1 // SPDX-License-Identifier: GPL-2.0-or-later
2 /* -*- mode: c; c-basic-offset: 8; -*-
3 * vim: noexpandtab sw=8 ts=8 sts=0:
4 *
5 * Copyright (C) 2004, 2005 Oracle. All rights reserved.
6 */
7
8 #include <linux/kernel.h>
9 #include <linux/sched.h>
10 #include <linux/jiffies.h>
11 #include <linux/module.h>
12 #include <linux/fs.h>
13 #include <linux/bio.h>
14 #include <linux/blkdev.h>
15 #include <linux/delay.h>
16 #include <linux/file.h>
17 #include <linux/kthread.h>
18 #include <linux/configfs.h>
19 #include <linux/random.h>
20 #include <linux/crc32.h>
21 #include <linux/time.h>
22 #include <linux/debugfs.h>
23 #include <linux/slab.h>
24 #include <linux/bitmap.h>
25 #include <linux/ktime.h>
26 #include "heartbeat.h"
27 #include "tcp.h"
28 #include "nodemanager.h"
29 #include "quorum.h"
30
31 #include "masklog.h"
32
33
34 /*
35 * The first heartbeat pass had one global thread that would serialize all hb
36 * callback calls. This global serializing sem should only be removed once
37 * we've made sure that all callees can deal with being called concurrently
38 * from multiple hb region threads.
39 */
40 static DECLARE_RWSEM(o2hb_callback_sem);
41
42 /*
43 * multiple hb threads are watching multiple regions. A node is live
44 * whenever any of the threads sees activity from the node in its region.
45 */
46 static DEFINE_SPINLOCK(o2hb_live_lock);
47 static struct list_head o2hb_live_slots[O2NM_MAX_NODES];
48 static unsigned long o2hb_live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
49 static LIST_HEAD(o2hb_node_events);
50 static DECLARE_WAIT_QUEUE_HEAD(o2hb_steady_queue);
51
52 /*
53 * In global heartbeat, we maintain a series of region bitmaps.
54 * - o2hb_region_bitmap allows us to limit the region number to max region.
55 * - o2hb_live_region_bitmap tracks live regions (seen steady iterations).
56 * - o2hb_quorum_region_bitmap tracks live regions that have seen all nodes
57 * heartbeat on it.
58 * - o2hb_failed_region_bitmap tracks the regions that have seen io timeouts.
59 */
60 static unsigned long o2hb_region_bitmap[BITS_TO_LONGS(O2NM_MAX_REGIONS)];
61 static unsigned long o2hb_live_region_bitmap[BITS_TO_LONGS(O2NM_MAX_REGIONS)];
62 static unsigned long o2hb_quorum_region_bitmap[BITS_TO_LONGS(O2NM_MAX_REGIONS)];
63 static unsigned long o2hb_failed_region_bitmap[BITS_TO_LONGS(O2NM_MAX_REGIONS)];
64
65 #define O2HB_DB_TYPE_LIVENODES 0
66 #define O2HB_DB_TYPE_LIVEREGIONS 1
67 #define O2HB_DB_TYPE_QUORUMREGIONS 2
68 #define O2HB_DB_TYPE_FAILEDREGIONS 3
69 #define O2HB_DB_TYPE_REGION_LIVENODES 4
70 #define O2HB_DB_TYPE_REGION_NUMBER 5
71 #define O2HB_DB_TYPE_REGION_ELAPSED_TIME 6
72 #define O2HB_DB_TYPE_REGION_PINNED 7
73 struct o2hb_debug_buf {
74 int db_type;
75 int db_size;
76 int db_len;
77 void *db_data;
78 };
79
80 static struct o2hb_debug_buf *o2hb_db_livenodes;
81 static struct o2hb_debug_buf *o2hb_db_liveregions;
82 static struct o2hb_debug_buf *o2hb_db_quorumregions;
83 static struct o2hb_debug_buf *o2hb_db_failedregions;
84
85 #define O2HB_DEBUG_DIR "o2hb"
86 #define O2HB_DEBUG_LIVENODES "livenodes"
87 #define O2HB_DEBUG_LIVEREGIONS "live_regions"
88 #define O2HB_DEBUG_QUORUMREGIONS "quorum_regions"
89 #define O2HB_DEBUG_FAILEDREGIONS "failed_regions"
90 #define O2HB_DEBUG_REGION_NUMBER "num"
91 #define O2HB_DEBUG_REGION_ELAPSED_TIME "elapsed_time_in_ms"
92 #define O2HB_DEBUG_REGION_PINNED "pinned"
93
94 static struct dentry *o2hb_debug_dir;
95
96 static LIST_HEAD(o2hb_all_regions);
97
98 static struct o2hb_callback {
99 struct list_head list;
100 } o2hb_callbacks[O2HB_NUM_CB];
101
102 static struct o2hb_callback *hbcall_from_type(enum o2hb_callback_type type);
103
104 #define O2HB_DEFAULT_BLOCK_BITS 9
105
106 enum o2hb_heartbeat_modes {
107 O2HB_HEARTBEAT_LOCAL = 0,
108 O2HB_HEARTBEAT_GLOBAL,
109 O2HB_HEARTBEAT_NUM_MODES,
110 };
111
112 static const char *o2hb_heartbeat_mode_desc[O2HB_HEARTBEAT_NUM_MODES] = {
113 "local", /* O2HB_HEARTBEAT_LOCAL */
114 "global", /* O2HB_HEARTBEAT_GLOBAL */
115 };
116
117 unsigned int o2hb_dead_threshold = O2HB_DEFAULT_DEAD_THRESHOLD;
118 static unsigned int o2hb_heartbeat_mode = O2HB_HEARTBEAT_LOCAL;
119
120 /*
121 * o2hb_dependent_users tracks the number of registered callbacks that depend
122 * on heartbeat. o2net and o2dlm are two entities that register this callback.
123 * However only o2dlm depends on the heartbeat. It does not want the heartbeat
124 * to stop while a dlm domain is still active.
125 */
126 static unsigned int o2hb_dependent_users;
127
128 /*
129 * In global heartbeat mode, all regions are pinned if there are one or more
130 * dependent users and the quorum region count is <= O2HB_PIN_CUT_OFF. All
131 * regions are unpinned if the region count exceeds the cut off or the number
132 * of dependent users falls to zero.
133 */
134 #define O2HB_PIN_CUT_OFF 3
135
136 /*
137 * In local heartbeat mode, we assume the dlm domain name to be the same as
138 * region uuid. This is true for domains created for the file system but not
139 * necessarily true for userdlm domains. This is a known limitation.
140 *
141 * In global heartbeat mode, we pin/unpin all o2hb regions. This solution
142 * works for both file system and userdlm domains.
143 */
144 static int o2hb_region_pin(const char *region_uuid);
145 static void o2hb_region_unpin(const char *region_uuid);
146
147 /* Only sets a new threshold if there are no active regions.
148 *
149 * No locking or otherwise interesting code is required for reading
150 * o2hb_dead_threshold as it can't change once regions are active and
151 * it's not interesting to anyone until then anyway. */
o2hb_dead_threshold_set(unsigned int threshold)152 static void o2hb_dead_threshold_set(unsigned int threshold)
153 {
154 if (threshold > O2HB_MIN_DEAD_THRESHOLD) {
155 spin_lock(&o2hb_live_lock);
156 if (list_empty(&o2hb_all_regions))
157 o2hb_dead_threshold = threshold;
158 spin_unlock(&o2hb_live_lock);
159 }
160 }
161
o2hb_global_heartbeat_mode_set(unsigned int hb_mode)162 static int o2hb_global_heartbeat_mode_set(unsigned int hb_mode)
163 {
164 int ret = -1;
165
166 if (hb_mode < O2HB_HEARTBEAT_NUM_MODES) {
167 spin_lock(&o2hb_live_lock);
168 if (list_empty(&o2hb_all_regions)) {
169 o2hb_heartbeat_mode = hb_mode;
170 ret = 0;
171 }
172 spin_unlock(&o2hb_live_lock);
173 }
174
175 return ret;
176 }
177
178 struct o2hb_node_event {
179 struct list_head hn_item;
180 enum o2hb_callback_type hn_event_type;
181 struct o2nm_node *hn_node;
182 int hn_node_num;
183 };
184
185 struct o2hb_disk_slot {
186 struct o2hb_disk_heartbeat_block *ds_raw_block;
187 u8 ds_node_num;
188 u64 ds_last_time;
189 u64 ds_last_generation;
190 u16 ds_equal_samples;
191 u16 ds_changed_samples;
192 struct list_head ds_live_item;
193 };
194
195 /* each thread owns a region.. when we're asked to tear down the region
196 * we ask the thread to stop, who cleans up the region */
197 struct o2hb_region {
198 struct config_item hr_item;
199
200 struct list_head hr_all_item;
201 unsigned hr_unclean_stop:1,
202 hr_aborted_start:1,
203 hr_item_pinned:1,
204 hr_item_dropped:1,
205 hr_node_deleted:1;
206
207 /* protected by the hr_callback_sem */
208 struct task_struct *hr_task;
209
210 unsigned int hr_blocks;
211 unsigned long long hr_start_block;
212
213 unsigned int hr_block_bits;
214 unsigned int hr_block_bytes;
215
216 unsigned int hr_slots_per_page;
217 unsigned int hr_num_pages;
218
219 struct page **hr_slot_data;
220 struct block_device *hr_bdev;
221 struct o2hb_disk_slot *hr_slots;
222
223 /* live node map of this region */
224 unsigned long hr_live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
225 unsigned int hr_region_num;
226
227 struct dentry *hr_debug_dir;
228 struct o2hb_debug_buf *hr_db_livenodes;
229 struct o2hb_debug_buf *hr_db_regnum;
230 struct o2hb_debug_buf *hr_db_elapsed_time;
231 struct o2hb_debug_buf *hr_db_pinned;
232
233 /* let the person setting up hb wait for it to return until it
234 * has reached a 'steady' state. This will be fixed when we have
235 * a more complete api that doesn't lead to this sort of fragility. */
236 atomic_t hr_steady_iterations;
237
238 /* terminate o2hb thread if it does not reach steady state
239 * (hr_steady_iterations == 0) within hr_unsteady_iterations */
240 atomic_t hr_unsteady_iterations;
241
242 char hr_dev_name[BDEVNAME_SIZE];
243
244 unsigned int hr_timeout_ms;
245
246 /* randomized as the region goes up and down so that a node
247 * recognizes a node going up and down in one iteration */
248 u64 hr_generation;
249
250 struct delayed_work hr_write_timeout_work;
251 unsigned long hr_last_timeout_start;
252
253 /* negotiate timer, used to negotiate extending hb timeout. */
254 struct delayed_work hr_nego_timeout_work;
255 unsigned long hr_nego_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
256
257 /* Used during o2hb_check_slot to hold a copy of the block
258 * being checked because we temporarily have to zero out the
259 * crc field. */
260 struct o2hb_disk_heartbeat_block *hr_tmp_block;
261
262 /* Message key for negotiate timeout message. */
263 unsigned int hr_key;
264 struct list_head hr_handler_list;
265
266 /* last hb status, 0 for success, other value for error. */
267 int hr_last_hb_status;
268 };
269
270 struct o2hb_bio_wait_ctxt {
271 atomic_t wc_num_reqs;
272 struct completion wc_io_complete;
273 int wc_error;
274 };
275
276 #define O2HB_NEGO_TIMEOUT_MS (O2HB_MAX_WRITE_TIMEOUT_MS/2)
277
278 enum {
279 O2HB_NEGO_TIMEOUT_MSG = 1,
280 O2HB_NEGO_APPROVE_MSG = 2,
281 };
282
283 struct o2hb_nego_msg {
284 u8 node_num;
285 };
286
o2hb_write_timeout(struct work_struct * work)287 static void o2hb_write_timeout(struct work_struct *work)
288 {
289 int failed, quorum;
290 struct o2hb_region *reg =
291 container_of(work, struct o2hb_region,
292 hr_write_timeout_work.work);
293
294 mlog(ML_ERROR, "Heartbeat write timeout to device %s after %u "
295 "milliseconds\n", reg->hr_dev_name,
296 jiffies_to_msecs(jiffies - reg->hr_last_timeout_start));
297
298 if (o2hb_global_heartbeat_active()) {
299 spin_lock(&o2hb_live_lock);
300 if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
301 set_bit(reg->hr_region_num, o2hb_failed_region_bitmap);
302 failed = bitmap_weight(o2hb_failed_region_bitmap,
303 O2NM_MAX_REGIONS);
304 quorum = bitmap_weight(o2hb_quorum_region_bitmap,
305 O2NM_MAX_REGIONS);
306 spin_unlock(&o2hb_live_lock);
307
308 mlog(ML_HEARTBEAT, "Number of regions %d, failed regions %d\n",
309 quorum, failed);
310
311 /*
312 * Fence if the number of failed regions >= half the number
313 * of quorum regions
314 */
315 if ((failed << 1) < quorum)
316 return;
317 }
318
319 o2quo_disk_timeout();
320 }
321
o2hb_arm_timeout(struct o2hb_region * reg)322 static void o2hb_arm_timeout(struct o2hb_region *reg)
323 {
324 /* Arm writeout only after thread reaches steady state */
325 if (atomic_read(®->hr_steady_iterations) != 0)
326 return;
327
328 mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n",
329 O2HB_MAX_WRITE_TIMEOUT_MS);
330
331 if (o2hb_global_heartbeat_active()) {
332 spin_lock(&o2hb_live_lock);
333 clear_bit(reg->hr_region_num, o2hb_failed_region_bitmap);
334 spin_unlock(&o2hb_live_lock);
335 }
336 cancel_delayed_work(®->hr_write_timeout_work);
337 schedule_delayed_work(®->hr_write_timeout_work,
338 msecs_to_jiffies(O2HB_MAX_WRITE_TIMEOUT_MS));
339
340 cancel_delayed_work(®->hr_nego_timeout_work);
341 /* negotiate timeout must be less than write timeout. */
342 schedule_delayed_work(®->hr_nego_timeout_work,
343 msecs_to_jiffies(O2HB_NEGO_TIMEOUT_MS));
344 memset(reg->hr_nego_node_bitmap, 0, sizeof(reg->hr_nego_node_bitmap));
345 }
346
o2hb_disarm_timeout(struct o2hb_region * reg)347 static void o2hb_disarm_timeout(struct o2hb_region *reg)
348 {
349 cancel_delayed_work_sync(®->hr_write_timeout_work);
350 cancel_delayed_work_sync(®->hr_nego_timeout_work);
351 }
352
o2hb_send_nego_msg(int key,int type,u8 target)353 static int o2hb_send_nego_msg(int key, int type, u8 target)
354 {
355 struct o2hb_nego_msg msg;
356 int status, ret;
357
358 msg.node_num = o2nm_this_node();
359 again:
360 ret = o2net_send_message(type, key, &msg, sizeof(msg),
361 target, &status);
362
363 if (ret == -EAGAIN || ret == -ENOMEM) {
364 msleep(100);
365 goto again;
366 }
367
368 return ret;
369 }
370
o2hb_nego_timeout(struct work_struct * work)371 static void o2hb_nego_timeout(struct work_struct *work)
372 {
373 unsigned long live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
374 int master_node, i, ret;
375 struct o2hb_region *reg;
376
377 reg = container_of(work, struct o2hb_region, hr_nego_timeout_work.work);
378 /* don't negotiate timeout if last hb failed since it is very
379 * possible io failed. Should let write timeout fence self.
