 AG |
4246 | wait_event(device->misc_wait, |
4247 | device->state.conn == C_WF_SYNC_UUID || | |
4248 | device->state.conn == C_BEHIND || | |
4249 | device->state.conn < C_CONNECTED || | |
4250 | device->state.disk < D_NEGOTIATING); | |
b411b363 | 4251 | |
0b0ba1ef | 4252 | /* D_ASSERT(device, device->state.conn == C_WF_SYNC_UUID ); */ |
b411b363 | 4253 | |
b411b363 PR |
4254 | /* Here the _drbd_uuid_ functions are right, current should |
4255 | _not_ be rotated into the history */ | |
b30ab791 AG |
4256 | if (get_ldev_if_state(device, D_NEGOTIATING)) { |
4257 | _drbd_uuid_set(device, UI_CURRENT, be64_to_cpu(p->uuid)); | |
4258 | _drbd_uuid_set(device, UI_BITMAP, 0UL); | |
b411b363 | 4259 | |
b30ab791 AG |
4260 | drbd_print_uuids(device, "updated sync uuid"); |
4261 | drbd_start_resync(device, C_SYNC_TARGET); | |
b411b363 | 4262 | |
b30ab791 | 4263 | put_ldev(device); |
b411b363 | 4264 | } else |
d0180171 | 4265 | drbd_err(device, "Ignoring SyncUUID packet!\n"); |
b411b363 | 4266 | |
82bc0194 | 4267 | return 0; |
b411b363 PR |
4268 | } |
4269 | ||
2c46407d AG |
4270 | /** |
4271 | * receive_bitmap_plain | |
4272 | * | |
4273 | * Return 0 when done, 1 when another iteration is needed, and a negative error | |
4274 | * code upon failure. | |
4275 | */ | |
4276 | static int | |
69a22773 | 4277 | receive_bitmap_plain(struct drbd_peer_device *peer_device, unsigned int size, |
e658983a | 4278 | unsigned long *p, struct bm_xfer_ctx *c) |
b411b363 | 4279 | { |
50d0b1ad | 4280 | unsigned int data_size = DRBD_SOCKET_BUFFER_SIZE - |
69a22773 | 4281 | drbd_header_size(peer_device->connection); |
e658983a | 4282 | unsigned int num_words = min_t(size_t, data_size / sizeof(*p), |
50d0b1ad | 4283 | c->bm_words - c->word_offset); |
e658983a | 4284 | unsigned int want = num_words * sizeof(*p); |
2c46407d | 4285 | int err; |
b411b363 | 4286 | |
50d0b1ad | 4287 | if (want != size) { |
69a22773 | 4288 | drbd_err(peer_device, "%s:want (%u) != size (%u)\n", __func__, want, size); |
2c46407d | 4289 | return -EIO; |
b411b363 PR |
4290 | } |
4291 | if (want == 0) | |
2c46407d | 4292 | return 0; |
69a22773 | 4293 | err = drbd_recv_all(peer_device->connection, p, want); |
82bc0194 | 4294 | if (err) |
2c46407d | 4295 | return err; |
b411b363 | 4296 | |
69a22773 | 4297 | drbd_bm_merge_lel(peer_device->device, c->word_offset, num_words, p); |
b411b363 PR |
4298 | |
4299 | c->word_offset += num_words; | |
4300 | c->bit_offset = c->word_offset * BITS_PER_LONG; | |
4301 | if (c->bit_offset > c->bm_bits) | |
4302 | c->bit_offset = c->bm_bits; | |
4303 | ||
2c46407d | 4304 | return 1; |
b411b363 PR |
4305 | } |
4306 | ||
a02d1240 AG |
4307 | static enum drbd_bitmap_code dcbp_get_code(struct p_compressed_bm *p) |
4308 | { | |
4309 | return (enum drbd_bitmap_code)(p->encoding & 0x0f); | |
4310 | } | |
4311 | ||
4312 | static int dcbp_get_start(struct p_compressed_bm *p) | |
4313 | { | |
4314 | return (p->encoding & 0x80) != 0; | |
4315 | } | |
4316 | ||
4317 | static int dcbp_get_pad_bits(struct p_compressed_bm *p) | |
4318 | { | |
4319 | return (p->encoding >> 4) & 0x7; | |
4320 | } | |
4321 | ||
2c46407d AG |
4322 | /** |
4323 | * recv_bm_rle_bits | |
4324 | * | |
4325 | * Return 0 when done, 1 when another iteration is needed, and a negative error | |
4326 | * code upon failure. | |
4327 | */ | |
4328 | static int | |
69a22773 | 4329 | recv_bm_rle_bits(struct drbd_peer_device *peer_device, |
b411b363 | 4330 | struct p_compressed_bm *p, |
c6d25cfe PR |
4331 | struct bm_xfer_ctx *c, |
4332 | unsigned int len) | |
b411b363 PR |
4333 | { |
4334 | struct bitstream bs; | |
4335 | u64 look_ahead; | |
4336 | u64 rl; | |
4337 | u64 tmp; | |
4338 | unsigned long s = c->bit_offset; | |
4339 | unsigned long e; | |
a02d1240 | 4340 | int toggle = dcbp_get_start(p); |
b411b363 PR |
4341 | int have; |
4342 | int bits; | |
4343 | ||
a02d1240 | 4344 | bitstream_init(&bs, p->code, len, dcbp_get_pad_bits(p)); |
b411b363 PR |
4345 | |
4346 | bits = bitstream_get_bits(&bs, &look_ahead, 64); | |
4347 | if (bits < 0) | |
2c46407d | 4348 | return -EIO; |
b411b363 PR |
4349 | |
4350 | for (have = bits; have > 0; s += rl, toggle = !toggle) { | |
4351 | bits = vli_decode_bits(&rl, look_ahead); | |
4352 | if (bits <= 0) | |
2c46407d | 4353 | return -EIO; |
b411b363 PR |
4354 | |
4355 | if (toggle) { | |
4356 | e = s + rl -1; | |
4357 | if (e >= c->bm_bits) { | |
69a22773 | 4358 | drbd_err(peer_device, "bitmap overflow (e:%lu) while decoding bm RLE packet\n", e); |
2c46407d | 4359 | return -EIO; |
b411b363 | 4360 | } |
69a22773 | 4361 | _drbd_bm_set_bits(peer_device->device, s, e); |
b411b363 PR |
4362 | } |
4363 | ||
4364 | if (have < bits) { | |
69a22773 | 4365 | drbd_err(peer_device, "bitmap decoding error: h:%d b:%d la:0x%08llx l:%u/%u\n", |
b411b363 PR |
4366 | have, bits, look_ahead, |
4367 | (unsigned int)(bs.cur.b - p->code), | |
4368 | (unsigned int)bs.buf_len); | |
2c46407d | 4369 | return -EIO; |
b411b363 | 4370 | } |
d2da5b0c LE |
4371 | /* if we consumed all 64 bits, assign 0; >> 64 is "undefined"; */ |
4372 | if (likely(bits < 64)) | |
4373 | look_ahead >>= bits; | |
4374 | else | |
4375 | look_ahead = 0; | |
b411b363 PR |
4376 | have -= bits; |
4377 | ||
4378 | bits = bitstream_get_bits(&bs, &tmp, 64 - have); | |
4379 | if (bits < 0) | |
2c46407d | 4380 | return -EIO; |
b411b363 PR |
4381 | look_ahead |= tmp << have; |
4382 | have += bits; | |
4383 | } | |
4384 | ||
4385 | c->bit_offset = s; | |
4386 | bm_xfer_ctx_bit_to_word_offset(c); | |
4387 | ||
2c46407d | 4388 | return (s != c->bm_bits); |
b411b363 PR |
4389 | } |
4390 | ||
2c46407d AG |
4391 | /** |
4392 | * decode_bitmap_c | |
4393 | * | |
4394 | * Return 0 when done, 1 when another iteration is needed, and a negative error | |
4395 | * code upon failure. | |
4396 | */ | |
4397 | static int | |
69a22773 | 4398 | decode_bitmap_c(struct drbd_peer_device *peer_device, |
b411b363 | 4399 | struct p_compressed_bm *p, |
c6d25cfe PR |
4400 | struct bm_xfer_ctx *c, |
4401 | unsigned int len) | |
b411b363 | 4402 | { |
a02d1240 | 4403 | if (dcbp_get_code(p) == RLE_VLI_Bits) |
69a22773 | 4404 | return recv_bm_rle_bits(peer_device, p, c, len - sizeof(*p)); |
b411b363 PR |
4405 | |
4406 | /* other variants had been implemented for evaluation, | |
4407 | * but have been dropped as this one turned out to be "best" | |
4408 | * during all our tests. */ | |
4409 | ||
69a22773 AG |
4410 | drbd_err(peer_device, "receive_bitmap_c: unknown encoding %u\n", p->encoding); |
4411 | conn_request_state(peer_device->connection, NS(conn, C_PROTOCOL_ERROR), CS_HARD); | |
2c46407d | 4412 | return -EIO; |
b411b363 PR |
4413 | } |
4414 | ||
b30ab791 | 4415 | void INFO_bm_xfer_stats(struct drbd_device *device, |
b411b363 PR |
4416 | const char *direction, struct bm_xfer_ctx *c) |
4417 | { | |
4418 | /* what would it take to transfer it "plaintext" */ | |
a6b32bc3 | 4419 | unsigned int header_size = drbd_header_size(first_peer_device(device)->connection); |
50d0b1ad AG |
4420 | unsigned int data_size = DRBD_SOCKET_BUFFER_SIZE - header_size; |
4421 | unsigned int plain = | |
4422 | header_size * (DIV_ROUND_UP(c->bm_words, data_size) + 1) + | |
4423 | c->bm_words * sizeof(unsigned long); | |
4424 | unsigned int total = c->bytes[0] + c->bytes[1]; | |
4425 | unsigned int r; | |
b411b363 PR |
4426 | |
4427 | /* total can not be zero. but just in case: */ | |
4428 | if (total == 0) | |
4429 | return; | |
4430 | ||
4431 | /* don't report if not compressed */ | |
4432 | if (total >= plain) | |
4433 | return; | |
4434 | ||
4435 | /* total < plain. check for overflow, still */ | |
4436 | r = (total > UINT_MAX/1000) ? (total / (plain/1000)) | |
4437 | : (1000 * total / plain); | |
4438 | ||
4439 | if (r > 1000) | |
4440 | r = 1000; | |
4441 | ||
4442 | r = 1000 - r; | |
