lock.c 118 KB
Newer Older
1 2 3
/******************************************************************************
*******************************************************************************
**
4
**  Copyright (C) 2005-2008 Red Hat, Inc.  All rights reserved.
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
**
**  This copyrighted material is made available to anyone wishing to use,
**  modify, copy, or redistribute it subject to the terms and conditions
**  of the GNU General Public License v.2.
**
*******************************************************************************
******************************************************************************/

/* Central locking logic has four stages:

   dlm_lock()
   dlm_unlock()

   request_lock(ls, lkb)
   convert_lock(ls, lkb)
   unlock_lock(ls, lkb)
   cancel_lock(ls, lkb)

   _request_lock(r, lkb)
   _convert_lock(r, lkb)
   _unlock_lock(r, lkb)
   _cancel_lock(r, lkb)

   do_request(r, lkb)
   do_convert(r, lkb)
   do_unlock(r, lkb)
   do_cancel(r, lkb)

   Stage 1 (lock, unlock) is mainly about checking input args and
   splitting into one of the four main operations:

       dlm_lock          = request_lock
       dlm_lock+CONVERT  = convert_lock
       dlm_unlock        = unlock_lock
       dlm_unlock+CANCEL = cancel_lock

   Stage 2, xxxx_lock(), just finds and locks the relevant rsb which is
   provided to the next stage.

   Stage 3, _xxxx_lock(), determines if the operation is local or remote.
   When remote, it calls send_xxxx(), when local it calls do_xxxx().

   Stage 4, do_xxxx(), is the guts of the operation.  It manipulates the
   given rsb and lkb and queues callbacks.

   For remote operations, send_xxxx() results in the corresponding do_xxxx()
   function being executed on the remote node.  The connecting send/receive
   calls on local (L) and remote (R) nodes:

   L: send_xxxx()              ->  R: receive_xxxx()
                                   R: do_xxxx()
   L: receive_xxxx_reply()     <-  R: send_xxxx_reply()
*/
David Teigland's avatar
David Teigland committed
58
#include <linux/types.h>
59
#include "dlm_internal.h"
David Teigland's avatar
David Teigland committed
60
#include <linux/dlm_device.h>
61 62 63 64 65 66 67 68 69 70 71 72
#include "memory.h"
#include "lowcomms.h"
#include "requestqueue.h"
#include "util.h"
#include "dir.h"
#include "member.h"
#include "lockspace.h"
#include "ast.h"
#include "lock.h"
#include "rcom.h"
#include "recover.h"
#include "lvb_table.h"
David Teigland's avatar
David Teigland committed
73
#include "user.h"
74 75 76 77 78 79 80 81 82 83 84
#include "config.h"

static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode);
static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb);
static int send_remove(struct dlm_rsb *r);
static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
85
static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb);
86 87 88
static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
				    struct dlm_message *ms);
static int receive_extralen(struct dlm_message *ms);
89
static void do_purge(struct dlm_ls *ls, int nodeid, int pid);
90
static void del_timeout(struct dlm_lkb *lkb);
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158

/*
 * Lock compatibilty matrix - thanks Steve
 * UN = Unlocked state. Not really a state, used as a flag
 * PD = Padding. Used to make the matrix a nice power of two in size
 * Other states are the same as the VMS DLM.
 * Usage: matrix[grmode+1][rqmode+1]  (although m[rq+1][gr+1] is the same)
 */

static const int __dlm_compat_matrix[8][8] = {
      /* UN NL CR CW PR PW EX PD */
        {1, 1, 1, 1, 1, 1, 1, 0},       /* UN */
        {1, 1, 1, 1, 1, 1, 1, 0},       /* NL */
        {1, 1, 1, 1, 1, 1, 0, 0},       /* CR */
        {1, 1, 1, 1, 0, 0, 0, 0},       /* CW */
        {1, 1, 1, 0, 1, 0, 0, 0},       /* PR */
        {1, 1, 1, 0, 0, 0, 0, 0},       /* PW */
        {1, 1, 0, 0, 0, 0, 0, 0},       /* EX */
        {0, 0, 0, 0, 0, 0, 0, 0}        /* PD */
};

/*
 * This defines the direction of transfer of LVB data.
 * Granted mode is the row; requested mode is the column.
 * Usage: matrix[grmode+1][rqmode+1]
 * 1 = LVB is returned to the caller
 * 0 = LVB is written to the resource
 * -1 = nothing happens to the LVB
 */

const int dlm_lvb_operations[8][8] = {
        /* UN   NL  CR  CW  PR  PW  EX  PD*/
        {  -1,  1,  1,  1,  1,  1,  1, -1 }, /* UN */
        {  -1,  1,  1,  1,  1,  1,  1,  0 }, /* NL */
        {  -1, -1,  1,  1,  1,  1,  1,  0 }, /* CR */
        {  -1, -1, -1,  1,  1,  1,  1,  0 }, /* CW */
        {  -1, -1, -1, -1,  1,  1,  1,  0 }, /* PR */
        {  -1,  0,  0,  0,  0,  0,  1,  0 }, /* PW */
        {  -1,  0,  0,  0,  0,  0,  0,  0 }, /* EX */
        {  -1,  0,  0,  0,  0,  0,  0,  0 }  /* PD */
};

#define modes_compat(gr, rq) \
	__dlm_compat_matrix[(gr)->lkb_grmode + 1][(rq)->lkb_rqmode + 1]

int dlm_modes_compat(int mode1, int mode2)
{
	return __dlm_compat_matrix[mode1 + 1][mode2 + 1];
}

/*
 * Compatibility matrix for conversions with QUECVT set.
 * Granted mode is the row; requested mode is the column.
 * Usage: matrix[grmode+1][rqmode+1]
 */

static const int __quecvt_compat_matrix[8][8] = {
      /* UN NL CR CW PR PW EX PD */
        {0, 0, 0, 0, 0, 0, 0, 0},       /* UN */
        {0, 0, 1, 1, 1, 1, 1, 0},       /* NL */
        {0, 0, 0, 1, 1, 1, 1, 0},       /* CR */
        {0, 0, 0, 0, 1, 1, 1, 0},       /* CW */
        {0, 0, 0, 1, 0, 1, 1, 0},       /* PR */
        {0, 0, 0, 0, 0, 0, 1, 0},       /* PW */
        {0, 0, 0, 0, 0, 0, 0, 0},       /* EX */
        {0, 0, 0, 0, 0, 0, 0, 0}        /* PD */
};

David Teigland's avatar
David Teigland committed
159
void dlm_print_lkb(struct dlm_lkb *lkb)
160 161 162 163 164 165 166 167
{
	printk(KERN_ERR "lkb: nodeid %d id %x remid %x exflags %x flags %x\n"
	       "     status %d rqmode %d grmode %d wait_type %d ast_type %d\n",
	       lkb->lkb_nodeid, lkb->lkb_id, lkb->lkb_remid, lkb->lkb_exflags,
	       lkb->lkb_flags, lkb->lkb_status, lkb->lkb_rqmode,
	       lkb->lkb_grmode, lkb->lkb_wait_type, lkb->lkb_ast_type);
}

168
static void dlm_print_rsb(struct dlm_rsb *r)
169 170 171 172 173 174
{
	printk(KERN_ERR "rsb: nodeid %d flags %lx first %x rlc %d name %s\n",
	       r->res_nodeid, r->res_flags, r->res_first_lkid,
	       r->res_recover_locks_count, r->res_name);
}

175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
void dlm_dump_rsb(struct dlm_rsb *r)
{
	struct dlm_lkb *lkb;

	dlm_print_rsb(r);

	printk(KERN_ERR "rsb: root_list empty %d recover_list empty %d\n",
	       list_empty(&r->res_root_list), list_empty(&r->res_recover_list));
	printk(KERN_ERR "rsb lookup list\n");
	list_for_each_entry(lkb, &r->res_lookup, lkb_rsb_lookup)
		dlm_print_lkb(lkb);
	printk(KERN_ERR "rsb grant queue:\n");
	list_for_each_entry(lkb, &r->res_grantqueue, lkb_statequeue)
		dlm_print_lkb(lkb);
	printk(KERN_ERR "rsb convert queue:\n");
	list_for_each_entry(lkb, &r->res_convertqueue, lkb_statequeue)
		dlm_print_lkb(lkb);
	printk(KERN_ERR "rsb wait queue:\n");
	list_for_each_entry(lkb, &r->res_waitqueue, lkb_statequeue)
		dlm_print_lkb(lkb);
}

197 198
/* Threads cannot use the lockspace while it's being recovered */

199
static inline void dlm_lock_recovery(struct dlm_ls *ls)
200 201 202 203
{
	down_read(&ls->ls_in_recovery);
}