380 */
381 if (reg->hr_last_hb_status)
382 return;
383
384 o2hb_fill_node_map(live_node_bitmap, sizeof(live_node_bitmap));
385 /* lowest node as master node to make negotiate decision. */
386 master_node = find_next_bit(live_node_bitmap, O2NM_MAX_NODES, 0);
387
388 if (master_node == o2nm_this_node()) {
389 if (!test_bit(master_node, reg->hr_nego_node_bitmap)) {
390 printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%s).\n",
391 o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000,
392 config_item_name(®->hr_item), reg->hr_dev_name);
393 set_bit(master_node, reg->hr_nego_node_bitmap);
394 }
395 if (memcmp(reg->hr_nego_node_bitmap, live_node_bitmap,
396 sizeof(reg->hr_nego_node_bitmap))) {
397 /* check negotiate bitmap every second to do timeout
398 * approve decision.
399 */
400 schedule_delayed_work(®->hr_nego_timeout_work,
401 msecs_to_jiffies(1000));
402
403 return;
404 }
405
406 printk(KERN_NOTICE "o2hb: all nodes hb write hung, maybe region %s (%s) is down.\n",
407 config_item_name(®->hr_item), reg->hr_dev_name);
408 /* approve negotiate timeout request. */
409 o2hb_arm_timeout(reg);
410
411 i = -1;
412 while ((i = find_next_bit(live_node_bitmap,
413 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
414 if (i == master_node)
415 continue;
416
417 mlog(ML_HEARTBEAT, "send NEGO_APPROVE msg to node %d\n", i);
418 ret = o2hb_send_nego_msg(reg->hr_key,
419 O2HB_NEGO_APPROVE_MSG, i);
420 if (ret)
421 mlog(ML_ERROR, "send NEGO_APPROVE msg to node %d fail %d\n",
422 i, ret);
423 }
424 } else {
425 /* negotiate timeout with master node. */
426 printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%s), negotiate timeout with node %d.\n",
427 o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000, config_item_name(®->hr_item),
428 reg->hr_dev_name, master_node);
429 ret = o2hb_send_nego_msg(reg->hr_key, O2HB_NEGO_TIMEOUT_MSG,
430 master_node);
431 if (ret)
432 mlog(ML_ERROR, "send NEGO_TIMEOUT msg to node %d fail %d\n",
433 master_node, ret);
434 }
435 }
436
o2hb_nego_timeout_handler(struct o2net_msg * msg,u32 len,void * data,void ** ret_data)437 static int o2hb_nego_timeout_handler(struct o2net_msg *msg, u32 len, void *data,
438 void **ret_data)
439 {
440 struct o2hb_region *reg = data;
441 struct o2hb_nego_msg *nego_msg;
442
443 nego_msg = (struct o2hb_nego_msg *)msg->buf;
444 printk(KERN_NOTICE "o2hb: receive negotiate timeout message from node %d on region %s (%s).\n",
445 nego_msg->node_num, config_item_name(®->hr_item), reg->hr_dev_name);
446 if (nego_msg->node_num < O2NM_MAX_NODES)
447 set_bit(nego_msg->node_num, reg->hr_nego_node_bitmap);
448 else
449 mlog(ML_ERROR, "got nego timeout message from bad node.\n");
450
451 return 0;
452 }
453
o2hb_nego_approve_handler(struct o2net_msg * msg,u32 len,void * data,void ** ret_data)454 static int o2hb_nego_approve_handler(struct o2net_msg *msg, u32 len, void *data,
455 void **ret_data)
456 {
457 struct o2hb_region *reg = data;
458
459 printk(KERN_NOTICE "o2hb: negotiate timeout approved by master node on region %s (%s).\n",
460 config_item_name(®->hr_item), reg->hr_dev_name);
461 o2hb_arm_timeout(reg);
462 return 0;
463 }
464
o2hb_bio_wait_init(struct o2hb_bio_wait_ctxt * wc)465 static inline void o2hb_bio_wait_init(struct o2hb_bio_wait_ctxt *wc)
466 {
467 atomic_set(&wc->wc_num_reqs, 1);
468 init_completion(&wc->wc_io_complete);
469 wc->wc_error = 0;
470 }
471
472 /* Used in error paths too */
o2hb_bio_wait_dec(struct o2hb_bio_wait_ctxt * wc,unsigned int num)473 static inline void o2hb_bio_wait_dec(struct o2hb_bio_wait_ctxt *wc,
474 unsigned int num)
475 {
476 /* sadly atomic_sub_and_test() isn't available on all platforms. The
477 * good news is that the fast path only completes one at a time */
478 while(num--) {
479 if (atomic_dec_and_test(&wc->wc_num_reqs)) {
480 BUG_ON(num > 0);
481 complete(&wc->wc_io_complete);
482 }
483 }
484 }
485
o2hb_wait_on_io(struct o2hb_bio_wait_ctxt * wc)486 static void o2hb_wait_on_io(struct o2hb_bio_wait_ctxt *wc)
487 {
488 o2hb_bio_wait_dec(wc, 1);
489 wait_for_completion(&wc->wc_io_complete);
490 }
491
o2hb_bio_end_io(struct bio * bio)492 static void o2hb_bio_end_io(struct bio *bio)
493 {
494 struct o2hb_bio_wait_ctxt *wc = bio->bi_private;
495
496 if (bio->bi_status) {
497 mlog(ML_ERROR, "IO Error %d\n", bio->bi_status);
498 wc->wc_error = blk_status_to_errno(bio->bi_status);
499 }
500
501 o2hb_bio_wait_dec(wc, 1);
502 bio_put(bio);
503 }
504
505 /* Setup a Bio to cover I/O against num_slots slots starting at
506 * start_slot. */
o2hb_setup_one_bio(struct o2hb_region * reg,struct o2hb_bio_wait_ctxt * wc,unsigned int * current_slot,unsigned int max_slots,int op,int op_flags)507 static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg,
508 struct o2hb_bio_wait_ctxt *wc,
509 unsigned int *current_slot,
510 unsigned int max_slots, int op,
511 int op_flags)
512 {
513 int len, current_page;
514 unsigned int vec_len, vec_start;
515 unsigned int bits = reg->hr_block_bits;
516 unsigned int spp = reg->hr_slots_per_page;
517 unsigned int cs = *current_slot;
518 struct bio *bio;
519 struct page *page;
520
521 /* Testing has shown this allocation to take long enough under
522 * GFP_KERNEL that the local node can get fenced. It would be
523 * nicest if we could pre-allocate these bios and avoid this
524 * all together. */
525 bio = bio_alloc(GFP_ATOMIC, 16);
526 if (!bio) {
527 mlog(ML_ERROR, "Could not alloc slots BIO!\n");
528 bio = ERR_PTR(-ENOMEM);
529 goto bail;
530 }
531
532 /* Must put everything in 512 byte sectors for the bio... */
533 bio->bi_iter.bi_sector = (reg->hr_start_block + cs) << (bits - 9);
534 bio_set_dev(bio, reg->hr_bdev);
535 bio->bi_private = wc;
536 bio->bi_end_io = o2hb_bio_end_io;
537 bio_set_op_attrs(bio, op, op_flags);
538
539 vec_start = (cs << bits) % PAGE_SIZE;
540 while(cs < max_slots) {
541 current_page = cs / spp;
542 page = reg->hr_slot_data[current_page];
543
544 vec_len = min(PAGE_SIZE - vec_start,
545 (max_slots-cs) * (PAGE_SIZE/spp) );
546
547 mlog(ML_HB_BIO, "page %d, vec_len = %u, vec_start = %u\n",
548 current_page, vec_len, vec_start);
549
550 len = bio_add_page(bio, page, vec_len, vec_start);
551 if (len != vec_len) break;
552
553 cs += vec_len / (PAGE_SIZE/spp);
554 vec_start = 0;
555 }
556
557 bail:
558 *current_slot = cs;
559 return bio;
560 }
561
o2hb_read_slots(struct o2hb_region * reg,unsigned int begin_slot,unsigned int max_slots)562 static int o2hb_read_slots(struct o2hb_region *reg,
563 unsigned int begin_slot,
564 unsigned int max_slots)
565 {
566 unsigned int current_slot = begin_slot;
567 int status;
568 struct o2hb_bio_wait_ctxt wc;
569 struct bio *bio;
570
571 o2hb_bio_wait_init(&wc);
572
573 while(current_slot < max_slots) {
574 bio = o2hb_setup_one_bio(reg, &wc, ¤t_slot, max_slots,
575 REQ_OP_READ, 0);
576 if (IS_ERR(bio)) {
577 status = PTR_ERR(bio);
578 mlog_errno(status);
579 goto bail_and_wait;
580 }
581
582 atomic_inc(&wc.wc_num_reqs);
583 submit_bio(bio);
584 }
585
586 status = 0;
587
588 bail_and_wait:
589 o2hb_wait_on_io(&wc);
590 if (wc.wc_error && !status)
591 status = wc.wc_error;
592
593 return status;
594 }
595
o2hb_issue_node_write(struct o2hb_region * reg,struct o2hb_bio_wait_ctxt * write_wc)596 static int o2hb_issue_node_write(struct o2hb_region *reg,
597 struct o2hb_bio_wait_ctxt *write_wc)
598 {
599 int status;
600 unsigned int slot;
601 struct bio *bio;
602
603 o2hb_bio_wait_init(write_wc);
604
605 slot = o2nm_this_node();
606
607 bio = o2hb_setup_one_bio(reg, write_wc, &slot, slot+1, REQ_OP_WRITE,
608 REQ_SYNC);
609 if (IS_ERR(bio)) {
610 status = PTR_ERR(bio);
611 mlog_errno(status);
612 goto bail;
613 }
614
615 atomic_inc(&write_wc->wc_num_reqs);
616 submit_bio(bio);
617
618 status = 0;
619 bail:
620 return status;
621 }
622
o2hb_compute_block_crc_le(struct o2hb_region * reg,struct o2hb_disk_heartbeat_block * hb_block)623 static u32 o2hb_compute_block_crc_le(struct o2hb_region *reg,
624 struct o2hb_disk_heartbeat_block *hb_block)
625 {
626 __le32 old_cksum;
627 u32 ret;
628
629 /* We want to compute the block crc with a 0 value in the
630 * hb_cksum field. Save it off here and replace after the
631 * crc. */
632 old_cksum = hb_block->hb_cksum;
633 hb_block->hb_cksum = 0;
634
635 ret = crc32_le(0, (unsigned char *) hb_block, reg->hr_block_bytes);
636
637 hb_block->hb_cksum = old_cksum;
638
639 return ret;
640 }
641
o2hb_dump_slot(struct o2hb_disk_heartbeat_block * hb_block)642 static void o2hb_dump_slot(struct o2hb_disk_heartbeat_block *hb_block)
643 {
644 mlog(ML_ERROR, "Dump slot information: seq = 0x%llx, node = %u, "
645 "cksum = 0x%x, generation 0x%llx\n",
646 (long long)le64_to_cpu(hb_block->hb_seq),
647 hb_block->hb_node, le32_to_cpu(hb_block->hb_cksum),
648 (long long)le64_to_cpu(hb_block->hb_generation));
649 }
650
o2hb_verify_crc(struct o2hb_region * reg,struct o2hb_disk_heartbeat_block * hb_block)651 static int o2hb_verify_crc(struct o2hb_region *reg,
652 struct o2hb_disk_heartbeat_block *hb_block)
653 {
654 u32 read, computed;
655
656 read = le32_to_cpu(hb_block->hb_cksum);
657 computed = o2hb_compute_block_crc_le(reg, hb_block);
658
659 return read == computed;
660 }
661
662 /*
663 * Compare the slot data with what we wrote in the last iteration.