d0180171 | 4443 | drbd_info(device, "%s bitmap stats [Bytes(packets)]: plain %u(%u), RLE %u(%u), " |
b411b363 PR |
4444 | "total %u; compression: %u.%u%%\n", |
4445 | direction, | |
4446 | c->bytes[1], c->packets[1], | |
4447 | c->bytes[0], c->packets[0], | |
4448 | total, r/10, r % 10); | |
4449 | } | |
4450 | ||
4451 | /* Since we are processing the bitfield from lower addresses to higher, | |
4452 | it does not matter if the process it in 32 bit chunks or 64 bit | |
4453 | chunks as long as it is little endian. (Understand it as byte stream, | |
4454 | beginning with the lowest byte...) If we would use big endian | |
4455 | we would need to process it from the highest address to the lowest, | |
4456 | in order to be agnostic to the 32 vs 64 bits issue. | |
4457 | ||
4458 | returns 0 on failure, 1 if we successfully received it. */ | |
bde89a9e | 4459 | static int receive_bitmap(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4460 | { |
9f4fe9ad | 4461 | struct drbd_peer_device *peer_device; |
b30ab791 | 4462 | struct drbd_device *device; |
b411b363 | 4463 | struct bm_xfer_ctx c; |
2c46407d | 4464 | int err; |
4a76b161 | 4465 | |
9f4fe9ad AG |
4466 | peer_device = conn_peer_device(connection, pi->vnr); |
4467 | if (!peer_device) | |
4a76b161 | 4468 | return -EIO; |
9f4fe9ad | 4469 | device = peer_device->device; |
b411b363 | 4470 | |
b30ab791 | 4471 | drbd_bm_lock(device, "receive bitmap", BM_LOCKED_SET_ALLOWED); |
20ceb2b2 LE |
4472 | /* you are supposed to send additional out-of-sync information |
4473 | * if you actually set bits during this phase */ | |
b411b363 | 4474 | |
b411b363 | 4475 | c = (struct bm_xfer_ctx) { |
b30ab791 AG |
4476 | .bm_bits = drbd_bm_bits(device), |
4477 | .bm_words = drbd_bm_words(device), | |
b411b363 PR |
4478 | }; |
4479 | ||
2c46407d | 4480 | for(;;) { |
e658983a | 4481 | if (pi->cmd == P_BITMAP) |
69a22773 | 4482 | err = receive_bitmap_plain(peer_device, pi->size, pi->data, &c); |
e658983a | 4483 | else if (pi->cmd == P_COMPRESSED_BITMAP) { |
b411b363 PR |
4484 | /* MAYBE: sanity check that we speak proto >= 90, |
4485 | * and the feature is enabled! */ | |
e658983a | 4486 | struct p_compressed_bm *p = pi->data; |
b411b363 | 4487 | |
bde89a9e | 4488 | if (pi->size > DRBD_SOCKET_BUFFER_SIZE - drbd_header_size(connection)) { |
d0180171 | 4489 | drbd_err(device, "ReportCBitmap packet too large\n"); |
82bc0194 | 4490 | err = -EIO; |
b411b363 PR |
4491 | goto out; |
4492 | } | |
e658983a | 4493 | if (pi->size <= sizeof(*p)) { |
d0180171 | 4494 | drbd_err(device, "ReportCBitmap packet too small (l:%u)\n", pi->size); |
82bc0194 | 4495 | err = -EIO; |
78fcbdae | 4496 | goto out; |
b411b363 | 4497 | } |
9f4fe9ad | 4498 | err = drbd_recv_all(peer_device->connection, p, pi->size); |
e658983a AG |
4499 | if (err) |
4500 | goto out; | |
69a22773 | 4501 | err = decode_bitmap_c(peer_device, p, &c, pi->size); |
b411b363 | 4502 | } else { |
d0180171 | 4503 | drbd_warn(device, "receive_bitmap: cmd neither ReportBitMap nor ReportCBitMap (is 0x%x)", pi->cmd); |
82bc0194 | 4504 | err = -EIO; |
b411b363 PR |
4505 | goto out; |
4506 | } | |
4507 | ||
e2857216 | 4508 | c.packets[pi->cmd == P_BITMAP]++; |
bde89a9e | 4509 | c.bytes[pi->cmd == P_BITMAP] += drbd_header_size(connection) + pi->size; |
b411b363 | 4510 | |
2c46407d AG |
4511 | if (err <= 0) { |
4512 | if (err < 0) | |
4513 | goto out; | |
b411b363 | 4514 | break; |
2c46407d | 4515 | } |
9f4fe9ad | 4516 | err = drbd_recv_header(peer_device->connection, pi); |
82bc0194 | 4517 | if (err) |
b411b363 | 4518 | goto out; |
2c46407d | 4519 | } |
b411b363 | 4520 | |
b30ab791 | 4521 | INFO_bm_xfer_stats(device, "receive", &c); |
b411b363 | 4522 | |
b30ab791 | 4523 | if (device->state.conn == C_WF_BITMAP_T) { |
de1f8e4a AG |
4524 | enum drbd_state_rv rv; |
4525 | ||
b30ab791 | 4526 | err = drbd_send_bitmap(device); |
82bc0194 | 4527 | if (err) |
b411b363 PR |
4528 | goto out; |
4529 | /* Omit CS_ORDERED with this state transition to avoid deadlocks. */ | |
b30ab791 | 4530 | rv = _drbd_request_state(device, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE); |
0b0ba1ef | 4531 | D_ASSERT(device, rv == SS_SUCCESS); |
b30ab791 | 4532 | } else if (device->state.conn != C_WF_BITMAP_S) { |
b411b363 PR |
4533 | /* admin may have requested C_DISCONNECTING, |
4534 | * other threads may have noticed network errors */ | |
d0180171 | 4535 | drbd_info(device, "unexpected cstate (%s) in receive_bitmap\n", |
b30ab791 | 4536 | drbd_conn_str(device->state.conn)); |
b411b363 | 4537 | } |
82bc0194 | 4538 | err = 0; |
b411b363 | 4539 | |
b411b363 | 4540 | out: |
b30ab791 AG |
4541 | drbd_bm_unlock(device); |
4542 | if (!err && device->state.conn == C_WF_BITMAP_S) | |
4543 | drbd_start_resync(device, C_SYNC_SOURCE); | |
82bc0194 | 4544 | return err; |
b411b363 PR |
4545 | } |
4546 | ||
bde89a9e | 4547 | static int receive_skip(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 4548 | { |
1ec861eb | 4549 | drbd_warn(connection, "skipping unknown optional packet type %d, l: %d!\n", |
e2857216 | 4550 | pi->cmd, pi->size); |
b411b363 | 4551 | |
bde89a9e | 4552 | return ignore_remaining_packet(connection, pi); |
b411b363 PR |
4553 | } |
4554 | ||
bde89a9e | 4555 | static int receive_UnplugRemote(struct drbd_connection *connection, struct packet_info *pi) |
0ced55a3 | 4556 | { |
e7f52dfb LE |
4557 | /* Make sure we've acked all the TCP data associated |
4558 | * with the data requests being unplugged */ | |
bde89a9e | 4559 | drbd_tcp_quickack(connection->data.socket); |
0ced55a3 | 4560 | |
82bc0194 | 4561 | return 0; |
0ced55a3 PR |
4562 | } |
4563 | ||
bde89a9e | 4564 | static int receive_out_of_sync(struct drbd_connection *connection, struct packet_info *pi) |
73a01a18 | 4565 | { |
9f4fe9ad | 4566 | struct drbd_peer_device *peer_device; |
b30ab791 | 4567 | struct drbd_device *device; |
e658983a | 4568 | struct p_block_desc *p = pi->data; |
4a76b161 | 4569 | |
9f4fe9ad AG |
4570 | peer_device = conn_peer_device(connection, pi->vnr); |
4571 | if (!peer_device) | |
4a76b161 | 4572 | return -EIO; |
9f4fe9ad | 4573 | device = peer_device->device; |
73a01a18 | 4574 | |
b30ab791 | 4575 | switch (device->state.conn) { |
f735e363 LE |
4576 | case C_WF_SYNC_UUID: |
4577 | case C_WF_BITMAP_T: | |
4578 | case C_BEHIND: | |
4579 | break; | |
4580 | default: | |
d0180171 | 4581 | drbd_err(device, "ASSERT FAILED cstate = %s, expected: WFSyncUUID|WFBitMapT|Behind\n", |
b30ab791 | 4582 | drbd_conn_str(device->state.conn)); |
f735e363 LE |
4583 | } |
4584 | ||
b30ab791 | 4585 | drbd_set_out_of_sync(device, be64_to_cpu(p->sector), be32_to_cpu(p->blksize)); |
73a01a18 | 4586 | |
82bc0194 | 4587 | return 0; |
73a01a18 PR |
4588 | } |
4589 | ||
02918be2 PR |
4590 | struct data_cmd { |
4591 | int expect_payload; | |
4592 | size_t pkt_size; | |
bde89a9e | 4593 | int (*fn)(struct drbd_connection *, struct packet_info *); |
02918be2 PR |
4594 | }; |
4595 | ||
4596 | static struct data_cmd drbd_cmd_handler[] = { | |
4597 | [P_DATA] = { 1, sizeof(struct p_data), receive_Data }, | |
4598 | [P_DATA_REPLY] = { 1, sizeof(struct p_data), receive_DataReply }, | |
4599 | [P_RS_DATA_REPLY] = { 1, sizeof(struct p_data), receive_RSDataReply } , | |
4600 | [P_BARRIER] = { 0, sizeof(struct p_barrier), receive_Barrier } , | |
e658983a AG |
4601 | [P_BITMAP] = { 1, 0, receive_bitmap } , |
4602 | [P_COMPRESSED_BITMAP] = { 1, 0, receive_bitmap } , | |
4603 | [P_UNPLUG_REMOTE] = { 0, 0, receive_UnplugRemote }, | |
02918be2 PR |
4604 | [P_DATA_REQUEST] = { 0, sizeof(struct p_block_req), receive_DataRequest }, |
4605 | [P_RS_DATA_REQUEST] = { 0, sizeof(struct p_block_req), receive_DataRequest }, | |
e658983a AG |
4606 | [P_SYNC_PARAM] = { 1, 0, receive_SyncParam }, |
4607 | [P_SYNC_PARAM89] = { 1, 0, receive_SyncParam }, | |
02918be2 PR |