204
void dlm_unlock_recovery(struct dlm_ls *ls)
205 206 207 208
{
	up_read(&ls->ls_in_recovery);
}

209
int dlm_lock_recovery_try(struct dlm_ls *ls)
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
{
	return down_read_trylock(&ls->ls_in_recovery);
}

static inline int can_be_queued(struct dlm_lkb *lkb)
{
	return !(lkb->lkb_exflags & DLM_LKF_NOQUEUE);
}

static inline int force_blocking_asts(struct dlm_lkb *lkb)
{
	return (lkb->lkb_exflags & DLM_LKF_NOQUEUEBAST);
}

static inline int is_demoted(struct dlm_lkb *lkb)
{
	return (lkb->lkb_sbflags & DLM_SBF_DEMOTED);
}

David Teigland's avatar
David Teigland committed
229 230 231 232 233 234 235 236 237 238
static inline int is_altmode(struct dlm_lkb *lkb)
{
	return (lkb->lkb_sbflags & DLM_SBF_ALTMODE);
}

static inline int is_granted(struct dlm_lkb *lkb)
{
	return (lkb->lkb_status == DLM_LKSTS_GRANTED);
}

239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
static inline int is_remote(struct dlm_rsb *r)
{
	DLM_ASSERT(r->res_nodeid >= 0, dlm_print_rsb(r););
	return !!r->res_nodeid;
}

static inline int is_process_copy(struct dlm_lkb *lkb)
{
	return (lkb->lkb_nodeid && !(lkb->lkb_flags & DLM_IFL_MSTCPY));
}

static inline int is_master_copy(struct dlm_lkb *lkb)
{
	if (lkb->lkb_flags & DLM_IFL_MSTCPY)
		DLM_ASSERT(lkb->lkb_nodeid, dlm_print_lkb(lkb););
254
	return (lkb->lkb_flags & DLM_IFL_MSTCPY) ? 1 : 0;
255 256 257 258 259 260
}

static inline int middle_conversion(struct dlm_lkb *lkb)
{
	if ((lkb->lkb_grmode==DLM_LOCK_PR && lkb->lkb_rqmode==DLM_LOCK_CW) ||
	    (lkb->lkb_rqmode==DLM_LOCK_PR && lkb->lkb_grmode==DLM_LOCK_CW))
261 262
		return 1;
	return 0;
263 264 265 266 267 268 269
}

static inline int down_conversion(struct dlm_lkb *lkb)
{
	return (!middle_conversion(lkb) && lkb->lkb_rqmode < lkb->lkb_grmode);
}

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
static inline int is_overlap_unlock(struct dlm_lkb *lkb)
{
	return lkb->lkb_flags & DLM_IFL_OVERLAP_UNLOCK;
}

static inline int is_overlap_cancel(struct dlm_lkb *lkb)
{
	return lkb->lkb_flags & DLM_IFL_OVERLAP_CANCEL;
}

static inline int is_overlap(struct dlm_lkb *lkb)
{
	return (lkb->lkb_flags & (DLM_IFL_OVERLAP_UNLOCK |
				  DLM_IFL_OVERLAP_CANCEL));
}

286 287 288 289 290
static void queue_cast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
	if (is_master_copy(lkb))
		return;

291 292
	del_timeout(lkb);

293 294
	DLM_ASSERT(lkb->lkb_lksb, dlm_print_lkb(lkb););

295 296 297 298 299 300 301
	/* if the operation was a cancel, then return -DLM_ECANCEL, if a
	   timeout caused the cancel then return -ETIMEDOUT */
	if (rv == -DLM_ECANCEL && (lkb->lkb_flags & DLM_IFL_TIMEOUT_CANCEL)) {
		lkb->lkb_flags &= ~DLM_IFL_TIMEOUT_CANCEL;
		rv = -ETIMEDOUT;
	}

302 303 304 305 306
	if (rv == -DLM_ECANCEL && (lkb->lkb_flags & DLM_IFL_DEADLOCK_CANCEL)) {
		lkb->lkb_flags &= ~DLM_IFL_DEADLOCK_CANCEL;
		rv = -EDEADLK;
	}

307 308 309
	lkb->lkb_lksb->sb_status = rv;
	lkb->lkb_lksb->sb_flags = lkb->lkb_sbflags;

310
	dlm_add_ast(lkb, AST_COMP, 0);
311 312
}

313 314 315 316 317 318
static inline void queue_cast_overlap(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	queue_cast(r, lkb,
		   is_overlap_unlock(lkb) ? -DLM_EUNLOCK : -DLM_ECANCEL);
}

319 320
static void queue_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int rqmode)
{
321 322
	lkb->lkb_time_bast = ktime_get();

323 324
	if (is_master_copy(lkb))
		send_bast(r, lkb, rqmode);
325 326
	else
		dlm_add_ast(lkb, AST_BAST, rqmode);
327 328 329 330 331 332 333 334 335 336
}

/*
 * Basic operations on rsb's and lkb's
 */

static struct dlm_rsb *create_rsb(struct dlm_ls *ls, char *name, int len)
{
	struct dlm_rsb *r;

337
	r = dlm_allocate_rsb(ls, len);
338 339 340 341 342 343
	if (!r)
		return NULL;

	r->res_ls = ls;
	r->res_length = len;
	memcpy(r->res_name, name, len);
344
	mutex_init(&r->res_mutex);
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365

	INIT_LIST_HEAD(&r->res_lookup);
	INIT_LIST_HEAD(&r->res_grantqueue);
	INIT_LIST_HEAD(&r->res_convertqueue);
	INIT_LIST_HEAD(&r->res_waitqueue);
	INIT_LIST_HEAD(&r->res_root_list);
	INIT_LIST_HEAD(&r->res_recover_list);

	return r;
}

static int search_rsb_list(struct list_head *head, char *name, int len,
			   unsigned int flags, struct dlm_rsb **r_ret)
{
	struct dlm_rsb *r;
	int error = 0;

	list_for_each_entry(r, head, res_hashchain) {
		if (len == r->res_length && !memcmp(name, r->res_name, len))
			goto found;
	}
366
	*r_ret = NULL;
David Teigland's avatar
David Teigland committed
367
	return -EBADR;
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414

 found:
	if (r->res_nodeid && (flags & R_MASTER))
		error = -ENOTBLK;
	*r_ret = r;
	return error;
}

static int _search_rsb(struct dlm_ls *ls, char *name, int len, int b,
		       unsigned int flags, struct dlm_rsb **r_ret)
{
	struct dlm_rsb *r;
	int error;

	error = search_rsb_list(&ls->ls_rsbtbl[b].list, name, len, flags, &r);
	if (!error) {
		kref_get(&r->res_ref);
		goto out;
	}
	error = search_rsb_list(&ls->ls_rsbtbl[b].toss, name, len, flags, &r);
	if (error)
		goto out;

	list_move(&r->res_hashchain, &ls->ls_rsbtbl[b].list);

	if (dlm_no_directory(ls))
		goto out;

	if (r->res_nodeid == -1) {
		rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
		r->res_first_lkid = 0;
	} else if (r->res_nodeid > 0) {
		rsb_set_flag(r, RSB_MASTER_UNCERTAIN);
		r->res_first_lkid = 0;
	} else {
		DLM_ASSERT(r->res_nodeid == 0, dlm_print_rsb(r););
		DLM_ASSERT(!rsb_flag(r, RSB_MASTER_UNCERTAIN),);
	}
 out:
	*r_ret = r;
	return error;
}

static int search_rsb(struct dlm_ls *ls, char *name, int len, int b,
		      unsigned int flags, struct dlm_rsb **r_ret)
{
	int error;
415
	spin_lock(&ls->ls_rsbtbl[b].lock);
416
	error = _search_rsb(ls, name, len, b, flags, r_ret);
417
	spin_unlock(&ls->ls_rsbtbl[b].lock);
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
	return error;
}

/*
 * Find rsb in rsbtbl and potentially create/add one
 *
 * Delaying the release of rsb's has a similar benefit to applications keeping
 * NL locks on an rsb, but without the guarantee that the cached master value
 * will still be valid when the rsb is reused.  Apps aren't always smart enough
 * to keep NL locks on an rsb that they may lock again shortly; this can lead
 * to excessive master lookups and removals if we don't delay the release.
 *
 * Searching for an rsb means looking through both the normal list and toss
 * list.  When found on the toss list the rsb is moved to the normal list with
 * ref count of 1; when found on normal list the ref count is incremented.
 */

static int find_rsb(struct dlm_ls *ls, char *name, int namelen,
		    unsigned int flags, struct dlm_rsb **r_ret)
{
	struct dlm_rsb *r, *tmp;
	uint32_t hash, bucket;
440 441 442 443
	int error = -EINVAL;

	if (namelen > DLM_RESNAME_MAXLEN)
		goto out;
444 445 446 447

	if (dlm_no_directory(ls))
		flags |= R_CREATE;