664 * If the match fails, print an appropriate error message. This is to
665 * detect errors like... another node hearting on the same slot,
666 * flaky device that is losing writes, etc.
667 * Returns 1 if check succeeds, 0 otherwise.
668 */
o2hb_check_own_slot(struct o2hb_region * reg)669 static int o2hb_check_own_slot(struct o2hb_region *reg)
670 {
671 struct o2hb_disk_slot *slot;
672 struct o2hb_disk_heartbeat_block *hb_block;
673 char *errstr;
674
675 slot = ®->hr_slots[o2nm_this_node()];
676 /* Don't check on our 1st timestamp */
677 if (!slot->ds_last_time)
678 return 0;
679
680 hb_block = slot->ds_raw_block;
681 if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time &&
682 le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation &&
683 hb_block->hb_node == slot->ds_node_num)
684 return 1;
685
686 #define ERRSTR1 "Another node is heartbeating on device"
687 #define ERRSTR2 "Heartbeat generation mismatch on device"
688 #define ERRSTR3 "Heartbeat sequence mismatch on device"
689
690 if (hb_block->hb_node != slot->ds_node_num)
691 errstr = ERRSTR1;
692 else if (le64_to_cpu(hb_block->hb_generation) !=
693 slot->ds_last_generation)
694 errstr = ERRSTR2;
695 else
696 errstr = ERRSTR3;
697
698 mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), "
699 "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name,
700 slot->ds_node_num, (unsigned long long)slot->ds_last_generation,
701 (unsigned long long)slot->ds_last_time, hb_block->hb_node,
702 (unsigned long long)le64_to_cpu(hb_block->hb_generation),
703 (unsigned long long)le64_to_cpu(hb_block->hb_seq));
704
705 return 0;
706 }
707
o2hb_prepare_block(struct o2hb_region * reg,u64 generation)708 static inline void o2hb_prepare_block(struct o2hb_region *reg,
709 u64 generation)
710 {
711 int node_num;
712 u64 cputime;
713 struct o2hb_disk_slot *slot;
714 struct o2hb_disk_heartbeat_block *hb_block;
715
716 node_num = o2nm_this_node();
717 slot = ®->hr_slots[node_num];
718
719 hb_block = (struct o2hb_disk_heartbeat_block *)slot->ds_raw_block;
720 memset(hb_block, 0, reg->hr_block_bytes);
721 /* TODO: time stuff */
722 cputime = ktime_get_real_seconds();
723 if (!cputime)
724 cputime = 1;
725
726 hb_block->hb_seq = cpu_to_le64(cputime);
727 hb_block->hb_node = node_num;
728 hb_block->hb_generation = cpu_to_le64(generation);
729 hb_block->hb_dead_ms = cpu_to_le32(o2hb_dead_threshold * O2HB_REGION_TIMEOUT_MS);
730
731 /* This step must always happen last! */
732 hb_block->hb_cksum = cpu_to_le32(o2hb_compute_block_crc_le(reg,
733 hb_block));
734
735 mlog(ML_HB_BIO, "our node generation = 0x%llx, cksum = 0x%x\n",
736 (long long)generation,
737 le32_to_cpu(hb_block->hb_cksum));
738 }
739
o2hb_fire_callbacks(struct o2hb_callback * hbcall,struct o2nm_node * node,int idx)740 static void o2hb_fire_callbacks(struct o2hb_callback *hbcall,
741 struct o2nm_node *node,
742 int idx)
743 {
744 struct o2hb_callback_func *f;
745
746 list_for_each_entry(f, &hbcall->list, hc_item) {
747 mlog(ML_HEARTBEAT, "calling funcs %p\n", f);
748 (f->hc_func)(node, idx, f->hc_data);
749 }
750 }
751
752 /* Will run the list in order until we process the passed event */
o2hb_run_event_list(struct o2hb_node_event * queued_event)753 static void o2hb_run_event_list(struct o2hb_node_event *queued_event)
754 {
755 struct o2hb_callback *hbcall;
756 struct o2hb_node_event *event;
757
758 /* Holding callback sem assures we don't alter the callback
759 * lists when doing this, and serializes ourselves with other
760 * processes wanting callbacks. */
761 down_write(&o2hb_callback_sem);
762
763 spin_lock(&o2hb_live_lock);
764 while (!list_empty(&o2hb_node_events)
765 && !list_empty(&queued_event->hn_item)) {
766 event = list_entry(o2hb_node_events.next,
767 struct o2hb_node_event,
768 hn_item);
769 list_del_init(&event->hn_item);
770 spin_unlock(&o2hb_live_lock);
771
772 mlog(ML_HEARTBEAT, "Node %s event for %d\n",
773 event->hn_event_type == O2HB_NODE_UP_CB ? "UP" : "DOWN",
774 event->hn_node_num);
775
776 hbcall = hbcall_from_type(event->hn_event_type);
777
778 /* We should *never* have gotten on to the list with a
779 * bad type... This isn't something that we should try
780 * to recover from. */
781 BUG_ON(IS_ERR(hbcall));
782
783 o2hb_fire_callbacks(hbcall, event->hn_node, event->hn_node_num);
784
785 spin_lock(&o2hb_live_lock);
786 }
787 spin_unlock(&o2hb_live_lock);
788
789 up_write(&o2hb_callback_sem);
790 }
791
o2hb_queue_node_event(struct o2hb_node_event * event,enum o2hb_callback_type type,struct o2nm_node * node,int node_num)792 static void o2hb_queue_node_event(struct o2hb_node_event *event,
793 enum o2hb_callback_type type,
794 struct o2nm_node *node,
795 int node_num)
796 {
797 assert_spin_locked(&o2hb_live_lock);
798
799 BUG_ON((!node) && (type != O2HB_NODE_DOWN_CB));
800
801 event->hn_event_type = type;
802 event->hn_node = node;
803 event->hn_node_num = node_num;
804
805 mlog(ML_HEARTBEAT, "Queue node %s event for node %d\n",
806 type == O2HB_NODE_UP_CB ? "UP" : "DOWN", node_num);
807
808 list_add_tail(&event->hn_item, &o2hb_node_events);
809 }
810
o2hb_shutdown_slot(struct o2hb_disk_slot * slot)811 static void o2hb_shutdown_slot(struct o2hb_disk_slot *slot)
812 {
813 struct o2hb_node_event event =
814 { .hn_item = LIST_HEAD_INIT(event.hn_item), };
815 struct o2nm_node *node;
816 int queued = 0;
817
818 node = o2nm_get_node_by_num(slot->ds_node_num);
819 if (!node)
820 return;
821
822 spin_lock(&o2hb_live_lock);
823 if (!list_empty(&slot->ds_live_item)) {
824 mlog(ML_HEARTBEAT, "Shutdown, node %d leaves region\n",
825 slot->ds_node_num);
826
827 list_del_init(&slot->ds_live_item);
828
829 if (list_empty(&o2hb_live_slots[slot->ds_node_num])) {
830 clear_bit(slot->ds_node_num, o2hb_live_node_bitmap);
831
832 o2hb_queue_node_event(&event, O2HB_NODE_DOWN_CB, node,
833 slot->ds_node_num);
834 queued = 1;
835 }
836 }
837 spin_unlock(&o2hb_live_lock);
838
839 if (queued)
840 o2hb_run_event_list(&event);
841
842 o2nm_node_put(node);
843 }
844
o2hb_set_quorum_device(struct o2hb_region * reg)845 static void o2hb_set_quorum_device(struct o2hb_region *reg)
846 {
847 if (!o2hb_global_heartbeat_active())
848 return;
849
850 /* Prevent race with o2hb_heartbeat_group_drop_item() */
851 if (kthread_should_stop())
852 return;
853
854 /* Tag region as quorum only after thread reaches steady state */
855 if (atomic_read(®->hr_steady_iterations) != 0)
856 return;
857
858 spin_lock(&o2hb_live_lock);
859
860 if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
861 goto unlock;
862
863 /*
864 * A region can be added to the quorum only when it sees all
865 * live nodes heartbeat on it. In other words, the region has been
866 * added to all nodes.
867 */
868 if (memcmp(reg->hr_live_node_bitmap, o2hb_live_node_bitmap,
869 sizeof(o2hb_live_node_bitmap)))
870 goto unlock;
871
872 printk(KERN_NOTICE "o2hb: Region %s (%s) is now a quorum device\n",
873 config_item_name(®->hr_item), reg->hr_dev_name);
874
875 set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
876
877 /*
878 * If global heartbeat active, unpin all regions if the
879 * region count > CUT_OFF
880 */
881 if (bitmap_weight(o2hb_quorum_region_bitmap,
882 O2NM_MAX_REGIONS) > O2HB_PIN_CUT_OFF)
883 o2hb_region_unpin(NULL);
884 unlock:
885 spin_unlock(&o2hb_live_lock);
886 }
887
o2hb_check_slot(struct o2hb_region * reg,struct o2hb_disk_slot * slot)888 static int o2hb_check_slot(struct o2hb_region *reg,
889 struct o2hb_disk_slot *slot)
890 {
891 int changed = 0, gen_changed = 0;
892 struct o2hb_node_event event =
893 { .hn_item = LIST_HEAD_INIT(event.hn_item), };
894 struct o2nm_node *node;
895 struct o2hb_disk_heartbeat_block *hb_block = reg->hr_tmp_block;
896 u64 cputime;
897 unsigned int dead_ms = o2hb_dead_threshold * O2HB_REGION_TIMEOUT_MS;
898 unsigned int slot_dead_ms;
899 int tmp;
900 int queued = 0;
901
902 memcpy(hb_block, slot->ds_raw_block, reg->hr_block_bytes);
903
904 /*
905 * If a node is no longer configured but is still in the livemap, we
906 * may need to clear that bit from the livemap.