4608 | [P_PROTOCOL] = { 1, sizeof(struct p_protocol), receive_protocol }, |
4609 | [P_UUIDS] = { 0, sizeof(struct p_uuids), receive_uuids }, | |
4610 | [P_SIZES] = { 0, sizeof(struct p_sizes), receive_sizes }, | |
4611 | [P_STATE] = { 0, sizeof(struct p_state), receive_state }, | |
4612 | [P_STATE_CHG_REQ] = { 0, sizeof(struct p_req_state), receive_req_state }, | |
4613 | [P_SYNC_UUID] = { 0, sizeof(struct p_rs_uuid), receive_sync_uuid }, | |
4614 | [P_OV_REQUEST] = { 0, sizeof(struct p_block_req), receive_DataRequest }, | |
4615 | [P_OV_REPLY] = { 1, sizeof(struct p_block_req), receive_DataRequest }, | |
4616 | [P_CSUM_RS_REQUEST] = { 1, sizeof(struct p_block_req), receive_DataRequest }, | |
4617 | [P_DELAY_PROBE] = { 0, sizeof(struct p_delay_probe93), receive_skip }, | |
73a01a18 | 4618 | [P_OUT_OF_SYNC] = { 0, sizeof(struct p_block_desc), receive_out_of_sync }, |
4a76b161 | 4619 | [P_CONN_ST_CHG_REQ] = { 0, sizeof(struct p_req_state), receive_req_conn_state }, |
036b17ea | 4620 | [P_PROTOCOL_UPDATE] = { 1, sizeof(struct p_protocol), receive_protocol }, |
a0fb3c47 | 4621 | [P_TRIM] = { 0, sizeof(struct p_trim), receive_Data }, |
b411b363 PR |
4622 | }; |
4623 | ||
bde89a9e | 4624 | static void drbdd(struct drbd_connection *connection) |
b411b363 | 4625 | { |
77351055 | 4626 | struct packet_info pi; |
02918be2 | 4627 | size_t shs; /* sub header size */ |
82bc0194 | 4628 | int err; |
b411b363 | 4629 | |
bde89a9e | 4630 | while (get_t_state(&connection->receiver) == RUNNING) { |
deebe195 | 4631 | struct data_cmd *cmd; |
b411b363 | 4632 | |
bde89a9e | 4633 | drbd_thread_current_set_cpu(&connection->receiver); |
944410e9 | 4634 | update_receiver_timing_details(connection, drbd_recv_header); |
bde89a9e | 4635 | if (drbd_recv_header(connection, &pi)) |
02918be2 | 4636 | goto err_out; |
b411b363 | 4637 | |
deebe195 | 4638 | cmd = &drbd_cmd_handler[pi.cmd]; |
4a76b161 | 4639 | if (unlikely(pi.cmd >= ARRAY_SIZE(drbd_cmd_handler) || !cmd->fn)) { |
1ec861eb | 4640 | drbd_err(connection, "Unexpected data packet %s (0x%04x)", |
2fcb8f30 | 4641 | cmdname(pi.cmd), pi.cmd); |
02918be2 | 4642 | goto err_out; |
0b33a916 | 4643 | } |
b411b363 | 4644 | |
e658983a AG |
4645 | shs = cmd->pkt_size; |
4646 | if (pi.size > shs && !cmd->expect_payload) { | |
1ec861eb | 4647 | drbd_err(connection, "No payload expected %s l:%d\n", |
2fcb8f30 | 4648 | cmdname(pi.cmd), pi.size); |
02918be2 | 4649 | goto err_out; |
b411b363 | 4650 | } |
b411b363 | 4651 | |
c13f7e1a | 4652 | if (shs) { |
944410e9 | 4653 | update_receiver_timing_details(connection, drbd_recv_all_warn); |
bde89a9e | 4654 | err = drbd_recv_all_warn(connection, pi.data, shs); |
a5c31904 | 4655 | if (err) |
c13f7e1a | 4656 | goto err_out; |
e2857216 | 4657 | pi.size -= shs; |
c13f7e1a LE |
4658 | } |
4659 | ||
944410e9 | 4660 | update_receiver_timing_details(connection, cmd->fn); |
bde89a9e | 4661 | err = cmd->fn(connection, &pi); |
4a76b161 | 4662 | if (err) { |
1ec861eb | 4663 | drbd_err(connection, "error receiving %s, e: %d l: %d!\n", |
9f5bdc33 | 4664 | cmdname(pi.cmd), err, pi.size); |
02918be2 | 4665 | goto err_out; |
b411b363 PR |
4666 | } |
4667 | } | |
82bc0194 | 4668 | return; |
b411b363 | 4669 | |
82bc0194 | 4670 | err_out: |
bde89a9e | 4671 | conn_request_state(connection, NS(conn, C_PROTOCOL_ERROR), CS_HARD); |
b411b363 PR |
4672 | } |
4673 | ||
bde89a9e | 4674 | static void conn_disconnect(struct drbd_connection *connection) |
b411b363 | 4675 | { |
c06ece6b | 4676 | struct drbd_peer_device *peer_device; |
bbeb641c | 4677 | enum drbd_conns oc; |
376694a0 | 4678 | int vnr; |
b411b363 | 4679 | |
bde89a9e | 4680 | if (connection->cstate == C_STANDALONE) |
b411b363 | 4681 | return; |
b411b363 | 4682 | |
545752d5 LE |
4683 | /* We are about to start the cleanup after connection loss. |
4684 | * Make sure drbd_make_request knows about that. | |
4685 | * Usually we should be in some network failure state already, | |
4686 | * but just in case we are not, we fix it up here. | |
4687 | */ | |
bde89a9e | 4688 | conn_request_state(connection, NS(conn, C_NETWORK_FAILURE), CS_HARD); |
545752d5 | 4689 | |
668700b4 | 4690 | /* ack_receiver does not clean up anything. it must not interfere, either */ |
1c03e520 | 4691 | drbd_thread_stop(&connection->ack_receiver); |
668700b4 PR |
4692 | if (connection->ack_sender) { |
4693 | destroy_workqueue(connection->ack_sender); | |
4694 | connection->ack_sender = NULL; | |
4695 | } | |
bde89a9e | 4696 | drbd_free_sock(connection); |
360cc740 | 4697 | |
c141ebda | 4698 | rcu_read_lock(); |
c06ece6b AG |
4699 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { |
4700 | struct drbd_device *device = peer_device->device; | |
b30ab791 | 4701 | kref_get(&device->kref); |
c141ebda | 4702 | rcu_read_unlock(); |
69a22773 | 4703 | drbd_disconnected(peer_device); |
c06ece6b | 4704 | kref_put(&device->kref, drbd_destroy_device); |
c141ebda PR |
4705 | rcu_read_lock(); |
4706 | } | |
4707 | rcu_read_unlock(); | |
4708 | ||
bde89a9e | 4709 | if (!list_empty(&connection->current_epoch->list)) |
1ec861eb | 4710 | drbd_err(connection, "ASSERTION FAILED: connection->current_epoch->list not empty\n"); |
12038a3a | 4711 | /* ok, no more ee's on the fly, it is safe to reset the epoch_size */ |
bde89a9e AG |
4712 | atomic_set(&connection->current_epoch->epoch_size, 0); |
4713 | connection->send.seen_any_write_yet = false; | |
12038a3a | 4714 | |
1ec861eb | 4715 | drbd_info(connection, "Connection closed\n"); |
360cc740 | 4716 | |
bde89a9e AG |
4717 | if (conn_highest_role(connection) == R_PRIMARY && conn_highest_pdsk(connection) >= D_UNKNOWN) |
4718 | conn_try_outdate_peer_async(connection); | |
cb703454 | 4719 | |
0500813f | 4720 | spin_lock_irq(&connection->resource->req_lock); |
bde89a9e | 4721 | oc = connection->cstate; |
bbeb641c | 4722 | if (oc >= C_UNCONNECTED) |
bde89a9e | 4723 | _conn_request_state(connection, NS(conn, C_UNCONNECTED), CS_VERBOSE); |
bbeb641c | 4724 | |
0500813f | 4725 | spin_unlock_irq(&connection->resource->req_lock); |
360cc740 | 4726 | |
f3dfa40a | 4727 | if (oc == C_DISCONNECTING) |
bde89a9e | 4728 | conn_request_state(connection, NS(conn, C_STANDALONE), CS_VERBOSE | CS_HARD); |
360cc740 PR |
4729 | } |
4730 | ||
69a22773 | 4731 | static int drbd_disconnected(struct drbd_peer_device *peer_device) |
360cc740 | 4732 | { |
69a22773 | 4733 | struct drbd_device *device = peer_device->device; |
360cc740 | 4734 | unsigned int i; |
b411b363 | 4735 | |
85719573 | 4736 | /* wait for current activity to cease. */ |
0500813f | 4737 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 AG |
4738 | _drbd_wait_ee_list_empty(device, &device->active_ee); |
4739 | _drbd_wait_ee_list_empty(device, &device->sync_ee); | |
4740 | _drbd_wait_ee_list_empty(device, &device->read_ee); | |
0500813f | 4741 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 PR |
4742 | |
4743 | /* We do not have data structures that would allow us to | |
4744 | * get the rs_pending_cnt down to 0 again. | |
4745 | * * On C_SYNC_TARGET we do not have any data structures describing | |
4746 | * the pending RSDataRequest's we have sent. | |
4747 | * * On C_SYNC_SOURCE there is no data structure that tracks | |
4748 | * the P_RS_DATA_REPLY blocks that we sent to the SyncTarget. | |
4749 | * And no, it is not the sum of the reference counts in the | |
4750 | * resync_LRU. The resync_LRU tracks the whole operation including | |
4751 | * the disk-IO, while the rs_pending_cnt only tracks the blocks | |
4752 | * on the fly. */ | |
b30ab791 AG |
4753 | drbd_rs_cancel_all(device); |
4754 | device->rs_total = 0; | |
4755 | device->rs_failed = 0; | |
4756 | atomic_set(&device->rs_pending_cnt, 0); | |
4757 | wake_up(&device->misc_wait); | |
b411b363 | 4758 | |
b30ab791 AG |
4759 | del_timer_sync(&device->resync_timer); |