448
	error = 0;
449 450 451 452 453 454 455
	hash = jhash(name, namelen, 0);
	bucket = hash & (ls->ls_rsbtbl_size - 1);

	error = search_rsb(ls, name, namelen, bucket, flags, &r);
	if (!error)
		goto out;

David Teigland's avatar
David Teigland committed
456
	if (error == -EBADR && !(flags & R_CREATE))
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
		goto out;

	/* the rsb was found but wasn't a master copy */
	if (error == -ENOTBLK)
		goto out;

	error = -ENOMEM;
	r = create_rsb(ls, name, namelen);
	if (!r)
		goto out;

	r->res_hash = hash;
	r->res_bucket = bucket;
	r->res_nodeid = -1;
	kref_init(&r->res_ref);

	/* With no directory, the master can be set immediately */
	if (dlm_no_directory(ls)) {
		int nodeid = dlm_dir_nodeid(r);
		if (nodeid == dlm_our_nodeid())
			nodeid = 0;
		r->res_nodeid = nodeid;
	}

481
	spin_lock(&ls->ls_rsbtbl[bucket].lock);
482 483
	error = _search_rsb(ls, name, namelen, bucket, 0, &tmp);
	if (!error) {
484
		spin_unlock(&ls->ls_rsbtbl[bucket].lock);
485
		dlm_free_rsb(r);
486 487 488 489
		r = tmp;
		goto out;
	}
	list_add(&r->res_hashchain, &ls->ls_rsbtbl[bucket].list);
490
	spin_unlock(&ls->ls_rsbtbl[bucket].lock);
491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519
	error = 0;
 out:
	*r_ret = r;
	return error;
}

/* This is only called to add a reference when the code already holds
   a valid reference to the rsb, so there's no need for locking. */

static inline void hold_rsb(struct dlm_rsb *r)
{
	kref_get(&r->res_ref);
}

void dlm_hold_rsb(struct dlm_rsb *r)
{
	hold_rsb(r);
}

static void toss_rsb(struct kref *kref)
{
	struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);
	struct dlm_ls *ls = r->res_ls;

	DLM_ASSERT(list_empty(&r->res_root_list), dlm_print_rsb(r););
	kref_init(&r->res_ref);
	list_move(&r->res_hashchain, &ls->ls_rsbtbl[r->res_bucket].toss);
	r->res_toss_time = jiffies;
	if (r->res_lvbptr) {
520
		dlm_free_lvb(r->res_lvbptr);
521 522 523 524 525 526 527 528 529 530 531 532
		r->res_lvbptr = NULL;
	}
}

/* When all references to the rsb are gone it's transfered to
   the tossed list for later disposal. */

static void put_rsb(struct dlm_rsb *r)
{
	struct dlm_ls *ls = r->res_ls;
	uint32_t bucket = r->res_bucket;

533
	spin_lock(&ls->ls_rsbtbl[bucket].lock);
534
	kref_put(&r->res_ref, toss_rsb);
535
	spin_unlock(&ls->ls_rsbtbl[bucket].lock);
536 537 538 539 540 541 542 543 544 545 546 547 548
}

void dlm_put_rsb(struct dlm_rsb *r)
{
	put_rsb(r);
}

/* See comment for unhold_lkb */

static void unhold_rsb(struct dlm_rsb *r)
{
	int rv;
	rv = kref_put(&r->res_ref, toss_rsb);
549
	DLM_ASSERT(!rv, dlm_dump_rsb(r););
550 551 552 553 554 555 556 557 558
}

static void kill_rsb(struct kref *kref)
{
	struct dlm_rsb *r = container_of(kref, struct dlm_rsb, res_ref);

	/* All work is done after the return from kref_put() so we
	   can release the write_lock before the remove and free. */

559 560 561 562 563 564
	DLM_ASSERT(list_empty(&r->res_lookup), dlm_dump_rsb(r););
	DLM_ASSERT(list_empty(&r->res_grantqueue), dlm_dump_rsb(r););
	DLM_ASSERT(list_empty(&r->res_convertqueue), dlm_dump_rsb(r););
	DLM_ASSERT(list_empty(&r->res_waitqueue), dlm_dump_rsb(r););
	DLM_ASSERT(list_empty(&r->res_root_list), dlm_dump_rsb(r););
	DLM_ASSERT(list_empty(&r->res_recover_list), dlm_dump_rsb(r););
565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589
}

/* Attaching/detaching lkb's from rsb's is for rsb reference counting.
   The rsb must exist as long as any lkb's for it do. */

static void attach_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	hold_rsb(r);
	lkb->lkb_resource = r;
}

static void detach_lkb(struct dlm_lkb *lkb)
{
	if (lkb->lkb_resource) {
		put_rsb(lkb->lkb_resource);
		lkb->lkb_resource = NULL;
	}
}

static int create_lkb(struct dlm_ls *ls, struct dlm_lkb **lkb_ret)
{
	struct dlm_lkb *lkb, *tmp;
	uint32_t lkid = 0;
	uint16_t bucket;

590
	lkb = dlm_allocate_lkb(ls);
591 592 593 594 595 596
	if (!lkb)
		return -ENOMEM;

	lkb->lkb_nodeid = -1;
	lkb->lkb_grmode = DLM_LOCK_IV;
	kref_init(&lkb->lkb_ref);
597
	INIT_LIST_HEAD(&lkb->lkb_ownqueue);
598
	INIT_LIST_HEAD(&lkb->lkb_rsb_lookup);
599
	INIT_LIST_HEAD(&lkb->lkb_time_list);
600 601 602 603 604 605 606 607 608

	get_random_bytes(&bucket, sizeof(bucket));
	bucket &= (ls->ls_lkbtbl_size - 1);

	write_lock(&ls->ls_lkbtbl[bucket].lock);

	/* counter can roll over so we must verify lkid is not in use */

	while (lkid == 0) {
David Teigland's avatar
David Teigland committed
609
		lkid = (bucket << 16) | ls->ls_lkbtbl[bucket].counter++;
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630

		list_for_each_entry(tmp, &ls->ls_lkbtbl[bucket].list,
				    lkb_idtbl_list) {
			if (tmp->lkb_id != lkid)
				continue;
			lkid = 0;
			break;
		}
	}

	lkb->lkb_id = lkid;
	list_add(&lkb->lkb_idtbl_list, &ls->ls_lkbtbl[bucket].list);
	write_unlock(&ls->ls_lkbtbl[bucket].lock);

	*lkb_ret = lkb;
	return 0;
}

static struct dlm_lkb *__find_lkb(struct dlm_ls *ls, uint32_t lkid)
{
	struct dlm_lkb *lkb;
David Teigland's avatar
David Teigland committed
631
	uint16_t bucket = (lkid >> 16);
632 633 634 635 636 637 638 639 640 641 642

	list_for_each_entry(lkb, &ls->ls_lkbtbl[bucket].list, lkb_idtbl_list) {
		if (lkb->lkb_id == lkid)
			return lkb;
	}
	return NULL;
}

static int find_lkb(struct dlm_ls *ls, uint32_t lkid, struct dlm_lkb **lkb_ret)
{
	struct dlm_lkb *lkb;
David Teigland's avatar
David Teigland committed
643
	uint16_t bucket = (lkid >> 16);
644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667

	if (bucket >= ls->ls_lkbtbl_size)
		return -EBADSLT;

	read_lock(&ls->ls_lkbtbl[bucket].lock);
	lkb = __find_lkb(ls, lkid);
	if (lkb)
		kref_get(&lkb->lkb_ref);
	read_unlock(&ls->ls_lkbtbl[bucket].lock);

	*lkb_ret = lkb;
	return lkb ? 0 : -ENOENT;
}

static void kill_lkb(struct kref *kref)
{
	struct dlm_lkb *lkb = container_of(kref, struct dlm_lkb, lkb_ref);

	/* All work is done after the return from kref_put() so we
	   can release the write_lock before the detach_lkb */

	DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););
}

668 669 670 671
/* __put_lkb() is used when an lkb may not have an rsb attached to
   it so we need to provide the lockspace explicitly */

static int __put_lkb(struct dlm_ls *ls, struct dlm_lkb *lkb)
672
{
David Teigland's avatar
David Teigland committed
673
	uint16_t bucket = (lkb->lkb_id >> 16);
674 675 676 677 678 679 680 681 682 683

	write_lock(&ls->ls_lkbtbl[bucket].lock);
	if (kref_put(&lkb->lkb_ref, kill_lkb)) {
		list_del(&lkb->lkb_idtbl_list);
		write_unlock(&ls->ls_lkbtbl[bucket].lock);

		detach_lkb(lkb);