907 */
908 node = o2nm_get_node_by_num(slot->ds_node_num);
909 if (!node) {
910 spin_lock(&o2hb_live_lock);
911 tmp = test_bit(slot->ds_node_num, o2hb_live_node_bitmap);
912 spin_unlock(&o2hb_live_lock);
913 if (!tmp)
914 return 0;
915 }
916
917 if (!o2hb_verify_crc(reg, hb_block)) {
918 /* all paths from here will drop o2hb_live_lock for
919 * us. */
920 spin_lock(&o2hb_live_lock);
921
922 /* Don't print an error on the console in this case -
923 * a freshly formatted heartbeat area will not have a
924 * crc set on it. */
925 if (list_empty(&slot->ds_live_item))
926 goto out;
927
928 /* The node is live but pushed out a bad crc. We
929 * consider it a transient miss but don't populate any
930 * other values as they may be junk. */
931 mlog(ML_ERROR, "Node %d has written a bad crc to %s\n",
932 slot->ds_node_num, reg->hr_dev_name);
933 o2hb_dump_slot(hb_block);
934
935 slot->ds_equal_samples++;
936 goto fire_callbacks;
937 }
938
939 /* we don't care if these wrap.. the state transitions below
940 * clear at the right places */
941 cputime = le64_to_cpu(hb_block->hb_seq);
942 if (slot->ds_last_time != cputime)
943 slot->ds_changed_samples++;
944 else
945 slot->ds_equal_samples++;
946 slot->ds_last_time = cputime;
947
948 /* The node changed heartbeat generations. We assume this to
949 * mean it dropped off but came back before we timed out. We
950 * want to consider it down for the time being but don't want
951 * to lose any changed_samples state we might build up to
952 * considering it live again. */
953 if (slot->ds_last_generation != le64_to_cpu(hb_block->hb_generation)) {
954 gen_changed = 1;
955 slot->ds_equal_samples = 0;
956 mlog(ML_HEARTBEAT, "Node %d changed generation (0x%llx "
957 "to 0x%llx)\n", slot->ds_node_num,
958 (long long)slot->ds_last_generation,
959 (long long)le64_to_cpu(hb_block->hb_generation));
960 }
961
962 slot->ds_last_generation = le64_to_cpu(hb_block->hb_generation);
963
964 mlog(ML_HEARTBEAT, "Slot %d gen 0x%llx cksum 0x%x "
965 "seq %llu last %llu changed %u equal %u\n",
966 slot->ds_node_num, (long long)slot->ds_last_generation,
967 le32_to_cpu(hb_block->hb_cksum),
968 (unsigned long long)le64_to_cpu(hb_block->hb_seq),
969 (unsigned long long)slot->ds_last_time, slot->ds_changed_samples,
970 slot->ds_equal_samples);
971
972 spin_lock(&o2hb_live_lock);
973
974 fire_callbacks:
975 /* dead nodes only come to life after some number of
976 * changes at any time during their dead time */
977 if (list_empty(&slot->ds_live_item) &&
978 slot->ds_changed_samples >= O2HB_LIVE_THRESHOLD) {
979 mlog(ML_HEARTBEAT, "Node %d (id 0x%llx) joined my region\n",
980 slot->ds_node_num, (long long)slot->ds_last_generation);
981
982 set_bit(slot->ds_node_num, reg->hr_live_node_bitmap);
983
984 /* first on the list generates a callback */
985 if (list_empty(&o2hb_live_slots[slot->ds_node_num])) {
986 mlog(ML_HEARTBEAT, "o2hb: Add node %d to live nodes "
987 "bitmap\n", slot->ds_node_num);
988 set_bit(slot->ds_node_num, o2hb_live_node_bitmap);
989
990 o2hb_queue_node_event(&event, O2HB_NODE_UP_CB, node,
991 slot->ds_node_num);
992
993 changed = 1;
994 queued = 1;
995 }
996
997 list_add_tail(&slot->ds_live_item,
998 &o2hb_live_slots[slot->ds_node_num]);
999
1000 slot->ds_equal_samples = 0;
1001
1002 /* We want to be sure that all nodes agree on the
1003 * number of milliseconds before a node will be
1004 * considered dead. The self-fencing timeout is
1005 * computed from this value, and a discrepancy might
1006 * result in heartbeat calling a node dead when it
1007 * hasn't self-fenced yet. */
1008 slot_dead_ms = le32_to_cpu(hb_block->hb_dead_ms);
1009 if (slot_dead_ms && slot_dead_ms != dead_ms) {
1010 /* TODO: Perhaps we can fail the region here. */
1011 mlog(ML_ERROR, "Node %d on device %s has a dead count "
1012 "of %u ms, but our count is %u ms.\n"
1013 "Please double check your configuration values "
1014 "for 'O2CB_HEARTBEAT_THRESHOLD'\n",
1015 slot->ds_node_num, reg->hr_dev_name, slot_dead_ms,
1016 dead_ms);
1017 }
1018 goto out;
1019 }
1020
1021 /* if the list is dead, we're done.. */
1022 if (list_empty(&slot->ds_live_item))
1023 goto out;
1024
1025 /* live nodes only go dead after enough consequtive missed
1026 * samples.. reset the missed counter whenever we see
1027 * activity */
1028 if (slot->ds_equal_samples >= o2hb_dead_threshold || gen_changed) {
1029 mlog(ML_HEARTBEAT, "Node %d left my region\n",
1030 slot->ds_node_num);
1031
1032 clear_bit(slot->ds_node_num, reg->hr_live_node_bitmap);
1033
1034 /* last off the live_slot generates a callback */
1035 list_del_init(&slot->ds_live_item);
1036 if (list_empty(&o2hb_live_slots[slot->ds_node_num])) {
1037 mlog(ML_HEARTBEAT, "o2hb: Remove node %d from live "
1038 "nodes bitmap\n", slot->ds_node_num);
1039 clear_bit(slot->ds_node_num, o2hb_live_node_bitmap);
1040
1041 /* node can be null */
1042 o2hb_queue_node_event(&event, O2HB_NODE_DOWN_CB,
1043 node, slot->ds_node_num);
1044
1045 changed = 1;
1046 queued = 1;
1047 }
1048
1049 /* We don't clear this because the node is still
1050 * actually writing new blocks. */
1051 if (!gen_changed)
1052 slot->ds_changed_samples = 0;
1053 goto out;
1054 }
1055 if (slot->ds_changed_samples) {
1056 slot->ds_changed_samples = 0;
1057 slot->ds_equal_samples = 0;
1058 }
1059 out:
1060 spin_unlock(&o2hb_live_lock);
1061
1062 if (queued)
1063 o2hb_run_event_list(&event);
1064
1065 if (node)
1066 o2nm_node_put(node);
1067 return changed;
1068 }
1069
o2hb_highest_node(unsigned long * nodes,int numbits)1070 static int o2hb_highest_node(unsigned long *nodes, int numbits)
1071 {
1072 return find_last_bit(nodes, numbits);
1073 }
1074
o2hb_lowest_node(unsigned long * nodes,int numbits)1075 static int o2hb_lowest_node(unsigned long *nodes, int numbits)
1076 {
1077 return find_first_bit(nodes, numbits);
1078 }
1079
o2hb_do_disk_heartbeat(struct o2hb_region * reg)1080 static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
1081 {
1082 int i, ret, highest_node, lowest_node;
1083 int membership_change = 0, own_slot_ok = 0;
1084 unsigned long configured_nodes[BITS_TO_LONGS(O2NM_MAX_NODES)];
1085 unsigned long live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
1086 struct o2hb_bio_wait_ctxt write_wc;
1087
1088 ret = o2nm_configured_node_map(configured_nodes,
1089 sizeof(configured_nodes));
1090 if (ret) {
1091 mlog_errno(ret);
1092 goto bail;
1093 }
1094
1095 /*
1096 * If a node is not configured but is in the livemap, we still need
1097 * to read the slot so as to be able to remove it from the livemap.
1098 */
1099 o2hb_fill_node_map(live_node_bitmap, sizeof(live_node_bitmap));
1100 i = -1;
1101 while ((i = find_next_bit(live_node_bitmap,
1102 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
1103 set_bit(i, configured_nodes);
1104 }
1105
1106 highest_node = o2hb_highest_node(configured_nodes, O2NM_MAX_NODES);
1107 lowest_node = o2hb_lowest_node(configured_nodes, O2NM_MAX_NODES);
1108 if (highest_node >= O2NM_MAX_NODES || lowest_node >= O2NM_MAX_NODES) {
1109 mlog(ML_NOTICE, "o2hb: No configured nodes found!\n");
1110 ret = -EINVAL;
1111 goto bail;
1112 }
1113
1114 /* No sense in reading the slots of nodes that don't exist
1115 * yet. Of course, if the node definitions have holes in them
1116 * then we're reading an empty slot anyway... Consider this
1117 * best-effort. */
1118 ret = o2hb_read_slots(reg, lowest_node, highest_node + 1);
1119 if (ret < 0) {
1120 mlog_errno(ret);
1121 goto bail;
1122 }
1123
1124 /* With an up to date view of the slots, we can check that no
1125 * other node has been improperly configured to heartbeat in
1126 * our slot. */
1127 own_slot_ok = o2hb_check_own_slot(reg);
1128
1129 /* fill in the proper info for our next heartbeat */
1130 o2hb_prepare_block(reg, reg->hr_generation);
1131
1132 ret = o2hb_issue_node_write(reg, &write_wc);
1133 if (ret < 0) {
1134 mlog_errno(ret);
1135 goto bail;
1136 }
1137
1138 i = -1;
1139 while((i = find_next_bit(configured_nodes,
1140 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
1141 membership_change |= o2hb_check_slot(reg, ®->hr_slots[i]);
1142 }
1143
1144 /*
1145 * We have to be sure we've advertised ourselves on disk
1146 * before we can go to steady state. This ensures that
1147 * people we find in our steady state have seen us.
1148 */
1149 o2hb_wait_on_io(&write_wc);
1150 if (write_wc.wc_error) {
1151 /* Do not re-arm the write timeout on I/O error - we
1152 * can't be sure that the new block ever made it to
1153 * disk */
1154 mlog(ML_ERROR, "Write error %d on device \"%s\"\n",
1155 write_wc.wc_error, reg->hr_dev_name);
1156 ret = write_wc.wc_error;
1157 goto bail;
1158 }
1159
1160 /* Skip disarming the timeout if own slot has stale/bad data */
1161 if (own_slot_ok) {
1162 o2hb_set_quorum_device(reg);
1163 o2hb_arm_timeout(reg);
1164 reg->hr_last_timeout_start = jiffies;
1165 }
1166
1167 bail:
1168 /* let the person who launched us know when things are steady */
1169 if (atomic_read(®->hr_steady_iterations) != 0) {
1170 if (!ret && own_slot_ok && !membership_change) {
1171 if (atomic_dec_and_test(®->hr_steady_iterations))
1172 wake_up(&o2hb_steady_queue);
1173 }
1174 }
1175
1176 if (atomic_read(®->hr_steady_iterations) != 0) {
1177 if (atomic_dec_and_test(®->hr_unsteady_iterations)) {
1178 printk(KERN_NOTICE "o2hb: Unable to stabilize "
1179 "heartbeat on region %s (%s)\n",
1180 config_item_name(®->hr_item),
1181 reg->hr_dev_name);
1182 atomic_set(®->hr_steady_iterations, 0);
1183 reg->hr_aborted_start = 1;
1184 wake_up(&o2hb_steady_queue);
1185 ret = -EIO;
1186 }
1187 }
1188
1189 return ret;
1190 }
1191
1192 /*
1193 * we ride the region ref that the region dir holds. before the region
1194 * dir is removed and drops it ref it will wait to tear down this
1195 * thread.
1196 */
o2hb_thread(void * data)1197 static int o2hb_thread(void *data)
1198 {
1199 int i, ret;
1200 struct o2hb_region *reg = data;
1201 struct o2hb_bio_wait_ctxt write_wc;
1202 ktime_t before_hb, after_hb;
1203 unsigned int elapsed_msec;
1204
1205 mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread running\n");
1206
1207 set_user_nice(current, MIN_NICE);
1208
1209 /* Pin node */
1210 ret = o2nm_depend_this_node();
1211 if (ret) {
1212 mlog(ML_ERROR, "Node has been deleted, ret = %d\n", ret);
1213 reg->hr_node_deleted = 1;
1214 wake_up(&o2hb_steady_queue);
1215 return 0;
1216 }
1217
1218 while (!kthread_should_stop() &&
1219 !reg->hr_unclean_stop && !reg->hr_aborted_start) {
1220 /* We track the time spent inside
1221 * o2hb_do_disk_heartbeat so that we avoid more than
1222 * hr_timeout_ms between disk writes. On busy systems
1223 * this should result in a heartbeat which is less
1224 * likely to time itself out. */
1225 before_hb = ktime_get_real();
1226
1227 ret = o2hb_do_disk_heartbeat(reg);
1228 reg->hr_last_hb_status = ret;
1229
1230 after_hb = ktime_get_real();
1231
1232 elapsed_msec = (unsigned int)
1233 ktime_ms_delta(after_hb, before_hb);
1234
1235 mlog(ML_HEARTBEAT,
1236 "start = %lld, end = %lld, msec = %u, ret = %d\n",
1237 before_hb, after_hb, elapsed_msec, ret);
1238
1239 if (!kthread_should_stop() &&
1240 elapsed_msec < reg->hr_timeout_ms) {
1241 /* the kthread api has blocked signals for us so no
1242 * need to record the return value. */
1243 msleep_interruptible(reg->hr_timeout_ms - elapsed_msec);
1244 }
1245 }
1246
1247 o2hb_disarm_timeout(reg);
1248
1249 /* unclean stop is only used in very bad situation */
1250 for(i = 0; !reg->hr_unclean_stop && i < reg->hr_blocks; i++)
1251 o2hb_shutdown_slot(®->hr_slots[i]);
1252
1253 /* Explicit down notification - avoid forcing the other nodes
1254 * to timeout on this region when we could just as easily
1255 * write a clear generation - thus indicating to them that
1256 * this node has left this region.