4760 | resync_timer_fn((unsigned long)device); | |
b411b363 | 4761 | |
b411b363 PR |
4762 | /* wait for all w_e_end_data_req, w_e_end_rsdata_req, w_send_barrier, |
4763 | * w_make_resync_request etc. which may still be on the worker queue | |
4764 | * to be "canceled" */ | |
b5043c5e | 4765 | drbd_flush_workqueue(&peer_device->connection->sender_work); |
b411b363 | 4766 | |
b30ab791 | 4767 | drbd_finish_peer_reqs(device); |
b411b363 | 4768 | |
d10b4ea3 PR |
4769 | /* This second workqueue flush is necessary, since drbd_finish_peer_reqs() |
4770 | might have issued a work again. The one before drbd_finish_peer_reqs() is | |
4771 | necessary to reclain net_ee in drbd_finish_peer_reqs(). */ | |
b5043c5e | 4772 | drbd_flush_workqueue(&peer_device->connection->sender_work); |
d10b4ea3 | 4773 | |
08332d73 LE |
4774 | /* need to do it again, drbd_finish_peer_reqs() may have populated it |
4775 | * again via drbd_try_clear_on_disk_bm(). */ | |
b30ab791 | 4776 | drbd_rs_cancel_all(device); |
b411b363 | 4777 | |
b30ab791 AG |
4778 | kfree(device->p_uuid); |
4779 | device->p_uuid = NULL; | |
b411b363 | 4780 | |
b30ab791 | 4781 | if (!drbd_suspended(device)) |
69a22773 | 4782 | tl_clear(peer_device->connection); |
b411b363 | 4783 | |
b30ab791 | 4784 | drbd_md_sync(device); |
b411b363 | 4785 | |
20ceb2b2 LE |
4786 | /* serialize with bitmap writeout triggered by the state change, |
4787 | * if any. */ | |
b30ab791 | 4788 | wait_event(device->misc_wait, !test_bit(BITMAP_IO, &device->flags)); |
20ceb2b2 | 4789 | |
b411b363 PR |
4790 | /* tcp_close and release of sendpage pages can be deferred. I don't |
4791 | * want to use SO_LINGER, because apparently it can be deferred for | |
4792 | * more than 20 seconds (longest time I checked). | |
4793 | * | |
4794 | * Actually we don't care for exactly when the network stack does its | |
4795 | * put_page(), but release our reference on these pages right here. | |
4796 | */ | |
b30ab791 | 4797 | i = drbd_free_peer_reqs(device, &device->net_ee); |
b411b363 | 4798 | if (i) |
d0180171 | 4799 | drbd_info(device, "net_ee not empty, killed %u entries\n", i); |
b30ab791 | 4800 | i = atomic_read(&device->pp_in_use_by_net); |
435f0740 | 4801 | if (i) |
d0180171 | 4802 | drbd_info(device, "pp_in_use_by_net = %d, expected 0\n", i); |
b30ab791 | 4803 | i = atomic_read(&device->pp_in_use); |
b411b363 | 4804 | if (i) |
d0180171 | 4805 | drbd_info(device, "pp_in_use = %d, expected 0\n", i); |
b411b363 | 4806 | |
0b0ba1ef AG |
4807 | D_ASSERT(device, list_empty(&device->read_ee)); |
4808 | D_ASSERT(device, list_empty(&device->active_ee)); | |
4809 | D_ASSERT(device, list_empty(&device->sync_ee)); | |
4810 | D_ASSERT(device, list_empty(&device->done_ee)); | |
b411b363 | 4811 | |
360cc740 | 4812 | return 0; |
b411b363 PR |
4813 | } |
4814 | ||
4815 | /* | |
4816 | * We support PRO_VERSION_MIN to PRO_VERSION_MAX. The protocol version | |
4817 | * we can agree on is stored in agreed_pro_version. | |
4818 | * | |
4819 | * feature flags and the reserved array should be enough room for future | |
4820 | * enhancements of the handshake protocol, and possible plugins... | |
4821 | * | |
4822 | * for now, they are expected to be zero, but ignored. | |
4823 | */ | |
bde89a9e | 4824 | static int drbd_send_features(struct drbd_connection *connection) |
b411b363 | 4825 | { |
9f5bdc33 AG |
4826 | struct drbd_socket *sock; |
4827 | struct p_connection_features *p; | |
b411b363 | 4828 | |
bde89a9e AG |
4829 | sock = &connection->data; |
4830 | p = conn_prepare_command(connection, sock); | |
9f5bdc33 | 4831 | if (!p) |
e8d17b01 | 4832 | return -EIO; |
b411b363 PR |
4833 | memset(p, 0, sizeof(*p)); |
4834 | p->protocol_min = cpu_to_be32(PRO_VERSION_MIN); | |
4835 | p->protocol_max = cpu_to_be32(PRO_VERSION_MAX); | |
20c68fde | 4836 | p->feature_flags = cpu_to_be32(PRO_FEATURES); |
bde89a9e | 4837 | return conn_send_command(connection, sock, P_CONNECTION_FEATURES, sizeof(*p), NULL, 0); |
b411b363 PR |
4838 | } |
4839 | ||
4840 | /* | |
4841 | * return values: | |
4842 | * 1 yes, we have a valid connection | |
4843 | * 0 oops, did not work out, please try again | |
4844 | * -1 peer talks different language, | |
4845 | * no point in trying again, please go standalone. | |
4846 | */ | |
bde89a9e | 4847 | static int drbd_do_features(struct drbd_connection *connection) |
b411b363 | 4848 | { |
bde89a9e | 4849 | /* ASSERT current == connection->receiver ... */ |
e658983a AG |
4850 | struct p_connection_features *p; |
4851 | const int expect = sizeof(struct p_connection_features); | |
77351055 | 4852 | struct packet_info pi; |
a5c31904 | 4853 | int err; |
b411b363 | 4854 | |
bde89a9e | 4855 | err = drbd_send_features(connection); |
e8d17b01 | 4856 | if (err) |
b411b363 PR |
4857 | return 0; |
4858 | ||
bde89a9e | 4859 | err = drbd_recv_header(connection, &pi); |
69bc7bc3 | 4860 | if (err) |
b411b363 PR |
4861 | return 0; |
4862 | ||
6038178e | 4863 | if (pi.cmd != P_CONNECTION_FEATURES) { |
1ec861eb | 4864 | drbd_err(connection, "expected ConnectionFeatures packet, received: %s (0x%04x)\n", |
2fcb8f30 | 4865 | cmdname(pi.cmd), pi.cmd); |
b411b363 PR |
4866 | return -1; |
4867 | } | |
4868 | ||
77351055 | 4869 | if (pi.size != expect) { |
1ec861eb | 4870 | drbd_err(connection, "expected ConnectionFeatures length: %u, received: %u\n", |
77351055 | 4871 | expect, pi.size); |
b411b363 PR |
4872 | return -1; |
4873 | } | |
4874 | ||
e658983a | 4875 | p = pi.data; |
bde89a9e | 4876 | err = drbd_recv_all_warn(connection, p, expect); |
a5c31904 | 4877 | if (err) |
b411b363 | 4878 | return 0; |
b411b363 | 4879 | |
b411b363 PR |
4880 | p->protocol_min = be32_to_cpu(p->protocol_min); |
4881 | p->protocol_max = be32_to_cpu(p->protocol_max); | |
4882 | if (p->protocol_max == 0) | |
4883 | p->protocol_max = p->protocol_min; | |
4884 | ||
4885 | if (PRO_VERSION_MAX < p->protocol_min || | |
4886 | PRO_VERSION_MIN > p->protocol_max) | |
4887 | goto incompat; | |
4888 | ||
bde89a9e | 4889 | connection->agreed_pro_version = min_t(int, PRO_VERSION_MAX, p->protocol_max); |
20c68fde | 4890 | connection->agreed_features = PRO_FEATURES & be32_to_cpu(p->feature_flags); |
b411b363 | 4891 | |
1ec861eb | 4892 | drbd_info(connection, "Handshake successful: " |
bde89a9e | 4893 | "Agreed network protocol version %d\n", connection->agreed_pro_version); |
b411b363 | 4894 | |
20c68fde LE |
4895 | drbd_info(connection, "Agreed to%ssupport TRIM on protocol level\n", |
4896 | connection->agreed_features & FF_TRIM ? " " : " not "); | |
4897 | ||
b411b363 PR |
4898 | return 1; |
4899 | ||
4900 | incompat: | |
1ec861eb | 4901 | drbd_err(connection, "incompatible DRBD dialects: " |
b411b363 PR |
4902 | "I support %d-%d, peer supports %d-%d\n", |
4903 | PRO_VERSION_MIN, PRO_VERSION_MAX, | |
4904 | p->protocol_min, p->protocol_max); | |
4905 | return -1; | |
4906 | } | |
4907 | ||
4908 | #if !defined(CONFIG_CRYPTO_HMAC) && !defined(CONFIG_CRYPTO_HMAC_MODULE) | |
bde89a9e | 4909 | static int drbd_do_auth(struct drbd_connection *connection) |
b411b363 | 4910 | { |
1ec861eb AG |
4911 | drbd_err(connection, "This kernel was build without CONFIG_CRYPTO_HMAC.\n"); |
4912 | drbd_err(connection, "You need to disable 'cram-hmac-alg' in drbd.conf.\n"); | |
b10d96cb | 4913 | return -1; |
b411b363 PR |
4914 | } |
4915 | #else | |
4916 | #define CHALLENGE_LEN 64 | |
b10d96cb JT |
4917 | |
4918 | /* Return value: | |
4919 | 1 - auth succeeded, | |
4920 | 0 - failed, try again (network error), | |
4921 | -1 - auth failed, don't try again. | |
4922 | */ | |
4923 | ||
bde89a9e | 4924 | static int drbd_do_auth(struct drbd_connection *connection) |
b411b363 | 4925 | { |
9f5bdc33 | 4926 | struct drbd_socket *sock; |
b411b363 PR |
4927 | char my_challenge[CHALLENGE_LEN]; /* 64 Bytes... */ |
4928 | struct scatterlist sg; | |