		/* for local/process lkbs, lvbptr points to caller's lksb */
		if (lkb->lkb_lvbptr && is_master_copy(lkb))
684 685
			dlm_free_lvb(lkb->lkb_lvbptr);
		dlm_free_lkb(lkb);
686 687 688 689 690 691 692 693 694
		return 1;
	} else {
		write_unlock(&ls->ls_lkbtbl[bucket].lock);
		return 0;
	}
}

int dlm_put_lkb(struct dlm_lkb *lkb)
{
695 696 697 698 699 700 701
	struct dlm_ls *ls;

	DLM_ASSERT(lkb->lkb_resource, dlm_print_lkb(lkb););
	DLM_ASSERT(lkb->lkb_resource->res_ls, dlm_print_lkb(lkb););

	ls = lkb->lkb_resource->res_ls;
	return __put_lkb(ls, lkb);
702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
}

/* This is only called to add a reference when the code already holds
   a valid reference to the lkb, so there's no need for locking. */

static inline void hold_lkb(struct dlm_lkb *lkb)
{
	kref_get(&lkb->lkb_ref);
}

/* This is called when we need to remove a reference and are certain
   it's not the last ref.  e.g. del_lkb is always called between a
   find_lkb/put_lkb and is always the inverse of a previous add_lkb.
   put_lkb would work fine, but would involve unnecessary locking */

static inline void unhold_lkb(struct dlm_lkb *lkb)
{
	int rv;
	rv = kref_put(&lkb->lkb_ref, kill_lkb);
	DLM_ASSERT(!rv, dlm_print_lkb(lkb););
}

static void lkb_add_ordered(struct list_head *new, struct list_head *head,
			    int mode)
{
	struct dlm_lkb *lkb = NULL;

	list_for_each_entry(lkb, head, lkb_statequeue)
		if (lkb->lkb_rqmode < mode)
			break;

	if (!lkb)
		list_add_tail(new, head);
	else
		__list_add(new, lkb->lkb_statequeue.prev, &lkb->lkb_statequeue);
}

/* add/remove lkb to rsb's grant/convert/wait queue */

static void add_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int status)
{
	kref_get(&lkb->lkb_ref);

	DLM_ASSERT(!lkb->lkb_status, dlm_print_lkb(lkb););

747 748
	lkb->lkb_timestamp = ktime_get();

749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
	lkb->lkb_status = status;

	switch (status) {
	case DLM_LKSTS_WAITING:
		if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
			list_add(&lkb->lkb_statequeue, &r->res_waitqueue);
		else
			list_add_tail(&lkb->lkb_statequeue, &r->res_waitqueue);
		break;
	case DLM_LKSTS_GRANTED:
		/* convention says granted locks kept in order of grmode */
		lkb_add_ordered(&lkb->lkb_statequeue, &r->res_grantqueue,
				lkb->lkb_grmode);
		break;
	case DLM_LKSTS_CONVERT:
		if (lkb->lkb_exflags & DLM_LKF_HEADQUE)
			list_add(&lkb->lkb_statequeue, &r->res_convertqueue);
		else
			list_add_tail(&lkb->lkb_statequeue,
				      &r->res_convertqueue);
		break;
	default:
		DLM_ASSERT(0, dlm_print_lkb(lkb); printk("sts=%d\n", status););
	}
}

static void del_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	lkb->lkb_status = 0;
	list_del(&lkb->lkb_statequeue);
	unhold_lkb(lkb);
}

static void move_lkb(struct dlm_rsb *r, struct dlm_lkb *lkb, int sts)
{
	hold_lkb(lkb);
	del_lkb(r, lkb);
	add_lkb(r, lkb, sts);
	unhold_lkb(lkb);
}

790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
static int msg_reply_type(int mstype)
{
	switch (mstype) {
	case DLM_MSG_REQUEST:
		return DLM_MSG_REQUEST_REPLY;
	case DLM_MSG_CONVERT:
		return DLM_MSG_CONVERT_REPLY;
	case DLM_MSG_UNLOCK:
		return DLM_MSG_UNLOCK_REPLY;
	case DLM_MSG_CANCEL:
		return DLM_MSG_CANCEL_REPLY;
	case DLM_MSG_LOOKUP:
		return DLM_MSG_LOOKUP_REPLY;
	}
	return -1;
}

807 808 809
/* add/remove lkb from global waiters list of lkb's waiting for
   a reply from a remote node */

810
static int add_to_waiters(struct dlm_lkb *lkb, int mstype)
811 812
{
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;
813
	int error = 0;
814

815
	mutex_lock(&ls->ls_waiters_mutex);
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840

	if (is_overlap_unlock(lkb) ||
	    (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL))) {
		error = -EINVAL;
		goto out;
	}

	if (lkb->lkb_wait_type || is_overlap_cancel(lkb)) {
		switch (mstype) {
		case DLM_MSG_UNLOCK:
			lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
			break;
		case DLM_MSG_CANCEL:
			lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
			break;
		default:
			error = -EBUSY;
			goto out;
		}
		lkb->lkb_wait_count++;
		hold_lkb(lkb);

		log_debug(ls, "add overlap %x cur %d new %d count %d flags %x",
			  lkb->lkb_id, lkb->lkb_wait_type, mstype,
			  lkb->lkb_wait_count, lkb->lkb_flags);
841 842
		goto out;
	}
843 844 845 846 847 848

	DLM_ASSERT(!lkb->lkb_wait_count,
		   dlm_print_lkb(lkb);
		   printk("wait_count %d\n", lkb->lkb_wait_count););

	lkb->lkb_wait_count++;
849
	lkb->lkb_wait_type = mstype;
850
	hold_lkb(lkb);
851 852
	list_add(&lkb->lkb_wait_reply, &ls->ls_waiters);
 out:
853 854 855 856
	if (error)
		log_error(ls, "add_to_waiters %x error %d flags %x %d %d %s",
			  lkb->lkb_id, error, lkb->lkb_flags, mstype,
			  lkb->lkb_wait_type, lkb->lkb_resource->res_name);
857
	mutex_unlock(&ls->ls_waiters_mutex);
858
	return error;
859 860
}

861 862 863 864 865
/* We clear the RESEND flag because we might be taking an lkb off the waiters
   list as part of process_requestqueue (e.g. a lookup that has an optimized
   request reply on the requestqueue) between dlm_recover_waiters_pre() which
   set RESEND and dlm_recover_waiters_post() */

866
static int _remove_from_waiters(struct dlm_lkb *lkb, int mstype)
867
{
868 869
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;
	int overlap_done = 0;
870

871 872 873 874
	if (is_overlap_unlock(lkb) && (mstype == DLM_MSG_UNLOCK_REPLY)) {
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
		overlap_done = 1;
		goto out_del;
875
	}
876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909

	if (is_overlap_cancel(lkb) && (mstype == DLM_MSG_CANCEL_REPLY)) {
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
		overlap_done = 1;
		goto out_del;
	}

	/* N.B. type of reply may not always correspond to type of original
	   msg due to lookup->request optimization, verify others? */

	if (lkb->lkb_wait_type) {
		lkb->lkb_wait_type = 0;
		goto out_del;
	}

	log_error(ls, "remove_from_waiters lkid %x flags %x types %d %d",
		  lkb->lkb_id, lkb->lkb_flags, mstype, lkb->lkb_wait_type);
	return -1;

 out_del:
	/* the force-unlock/cancel has completed and we haven't recvd a reply
	   to the op that was in progress prior to the unlock/cancel; we
	   give up on any reply to the earlier op.  FIXME: not sure when/how
	   this would happen */

	if (overlap_done && lkb->lkb_wait_type) {
		log_error(ls, "remove_from_waiters %x reply %d give up on %d",
			  lkb->lkb_id, mstype, lkb->lkb_wait_type);
		lkb->lkb_wait_count--;
		lkb->lkb_wait_type = 0;
	}

	DLM_ASSERT(lkb->lkb_wait_count, dlm_print_lkb(lkb););

910
	lkb->lkb_flags &= ~DLM_IFL_RESEND;
911 912 913
	lkb->lkb_wait_count--;
	if (!lkb->lkb_wait_count)
		list_del_init(&lkb->lkb_wait_reply);
914
	unhold_lkb(lkb);
915
	return 0;
916 917
}

918
static int remove_from_waiters(struct dlm_lkb *lkb, int mstype)
919 920 921 922
{
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;
	int error;

923
	mutex_lock(&ls->ls_waiters_mutex);
924
	error = _remove_from_waiters(lkb, mstype);
925
	mutex_unlock(&ls->ls_waiters_mutex);
926 927 928
	return error;
}