1257 */
1258 if (!reg->hr_unclean_stop && !reg->hr_aborted_start) {
1259 o2hb_prepare_block(reg, 0);
1260 ret = o2hb_issue_node_write(reg, &write_wc);
1261 if (ret == 0)
1262 o2hb_wait_on_io(&write_wc);
1263 else
1264 mlog_errno(ret);
1265 }
1266
1267 /* Unpin node */
1268 o2nm_undepend_this_node();
1269
1270 mlog(ML_HEARTBEAT|ML_KTHREAD, "o2hb thread exiting\n");
1271
1272 return 0;
1273 }
1274
1275 #ifdef CONFIG_DEBUG_FS
o2hb_debug_open(struct inode * inode,struct file * file)1276 static int o2hb_debug_open(struct inode *inode, struct file *file)
1277 {
1278 struct o2hb_debug_buf *db = inode->i_private;
1279 struct o2hb_region *reg;
1280 unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)];
1281 unsigned long lts;
1282 char *buf = NULL;
1283 int i = -1;
1284 int out = 0;
1285
1286 /* max_nodes should be the largest bitmap we pass here */
1287 BUG_ON(sizeof(map) < db->db_size);
1288
1289 buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
1290 if (!buf)
1291 goto bail;
1292
1293 switch (db->db_type) {
1294 case O2HB_DB_TYPE_LIVENODES:
1295 case O2HB_DB_TYPE_LIVEREGIONS:
1296 case O2HB_DB_TYPE_QUORUMREGIONS:
1297 case O2HB_DB_TYPE_FAILEDREGIONS:
1298 spin_lock(&o2hb_live_lock);
1299 memcpy(map, db->db_data, db->db_size);
1300 spin_unlock(&o2hb_live_lock);
1301 break;
1302
1303 case O2HB_DB_TYPE_REGION_LIVENODES:
1304 spin_lock(&o2hb_live_lock);
1305 reg = (struct o2hb_region *)db->db_data;
1306 memcpy(map, reg->hr_live_node_bitmap, db->db_size);
1307 spin_unlock(&o2hb_live_lock);
1308 break;
1309
1310 case O2HB_DB_TYPE_REGION_NUMBER:
1311 reg = (struct o2hb_region *)db->db_data;
1312 out += snprintf(buf + out, PAGE_SIZE - out, "%d\n",
1313 reg->hr_region_num);
1314 goto done;
1315
1316 case O2HB_DB_TYPE_REGION_ELAPSED_TIME:
1317 reg = (struct o2hb_region *)db->db_data;
1318 lts = reg->hr_last_timeout_start;
1319 /* If 0, it has never been set before */
1320 if (lts)
1321 lts = jiffies_to_msecs(jiffies - lts);
1322 out += snprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts);
1323 goto done;
1324
1325 case O2HB_DB_TYPE_REGION_PINNED:
1326 reg = (struct o2hb_region *)db->db_data;
1327 out += snprintf(buf + out, PAGE_SIZE - out, "%u\n",
1328 !!reg->hr_item_pinned);
1329 goto done;
1330
1331 default:
1332 goto done;
1333 }
1334
1335 while ((i = find_next_bit(map, db->db_len, i + 1)) < db->db_len)
1336 out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i);
1337 out += snprintf(buf + out, PAGE_SIZE - out, "\n");
1338
1339 done:
1340 i_size_write(inode, out);
1341
1342 file->private_data = buf;
1343
1344 return 0;
1345 bail:
1346 return -ENOMEM;
1347 }
1348
o2hb_debug_release(struct inode * inode,struct file * file)1349 static int o2hb_debug_release(struct inode *inode, struct file *file)
1350 {
1351 kfree(file->private_data);
1352 return 0;
1353 }
1354
o2hb_debug_read(struct file * file,char __user * buf,size_t nbytes,loff_t * ppos)1355 static ssize_t o2hb_debug_read(struct file *file, char __user *buf,
1356 size_t nbytes, loff_t *ppos)
1357 {
1358 return simple_read_from_buffer(buf, nbytes, ppos, file->private_data,
1359 i_size_read(file->f_mapping->host));
1360 }
1361 #else
o2hb_debug_open(struct inode * inode,struct file * file)1362 static int o2hb_debug_open(struct inode *inode, struct file *file)
1363 {
1364 return 0;
1365 }
o2hb_debug_release(struct inode * inode,struct file * file)1366 static int o2hb_debug_release(struct inode *inode, struct file *file)
1367 {
1368 return 0;
1369 }
o2hb_debug_read(struct file * file,char __user * buf,size_t nbytes,loff_t * ppos)1370 static ssize_t o2hb_debug_read(struct file *file, char __user *buf,
1371 size_t nbytes, loff_t *ppos)
1372 {
1373 return 0;
1374 }
1375 #endif /* CONFIG_DEBUG_FS */
1376
1377 static const struct file_operations o2hb_debug_fops = {
1378 .open = o2hb_debug_open,
1379 .release = o2hb_debug_release,
1380 .read = o2hb_debug_read,
1381 .llseek = generic_file_llseek,
1382 };
1383
o2hb_exit(void)1384 void o2hb_exit(void)
1385 {
1386 debugfs_remove_recursive(o2hb_debug_dir);
1387 kfree(o2hb_db_livenodes);
1388 kfree(o2hb_db_liveregions);
1389 kfree(o2hb_db_quorumregions);
1390 kfree(o2hb_db_failedregions);
1391 }
1392
o2hb_debug_create(const char * name,struct dentry * dir,struct o2hb_debug_buf ** db,int db_len,int type,int size,int len,void * data)1393 static void o2hb_debug_create(const char *name, struct dentry *dir,
1394 struct o2hb_debug_buf **db, int db_len, int type,
1395 int size, int len, void *data)
1396 {
1397 *db = kmalloc(db_len, GFP_KERNEL);
1398 if (!*db)
1399 return;
1400
1401 (*db)->db_type = type;
1402 (*db)->db_size = size;
1403 (*db)->db_len = len;
1404 (*db)->db_data = data;
1405
1406 debugfs_create_file(name, S_IFREG|S_IRUSR, dir, *db, &o2hb_debug_fops);
1407 }
1408
o2hb_debug_init(void)1409 static void o2hb_debug_init(void)
1410 {
1411 o2hb_debug_dir = debugfs_create_dir(O2HB_DEBUG_DIR, NULL);
1412
1413 o2hb_debug_create(O2HB_DEBUG_LIVENODES, o2hb_debug_dir,
1414 &o2hb_db_livenodes, sizeof(*o2hb_db_livenodes),
1415 O2HB_DB_TYPE_LIVENODES, sizeof(o2hb_live_node_bitmap),
1416 O2NM_MAX_NODES, o2hb_live_node_bitmap);
1417
1418 o2hb_debug_create(O2HB_DEBUG_LIVEREGIONS, o2hb_debug_dir,
1419 &o2hb_db_liveregions, sizeof(*o2hb_db_liveregions),
1420 O2HB_DB_TYPE_LIVEREGIONS,
1421 sizeof(o2hb_live_region_bitmap), O2NM_MAX_REGIONS,
1422 o2hb_live_region_bitmap);
1423
1424 o2hb_debug_create(O2HB_DEBUG_QUORUMREGIONS, o2hb_debug_dir,
1425 &o2hb_db_quorumregions,
1426 sizeof(*o2hb_db_quorumregions),
1427 O2HB_DB_TYPE_QUORUMREGIONS,
1428 sizeof(o2hb_quorum_region_bitmap), O2NM_MAX_REGIONS,
1429 o2hb_quorum_region_bitmap);
1430
1431 o2hb_debug_create(O2HB_DEBUG_FAILEDREGIONS, o2hb_debug_dir,
1432 &o2hb_db_failedregions,
1433 sizeof(*o2hb_db_failedregions),
1434 O2HB_DB_TYPE_FAILEDREGIONS,
1435 sizeof(o2hb_failed_region_bitmap), O2NM_MAX_REGIONS,
1436 o2hb_failed_region_bitmap);
1437 }
1438
o2hb_init(void)1439 void o2hb_init(void)
1440 {
1441 int i;
1442
1443 for (i = 0; i < ARRAY_SIZE(o2hb_callbacks); i++)
1444 INIT_LIST_HEAD(&o2hb_callbacks[i].list);
1445
1446 for (i = 0; i < ARRAY_SIZE(o2hb_live_slots); i++)
1447 INIT_LIST_HEAD(&o2hb_live_slots[i]);
1448
1449 INIT_LIST_HEAD(&o2hb_node_events);
1450
1451 memset(o2hb_live_node_bitmap, 0, sizeof(o2hb_live_node_bitmap));
1452 memset(o2hb_region_bitmap, 0, sizeof(o2hb_region_bitmap));
1453 memset(o2hb_live_region_bitmap, 0, sizeof(o2hb_live_region_bitmap));
1454 memset(o2hb_quorum_region_bitmap, 0, sizeof(o2hb_quorum_region_bitmap));
1455 memset(o2hb_failed_region_bitmap, 0, sizeof(o2hb_failed_region_bitmap));
1456
1457 o2hb_dependent_users = 0;
1458
1459 o2hb_debug_init();
1460 }
1461
1462 /* if we're already in a callback then we're already serialized by the sem */
o2hb_fill_node_map_from_callback(unsigned long * map,unsigned bytes)1463 static void o2hb_fill_node_map_from_callback(unsigned long *map,
1464 unsigned bytes)
1465 {
1466 BUG_ON(bytes < (BITS_TO_LONGS(O2NM_MAX_NODES) * sizeof(unsigned long)));
1467
1468 memcpy(map, &o2hb_live_node_bitmap, bytes);
1469 }
1470
1471 /*
1472 * get a map of all nodes that are heartbeating in any regions
1473 */
o2hb_fill_node_map(unsigned long * map,unsigned bytes)1474 void o2hb_fill_node_map(unsigned long *map, unsigned bytes)
1475 {
1476 /* callers want to serialize this map and callbacks so that they
1477 * can trust that they don't miss nodes coming to the party */
1478 down_read(&o2hb_callback_sem);
1479 spin_lock(&o2hb_live_lock);
1480 o2hb_fill_node_map_from_callback(map, bytes);
1481 spin_unlock(&o2hb_live_lock);
1482 up_read(&o2hb_callback_sem);
1483 }
1484 EXPORT_SYMBOL_GPL(o2hb_fill_node_map);
1485
1486 /*
1487 * heartbeat configfs bits. The heartbeat set is a default set under
1488 * the cluster set in nodemanager.c.