4929 | char *response = NULL; | |
4930 | char *right_response = NULL; | |
4931 | char *peers_ch = NULL; | |
44ed167d PR |
4932 | unsigned int key_len; |
4933 | char secret[SHARED_SECRET_MAX]; /* 64 byte */ | |
b411b363 PR |
4934 | unsigned int resp_size; |
4935 | struct hash_desc desc; | |
77351055 | 4936 | struct packet_info pi; |
44ed167d | 4937 | struct net_conf *nc; |
69bc7bc3 | 4938 | int err, rv; |
b411b363 | 4939 | |
9f5bdc33 | 4940 | /* FIXME: Put the challenge/response into the preallocated socket buffer. */ |
b411b363 | 4941 | |
44ed167d | 4942 | rcu_read_lock(); |
bde89a9e | 4943 | nc = rcu_dereference(connection->net_conf); |
44ed167d PR |
4944 | key_len = strlen(nc->shared_secret); |
4945 | memcpy(secret, nc->shared_secret, key_len); | |
4946 | rcu_read_unlock(); | |
4947 | ||
bde89a9e | 4948 | desc.tfm = connection->cram_hmac_tfm; |
b411b363 PR |
4949 | desc.flags = 0; |
4950 | ||
bde89a9e | 4951 | rv = crypto_hash_setkey(connection->cram_hmac_tfm, (u8 *)secret, key_len); |
b411b363 | 4952 | if (rv) { |
1ec861eb | 4953 | drbd_err(connection, "crypto_hash_setkey() failed with %d\n", rv); |
b10d96cb | 4954 | rv = -1; |
b411b363 PR |
4955 | goto fail; |
4956 | } | |
4957 | ||
4958 | get_random_bytes(my_challenge, CHALLENGE_LEN); | |
4959 | ||
bde89a9e AG |
4960 | sock = &connection->data; |
4961 | if (!conn_prepare_command(connection, sock)) { | |
9f5bdc33 AG |
4962 | rv = 0; |
4963 | goto fail; | |
4964 | } | |
bde89a9e | 4965 | rv = !conn_send_command(connection, sock, P_AUTH_CHALLENGE, 0, |
9f5bdc33 | 4966 | my_challenge, CHALLENGE_LEN); |
b411b363 PR |
4967 | if (!rv) |
4968 | goto fail; | |
4969 | ||
bde89a9e | 4970 | err = drbd_recv_header(connection, &pi); |
69bc7bc3 AG |
4971 | if (err) { |
4972 | rv = 0; | |
b411b363 | 4973 | goto fail; |
69bc7bc3 | 4974 | } |
b411b363 | 4975 | |
77351055 | 4976 | if (pi.cmd != P_AUTH_CHALLENGE) { |
1ec861eb | 4977 | drbd_err(connection, "expected AuthChallenge packet, received: %s (0x%04x)\n", |
2fcb8f30 | 4978 | cmdname(pi.cmd), pi.cmd); |
b411b363 PR |
4979 | rv = 0; |
4980 | goto fail; | |
4981 | } | |
4982 | ||
77351055 | 4983 | if (pi.size > CHALLENGE_LEN * 2) { |
1ec861eb | 4984 | drbd_err(connection, "expected AuthChallenge payload too big.\n"); |
b10d96cb | 4985 | rv = -1; |
b411b363 PR |
4986 | goto fail; |
4987 | } | |
4988 | ||
67cca286 PR |
4989 | if (pi.size < CHALLENGE_LEN) { |
4990 | drbd_err(connection, "AuthChallenge payload too small.\n"); | |
4991 | rv = -1; | |
4992 | goto fail; | |
4993 | } | |
4994 | ||
77351055 | 4995 | peers_ch = kmalloc(pi.size, GFP_NOIO); |
b411b363 | 4996 | if (peers_ch == NULL) { |
1ec861eb | 4997 | drbd_err(connection, "kmalloc of peers_ch failed\n"); |
b10d96cb | 4998 | rv = -1; |
b411b363 PR |
4999 | goto fail; |
5000 | } | |
5001 | ||
bde89a9e | 5002 | err = drbd_recv_all_warn(connection, peers_ch, pi.size); |
a5c31904 | 5003 | if (err) { |
b411b363 PR |
5004 | rv = 0; |
5005 | goto fail; | |
5006 | } | |
5007 | ||
67cca286 PR |
5008 | if (!memcmp(my_challenge, peers_ch, CHALLENGE_LEN)) { |
5009 | drbd_err(connection, "Peer presented the same challenge!\n"); | |
5010 | rv = -1; | |
5011 | goto fail; | |
5012 | } | |
5013 | ||
bde89a9e | 5014 | resp_size = crypto_hash_digestsize(connection->cram_hmac_tfm); |
b411b363 PR |
5015 | response = kmalloc(resp_size, GFP_NOIO); |
5016 | if (response == NULL) { | |
1ec861eb | 5017 | drbd_err(connection, "kmalloc of response failed\n"); |
b10d96cb | 5018 | rv = -1; |
b411b363 PR |
5019 | goto fail; |
5020 | } | |
5021 | ||
5022 | sg_init_table(&sg, 1); | |
77351055 | 5023 | sg_set_buf(&sg, peers_ch, pi.size); |
b411b363 PR |
5024 | |
5025 | rv = crypto_hash_digest(&desc, &sg, sg.length, response); | |
5026 | if (rv) { | |
1ec861eb | 5027 | drbd_err(connection, "crypto_hash_digest() failed with %d\n", rv); |
b10d96cb | 5028 | rv = -1; |
b411b363 PR |
5029 | goto fail; |
5030 | } | |
5031 | ||
bde89a9e | 5032 | if (!conn_prepare_command(connection, sock)) { |
9f5bdc33 | 5033 | rv = 0; |
b411b363 | 5034 | goto fail; |
9f5bdc33 | 5035 | } |
bde89a9e | 5036 | rv = !conn_send_command(connection, sock, P_AUTH_RESPONSE, 0, |
9f5bdc33 | 5037 | response, resp_size); |
b411b363 PR |
5038 | if (!rv) |
5039 | goto fail; | |
5040 | ||
bde89a9e | 5041 | err = drbd_recv_header(connection, &pi); |
69bc7bc3 | 5042 | if (err) { |
b411b363 PR |
5043 | rv = 0; |
5044 | goto fail; | |
5045 | } | |
5046 | ||
77351055 | 5047 | if (pi.cmd != P_AUTH_RESPONSE) { |
1ec861eb | 5048 | drbd_err(connection, "expected AuthResponse packet, received: %s (0x%04x)\n", |
2fcb8f30 | 5049 | cmdname(pi.cmd), pi.cmd); |
b411b363 PR |
5050 | rv = 0; |
5051 | goto fail; | |
5052 | } | |
5053 | ||
77351055 | 5054 | if (pi.size != resp_size) { |
1ec861eb | 5055 | drbd_err(connection, "expected AuthResponse payload of wrong size\n"); |
b411b363 PR |
5056 | rv = 0; |
5057 | goto fail; | |
5058 | } | |
b411b363 | 5059 | |
bde89a9e | 5060 | err = drbd_recv_all_warn(connection, response , resp_size); |
a5c31904 | 5061 | if (err) { |
b411b363 PR |
5062 | rv = 0; |
5063 | goto fail; | |
5064 | } | |
5065 | ||
5066 | right_response = kmalloc(resp_size, GFP_NOIO); | |
2d1ee87d | 5067 | if (right_response == NULL) { |
1ec861eb | 5068 | drbd_err(connection, "kmalloc of right_response failed\n"); |
b10d96cb | 5069 | rv = -1; |
b411b363 PR |
5070 | goto fail; |
5071 | } | |
5072 | ||
5073 | sg_set_buf(&sg, my_challenge, CHALLENGE_LEN); | |
5074 | ||
5075 | rv = crypto_hash_digest(&desc, &sg, sg.length, right_response); | |
5076 | if (rv) { | |
1ec861eb | 5077 | drbd_err(connection, "crypto_hash_digest() failed with %d\n", rv); |
b10d96cb | 5078 | rv = -1; |
b411b363 PR |
5079 | goto fail; |
5080 | } | |
5081 | ||
5082 | rv = !memcmp(response, right_response, resp_size); | |
5083 | ||
5084 | if (rv) | |
1ec861eb | 5085 | drbd_info(connection, "Peer authenticated using %d bytes HMAC\n", |
44ed167d | 5086 | resp_size); |
b10d96cb JT |
5087 | else |
5088 | rv = -1; | |
b411b363 PR |
5089 | |
5090 | fail: | |
5091 | kfree(peers_ch); | |
5092 | kfree(response); | |
5093 | kfree(right_response); | |
5094 | ||
5095 | return rv; | |
5096 | } | |
5097 | #endif | |
5098 | ||
8fe60551 | 5099 | int drbd_receiver(struct drbd_thread *thi) |
b411b363 | 5100 | { |
bde89a9e | 5101 | struct drbd_connection *connection = thi->connection; |
b411b363 PR |
5102 | int h; |
5103 | ||
1ec861eb | 5104 | drbd_info(connection, "receiver (re)started\n"); |
b411b363 PR |
5105 | |
5106 | do { | |
bde89a9e | 5107 | h = conn_connect(connection); |
b411b363 | 5108 | if (h == 0) { |
bde89a9e | 5109 | conn_disconnect(connection); |
20ee6390 | 5110 | schedule_timeout_interruptible(HZ); |
b411b363 PR |
5111 | } |
5112 | if (h == -1) { | |
1ec861eb | 5113 | drbd_warn(connection, "Discarding network configuration.\n"); |
bde89a9e | 5114 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b411b363 PR |
5115 | } |
5116 | } while (h == 0); | |
5117 | ||
91fd4dad | 5118 | if (h > 0) |
bde89a9e | 5119 | drbdd(connection); |
b411b363 | 5120 | |
bde89a9e | 5121 | conn_disconnect(connection); |
b411b363 | 5122 | |
1ec861eb | 5123 | drbd_info(connection, "receiver terminated\n"); |
b411b363 PR |
5124 | return 0; |
5125 | } | |
5126 | ||
5127 | /* ********* acknowledge sender ******** */ | |
5128 | ||
bde89a9e | 5129 | static int got_conn_RqSReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5130 | { |
e658983a | 5131 | struct p_req_state_reply *p = pi->data; |
e4f78ede PR |
5132 | int retcode = be32_to_cpu(p->retcode); |
5133 | ||
5134 | if (retcode >= SS_SUCCESS) { | |
bde89a9e | 5135 | set_bit(CONN_WD_ST_CHG_OKAY, &connection->flags); |
e4f78ede | 5136 | } else { |
bde89a9e | 5137 | set_bit(CONN_WD_ST_CHG_FAIL, &connection->flags); |
1ec861eb | 5138 | drbd_err(connection, "Requested state change failed by peer: %s (%d)\n", |