929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944
/* Handles situations where we might be processing a "fake" or "stub" reply in
   which we can't try to take waiters_mutex again. */

static int remove_from_waiters_ms(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;
	int error;

	if (ms != &ls->ls_stub_ms)
		mutex_lock(&ls->ls_waiters_mutex);
	error = _remove_from_waiters(lkb, ms->m_type);
	if (ms != &ls->ls_stub_ms)
		mutex_unlock(&ls->ls_waiters_mutex);
	return error;
}

945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968
static void dir_remove(struct dlm_rsb *r)
{
	int to_nodeid;

	if (dlm_no_directory(r->res_ls))
		return;

	to_nodeid = dlm_dir_nodeid(r);
	if (to_nodeid != dlm_our_nodeid())
		send_remove(r);
	else
		dlm_dir_remove_entry(r->res_ls, to_nodeid,
				     r->res_name, r->res_length);
}

/* FIXME: shouldn't this be able to exit as soon as one non-due rsb is
   found since they are in order of newest to oldest? */

static int shrink_bucket(struct dlm_ls *ls, int b)
{
	struct dlm_rsb *r;
	int count = 0, found;

	for (;;) {
969
		found = 0;
970
		spin_lock(&ls->ls_rsbtbl[b].lock);
971 972 973
		list_for_each_entry_reverse(r, &ls->ls_rsbtbl[b].toss,
					    res_hashchain) {
			if (!time_after_eq(jiffies, r->res_toss_time +
974
					   dlm_config.ci_toss_secs * HZ))
975
				continue;
976
			found = 1;
977 978 979 980
			break;
		}

		if (!found) {
981
			spin_unlock(&ls->ls_rsbtbl[b].lock);
982 983 984 985 986
			break;
		}

		if (kref_put(&r->res_ref, kill_rsb)) {
			list_del(&r->res_hashchain);
987
			spin_unlock(&ls->ls_rsbtbl[b].lock);
988 989 990

			if (is_master(r))
				dir_remove(r);
991
			dlm_free_rsb(r);
992 993
			count++;
		} else {
994
			spin_unlock(&ls->ls_rsbtbl[b].lock);
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
			log_error(ls, "tossed rsb in use %s", r->res_name);
		}
	}

	return count;
}

void dlm_scan_rsbs(struct dlm_ls *ls)
{
	int i;

	for (i = 0; i < ls->ls_rsbtbl_size; i++) {
		shrink_bucket(ls, i);
1008 1009
		if (dlm_locking_stopped(ls))
			break;
1010 1011 1012 1013
		cond_resched();
	}
}

1014 1015 1016 1017
static void add_timeout(struct dlm_lkb *lkb)
{
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;

1018
	if (is_master_copy(lkb))
1019 1020 1021 1022 1023 1024 1025
		return;

	if (test_bit(LSFL_TIMEWARN, &ls->ls_flags) &&
	    !(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
		lkb->lkb_flags |= DLM_IFL_WATCH_TIMEWARN;
		goto add_it;
	}
David Teigland's avatar
David Teigland committed
1026 1027
	if (lkb->lkb_exflags & DLM_LKF_TIMEOUT)
		goto add_it;
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060
	return;

 add_it:
	DLM_ASSERT(list_empty(&lkb->lkb_time_list), dlm_print_lkb(lkb););
	mutex_lock(&ls->ls_timeout_mutex);
	hold_lkb(lkb);
	list_add_tail(&lkb->lkb_time_list, &ls->ls_timeout);
	mutex_unlock(&ls->ls_timeout_mutex);
}

static void del_timeout(struct dlm_lkb *lkb)
{
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;

	mutex_lock(&ls->ls_timeout_mutex);
	if (!list_empty(&lkb->lkb_time_list)) {
		list_del_init(&lkb->lkb_time_list);
		unhold_lkb(lkb);
	}
	mutex_unlock(&ls->ls_timeout_mutex);
}

/* FIXME: is it safe to look at lkb_exflags, lkb_flags, lkb_timestamp, and
   lkb_lksb_timeout without lock_rsb?  Note: we can't lock timeout_mutex
   and then lock rsb because of lock ordering in add_timeout.  We may need
   to specify some special timeout-related bits in the lkb that are just to
   be accessed under the timeout_mutex. */

void dlm_scan_timeout(struct dlm_ls *ls)
{
	struct dlm_rsb *r;
	struct dlm_lkb *lkb;
	int do_cancel, do_warn;
1061
	s64 wait_us;
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071

	for (;;) {
		if (dlm_locking_stopped(ls))
			break;

		do_cancel = 0;
		do_warn = 0;
		mutex_lock(&ls->ls_timeout_mutex);
		list_for_each_entry(lkb, &ls->ls_timeout, lkb_time_list) {

1072 1073 1074
			wait_us = ktime_to_us(ktime_sub(ktime_get(),
					      		lkb->lkb_timestamp));

1075
			if ((lkb->lkb_exflags & DLM_LKF_TIMEOUT) &&
1076
			    wait_us >= (lkb->lkb_timeout_cs * 10000))
1077 1078 1079
				do_cancel = 1;

			if ((lkb->lkb_flags & DLM_IFL_WATCH_TIMEWARN) &&
1080
			    wait_us >= dlm_config.ci_timewarn_cs * 10000)
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105
				do_warn = 1;

			if (!do_cancel && !do_warn)
				continue;
			hold_lkb(lkb);
			break;
		}
		mutex_unlock(&ls->ls_timeout_mutex);

		if (!do_cancel && !do_warn)
			break;

		r = lkb->lkb_resource;
		hold_rsb(r);
		lock_rsb(r);

		if (do_warn) {
			/* clear flag so we only warn once */
			lkb->lkb_flags &= ~DLM_IFL_WATCH_TIMEWARN;
			if (!(lkb->lkb_exflags & DLM_LKF_TIMEOUT))
				del_timeout(lkb);
			dlm_timeout_warn(lkb);
		}

		if (do_cancel) {
Steven Whitehouse's avatar
Steven Whitehouse committed
1106
			log_debug(ls, "timeout cancel %x node %d %s",
David Teigland's avatar
David Teigland committed
1107
				  lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
			lkb->lkb_flags &= ~DLM_IFL_WATCH_TIMEWARN;
			lkb->lkb_flags |= DLM_IFL_TIMEOUT_CANCEL;
			del_timeout(lkb);
			_cancel_lock(r, lkb);
		}

		unlock_rsb(r);
		unhold_rsb(r);
		dlm_put_lkb(lkb);
	}
}

/* This is only called by dlm_recoverd, and we rely on dlm_ls_stop() stopping
   dlm_recoverd before checking/setting ls_recover_begin. */

void dlm_adjust_timeouts(struct dlm_ls *ls)
{
	struct dlm_lkb *lkb;
1126
	u64 adj_us = jiffies_to_usecs(jiffies - ls->ls_recover_begin);
1127 1128 1129 1130

	ls->ls_recover_begin = 0;
	mutex_lock(&ls->ls_timeout_mutex);
	list_for_each_entry(lkb, &ls->ls_timeout, lkb_time_list)
1131
		lkb->lkb_timestamp = ktime_add_us(lkb->lkb_timestamp, adj_us);
1132 1133 1134
	mutex_unlock(&ls->ls_timeout_mutex);
}

1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
/* lkb is master or local copy */

static void set_lvb_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int b, len = r->res_ls->ls_lvblen;

	/* b=1 lvb returned to caller
	   b=0 lvb written to rsb or invalidated
	   b=-1 do nothing */

	b =  dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];

	if (b == 1) {
		if (!lkb->lkb_lvbptr)
			return;

		if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
			return;

		if (!r->res_lvbptr)
			return;

		memcpy(lkb->lkb_lvbptr, r->res_lvbptr, len);
		lkb->lkb_lvbseq = r->res_lvbseq;

	} else if (b == 0) {
		if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
			rsb_set_flag(r, RSB_VALNOTVALID);
			return;
		}

		if (!lkb->lkb_lvbptr)
			return;

		if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
			return;

		if (!r->res_lvbptr)
1173
			r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204

		if (!r->res_lvbptr)
			return;

		memcpy(r->res_lvbptr, lkb->lkb_lvbptr, len);
		r->res_lvbseq++;
		lkb->lkb_lvbseq = r->res_lvbseq;
		rsb_clear_flag(r, RSB_VALNOTVALID);
	}

	if (rsb_flag(r, RSB_VALNOTVALID))
		lkb->lkb_sbflags |= DLM_SBF_VALNOTVALID;
}

static void set_lvb_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	if (lkb->lkb_grmode < DLM_LOCK_PW)
		return;

	if (lkb->lkb_exflags & DLM_LKF_IVVALBLK) {
		rsb_set_flag(r, RSB_VALNOTVALID);
		return;
	}

	if (!lkb->lkb_lvbptr)
		return;

	if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
		return;

	if (!r->res_lvbptr)
1205
		r->res_lvbptr = dlm_allocate_lvb(r->res_ls);
1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227

	if (!r->res_lvbptr)
		return;

	memcpy(r->res_lvbptr, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
	r->res_lvbseq++;
	rsb_clear_flag(r, RSB_VALNOTVALID);
}