1489 */
1490
to_o2hb_region(struct config_item * item)1491 static struct o2hb_region *to_o2hb_region(struct config_item *item)
1492 {
1493 return item ? container_of(item, struct o2hb_region, hr_item) : NULL;
1494 }
1495
1496 /* drop_item only drops its ref after killing the thread, nothing should
1497 * be using the region anymore. this has to clean up any state that
1498 * attributes might have built up. */
o2hb_region_release(struct config_item * item)1499 static void o2hb_region_release(struct config_item *item)
1500 {
1501 int i;
1502 struct page *page;
1503 struct o2hb_region *reg = to_o2hb_region(item);
1504
1505 mlog(ML_HEARTBEAT, "hb region release (%s)\n", reg->hr_dev_name);
1506
1507 kfree(reg->hr_tmp_block);
1508
1509 if (reg->hr_slot_data) {
1510 for (i = 0; i < reg->hr_num_pages; i++) {
1511 page = reg->hr_slot_data[i];
1512 if (page)
1513 __free_page(page);
1514 }
1515 kfree(reg->hr_slot_data);
1516 }
1517
1518 if (reg->hr_bdev)
1519 blkdev_put(reg->hr_bdev, FMODE_READ|FMODE_WRITE);
1520
1521 kfree(reg->hr_slots);
1522
1523 debugfs_remove_recursive(reg->hr_debug_dir);
1524 kfree(reg->hr_db_livenodes);
1525 kfree(reg->hr_db_regnum);
1526 kfree(reg->hr_db_elapsed_time);
1527 kfree(reg->hr_db_pinned);
1528
1529 spin_lock(&o2hb_live_lock);
1530 list_del(®->hr_all_item);
1531 spin_unlock(&o2hb_live_lock);
1532
1533 o2net_unregister_handler_list(®->hr_handler_list);
1534 kfree(reg);
1535 }
1536
o2hb_read_block_input(struct o2hb_region * reg,const char * page,unsigned long * ret_bytes,unsigned int * ret_bits)1537 static int o2hb_read_block_input(struct o2hb_region *reg,
1538 const char *page,
1539 unsigned long *ret_bytes,
1540 unsigned int *ret_bits)
1541 {
1542 unsigned long bytes;
1543 char *p = (char *)page;
1544
1545 bytes = simple_strtoul(p, &p, 0);
1546 if (!p || (*p && (*p != '\n')))
1547 return -EINVAL;
1548
1549 /* Heartbeat and fs min / max block sizes are the same. */
1550 if (bytes > 4096 || bytes < 512)
1551 return -ERANGE;
1552 if (hweight16(bytes) != 1)
1553 return -EINVAL;
1554
1555 if (ret_bytes)
1556 *ret_bytes = bytes;
1557 if (ret_bits)
1558 *ret_bits = ffs(bytes) - 1;
1559
1560 return 0;
1561 }
1562
o2hb_region_block_bytes_show(struct config_item * item,char * page)1563 static ssize_t o2hb_region_block_bytes_show(struct config_item *item,
1564 char *page)
1565 {
1566 return sprintf(page, "%u\n", to_o2hb_region(item)->hr_block_bytes);
1567 }
1568
o2hb_region_block_bytes_store(struct config_item * item,const char * page,size_t count)1569 static ssize_t o2hb_region_block_bytes_store(struct config_item *item,
1570 const char *page,
1571 size_t count)
1572 {
1573 struct o2hb_region *reg = to_o2hb_region(item);
1574 int status;
1575 unsigned long block_bytes;
1576 unsigned int block_bits;
1577
1578 if (reg->hr_bdev)
1579 return -EINVAL;
1580
1581 status = o2hb_read_block_input(reg, page, &block_bytes,
1582 &block_bits);
1583 if (status)
1584 return status;
1585
1586 reg->hr_block_bytes = (unsigned int)block_bytes;
1587 reg->hr_block_bits = block_bits;
1588
1589 return count;
1590 }
1591
o2hb_region_start_block_show(struct config_item * item,char * page)1592 static ssize_t o2hb_region_start_block_show(struct config_item *item,
1593 char *page)
1594 {
1595 return sprintf(page, "%llu\n", to_o2hb_region(item)->hr_start_block);
1596 }
1597
o2hb_region_start_block_store(struct config_item * item,const char * page,size_t count)1598 static ssize_t o2hb_region_start_block_store(struct config_item *item,
1599 const char *page,
1600 size_t count)
1601 {
1602 struct o2hb_region *reg = to_o2hb_region(item);
1603 unsigned long long tmp;
1604 char *p = (char *)page;
1605
1606 if (reg->hr_bdev)
1607 return -EINVAL;
1608
1609 tmp = simple_strtoull(p, &p, 0);
1610 if (!p || (*p && (*p != '\n')))
1611 return -EINVAL;
1612
1613 reg->hr_start_block = tmp;
1614
1615 return count;
1616 }
1617
o2hb_region_blocks_show(struct config_item * item,char * page)1618 static ssize_t o2hb_region_blocks_show(struct config_item *item, char *page)
1619 {
1620 return sprintf(page, "%d\n", to_o2hb_region(item)->hr_blocks);
1621 }
1622
o2hb_region_blocks_store(struct config_item * item,const char * page,size_t count)1623 static ssize_t o2hb_region_blocks_store(struct config_item *item,
1624 const char *page,
1625 size_t count)
1626 {
1627 struct o2hb_region *reg = to_o2hb_region(item);
1628 unsigned long tmp;
1629 char *p = (char *)page;
1630
1631 if (reg->hr_bdev)
1632 return -EINVAL;
1633
1634 tmp = simple_strtoul(p, &p, 0);
1635 if (!p || (*p && (*p != '\n')))
1636 return -EINVAL;
1637
1638 if (tmp > O2NM_MAX_NODES || tmp == 0)
1639 return -ERANGE;
1640
1641 reg->hr_blocks = (unsigned int)tmp;
1642
1643 return count;
1644 }
1645
o2hb_region_dev_show(struct config_item * item,char * page)1646 static ssize_t o2hb_region_dev_show(struct config_item *item, char *page)
1647 {
1648 unsigned int ret = 0;
1649
1650 if (to_o2hb_region(item)->hr_bdev)
1651 ret = sprintf(page, "%s\n", to_o2hb_region(item)->hr_dev_name);
1652
1653 return ret;
1654 }
1655
o2hb_init_region_params(struct o2hb_region * reg)1656 static void o2hb_init_region_params(struct o2hb_region *reg)
1657 {
1658 reg->hr_slots_per_page = PAGE_SIZE >> reg->hr_block_bits;
1659 reg->hr_timeout_ms = O2HB_REGION_TIMEOUT_MS;
1660
1661 mlog(ML_HEARTBEAT, "hr_start_block = %llu, hr_blocks = %u\n",
1662 reg->hr_start_block, reg->hr_blocks);
1663 mlog(ML_HEARTBEAT, "hr_block_bytes = %u, hr_block_bits = %u\n",
1664 reg->hr_block_bytes, reg->hr_block_bits);
1665 mlog(ML_HEARTBEAT, "hr_timeout_ms = %u\n", reg->hr_timeout_ms);
1666 mlog(ML_HEARTBEAT, "dead threshold = %u\n", o2hb_dead_threshold);
1667 }
1668
o2hb_map_slot_data(struct o2hb_region * reg)1669 static int o2hb_map_slot_data(struct o2hb_region *reg)
1670 {
1671 int i, j;
1672 unsigned int last_slot;
1673 unsigned int spp = reg->hr_slots_per_page;
1674 struct page *page;
1675 char *raw;
1676 struct o2hb_disk_slot *slot;
1677
1678 reg->hr_tmp_block = kmalloc(reg->hr_block_bytes, GFP_KERNEL);
1679 if (reg->hr_tmp_block == NULL)
1680 return -ENOMEM;
1681
1682 reg->hr_slots = kcalloc(reg->hr_blocks,
1683 sizeof(struct o2hb_disk_slot), GFP_KERNEL);
1684 if (reg->hr_slots == NULL)
1685 return -ENOMEM;
1686
1687 for(i = 0; i < reg->hr_blocks; i++) {
1688 slot = ®->hr_slots[i];
1689 slot->ds_node_num = i;
1690 INIT_LIST_HEAD(&slot->ds_live_item);
1691 slot->ds_raw_block = NULL;
1692 }
1693
1694 reg->hr_num_pages = (reg->hr_blocks + spp - 1) / spp;
1695 mlog(ML_HEARTBEAT, "Going to require %u pages to cover %u blocks "
1696 "at %u blocks per page\n",
1697 reg->hr_num_pages, reg->hr_blocks, spp);
1698
1699 reg->hr_slot_data = kcalloc(reg->hr_num_pages, sizeof(struct page *),
1700 GFP_KERNEL);
1701 if (!reg->hr_slot_data)
1702 return -ENOMEM;
1703
1704 for(i = 0; i < reg->hr_num_pages; i++) {
1705 page = alloc_page(GFP_KERNEL);
1706 if (!page)
1707 return -ENOMEM;
1708
1709 reg->hr_slot_data[i] = page;
1710
1711 last_slot = i * spp;
1712 raw = page_address(page);
1713 for (j = 0;
1714 (j < spp) && ((j + last_slot) < reg->hr_blocks);
1715 j++) {
1716 BUG_ON((j + last_slot) >= reg->hr_blocks);
1717
1718 slot = ®->hr_slots[j + last_slot];
1719 slot->ds_raw_block =
1720 (struct o2hb_disk_heartbeat_block *) raw;
1721
1722 raw += reg->hr_block_bytes;
1723 }
1724 }
1725
1726 return 0;
1727 }
1728
1729 /* Read in all the slots available and populate the tracking
1730 * structures so that we can start with a baseline idea of what's
1731 * there. */
o2hb_populate_slot_data(struct o2hb_region * reg)1732 static int o2hb_populate_slot_data(struct o2hb_region *reg)
1733 {
1734 int ret, i;
1735 struct o2hb_disk_slot *slot;
1736 struct o2hb_disk_heartbeat_block *hb_block;
1737
1738 ret = o2hb_read_slots(reg, 0, reg->hr_blocks);
1739 if (ret)
1740 goto out;
1741
1742 /* We only want to get an idea of the values initially in each
1743 * slot, so we do no verification - o2hb_check_slot will
1744 * actually determine if each configured slot is valid and
1745 * whether any values have changed. */
1746 for(i = 0; i < reg->hr_blocks; i++) {
1747 slot = ®->hr_slots[i];
1748 hb_block = (struct o2hb_disk_heartbeat_block *) slot->ds_raw_block;
1749
1750 /* Only fill the values that o2hb_check_slot uses to
1751 * determine changing slots */
1752 slot->ds_last_time = le64_to_cpu(hb_block->hb_seq);
1753 slot->ds_last_generation = le64_to_cpu(hb_block->hb_generation);
1754 }
1755
1756 out:
1757 return ret;
1758 }
1759
1760 /* this is acting as commit; we set up all of hr_bdev and hr_task or nothing */
o2hb_region_dev_store(struct config_item * item,const char * page,size_t count)1761 static ssize_t o2hb_region_dev_store(struct config_item *item,
1762 const char *page,
1763 size_t count)
1764 {
1765 struct o2hb_region *reg = to_o2hb_region(item);
1766 struct task_struct *hb_task;
1767 long fd;
1768 int sectsize;
1769 char *p = (char *)page;
1770 struct fd f;
1771 struct inode *inode;
1772 ssize_t ret = -EINVAL;
1773 int live_threshold;
1774
1775 if (reg->hr_bdev)
1776 goto out;
1777
1778 /* We can't heartbeat without having had our node number
1779 * configured yet. */
1780 if (o2nm_this_node() == O2NM_MAX_NODES)
1781 goto out;
1782
1783 fd = simple_strtol(p, &p, 0);
1784 if (!p || (*p && (*p != '\n')))
1785 goto out;
1786
1787 if (fd < 0 || fd >= INT_MAX)
1788 goto out;
1789
1790 f = fdget(fd);
1791 if (f.file == NULL)
1792 goto out;
1793
1794 if (reg->hr_blocks == 0 || reg->hr_start_block == 0 ||
1795 reg->hr_block_bytes == 0)
1796 goto out2;
1797
1798 inode = igrab(f.file->f_mapping->host);
1799 if (inode == NULL)
1800 goto out2;
1801
1802 if (!S_ISBLK(inode->i_mode))
1803 goto out3;
1804
1805 reg->hr_bdev = I_BDEV(f.file->f_mapping->host);
1806 ret = blkdev_get(reg->hr_bdev, FMODE_WRITE | FMODE_READ, NULL);
1807 if (ret) {
1808 reg->hr_bdev = NULL;
1809 goto out3;
1810 }
1811 inode = NULL;
1812
1813 bdevname(reg->hr_bdev, reg->hr_dev_name);
1814
1815 sectsize = bdev_logical_block_size(reg->hr_bdev);
1816 if (sectsize != reg->hr_block_bytes) {
1817 mlog(ML_ERROR,
1818 "blocksize %u incorrect for device, expected %d",
1819 reg->hr_block_bytes, sectsize);
1820 ret = -EINVAL;
1821 goto out3;
1822 }
1823
1824 o2hb_init_region_params(reg);
1825
1826 /* Generation of zero is invalid */
1827 do {
1828 get_random_bytes(®->hr_generation,
1829 sizeof(reg->hr_generation));
1830 } while (reg->hr_generation == 0);
1831
1832 ret = o2hb_map_slot_data(reg);
1833 if (ret) {
1834 mlog_errno(ret);
1835 goto out3;
1836 }
1837
1838 ret = o2hb_populate_slot_data(reg);
1839 if (ret) {
1840 mlog_errno(ret);
1841 goto out3;
1842 }
1843
1844 INIT_DELAYED_WORK(®->hr_write_timeout_work, o2hb_write_timeout);
1845 INIT_DELAYED_WORK(®->hr_nego_timeout_work, o2hb_nego_timeout);
1846
1847 /*
1848 * A node is considered live after it has beat LIVE_THRESHOLD
1849 * times. We're not steady until we've given them a chance
1850 * _after_ our first read.
1851 * The default threshold is bare minimum so as to limit the delay
1852 * during mounts. For global heartbeat, the threshold doubled for the
1853 * first region.