e4f78ede PR |
5139 | drbd_set_st_err_str(retcode), retcode); |
5140 | } | |
bde89a9e | 5141 | wake_up(&connection->ping_wait); |
e4f78ede | 5142 | |
2735a594 | 5143 | return 0; |
e4f78ede | 5144 | } |
b411b363 | 5145 | |
bde89a9e | 5146 | static int got_RqSReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5147 | { |
9f4fe9ad | 5148 | struct drbd_peer_device *peer_device; |
b30ab791 | 5149 | struct drbd_device *device; |
e658983a | 5150 | struct p_req_state_reply *p = pi->data; |
b411b363 PR |
5151 | int retcode = be32_to_cpu(p->retcode); |
5152 | ||
9f4fe9ad AG |
5153 | peer_device = conn_peer_device(connection, pi->vnr); |
5154 | if (!peer_device) | |
2735a594 | 5155 | return -EIO; |
9f4fe9ad | 5156 | device = peer_device->device; |
1952e916 | 5157 | |
bde89a9e | 5158 | if (test_bit(CONN_WD_ST_CHG_REQ, &connection->flags)) { |
0b0ba1ef | 5159 | D_ASSERT(device, connection->agreed_pro_version < 100); |
bde89a9e | 5160 | return got_conn_RqSReply(connection, pi); |
4d0fc3fd PR |
5161 | } |
5162 | ||
b411b363 | 5163 | if (retcode >= SS_SUCCESS) { |
b30ab791 | 5164 | set_bit(CL_ST_CHG_SUCCESS, &device->flags); |
b411b363 | 5165 | } else { |
b30ab791 | 5166 | set_bit(CL_ST_CHG_FAIL, &device->flags); |
d0180171 | 5167 | drbd_err(device, "Requested state change failed by peer: %s (%d)\n", |
e4f78ede | 5168 | drbd_set_st_err_str(retcode), retcode); |
b411b363 | 5169 | } |
b30ab791 | 5170 | wake_up(&device->state_wait); |
b411b363 | 5171 | |
2735a594 | 5172 | return 0; |
b411b363 PR |
5173 | } |
5174 | ||
bde89a9e | 5175 | static int got_Ping(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5176 | { |
bde89a9e | 5177 | return drbd_send_ping_ack(connection); |
b411b363 PR |
5178 | |
5179 | } | |
5180 | ||
bde89a9e | 5181 | static int got_PingAck(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 PR |
5182 | { |
5183 | /* restore idle timeout */ | |
bde89a9e AG |
5184 | connection->meta.socket->sk->sk_rcvtimeo = connection->net_conf->ping_int*HZ; |
5185 | if (!test_and_set_bit(GOT_PING_ACK, &connection->flags)) | |
5186 | wake_up(&connection->ping_wait); | |
b411b363 | 5187 | |
2735a594 | 5188 | return 0; |
b411b363 PR |
5189 | } |
5190 | ||
bde89a9e | 5191 | static int got_IsInSync(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5192 | { |
9f4fe9ad | 5193 | struct drbd_peer_device *peer_device; |
b30ab791 | 5194 | struct drbd_device *device; |
e658983a | 5195 | struct p_block_ack *p = pi->data; |
b411b363 PR |
5196 | sector_t sector = be64_to_cpu(p->sector); |
5197 | int blksize = be32_to_cpu(p->blksize); | |
5198 | ||
9f4fe9ad AG |
5199 | peer_device = conn_peer_device(connection, pi->vnr); |
5200 | if (!peer_device) | |
2735a594 | 5201 | return -EIO; |
9f4fe9ad | 5202 | device = peer_device->device; |
1952e916 | 5203 | |
9f4fe9ad | 5204 | D_ASSERT(device, peer_device->connection->agreed_pro_version >= 89); |
b411b363 | 5205 | |
69a22773 | 5206 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
b411b363 | 5207 | |
b30ab791 AG |
5208 | if (get_ldev(device)) { |
5209 | drbd_rs_complete_io(device, sector); | |
5210 | drbd_set_in_sync(device, sector, blksize); | |
1d53f09e | 5211 | /* rs_same_csums is supposed to count in units of BM_BLOCK_SIZE */ |
b30ab791 AG |
5212 | device->rs_same_csum += (blksize >> BM_BLOCK_SHIFT); |
5213 | put_ldev(device); | |
1d53f09e | 5214 | } |
b30ab791 AG |
5215 | dec_rs_pending(device); |
5216 | atomic_add(blksize >> 9, &device->rs_sect_in); | |
b411b363 | 5217 | |
2735a594 | 5218 | return 0; |
b411b363 PR |
5219 | } |
5220 | ||
bc9c5c41 | 5221 | static int |
b30ab791 | 5222 | validate_req_change_req_state(struct drbd_device *device, u64 id, sector_t sector, |
bc9c5c41 AG |
5223 | struct rb_root *root, const char *func, |
5224 | enum drbd_req_event what, bool missing_ok) | |
b411b363 PR |
5225 | { |
5226 | struct drbd_request *req; | |
5227 | struct bio_and_error m; | |
5228 | ||
0500813f | 5229 | spin_lock_irq(&device->resource->req_lock); |
b30ab791 | 5230 | req = find_request(device, root, id, sector, missing_ok, func); |
b411b363 | 5231 | if (unlikely(!req)) { |
0500813f | 5232 | spin_unlock_irq(&device->resource->req_lock); |
85997675 | 5233 | return -EIO; |
b411b363 PR |
5234 | } |
5235 | __req_mod(req, what, &m); | |
0500813f | 5236 | spin_unlock_irq(&device->resource->req_lock); |
b411b363 PR |
5237 | |
5238 | if (m.bio) | |
b30ab791 | 5239 | complete_master_bio(device, &m); |
85997675 | 5240 | return 0; |
b411b363 PR |
5241 | } |
5242 | ||
bde89a9e | 5243 | static int got_BlockAck(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5244 | { |
9f4fe9ad | 5245 | struct drbd_peer_device *peer_device; |
b30ab791 | 5246 | struct drbd_device *device; |
e658983a | 5247 | struct p_block_ack *p = pi->data; |
b411b363 PR |
5248 | sector_t sector = be64_to_cpu(p->sector); |
5249 | int blksize = be32_to_cpu(p->blksize); | |
5250 | enum drbd_req_event what; | |
5251 | ||
9f4fe9ad AG |
5252 | peer_device = conn_peer_device(connection, pi->vnr); |
5253 | if (!peer_device) | |
2735a594 | 5254 | return -EIO; |
9f4fe9ad | 5255 | device = peer_device->device; |
1952e916 | 5256 | |
69a22773 | 5257 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
b411b363 | 5258 | |
579b57ed | 5259 | if (p->block_id == ID_SYNCER) { |
b30ab791 AG |
5260 | drbd_set_in_sync(device, sector, blksize); |
5261 | dec_rs_pending(device); | |
2735a594 | 5262 | return 0; |
b411b363 | 5263 | } |
e05e1e59 | 5264 | switch (pi->cmd) { |
b411b363 | 5265 | case P_RS_WRITE_ACK: |
8554df1c | 5266 | what = WRITE_ACKED_BY_PEER_AND_SIS; |
b411b363 PR |
5267 | break; |
5268 | case P_WRITE_ACK: | |
8554df1c | 5269 | what = WRITE_ACKED_BY_PEER; |
b411b363 PR |
5270 | break; |
5271 | case P_RECV_ACK: | |
8554df1c | 5272 | what = RECV_ACKED_BY_PEER; |
b411b363 | 5273 | break; |
d4dabbe2 LE |
5274 | case P_SUPERSEDED: |
5275 | what = CONFLICT_RESOLVED; | |
b411b363 | 5276 | break; |
7be8da07 | 5277 | case P_RETRY_WRITE: |
7be8da07 | 5278 | what = POSTPONE_WRITE; |
b411b363 PR |
5279 | break; |
5280 | default: | |
2735a594 | 5281 | BUG(); |
b411b363 PR |
5282 | } |
5283 | ||
b30ab791 AG |
5284 | return validate_req_change_req_state(device, p->block_id, sector, |
5285 | &device->write_requests, __func__, | |
2735a594 | 5286 | what, false); |
b411b363 PR |
5287 | } |
5288 | ||
bde89a9e | 5289 | static int got_NegAck(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5290 | { |
9f4fe9ad | 5291 | struct drbd_peer_device *peer_device; |
b30ab791 | 5292 | struct drbd_device *device; |
e658983a | 5293 | struct p_block_ack *p = pi->data; |
b411b363 | 5294 | sector_t sector = be64_to_cpu(p->sector); |
2deb8336 | 5295 | int size = be32_to_cpu(p->blksize); |
85997675 | 5296 | int err; |
b411b363 | 5297 | |
9f4fe9ad AG |
5298 | peer_device = conn_peer_device(connection, pi->vnr); |
5299 | if (!peer_device) | |
2735a594 | 5300 | return -EIO; |
9f4fe9ad | 5301 | device = peer_device->device; |
b411b363 | 5302 | |
69a22773 | 5303 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
b411b363 | 5304 | |
579b57ed | 5305 | if (p->block_id == ID_SYNCER) { |
b30ab791 AG |
5306 | dec_rs_pending(device); |
5307 | drbd_rs_failed_io(device, sector, size); | |
2735a594 | 5308 | return 0; |
b411b363 | 5309 | } |
2deb8336 | 5310 | |
b30ab791 AG |
5311 | err = validate_req_change_req_state(device, p->block_id, sector, |
5312 | &device->write_requests, __func__, | |
303d1448 | 5313 | NEG_ACKED, true); |
85997675 | 5314 | if (err) { |
c3afd8f5 AG |
5315 | /* Protocol A has no P_WRITE_ACKs, but has P_NEG_ACKs. |
5316 | The master bio might already be completed, therefore the | |
5317 | request is no longer in the collision hash. */ | |
5318 | /* In Protocol B we might already have got a P_RECV_ACK | |
5319 | but then get a P_NEG_ACK afterwards. */ | |