/* lkb is process copy (pc) */

static void set_lvb_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
			    struct dlm_message *ms)
{
	int b;

	if (!lkb->lkb_lvbptr)
		return;

	if (!(lkb->lkb_exflags & DLM_LKF_VALBLK))
		return;

David Teigland's avatar
David Teigland committed
1228
	b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1];
1229 1230
	if (b == 1) {
		int len = receive_extralen(ms);
1231 1232
		if (len > DLM_RESNAME_MAXLEN)
			len = DLM_RESNAME_MAXLEN;
1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
		memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
		lkb->lkb_lvbseq = ms->m_lvbseq;
	}
}

/* Manipulate lkb's on rsb's convert/granted/waiting queues
   remove_lock -- used for unlock, removes lkb from granted
   revert_lock -- used for cancel, moves lkb from convert to granted
   grant_lock  -- used for request and convert, adds lkb to granted or
                  moves lkb from convert or waiting to granted

   Each of these is used for master or local copy lkb's.  There is
   also a _pc() variation used to make the corresponding change on
   a process copy (pc) lkb. */

static void _remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	del_lkb(r, lkb);
	lkb->lkb_grmode = DLM_LOCK_IV;
	/* this unhold undoes the original ref from create_lkb()
	   so this leads to the lkb being freed */
	unhold_lkb(lkb);
}

static void remove_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	set_lvb_unlock(r, lkb);
	_remove_lock(r, lkb);
}

static void remove_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	_remove_lock(r, lkb);
}

1268 1269 1270 1271 1272
/* returns: 0 did nothing
	    1 moved lock to granted
	   -1 removed lock */

static int revert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
1273
{
1274 1275
	int rv = 0;

1276 1277 1278
	lkb->lkb_rqmode = DLM_LOCK_IV;

	switch (lkb->lkb_status) {
David Teigland's avatar
David Teigland committed
1279 1280
	case DLM_LKSTS_GRANTED:
		break;
1281 1282
	case DLM_LKSTS_CONVERT:
		move_lkb(r, lkb, DLM_LKSTS_GRANTED);
1283
		rv = 1;
1284 1285 1286 1287 1288 1289 1290
		break;
	case DLM_LKSTS_WAITING:
		del_lkb(r, lkb);
		lkb->lkb_grmode = DLM_LOCK_IV;
		/* this unhold undoes the original ref from create_lkb()
		   so this leads to the lkb being freed */
		unhold_lkb(lkb);
1291
		rv = -1;
1292 1293 1294 1295
		break;
	default:
		log_print("invalid status for revert %d", lkb->lkb_status);
	}
1296
	return rv;
1297 1298
}

1299
static int revert_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb)
1300
{
1301
	return revert_lock(r, lkb);
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343
}

static void _grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	if (lkb->lkb_grmode != lkb->lkb_rqmode) {
		lkb->lkb_grmode = lkb->lkb_rqmode;
		if (lkb->lkb_status)
			move_lkb(r, lkb, DLM_LKSTS_GRANTED);
		else
			add_lkb(r, lkb, DLM_LKSTS_GRANTED);
	}

	lkb->lkb_rqmode = DLM_LOCK_IV;
}

static void grant_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	set_lvb_lock(r, lkb);
	_grant_lock(r, lkb);
	lkb->lkb_highbast = 0;
}

static void grant_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb,
			  struct dlm_message *ms)
{
	set_lvb_lock_pc(r, lkb, ms);
	_grant_lock(r, lkb);
}

/* called by grant_pending_locks() which means an async grant message must
   be sent to the requesting node in addition to granting the lock if the
   lkb belongs to a remote node. */

static void grant_lock_pending(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	grant_lock(r, lkb);
	if (is_master_copy(lkb))
		send_grant(r, lkb);
	else
		queue_cast(r, lkb, 0);
}

David Teigland's avatar
David Teigland committed
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387
/* The special CONVDEADLK, ALTPR and ALTCW flags allow the master to
   change the granted/requested modes.  We're munging things accordingly in
   the process copy.
   CONVDEADLK: our grmode may have been forced down to NL to resolve a
   conversion deadlock
   ALTPR/ALTCW: our rqmode may have been changed to PR or CW to become
   compatible with other granted locks */

static void munge_demoted(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	if (ms->m_type != DLM_MSG_CONVERT_REPLY) {
		log_print("munge_demoted %x invalid reply type %d",
			  lkb->lkb_id, ms->m_type);
		return;
	}

	if (lkb->lkb_rqmode == DLM_LOCK_IV || lkb->lkb_grmode == DLM_LOCK_IV) {
		log_print("munge_demoted %x invalid modes gr %d rq %d",
			  lkb->lkb_id, lkb->lkb_grmode, lkb->lkb_rqmode);
		return;
	}

	lkb->lkb_grmode = DLM_LOCK_NL;
}

static void munge_altmode(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	if (ms->m_type != DLM_MSG_REQUEST_REPLY &&
	    ms->m_type != DLM_MSG_GRANT) {
		log_print("munge_altmode %x invalid reply type %d",
			  lkb->lkb_id, ms->m_type);
		return;
	}

	if (lkb->lkb_exflags & DLM_LKF_ALTPR)
		lkb->lkb_rqmode = DLM_LOCK_PR;
	else if (lkb->lkb_exflags & DLM_LKF_ALTCW)
		lkb->lkb_rqmode = DLM_LOCK_CW;
	else {
		log_print("munge_altmode invalid exflags %x", lkb->lkb_exflags);
		dlm_print_lkb(lkb);
	}
}

1388 1389 1390 1391 1392
static inline int first_in_list(struct dlm_lkb *lkb, struct list_head *head)
{
	struct dlm_lkb *first = list_entry(head->next, struct dlm_lkb,
					   lkb_statequeue);
	if (lkb->lkb_id == first->lkb_id)
1393
		return 1;
1394

1395
	return 0;
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406
}

/* Check if the given lkb conflicts with another lkb on the queue. */

static int queue_conflict(struct list_head *head, struct dlm_lkb *lkb)
{
	struct dlm_lkb *this;

	list_for_each_entry(this, head, lkb_statequeue) {
		if (this == lkb)
			continue;
1407
		if (!modes_compat(this, lkb))
1408
			return 1;
1409
	}
1410
	return 0;
1411 1412 1413 1414 1415 1416 1417
}

/*
 * "A conversion deadlock arises with a pair of lock requests in the converting
 * queue for one resource.  The granted mode of each lock blocks the requested
 * mode of the other lock."
 *
1418 1419
 * Part 2: if the granted mode of lkb is preventing an earlier lkb in the
 * convert queue from being granted, then deadlk/demote lkb.
1420 1421 1422 1423 1424 1425 1426 1427
 *
 * Example:
 * Granted Queue: empty
 * Convert Queue: NL->EX (first lock)
 *                PR->EX (second lock)
 *
 * The first lock can't be granted because of the granted mode of the second
 * lock and the second lock can't be granted because it's not first in the
1428 1429 1430
 * list.  We either cancel lkb's conversion (PR->EX) and return EDEADLK, or we
 * demote the granted mode of lkb (from PR to NL) if it has the CONVDEADLK
 * flag set and return DEMOTED in the lksb flags.
1431
 *
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
 * Originally, this function detected conv-deadlk in a more limited scope:
 * - if !modes_compat(lkb1, lkb2) && !modes_compat(lkb2, lkb1), or
 * - if lkb1 was the first entry in the queue (not just earlier), and was
 *   blocked by the granted mode of lkb2, and there was nothing on the
 *   granted queue preventing lkb1 from being granted immediately, i.e.
 *   lkb2 was the only thing preventing lkb1 from being granted.
 *
 * That second condition meant we'd only say there was conv-deadlk if
 * resolving it (by demotion) would lead to the first lock on the convert
 * queue being granted right away.  It allowed conversion deadlocks to exist
 * between locks on the convert queue while they couldn't be granted anyway.
 *
 * Now, we detect and take action on conversion deadlocks immediately when
 * they're created, even if they may not be immediately consequential.  If
 * lkb1 exists anywhere in the convert queue and lkb2 comes in with a granted
 * mode that would prevent lkb1's conversion from being granted, we do a
 * deadlk/demote on lkb2 right away and don't let it onto the convert queue.
 * I think this means that the lkb_is_ahead condition below should always
 * be zero, i.e. there will never be conv-deadlk between two locks that are
 * both already on the convert queue.
1452 1453
 */