1854 */
1855 live_threshold = O2HB_LIVE_THRESHOLD;
1856 if (o2hb_global_heartbeat_active()) {
1857 spin_lock(&o2hb_live_lock);
1858 if (bitmap_weight(o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1)
1859 live_threshold <<= 1;
1860 spin_unlock(&o2hb_live_lock);
1861 }
1862 ++live_threshold;
1863 atomic_set(®->hr_steady_iterations, live_threshold);
1864 /* unsteady_iterations is triple the steady_iterations */
1865 atomic_set(®->hr_unsteady_iterations, (live_threshold * 3));
1866
1867 hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s",
1868 reg->hr_item.ci_name);
1869 if (IS_ERR(hb_task)) {
1870 ret = PTR_ERR(hb_task);
1871 mlog_errno(ret);
1872 goto out3;
1873 }
1874
1875 spin_lock(&o2hb_live_lock);
1876 reg->hr_task = hb_task;
1877 spin_unlock(&o2hb_live_lock);
1878
1879 ret = wait_event_interruptible(o2hb_steady_queue,
1880 atomic_read(®->hr_steady_iterations) == 0 ||
1881 reg->hr_node_deleted);
1882 if (ret) {
1883 atomic_set(®->hr_steady_iterations, 0);
1884 reg->hr_aborted_start = 1;
1885 }
1886
1887 if (reg->hr_aborted_start) {
1888 ret = -EIO;
1889 goto out3;
1890 }
1891
1892 if (reg->hr_node_deleted) {
1893 ret = -EINVAL;
1894 goto out3;
1895 }
1896
1897 /* Ok, we were woken. Make sure it wasn't by drop_item() */
1898 spin_lock(&o2hb_live_lock);
1899 hb_task = reg->hr_task;
1900 if (o2hb_global_heartbeat_active())
1901 set_bit(reg->hr_region_num, o2hb_live_region_bitmap);
1902 spin_unlock(&o2hb_live_lock);
1903
1904 if (hb_task)
1905 ret = count;
1906 else
1907 ret = -EIO;
1908
1909 if (hb_task && o2hb_global_heartbeat_active())
1910 printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%s)\n",
1911 config_item_name(®->hr_item), reg->hr_dev_name);
1912
1913 out3:
1914 iput(inode);
1915 out2:
1916 fdput(f);
1917 out:
1918 if (ret < 0) {
1919 if (reg->hr_bdev) {
1920 blkdev_put(reg->hr_bdev, FMODE_READ|FMODE_WRITE);
1921 reg->hr_bdev = NULL;
1922 }
1923 }
1924 return ret;
1925 }
1926
o2hb_region_pid_show(struct config_item * item,char * page)1927 static ssize_t o2hb_region_pid_show(struct config_item *item, char *page)
1928 {
1929 struct o2hb_region *reg = to_o2hb_region(item);
1930 pid_t pid = 0;
1931
1932 spin_lock(&o2hb_live_lock);
1933 if (reg->hr_task)
1934 pid = task_pid_nr(reg->hr_task);
1935 spin_unlock(&o2hb_live_lock);
1936
1937 if (!pid)
1938 return 0;
1939
1940 return sprintf(page, "%u\n", pid);
1941 }
1942
1943 CONFIGFS_ATTR(o2hb_region_, block_bytes);
1944 CONFIGFS_ATTR(o2hb_region_, start_block);
1945 CONFIGFS_ATTR(o2hb_region_, blocks);
1946 CONFIGFS_ATTR(o2hb_region_, dev);
1947 CONFIGFS_ATTR_RO(o2hb_region_, pid);
1948
1949 static struct configfs_attribute *o2hb_region_attrs[] = {
1950 &o2hb_region_attr_block_bytes,
1951 &o2hb_region_attr_start_block,
1952 &o2hb_region_attr_blocks,
1953 &o2hb_region_attr_dev,
1954 &o2hb_region_attr_pid,
1955 NULL,
1956 };
1957
1958 static struct configfs_item_operations o2hb_region_item_ops = {
1959 .release = o2hb_region_release,
1960 };
1961
1962 static const struct config_item_type o2hb_region_type = {
1963 .ct_item_ops = &o2hb_region_item_ops,
1964 .ct_attrs = o2hb_region_attrs,
1965 .ct_owner = THIS_MODULE,
1966 };
1967
1968 /* heartbeat set */
1969
1970 struct o2hb_heartbeat_group {
1971 struct config_group hs_group;
1972 /* some stuff? */
1973 };
1974
to_o2hb_heartbeat_group(struct config_group * group)1975 static struct o2hb_heartbeat_group *to_o2hb_heartbeat_group(struct config_group *group)
1976 {
1977 return group ?
1978 container_of(group, struct o2hb_heartbeat_group, hs_group)
1979 : NULL;
1980 }
1981
o2hb_debug_region_init(struct o2hb_region * reg,struct dentry * parent)1982 static void o2hb_debug_region_init(struct o2hb_region *reg,
1983 struct dentry *parent)
1984 {
1985 struct dentry *dir;
1986
1987 dir = debugfs_create_dir(config_item_name(®->hr_item), parent);
1988 reg->hr_debug_dir = dir;
1989
1990 o2hb_debug_create(O2HB_DEBUG_LIVENODES, dir, &(reg->hr_db_livenodes),
1991 sizeof(*(reg->hr_db_livenodes)),
1992 O2HB_DB_TYPE_REGION_LIVENODES,
1993 sizeof(reg->hr_live_node_bitmap), O2NM_MAX_NODES,
1994 reg);
1995
1996 o2hb_debug_create(O2HB_DEBUG_REGION_NUMBER, dir, &(reg->hr_db_regnum),
1997 sizeof(*(reg->hr_db_regnum)),
1998 O2HB_DB_TYPE_REGION_NUMBER, 0, O2NM_MAX_NODES, reg);
1999
2000 o2hb_debug_create(O2HB_DEBUG_REGION_ELAPSED_TIME, dir,
2001 &(reg->hr_db_elapsed_time),
2002 sizeof(*(reg->hr_db_elapsed_time)),
2003 O2HB_DB_TYPE_REGION_ELAPSED_TIME, 0, 0, reg);
2004
2005 o2hb_debug_create(O2HB_DEBUG_REGION_PINNED, dir, &(reg->hr_db_pinned),
2006 sizeof(*(reg->hr_db_pinned)),
2007 O2HB_DB_TYPE_REGION_PINNED, 0, 0, reg);
2008
2009 }
2010
o2hb_heartbeat_group_make_item(struct config_group * group,const char * name)2011 static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *group,
2012 const char *name)
2013 {
2014 struct o2hb_region *reg = NULL;
2015 int ret;
2016
2017 reg = kzalloc(sizeof(struct o2hb_region), GFP_KERNEL);
2018 if (reg == NULL)
2019 return ERR_PTR(-ENOMEM);
2020
2021 if (strlen(name) > O2HB_MAX_REGION_NAME_LEN) {
2022 ret = -ENAMETOOLONG;
2023 goto free;
2024 }
2025
2026 spin_lock(&o2hb_live_lock);
2027 reg->hr_region_num = 0;
2028 if (o2hb_global_heartbeat_active()) {
2029 reg->hr_region_num = find_first_zero_bit(o2hb_region_bitmap,
2030 O2NM_MAX_REGIONS);
2031 if (reg->hr_region_num >= O2NM_MAX_REGIONS) {
2032 spin_unlock(&o2hb_live_lock);
2033 ret = -EFBIG;
2034 goto free;
2035 }
2036 set_bit(reg->hr_region_num, o2hb_region_bitmap);
2037 }
2038 list_add_tail(®->hr_all_item, &o2hb_all_regions);
2039 spin_unlock(&o2hb_live_lock);
2040
2041 config_item_init_type_name(®->hr_item, name, &o2hb_region_type);
2042
2043 /* this is the same way to generate msg key as dlm, for local heartbeat,
2044 * name is also the same, so make initial crc value different to avoid
2045 * message key conflict.
2046 */
2047 reg->hr_key = crc32_le(reg->hr_region_num + O2NM_MAX_REGIONS,
2048 name, strlen(name));
2049 INIT_LIST_HEAD(®->hr_handler_list);
2050 ret = o2net_register_handler(O2HB_NEGO_TIMEOUT_MSG, reg->hr_key,
2051 sizeof(struct o2hb_nego_msg),
2052 o2hb_nego_timeout_handler,
2053 reg, NULL, ®->hr_handler_list);
2054 if (ret)
2055 goto free;
2056
2057 ret = o2net_register_handler(O2HB_NEGO_APPROVE_MSG, reg->hr_key,
2058 sizeof(struct o2hb_nego_msg),
2059 o2hb_nego_approve_handler,
2060 reg, NULL, ®->hr_handler_list);
2061 if (ret)
2062 goto unregister_handler;
2063
2064 o2hb_debug_region_init(reg, o2hb_debug_dir);
2065
2066 return ®->hr_item;
2067
2068 unregister_handler:
2069 o2net_unregister_handler_list(®->hr_handler_list);
2070 free:
2071 kfree(reg);
2072 return ERR_PTR(ret);
2073 }
2074
o2hb_heartbeat_group_drop_item(struct config_group * group,struct config_item * item)2075 static void o2hb_heartbeat_group_drop_item(struct config_group *group,
2076 struct config_item *item)
2077 {
2078 struct task_struct *hb_task;
2079 struct o2hb_region *reg = to_o2hb_region(item);
2080 int quorum_region = 0;
2081
2082 /* stop the thread when the user removes the region dir */
2083 spin_lock(&o2hb_live_lock);
2084 hb_task = reg->hr_task;
2085 reg->hr_task = NULL;
2086 reg->hr_item_dropped = 1;
2087 spin_unlock(&o2hb_live_lock);
2088
2089 if (hb_task)
2090 kthread_stop(hb_task);
2091
2092 if (o2hb_global_heartbeat_active()) {
2093 spin_lock(&o2hb_live_lock);
2094 clear_bit(reg->hr_region_num, o2hb_region_bitmap);
2095 clear_bit(reg->hr_region_num, o2hb_live_region_bitmap);
2096 if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
2097 quorum_region = 1;
2098 clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
2099 spin_unlock(&o2hb_live_lock);
2100 printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%s)\n",
2101 ((atomic_read(®->hr_steady_iterations) == 0) ?
2102 "stopped" : "start aborted"), config_item_name(item),
2103 reg->hr_dev_name);
2104 }
2105
2106 /*
2107 * If we're racing a dev_write(), we need to wake them. They will
2108 * check reg->hr_task
2109 */
2110 if (atomic_read(®->hr_steady_iterations) != 0) {
2111 reg->hr_aborted_start = 1;
2112 atomic_set(®->hr_steady_iterations, 0);
2113 wake_up(&o2hb_steady_queue);
2114 }
2115
2116 config_item_put(item);
2117
2118 if (!o2hb_global_heartbeat_active() || !quorum_region)
2119 return;
2120
2121 /*
2122 * If global heartbeat active and there are dependent users,
2123 * pin all regions if quorum region count <= CUT_OFF
2124 */
2125 spin_lock(&o2hb_live_lock);
2126
2127 if (!o2hb_dependent_users)
2128 goto unlock;
2129
2130 if (bitmap_weight(o2hb_quorum_region_bitmap,
2131 O2NM_MAX_REGIONS) <= O2HB_PIN_CUT_OFF)
2132 o2hb_region_pin(NULL);
2133
2134 unlock:
2135 spin_unlock(&o2hb_live_lock);
2136 }
2137
o2hb_heartbeat_group_dead_threshold_show(struct config_item * item,char * page)2138 static ssize_t o2hb_heartbeat_group_dead_threshold_show(struct config_item *item,
2139 char *page)
2140 {
2141 return sprintf(page, "%u\n", o2hb_dead_threshold);
2142 }
2143
o2hb_heartbeat_group_dead_threshold_store(struct config_item * item,const char * page,size_t count)2144 static ssize_t o2hb_heartbeat_group_dead_threshold_store(struct config_item *item,
2145 const char *page, size_t count)
2146 {
2147 unsigned long tmp;
2148 char *p = (char *)page;
2149
2150 tmp = simple_strtoul(p, &p, 10);
2151 if (!p || (*p && (*p != '\n')))
2152 return -EINVAL;
2153
2154 /* this will validate ranges for us. */
2155 o2hb_dead_threshold_set((unsigned int) tmp);
2156
2157 return count;
2158 }
2159
o2hb_heartbeat_group_mode_show(struct config_item * item,char * page)2160 static ssize_t o2hb_heartbeat_group_mode_show(struct config_item *item,
2161 char *page)
2162 {
2163 return sprintf(page, "%s\n",
2164 o2hb_heartbeat_mode_desc[o2hb_heartbeat_mode]);
2165 }
2166
o2hb_heartbeat_group_mode_store(struct config_item * item,const char * page,size_t count)2167 static ssize_t o2hb_heartbeat_group_mode_store(struct config_item *item,
2168 const char *page, size_t count)
2169 {
2170 unsigned int i;
2171 int ret;
2172 size_t len;
2173
2174 len = (page[count - 1] == '\n') ? count - 1 : count;
2175 if (!len)
2176 return -EINVAL;
2177
2178 for (i = 0; i < O2HB_HEARTBEAT_NUM_MODES; ++i) {
2179 if (strncasecmp(page, o2hb_heartbeat_mode_desc[i], len))
2180 continue;
2181
2182 ret = o2hb_global_heartbeat_mode_set(i);
2183 if (!ret)
2184 printk(KERN_NOTICE "o2hb: Heartbeat mode set to %s\n",
2185 o2hb_heartbeat_mode_desc[i]);
2186 return count;
2187 }
2188
2189 return -EINVAL;
2190
2191 }
2192
2193 CONFIGFS_ATTR(o2hb_heartbeat_group_, dead_threshold);
2194 CONFIGFS_ATTR(o2hb_heartbeat_group_, mode);
2195
2196 static struct configfs_attribute *o2hb_heartbeat_group_attrs[] = {
2197 &o2hb_heartbeat_group_attr_dead_threshold,
2198 &o2hb_heartbeat_group_attr_mode,
2199 NULL,
2200 };
2201
2202 static struct configfs_group_operations o2hb_heartbeat_group_group_ops = {
2203 .make_item = o2hb_heartbeat_group_make_item,
2204 .drop_item = o2hb_heartbeat_group_drop_item,
2205 };
2206
2207 static const struct config_item_type o2hb_heartbeat_group_type = {
2208 .ct_group_ops = &o2hb_heartbeat_group_group_ops,
2209 .ct_attrs = o2hb_heartbeat_group_attrs,
2210 .ct_owner = THIS_MODULE,
2211 };
2212
2213 /* this is just here to avoid touching group in heartbeat.h which the
2214 * entire damn world #includes */
o2hb_alloc_hb_set(void)2215 struct config_group *o2hb_alloc_hb_set(void)
2216 {
2217 struct o2hb_heartbeat_group *hs = NULL;
2218 struct config_group *ret = NULL;
2219
2220 hs = kzalloc(sizeof(struct o2hb_heartbeat_group), GFP_KERNEL);
2221 if (hs == NULL)
2222 goto out;
2223
2224 config_group_init_type_name(&hs->hs_group, "heartbeat",
2225 &o2hb_heartbeat_group_type);
2226
2227 ret = &hs->hs_group;
2228 out:
2229 if (ret == NULL)
2230 kfree(hs);
2231 return ret;
2232 }
2233
o2hb_free_hb_set(struct config_group * group)2234 void o2hb_free_hb_set(struct config_group *group)
2235 {
2236 struct o2hb_heartbeat_group *hs = to_o2hb_heartbeat_group(group);
2237 kfree(hs);
2238 }
2239
2240 /* hb callback registration and issuing */
2241
hbcall_from_type(enum o2hb_callback_type type)2242 static struct o2hb_callback *hbcall_from_type(enum o2hb_callback_type type)
2243 {
2244 if (type == O2HB_NUM_CB)
2245 return ERR_PTR(-EINVAL);
2246
2247 return &o2hb_callbacks[type];
2248 }
2249
o2hb_setup_callback(struct o2hb_callback_func * hc,enum o2hb_callback_type type,o2hb_cb_func * func,void * data,int priority)2250 void o2hb_setup_callback(struct o2hb_callback_func *hc,
2251 enum o2hb_callback_type type,
2252 o2hb_cb_func *func,
2253 void *data,
2254 int priority)
2255 {
2256 INIT_LIST_HEAD(&hc->hc_item);
2257 hc->hc_func = func;
2258 hc->hc_data = data;
2259 hc->hc_priority = priority;
2260 hc->hc_type = type;
2261 hc->hc_magic = O2HB_CB_MAGIC;
2262 }
2263 EXPORT_SYMBOL_GPL(o2hb_setup_callback);
2264
2265 /*
2266 * In local heartbeat mode, region_uuid passed matches the dlm domain name.