b30ab791 | 5320 | drbd_set_out_of_sync(device, sector, size); |
2deb8336 | 5321 | } |
2735a594 | 5322 | return 0; |
b411b363 PR |
5323 | } |
5324 | ||
bde89a9e | 5325 | static int got_NegDReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5326 | { |
9f4fe9ad | 5327 | struct drbd_peer_device *peer_device; |
b30ab791 | 5328 | struct drbd_device *device; |
e658983a | 5329 | struct p_block_ack *p = pi->data; |
b411b363 PR |
5330 | sector_t sector = be64_to_cpu(p->sector); |
5331 | ||
9f4fe9ad AG |
5332 | peer_device = conn_peer_device(connection, pi->vnr); |
5333 | if (!peer_device) | |
2735a594 | 5334 | return -EIO; |
9f4fe9ad | 5335 | device = peer_device->device; |
1952e916 | 5336 | |
69a22773 | 5337 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
7be8da07 | 5338 | |
d0180171 | 5339 | drbd_err(device, "Got NegDReply; Sector %llus, len %u.\n", |
b411b363 PR |
5340 | (unsigned long long)sector, be32_to_cpu(p->blksize)); |
5341 | ||
b30ab791 AG |
5342 | return validate_req_change_req_state(device, p->block_id, sector, |
5343 | &device->read_requests, __func__, | |
2735a594 | 5344 | NEG_ACKED, false); |
b411b363 PR |
5345 | } |
5346 | ||
bde89a9e | 5347 | static int got_NegRSDReply(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5348 | { |
9f4fe9ad | 5349 | struct drbd_peer_device *peer_device; |
b30ab791 | 5350 | struct drbd_device *device; |
b411b363 PR |
5351 | sector_t sector; |
5352 | int size; | |
e658983a | 5353 | struct p_block_ack *p = pi->data; |
1952e916 | 5354 | |
9f4fe9ad AG |
5355 | peer_device = conn_peer_device(connection, pi->vnr); |
5356 | if (!peer_device) | |
2735a594 | 5357 | return -EIO; |
9f4fe9ad | 5358 | device = peer_device->device; |
b411b363 PR |
5359 | |
5360 | sector = be64_to_cpu(p->sector); | |
5361 | size = be32_to_cpu(p->blksize); | |
b411b363 | 5362 | |
69a22773 | 5363 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
b411b363 | 5364 | |
b30ab791 | 5365 | dec_rs_pending(device); |
b411b363 | 5366 | |
b30ab791 AG |
5367 | if (get_ldev_if_state(device, D_FAILED)) { |
5368 | drbd_rs_complete_io(device, sector); | |
e05e1e59 | 5369 | switch (pi->cmd) { |
d612d309 | 5370 | case P_NEG_RS_DREPLY: |
b30ab791 | 5371 | drbd_rs_failed_io(device, sector, size); |
d612d309 PR |
5372 | case P_RS_CANCEL: |
5373 | break; | |
5374 | default: | |
2735a594 | 5375 | BUG(); |
d612d309 | 5376 | } |
b30ab791 | 5377 | put_ldev(device); |
b411b363 PR |
5378 | } |
5379 | ||
2735a594 | 5380 | return 0; |
b411b363 PR |
5381 | } |
5382 | ||
bde89a9e | 5383 | static int got_BarrierAck(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5384 | { |
e658983a | 5385 | struct p_barrier_ack *p = pi->data; |
c06ece6b | 5386 | struct drbd_peer_device *peer_device; |
9ed57dcb | 5387 | int vnr; |
1952e916 | 5388 | |
bde89a9e | 5389 | tl_release(connection, p->barrier, be32_to_cpu(p->set_size)); |
b411b363 | 5390 | |
9ed57dcb | 5391 | rcu_read_lock(); |
c06ece6b AG |
5392 | idr_for_each_entry(&connection->peer_devices, peer_device, vnr) { |
5393 | struct drbd_device *device = peer_device->device; | |
5394 | ||
b30ab791 AG |
5395 | if (device->state.conn == C_AHEAD && |
5396 | atomic_read(&device->ap_in_flight) == 0 && | |
5397 | !test_and_set_bit(AHEAD_TO_SYNC_SOURCE, &device->flags)) { | |
5398 | device->start_resync_timer.expires = jiffies + HZ; | |
5399 | add_timer(&device->start_resync_timer); | |
9ed57dcb | 5400 | } |
c4752ef1 | 5401 | } |
9ed57dcb | 5402 | rcu_read_unlock(); |
c4752ef1 | 5403 | |
2735a594 | 5404 | return 0; |
b411b363 PR |
5405 | } |
5406 | ||
bde89a9e | 5407 | static int got_OVResult(struct drbd_connection *connection, struct packet_info *pi) |
b411b363 | 5408 | { |
9f4fe9ad | 5409 | struct drbd_peer_device *peer_device; |
b30ab791 | 5410 | struct drbd_device *device; |
e658983a | 5411 | struct p_block_ack *p = pi->data; |
84b8c06b | 5412 | struct drbd_device_work *dw; |
b411b363 PR |
5413 | sector_t sector; |
5414 | int size; | |
5415 | ||
9f4fe9ad AG |
5416 | peer_device = conn_peer_device(connection, pi->vnr); |
5417 | if (!peer_device) | |
2735a594 | 5418 | return -EIO; |
9f4fe9ad | 5419 | device = peer_device->device; |
1952e916 | 5420 | |
b411b363 PR |
5421 | sector = be64_to_cpu(p->sector); |
5422 | size = be32_to_cpu(p->blksize); | |
5423 | ||
69a22773 | 5424 | update_peer_seq(peer_device, be32_to_cpu(p->seq_num)); |
b411b363 PR |
5425 | |
5426 | if (be64_to_cpu(p->block_id) == ID_OUT_OF_SYNC) | |
b30ab791 | 5427 | drbd_ov_out_of_sync_found(device, sector, size); |
b411b363 | 5428 | else |
b30ab791 | 5429 | ov_out_of_sync_print(device); |
b411b363 | 5430 | |
b30ab791 | 5431 | if (!get_ldev(device)) |
2735a594 | 5432 | return 0; |
1d53f09e | 5433 | |
b30ab791 AG |
5434 | drbd_rs_complete_io(device, sector); |
5435 | dec_rs_pending(device); | |
b411b363 | 5436 | |
b30ab791 | 5437 | --device->ov_left; |
ea5442af LE |
5438 | |
5439 | /* let's advance progress step marks only for every other megabyte */ | |
b30ab791 AG |
5440 | if ((device->ov_left & 0x200) == 0x200) |
5441 | drbd_advance_rs_marks(device, device->ov_left); | |
ea5442af | 5442 | |
b30ab791 | 5443 | if (device->ov_left == 0) { |
84b8c06b AG |
5444 | dw = kmalloc(sizeof(*dw), GFP_NOIO); |
5445 | if (dw) { | |
5446 | dw->w.cb = w_ov_finished; | |
5447 | dw->device = device; | |
5448 | drbd_queue_work(&peer_device->connection->sender_work, &dw->w); | |
b411b363 | 5449 | } else { |
84b8c06b | 5450 | drbd_err(device, "kmalloc(dw) failed."); |
b30ab791 AG |
5451 | ov_out_of_sync_print(device); |
5452 | drbd_resync_finished(device); | |
b411b363 PR |
5453 | } |
5454 | } | |
b30ab791 | 5455 | put_ldev(device); |
2735a594 | 5456 | return 0; |
b411b363 PR |
5457 | } |
5458 | ||
bde89a9e | 5459 | static int got_skip(struct drbd_connection *connection, struct packet_info *pi) |
0ced55a3 | 5460 | { |
2735a594 | 5461 | return 0; |
b411b363 PR |
5462 | } |
5463 | ||
668700b4 PR |
5464 | struct meta_sock_cmd { |
5465 | size_t pkt_size; | |
5466 | int (*fn)(struct drbd_connection *connection, struct packet_info *); | |
5467 | }; | |
5468 | ||
5469 | static void set_rcvtimeo(struct drbd_connection *connection, bool ping_timeout) | |
0ced55a3 | 5470 | { |
668700b4 PR |
5471 | long t; |
5472 | struct net_conf *nc; | |
32862ec7 | 5473 | |
668700b4 PR |
5474 | rcu_read_lock(); |
5475 | nc = rcu_dereference(connection->net_conf); | |
5476 | t = ping_timeout ? nc->ping_timeo : nc->ping_int; | |
5477 | rcu_read_unlock(); | |
c141ebda | 5478 | |
668700b4 PR |
5479 | t *= HZ; |
5480 | if (ping_timeout) | |
5481 | t /= 10; | |
082a3439 | 5482 | |
668700b4 PR |
5483 | connection->meta.socket->sk->sk_rcvtimeo = t; |
5484 | } | |
32862ec7 | 5485 | |
668700b4 PR |
5486 | static void set_ping_timeout(struct drbd_connection *connection) |
5487 | { | |
5488 | set_rcvtimeo(connection, 1); | |
0ced55a3 PR |
5489 | } |
5490 | ||
668700b4 PR |
5491 | static void set_idle_timeout(struct drbd_connection *connection) |
5492 | { | |
5493 | set_rcvtimeo(connection, 0); | |
5494 | } | |
b411b363 | 5495 | |
668700b4 | 5496 | static struct meta_sock_cmd ack_receiver_tbl[] = { |
e658983a AG |
5497 | [P_PING] = { 0, got_Ping }, |
5498 | [P_PING_ACK] = { 0, got_PingAck }, | |
b411b363 PR |
5499 | [P_RECV_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, |
5500 | [P_WRITE_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, | |
5501 | [P_RS_WRITE_ACK] = { sizeof(struct p_block_ack), got_BlockAck }, | |
d4dabbe2 | 5502 | [P_SUPERSEDED] = { sizeof(struct p_block_ack), got_BlockAck }, |
b411b363 PR |
5503 | [P_NEG_ACK] = { sizeof(struct p_block_ack), got_NegAck }, |
5504 | [P_NEG_DREPLY] = { sizeof(struct p_block_ack), got_NegDReply }, | |
1952e916 | 5505 | [P_NEG_RS_DREPLY] = { sizeof(struct p_block_ack), got_NegRSDReply }, |
b411b363 PR |