1454
static int conversion_deadlock_detect(struct dlm_rsb *r, struct dlm_lkb *lkb2)
1455
{
1456 1457
	struct dlm_lkb *lkb1;
	int lkb_is_ahead = 0;
1458

1459 1460 1461
	list_for_each_entry(lkb1, &r->res_convertqueue, lkb_statequeue) {
		if (lkb1 == lkb2) {
			lkb_is_ahead = 1;
1462 1463 1464
			continue;
		}

1465 1466 1467 1468 1469 1470 1471 1472
		if (!lkb_is_ahead) {
			if (!modes_compat(lkb2, lkb1))
				return 1;
		} else {
			if (!modes_compat(lkb2, lkb1) &&
			    !modes_compat(lkb1, lkb2))
				return 1;
		}
1473
	}
1474
	return 0;
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513
}

/*
 * Return 1 if the lock can be granted, 0 otherwise.
 * Also detect and resolve conversion deadlocks.
 *
 * lkb is the lock to be granted
 *
 * now is 1 if the function is being called in the context of the
 * immediate request, it is 0 if called later, after the lock has been
 * queued.
 *
 * References are from chapter 6 of "VAXcluster Principles" by Roy Davis
 */

static int _can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now)
{
	int8_t conv = (lkb->lkb_grmode != DLM_LOCK_IV);

	/*
	 * 6-10: Version 5.4 introduced an option to address the phenomenon of
	 * a new request for a NL mode lock being blocked.
	 *
	 * 6-11: If the optional EXPEDITE flag is used with the new NL mode
	 * request, then it would be granted.  In essence, the use of this flag
	 * tells the Lock Manager to expedite theis request by not considering
	 * what may be in the CONVERTING or WAITING queues...  As of this
	 * writing, the EXPEDITE flag can be used only with new requests for NL
	 * mode locks.  This flag is not valid for conversion requests.
	 *
	 * A shortcut.  Earlier checks return an error if EXPEDITE is used in a
	 * conversion or used with a non-NL requested mode.  We also know an
	 * EXPEDITE request is always granted immediately, so now must always
	 * be 1.  The full condition to grant an expedite request: (now &&
	 * !conv && lkb->rqmode == DLM_LOCK_NL && (flags & EXPEDITE)) can
	 * therefore be shortened to just checking the flag.
	 */

	if (lkb->lkb_exflags & DLM_LKF_EXPEDITE)
1514
		return 1;
1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556

	/*
	 * A shortcut. Without this, !queue_conflict(grantqueue, lkb) would be
	 * added to the remaining conditions.
	 */

	if (queue_conflict(&r->res_grantqueue, lkb))
		goto out;

	/*
	 * 6-3: By default, a conversion request is immediately granted if the
	 * requested mode is compatible with the modes of all other granted
	 * locks
	 */

	if (queue_conflict(&r->res_convertqueue, lkb))
		goto out;

	/*
	 * 6-5: But the default algorithm for deciding whether to grant or
	 * queue conversion requests does not by itself guarantee that such
	 * requests are serviced on a "first come first serve" basis.  This, in
	 * turn, can lead to a phenomenon known as "indefinate postponement".
	 *
	 * 6-7: This issue is dealt with by using the optional QUECVT flag with
	 * the system service employed to request a lock conversion.  This flag
	 * forces certain conversion requests to be queued, even if they are
	 * compatible with the granted modes of other locks on the same
	 * resource.  Thus, the use of this flag results in conversion requests
	 * being ordered on a "first come first servce" basis.
	 *
	 * DCT: This condition is all about new conversions being able to occur
	 * "in place" while the lock remains on the granted queue (assuming
	 * nothing else conflicts.)  IOW if QUECVT isn't set, a conversion
	 * doesn't _have_ to go onto the convert queue where it's processed in
	 * order.  The "now" variable is necessary to distinguish converts
	 * being received and processed for the first time now, because once a
	 * convert is moved to the conversion queue the condition below applies
	 * requiring fifo granting.
	 */

	if (now && conv && !(lkb->lkb_exflags & DLM_LKF_QUECVT))
1557
		return 1;
1558 1559

	/*
1560 1561
	 * The NOORDER flag is set to avoid the standard vms rules on grant
	 * order.
1562 1563 1564
	 */

	if (lkb->lkb_exflags & DLM_LKF_NOORDER)
1565
		return 1;
1566 1567 1568 1569 1570 1571 1572 1573

	/*
	 * 6-3: Once in that queue [CONVERTING], a conversion request cannot be
	 * granted until all other conversion requests ahead of it are granted
	 * and/or canceled.
	 */

	if (!now && conv && first_in_list(lkb, &r->res_convertqueue))
1574
		return 1;
1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588

	/*
	 * 6-4: By default, a new request is immediately granted only if all
	 * three of the following conditions are satisfied when the request is
	 * issued:
	 * - The queue of ungranted conversion requests for the resource is
	 *   empty.
	 * - The queue of ungranted new requests for the resource is empty.
	 * - The mode of the new request is compatible with the most
	 *   restrictive mode of all granted locks on the resource.
	 */

	if (now && !conv && list_empty(&r->res_convertqueue) &&
	    list_empty(&r->res_waitqueue))
1589
		return 1;
1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600

	/*
	 * 6-4: Once a lock request is in the queue of ungranted new requests,
	 * it cannot be granted until the queue of ungranted conversion
	 * requests is empty, all ungranted new requests ahead of it are
	 * granted and/or canceled, and it is compatible with the granted mode
	 * of the most restrictive lock granted on the resource.
	 */

	if (!now && !conv && list_empty(&r->res_convertqueue) &&
	    first_in_list(lkb, &r->res_waitqueue))
1601
		return 1;
1602
 out:
1603
	return 0;
1604 1605
}

1606 1607
static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now,
			  int *err)
1608 1609 1610
{
	int rv;
	int8_t alt = 0, rqmode = lkb->lkb_rqmode;
1611 1612 1613 1614
	int8_t is_convert = (lkb->lkb_grmode != DLM_LOCK_IV);

	if (err)
		*err = 0;
1615 1616 1617 1618 1619

	rv = _can_be_granted(r, lkb, now);
	if (rv)
		goto out;

1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639
	/*
	 * The CONVDEADLK flag is non-standard and tells the dlm to resolve
	 * conversion deadlocks by demoting grmode to NL, otherwise the dlm
	 * cancels one of the locks.
	 */

	if (is_convert && can_be_queued(lkb) &&
	    conversion_deadlock_detect(r, lkb)) {
		if (lkb->lkb_exflags & DLM_LKF_CONVDEADLK) {
			lkb->lkb_grmode = DLM_LOCK_NL;
			lkb->lkb_sbflags |= DLM_SBF_DEMOTED;
		} else if (!(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
			if (err)
				*err = -EDEADLK;
			else {
				log_print("can_be_granted deadlock %x now %d",
					  lkb->lkb_id, now);
				dlm_dump_rsb(r);
			}
		}
1640
		goto out;
1641
	}
1642

1643 1644 1645 1646 1647 1648 1649 1650
	/*
	 * The ALTPR and ALTCW flags are non-standard and tell the dlm to try
	 * to grant a request in a mode other than the normal rqmode.  It's a
	 * simple way to provide a big optimization to applications that can
	 * use them.
	 */

	if (rqmode != DLM_LOCK_PR && (lkb->lkb_exflags & DLM_LKF_ALTPR))
1651
		alt = DLM_LOCK_PR;
1652
	else if (rqmode != DLM_LOCK_CW && (lkb->lkb_exflags & DLM_LKF_ALTCW))
1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666
		alt = DLM_LOCK_CW;

	if (alt) {
		lkb->lkb_rqmode = alt;
		rv = _can_be_granted(r, lkb, now);
		if (rv)
			lkb->lkb_sbflags |= DLM_SBF_ALTMODE;
		else
			lkb->lkb_rqmode = rqmode;
	}
 out:
	return rv;
}

1667 1668 1669 1670 1671 1672 1673
/* FIXME: I don't think that can_be_granted() can/will demote or find deadlock
   for locks pending on the convert list.  Once verified (watch for these
   log_prints), we should be able to just call _can_be_granted() and not
   bother with the demote/deadlk cases here (and there's no easy way to deal
   with a deadlk here, we'd have to generate something like grant_lock with
   the deadlk error.) */

1674 1675
/* Returns the highest requested mode of all blocked conversions; sets
   cw if there's a blocked conversion to DLM_LOCK_CW. */
1676