2267 * In global heartbeat mode, region_uuid passed is NULL.
2268 *
2269 * In local, we only pin the matching region. In global we pin all the active
2270 * regions.
2271 */
o2hb_region_pin(const char * region_uuid)2272 static int o2hb_region_pin(const char *region_uuid)
2273 {
2274 int ret = 0, found = 0;
2275 struct o2hb_region *reg;
2276 char *uuid;
2277
2278 assert_spin_locked(&o2hb_live_lock);
2279
2280 list_for_each_entry(reg, &o2hb_all_regions, hr_all_item) {
2281 if (reg->hr_item_dropped)
2282 continue;
2283
2284 uuid = config_item_name(®->hr_item);
2285
2286 /* local heartbeat */
2287 if (region_uuid) {
2288 if (strcmp(region_uuid, uuid))
2289 continue;
2290 found = 1;
2291 }
2292
2293 if (reg->hr_item_pinned || reg->hr_item_dropped)
2294 goto skip_pin;
2295
2296 /* Ignore ENOENT only for local hb (userdlm domain) */
2297 ret = o2nm_depend_item(®->hr_item);
2298 if (!ret) {
2299 mlog(ML_CLUSTER, "Pin region %s\n", uuid);
2300 reg->hr_item_pinned = 1;
2301 } else {
2302 if (ret == -ENOENT && found)
2303 ret = 0;
2304 else {
2305 mlog(ML_ERROR, "Pin region %s fails with %d\n",
2306 uuid, ret);
2307 break;
2308 }
2309 }
2310 skip_pin:
2311 if (found)
2312 break;
2313 }
2314
2315 return ret;
2316 }
2317
2318 /*
2319 * In local heartbeat mode, region_uuid passed matches the dlm domain name.
2320 * In global heartbeat mode, region_uuid passed is NULL.
2321 *
2322 * In local, we only unpin the matching region. In global we unpin all the
2323 * active regions.
2324 */
o2hb_region_unpin(const char * region_uuid)2325 static void o2hb_region_unpin(const char *region_uuid)
2326 {
2327 struct o2hb_region *reg;
2328 char *uuid;
2329 int found = 0;
2330
2331 assert_spin_locked(&o2hb_live_lock);
2332
2333 list_for_each_entry(reg, &o2hb_all_regions, hr_all_item) {
2334 if (reg->hr_item_dropped)
2335 continue;
2336
2337 uuid = config_item_name(®->hr_item);
2338 if (region_uuid) {
2339 if (strcmp(region_uuid, uuid))
2340 continue;
2341 found = 1;
2342 }
2343
2344 if (reg->hr_item_pinned) {
2345 mlog(ML_CLUSTER, "Unpin region %s\n", uuid);
2346 o2nm_undepend_item(®->hr_item);
2347 reg->hr_item_pinned = 0;
2348 }
2349 if (found)
2350 break;
2351 }
2352 }
2353
o2hb_region_inc_user(const char * region_uuid)2354 static int o2hb_region_inc_user(const char *region_uuid)
2355 {
2356 int ret = 0;
2357
2358 spin_lock(&o2hb_live_lock);
2359
2360 /* local heartbeat */
2361 if (!o2hb_global_heartbeat_active()) {
2362 ret = o2hb_region_pin(region_uuid);
2363 goto unlock;
2364 }
2365
2366 /*
2367 * if global heartbeat active and this is the first dependent user,
2368 * pin all regions if quorum region count <= CUT_OFF
2369 */
2370 o2hb_dependent_users++;
2371 if (o2hb_dependent_users > 1)
2372 goto unlock;
2373
2374 if (bitmap_weight(o2hb_quorum_region_bitmap,
2375 O2NM_MAX_REGIONS) <= O2HB_PIN_CUT_OFF)
2376 ret = o2hb_region_pin(NULL);
2377
2378 unlock:
2379 spin_unlock(&o2hb_live_lock);
2380 return ret;
2381 }
2382
o2hb_region_dec_user(const char * region_uuid)2383 static void o2hb_region_dec_user(const char *region_uuid)
2384 {
2385 spin_lock(&o2hb_live_lock);
2386
2387 /* local heartbeat */
2388 if (!o2hb_global_heartbeat_active()) {
2389 o2hb_region_unpin(region_uuid);
2390 goto unlock;
2391 }
2392
2393 /*
2394 * if global heartbeat active and there are no dependent users,
2395 * unpin all quorum regions
2396 */
2397 o2hb_dependent_users--;
2398 if (!o2hb_dependent_users)
2399 o2hb_region_unpin(NULL);
2400
2401 unlock:
2402 spin_unlock(&o2hb_live_lock);
2403 }
2404
o2hb_register_callback(const char * region_uuid,struct o2hb_callback_func * hc)2405 int o2hb_register_callback(const char *region_uuid,
2406 struct o2hb_callback_func *hc)
2407 {
2408 struct o2hb_callback_func *f;
2409 struct o2hb_callback *hbcall;
2410 int ret;
2411
2412 BUG_ON(hc->hc_magic != O2HB_CB_MAGIC);
2413 BUG_ON(!list_empty(&hc->hc_item));
2414
2415 hbcall = hbcall_from_type(hc->hc_type);
2416 if (IS_ERR(hbcall)) {
2417 ret = PTR_ERR(hbcall);
2418 goto out;
2419 }
2420
2421 if (region_uuid) {
2422 ret = o2hb_region_inc_user(region_uuid);
2423 if (ret) {
2424 mlog_errno(ret);
2425 goto out;
2426 }
2427 }
2428
2429 down_write(&o2hb_callback_sem);
2430
2431 list_for_each_entry(f, &hbcall->list, hc_item) {
2432 if (hc->hc_priority < f->hc_priority) {
2433 list_add_tail(&hc->hc_item, &f->hc_item);
2434 break;
2435 }
2436 }
2437 if (list_empty(&hc->hc_item))
2438 list_add_tail(&hc->hc_item, &hbcall->list);
2439
2440 up_write(&o2hb_callback_sem);
2441 ret = 0;
2442 out:
2443 mlog(ML_CLUSTER, "returning %d on behalf of %p for funcs %p\n",
2444 ret, __builtin_return_address(0), hc);
2445 return ret;
2446 }
2447 EXPORT_SYMBOL_GPL(o2hb_register_callback);
2448
o2hb_unregister_callback(const char * region_uuid,struct o2hb_callback_func * hc)2449 void o2hb_unregister_callback(const char *region_uuid,
2450 struct o2hb_callback_func *hc)
2451 {
2452 BUG_ON(hc->hc_magic != O2HB_CB_MAGIC);
2453
2454 mlog(ML_CLUSTER, "on behalf of %p for funcs %p\n",
2455 __builtin_return_address(0), hc);
2456
2457 /* XXX Can this happen _with_ a region reference? */
2458 if (list_empty(&hc->hc_item))
2459 return;
2460
2461 if (region_uuid)
2462 o2hb_region_dec_user(region_uuid);
2463
2464 down_write(&o2hb_callback_sem);
2465
2466 list_del_init(&hc->hc_item);
2467
2468 up_write(&o2hb_callback_sem);
2469 }
2470 EXPORT_SYMBOL_GPL(o2hb_unregister_callback);
2471
o2hb_check_node_heartbeating_no_sem(u8 node_num)2472 int o2hb_check_node_heartbeating_no_sem(u8 node_num)
2473 {
2474 unsigned long testing_map[BITS_TO_LONGS(O2NM_MAX_NODES)];
2475
2476 spin_lock(&o2hb_live_lock);
2477 o2hb_fill_node_map_from_callback(testing_map, sizeof(testing_map));
2478 spin_unlock(&o2hb_live_lock);
2479 if (!test_bit(node_num, testing_map)) {
2480 mlog(ML_HEARTBEAT,
2481 "node (%u) does not have heartbeating enabled.\n",
2482 node_num);
2483 return 0;
2484 }
2485
2486 return 1;
2487 }
2488 EXPORT_SYMBOL_GPL(o2hb_check_node_heartbeating_no_sem);
2489
o2hb_check_node_heartbeating_from_callback(u8 node_num)2490 int o2hb_check_node_heartbeating_from_callback(u8 node_num)
2491 {
2492 unsigned long testing_map[BITS_TO_LONGS(O2NM_MAX_NODES)];
2493
2494 o2hb_fill_node_map_from_callback(testing_map, sizeof(testing_map));
2495 if (!test_bit(node_num, testing_map)) {
2496 mlog(ML_HEARTBEAT,
2497 "node (%u) does not have heartbeating enabled.\n",
2498 node_num);
2499 return 0;
2500 }
2501
2502 return 1;
2503 }
2504 EXPORT_SYMBOL_GPL(o2hb_check_node_heartbeating_from_callback);
2505
2506 /*
2507 * this is just a hack until we get the plumbing which flips file systems
2508 * read only and drops the hb ref instead of killing the node dead.
2509 */
o2hb_stop_all_regions(void)2510 void o2hb_stop_all_regions(void)
2511 {
2512 struct o2hb_region *reg;
2513
2514 mlog(ML_ERROR, "stopping heartbeat on all active regions.\n");
2515
2516 spin_lock(&o2hb_live_lock);
2517
2518 list_for_each_entry(reg, &o2hb_all_regions, hr_all_item)
2519 reg->hr_unclean_stop = 1;
2520
2521 spin_unlock(&o2hb_live_lock);
2522 }
2523 EXPORT_SYMBOL_GPL(o2hb_stop_all_regions);
2524
o2hb_get_all_regions(char * region_uuids,u8 max_regions)2525 int o2hb_get_all_regions(char *region_uuids, u8 max_regions)
2526 {
2527 struct o2hb_region *reg;
2528 int numregs = 0;
2529 char *p;
2530
2531 spin_lock(&o2hb_live_lock);
2532
2533 p = region_uuids;
2534 list_for_each_entry(reg, &o2hb_all_regions, hr_all_item) {
2535 if (reg->hr_item_dropped)
2536 continue;
2537
2538 mlog(0, "Region: %s\n", config_item_name(®->hr_item));
2539 if (numregs < max_regions) {
2540 memcpy(p, config_item_name(®->hr_item),
2541 O2HB_MAX_REGION_NAME_LEN);
2542 p += O2HB_MAX_REGION_NAME_LEN;
2543 }
2544 numregs++;
2545 }
2546
2547 spin_unlock(&o2hb_live_lock);
2548
2549 return numregs;
2550 }
2551 EXPORT_SYMBOL_GPL(o2hb_get_all_regions);
2552
o2hb_global_heartbeat_active(void)2553 int o2hb_global_heartbeat_active(void)
2554 {
2555 return (o2hb_heartbeat_mode == O2HB_HEARTBEAT_GLOBAL);
2556 }
2557 EXPORT_SYMBOL(o2hb_global_heartbeat_active);
2558