5506 | [P_OV_RESULT] = { sizeof(struct p_block_ack), got_OVResult }, |
5507 | [P_BARRIER_ACK] = { sizeof(struct p_barrier_ack), got_BarrierAck }, | |
5508 | [P_STATE_CHG_REPLY] = { sizeof(struct p_req_state_reply), got_RqSReply }, | |
5509 | [P_RS_IS_IN_SYNC] = { sizeof(struct p_block_ack), got_IsInSync }, | |
02918be2 | 5510 | [P_DELAY_PROBE] = { sizeof(struct p_delay_probe93), got_skip }, |
1952e916 AG |
5511 | [P_RS_CANCEL] = { sizeof(struct p_block_ack), got_NegRSDReply }, |
5512 | [P_CONN_ST_CHG_REPLY]={ sizeof(struct p_req_state_reply), got_conn_RqSReply }, | |
5513 | [P_RETRY_WRITE] = { sizeof(struct p_block_ack), got_BlockAck }, | |
7201b972 | 5514 | }; |
b411b363 | 5515 | |
1c03e520 | 5516 | int drbd_ack_receiver(struct drbd_thread *thi) |
b411b363 | 5517 | { |
bde89a9e | 5518 | struct drbd_connection *connection = thi->connection; |
668700b4 | 5519 | struct meta_sock_cmd *cmd = NULL; |
77351055 | 5520 | struct packet_info pi; |
668700b4 | 5521 | unsigned long pre_recv_jif; |
257d0af6 | 5522 | int rv; |
bde89a9e | 5523 | void *buf = connection->meta.rbuf; |
b411b363 | 5524 | int received = 0; |
bde89a9e | 5525 | unsigned int header_size = drbd_header_size(connection); |
52b061a4 | 5526 | int expect = header_size; |
44ed167d | 5527 | bool ping_timeout_active = false; |
3990e04d | 5528 | struct sched_param param = { .sched_priority = 2 }; |
b411b363 | 5529 | |
3990e04d PR |
5530 | rv = sched_setscheduler(current, SCHED_RR, ¶m); |
5531 | if (rv < 0) | |
668700b4 | 5532 | drbd_err(connection, "drbd_ack_receiver: ERROR set priority, ret=%d\n", rv); |
b411b363 | 5533 | |
e77a0a5c | 5534 | while (get_t_state(thi) == RUNNING) { |
80822284 | 5535 | drbd_thread_current_set_cpu(thi); |
b411b363 | 5536 | |
668700b4 | 5537 | conn_reclaim_net_peer_reqs(connection); |
44ed167d | 5538 | |
bde89a9e AG |
5539 | if (test_and_clear_bit(SEND_PING, &connection->flags)) { |
5540 | if (drbd_send_ping(connection)) { | |
1ec861eb | 5541 | drbd_err(connection, "drbd_send_ping has failed\n"); |
b411b363 | 5542 | goto reconnect; |
841ce241 | 5543 | } |
668700b4 | 5544 | set_ping_timeout(connection); |
44ed167d | 5545 | ping_timeout_active = true; |
b411b363 PR |
5546 | } |
5547 | ||
668700b4 | 5548 | pre_recv_jif = jiffies; |
bde89a9e | 5549 | rv = drbd_recv_short(connection->meta.socket, buf, expect-received, 0); |
b411b363 PR |
5550 | |
5551 | /* Note: | |
5552 | * -EINTR (on meta) we got a signal | |
5553 | * -EAGAIN (on meta) rcvtimeo expired | |
5554 | * -ECONNRESET other side closed the connection | |
5555 | * -ERESTARTSYS (on data) we got a signal | |
5556 | * rv < 0 other than above: unexpected error! | |
5557 | * rv == expected: full header or command | |
5558 | * rv < expected: "woken" by signal during receive | |
5559 | * rv == 0 : "connection shut down by peer" | |
5560 | */ | |
5561 | if (likely(rv > 0)) { | |
5562 | received += rv; | |
5563 | buf += rv; | |
5564 | } else if (rv == 0) { | |
bde89a9e | 5565 | if (test_bit(DISCONNECT_SENT, &connection->flags)) { |
b66623e3 PR |
5566 | long t; |
5567 | rcu_read_lock(); | |
bde89a9e | 5568 | t = rcu_dereference(connection->net_conf)->ping_timeo * HZ/10; |
b66623e3 PR |
5569 | rcu_read_unlock(); |
5570 | ||
bde89a9e AG |
5571 | t = wait_event_timeout(connection->ping_wait, |
5572 | connection->cstate < C_WF_REPORT_PARAMS, | |
b66623e3 | 5573 | t); |
599377ac PR |
5574 | if (t) |
5575 | break; | |
5576 | } | |
1ec861eb | 5577 | drbd_err(connection, "meta connection shut down by peer.\n"); |
b411b363 PR |
5578 | goto reconnect; |
5579 | } else if (rv == -EAGAIN) { | |
cb6518cb LE |
5580 | /* If the data socket received something meanwhile, |
5581 | * that is good enough: peer is still alive. */ | |
668700b4 | 5582 | if (time_after(connection->last_received, pre_recv_jif)) |
cb6518cb | 5583 | continue; |
f36af18c | 5584 | if (ping_timeout_active) { |
1ec861eb | 5585 | drbd_err(connection, "PingAck did not arrive in time.\n"); |
b411b363 PR |
5586 | goto reconnect; |
5587 | } | |
bde89a9e | 5588 | set_bit(SEND_PING, &connection->flags); |
b411b363 PR |
5589 | continue; |
5590 | } else if (rv == -EINTR) { | |
668700b4 PR |
5591 | /* maybe drbd_thread_stop(): the while condition will notice. |
5592 | * maybe woken for send_ping: we'll send a ping above, | |
5593 | * and change the rcvtimeo */ | |
5594 | flush_signals(current); | |
b411b363 PR |
5595 | continue; |
5596 | } else { | |
1ec861eb | 5597 | drbd_err(connection, "sock_recvmsg returned %d\n", rv); |
b411b363 PR |
5598 | goto reconnect; |
5599 | } | |
5600 | ||
5601 | if (received == expect && cmd == NULL) { | |
bde89a9e | 5602 | if (decode_header(connection, connection->meta.rbuf, &pi)) |
b411b363 | 5603 | goto reconnect; |
668700b4 PR |
5604 | cmd = &ack_receiver_tbl[pi.cmd]; |
5605 | if (pi.cmd >= ARRAY_SIZE(ack_receiver_tbl) || !cmd->fn) { | |
1ec861eb | 5606 | drbd_err(connection, "Unexpected meta packet %s (0x%04x)\n", |
2fcb8f30 | 5607 | cmdname(pi.cmd), pi.cmd); |
b411b363 PR |
5608 | goto disconnect; |
5609 | } | |
e658983a | 5610 | expect = header_size + cmd->pkt_size; |
52b061a4 | 5611 | if (pi.size != expect - header_size) { |
1ec861eb | 5612 | drbd_err(connection, "Wrong packet size on meta (c: %d, l: %d)\n", |
77351055 | 5613 | pi.cmd, pi.size); |
b411b363 | 5614 | goto reconnect; |
257d0af6 | 5615 | } |
b411b363 PR |
5616 | } |
5617 | if (received == expect) { | |
2735a594 | 5618 | bool err; |
a4fbda8e | 5619 | |
bde89a9e | 5620 | err = cmd->fn(connection, &pi); |
2735a594 | 5621 | if (err) { |
1ec861eb | 5622 | drbd_err(connection, "%pf failed\n", cmd->fn); |
b411b363 | 5623 | goto reconnect; |
1952e916 | 5624 | } |
b411b363 | 5625 | |
bde89a9e | 5626 | connection->last_received = jiffies; |
f36af18c | 5627 | |
668700b4 PR |
5628 | if (cmd == &ack_receiver_tbl[P_PING_ACK]) { |
5629 | set_idle_timeout(connection); | |
44ed167d PR |
5630 | ping_timeout_active = false; |
5631 | } | |
f36af18c | 5632 | |
bde89a9e | 5633 | buf = connection->meta.rbuf; |
b411b363 | 5634 | received = 0; |
52b061a4 | 5635 | expect = header_size; |
b411b363 PR |
5636 | cmd = NULL; |
5637 | } | |
5638 | } | |
5639 | ||
5640 | if (0) { | |
5641 | reconnect: | |
bde89a9e AG |
5642 | conn_request_state(connection, NS(conn, C_NETWORK_FAILURE), CS_HARD); |
5643 | conn_md_sync(connection); | |
b411b363 PR |
5644 | } |
5645 | if (0) { | |
5646 | disconnect: | |
bde89a9e | 5647 | conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD); |
b411b363 | 5648 | } |
b411b363 | 5649 | |
668700b4 | 5650 | drbd_info(connection, "ack_receiver terminated\n"); |
b411b363 PR |
5651 | |
5652 | return 0; | |
5653 | } | |
668700b4 PR |
5654 | |
5655 | void drbd_send_acks_wf(struct work_struct *ws) | |
5656 | { | |
5657 | struct drbd_peer_device *peer_device = | |
5658 | container_of(ws, struct drbd_peer_device, send_acks_work); | |
5659 | struct drbd_connection *connection = peer_device->connection; | |
5660 | struct drbd_device *device = peer_device->device; | |
5661 | struct net_conf *nc; | |
5662 | int tcp_cork, err; | |
5663 | ||
5664 | rcu_read_lock(); | |
5665 | nc = rcu_dereference(connection->net_conf); | |
5666 | tcp_cork = nc->tcp_cork; | |
5667 | rcu_read_unlock(); | |
5668 | ||
5669 | if (tcp_cork) | |
5670 | drbd_tcp_cork(connection->meta.socket); | |
5671 | ||
5672 | err = drbd_finish_peer_reqs(device); | |
5673 | kref_put(&device->kref, drbd_destroy_device); | |
5674 | /* get is in drbd_endio_write_sec_final(). That is necessary to keep the | |
5675 | struct work_struct send_acks_work alive, which is in the peer_device object */ | |
5676 | ||
5677 | if (err) { | |
5678 | conn_request_state(connection, NS(conn, C_NETWORK_FAILURE), CS_HARD); | |
5679 | return; | |
5680 | } | |
5681 | ||
5682 | if (tcp_cork) | |
5683 | drbd_tcp_uncork(connection->meta.socket); | |
5684 | ||
5685 | return; | |
5686 | } |