1677
static int grant_pending_convert(struct dlm_rsb *r, int high, int *cw)
1678 1679 1680
{
	struct dlm_lkb *lkb, *s;
	int hi, demoted, quit, grant_restart, demote_restart;
1681
	int deadlk;
1682 1683 1684 1685 1686 1687 1688 1689 1690

	quit = 0;
 restart:
	grant_restart = 0;
	demote_restart = 0;
	hi = DLM_LOCK_IV;

	list_for_each_entry_safe(lkb, s, &r->res_convertqueue, lkb_statequeue) {
		demoted = is_demoted(lkb);
1691 1692 1693
		deadlk = 0;

		if (can_be_granted(r, lkb, 0, &deadlk)) {
1694 1695
			grant_lock_pending(r, lkb);
			grant_restart = 1;
1696
			continue;
1697
		}
1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713

		if (!demoted && is_demoted(lkb)) {
			log_print("WARN: pending demoted %x node %d %s",
				  lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
			demote_restart = 1;
			continue;
		}

		if (deadlk) {
			log_print("WARN: pending deadlock %x node %d %s",
				  lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
			dlm_dump_rsb(r);
			continue;
		}

		hi = max_t(int, lkb->lkb_rqmode, hi);
1714 1715 1716

		if (cw && lkb->lkb_rqmode == DLM_LOCK_CW)
			*cw = 1;
1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728
	}

	if (grant_restart)
		goto restart;
	if (demote_restart && !quit) {
		quit = 1;
		goto restart;
	}

	return max_t(int, high, hi);
}

1729
static int grant_pending_wait(struct dlm_rsb *r, int high, int *cw)
1730 1731 1732 1733
{
	struct dlm_lkb *lkb, *s;

	list_for_each_entry_safe(lkb, s, &r->res_waitqueue, lkb_statequeue) {
1734
		if (can_be_granted(r, lkb, 0, NULL))
1735
			grant_lock_pending(r, lkb);
1736
                else {
1737
			high = max_t(int, lkb->lkb_rqmode, high);
1738 1739 1740
			if (lkb->lkb_rqmode == DLM_LOCK_CW)
				*cw = 1;
		}
1741 1742 1743 1744 1745
	}

	return high;
}

1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764
/* cw of 1 means there's a lock with a rqmode of DLM_LOCK_CW that's blocked
   on either the convert or waiting queue.
   high is the largest rqmode of all locks blocked on the convert or
   waiting queue. */

static int lock_requires_bast(struct dlm_lkb *gr, int high, int cw)
{
	if (gr->lkb_grmode == DLM_LOCK_PR && cw) {
		if (gr->lkb_highbast < DLM_LOCK_EX)
			return 1;
		return 0;
	}

	if (gr->lkb_highbast < high &&
	    !__dlm_compat_matrix[gr->lkb_grmode+1][high+1])
		return 1;
	return 0;
}

1765 1766 1767 1768
static void grant_pending_locks(struct dlm_rsb *r)
{
	struct dlm_lkb *lkb, *s;
	int high = DLM_LOCK_IV;
1769
	int cw = 0;
1770

1771
	DLM_ASSERT(is_master(r), dlm_dump_rsb(r););
1772

1773 1774
	high = grant_pending_convert(r, high, &cw);
	high = grant_pending_wait(r, high, &cw);
1775 1776 1777 1778 1779 1780 1781

	if (high == DLM_LOCK_IV)
		return;

	/*
	 * If there are locks left on the wait/convert queue then send blocking
	 * ASTs to granted locks based on the largest requested mode (high)
1782
	 * found above.
1783 1784 1785
	 */

	list_for_each_entry_safe(lkb, s, &r->res_grantqueue, lkb_statequeue) {
1786
		if (lkb->lkb_bastfn && lock_requires_bast(lkb, high, cw)) {
1787 1788
			if (cw && high == DLM_LOCK_PR &&
			    lkb->lkb_grmode == DLM_LOCK_PR)
1789 1790 1791
				queue_bast(r, lkb, DLM_LOCK_CW);
			else
				queue_bast(r, lkb, high);
1792 1793 1794 1795 1796
			lkb->lkb_highbast = high;
		}
	}
}

1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810
static int modes_require_bast(struct dlm_lkb *gr, struct dlm_lkb *rq)
{
	if ((gr->lkb_grmode == DLM_LOCK_PR && rq->lkb_rqmode == DLM_LOCK_CW) ||
	    (gr->lkb_grmode == DLM_LOCK_CW && rq->lkb_rqmode == DLM_LOCK_PR)) {
		if (gr->lkb_highbast < DLM_LOCK_EX)
			return 1;
		return 0;
	}

	if (gr->lkb_highbast < rq->lkb_rqmode && !modes_compat(gr, rq))
		return 1;
	return 0;
}

1811 1812 1813 1814 1815 1816
static void send_bast_queue(struct dlm_rsb *r, struct list_head *head,
			    struct dlm_lkb *lkb)
{
	struct dlm_lkb *gr;

	list_for_each_entry(gr, head, lkb_statequeue) {
1817
		if (gr->lkb_bastfn && modes_require_bast(gr, lkb)) {
1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
			queue_bast(r, gr, lkb->lkb_rqmode);
			gr->lkb_highbast = lkb->lkb_rqmode;
		}
	}
}

static void send_blocking_asts(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	send_bast_queue(r, &r->res_grantqueue, lkb);
}

static void send_blocking_asts_all(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	send_bast_queue(r, &r->res_grantqueue, lkb);
	send_bast_queue(r, &r->res_convertqueue, lkb);
}

/* set_master(r, lkb) -- set the master nodeid of a resource

   The purpose of this function is to set the nodeid field in the given
   lkb using the nodeid field in the given rsb.  If the rsb's nodeid is
   known, it can just be copied to the lkb and the function will return
   0.  If the rsb's nodeid is _not_ known, it needs to be looked up
   before it can be copied to the lkb.

   When the rsb nodeid is being looked up remotely, the initial lkb
   causing the lookup is kept on the ls_waiters list waiting for the
   lookup reply.  Other lkb's waiting for the same rsb lookup are kept
   on the rsb's res_lookup list until the master is verified.

   Return values:
   0: nodeid is set in rsb/lkb and the caller should go ahead and use it
   1: the rsb master is not available and the lkb has been placed on
      a wait queue
*/

static int set_master(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	struct dlm_ls *ls = r->res_ls;
David Teigland's avatar
David Teigland committed
1857
	int i, error, dir_nodeid, ret_nodeid, our_nodeid = dlm_our_nodeid();
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880

	if (rsb_flag(r, RSB_MASTER_UNCERTAIN)) {
		rsb_clear_flag(r, RSB_MASTER_UNCERTAIN);
		r->res_first_lkid = lkb->lkb_id;
		lkb->lkb_nodeid = r->res_nodeid;
		return 0;
	}

	if (r->res_first_lkid && r->res_first_lkid != lkb->lkb_id) {
		list_add_tail(&lkb->lkb_rsb_lookup, &r->res_lookup);
		return 1;
	}

	if (r->res_nodeid == 0) {
		lkb->lkb_nodeid = 0;
		return 0;
	}

	if (r->res_nodeid > 0) {
		lkb->lkb_nodeid = r->res_nodeid;
		return 0;
	}

1881
	DLM_ASSERT(r->res_nodeid == -1, dlm_dump_rsb(r););
1882 1883 1884 1885 1886 1887 1888 1889 1890

	dir_nodeid = dlm_dir_nodeid(r);

	if (dir_nodeid != our_nodeid) {
		r->res_first_lkid = lkb->lkb_id;
		send_lookup(r, lkb);
		return 1;
	}

David Teigland's avatar
David Teigland committed
1891
	for (i = 0; i < 2; i++) {
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904
		/* It's possible for dlm_scand to remove an old rsb for
		   this same resource from the toss list, us to create
		   a new one, look up the master locally, and find it
		   already exists just before dlm_scand does the
		   dir_remove() on the previous rsb. */

		error = dlm_dir_lookup(ls, our_nodeid, r->res_name,
				       r->res_length, &ret_nodeid);
		if (!error)
			break;
		log_debug(ls, "dir_lookup error %d %s", error, r->res_name);
		schedule();
	}
David Teigland's avatar
David Teigland committed
1905 1906
	if (error && error != -EEXIST)
		return error;
1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924

	if (ret_nodeid == our_nodeid) {
		r->res_first_lkid = 0;
		r->res_nodeid = 0;
		lkb->lkb_nodeid = 0;
	} else {
		r->res_first_lkid = lkb->lkb_id;
		r->res_nodeid = ret_nodeid;
		lkb->lkb_nodeid = ret_nodeid;
	}
	return 0;
}

static void process_lookup_list(struct dlm_rsb *r)
{
	struct dlm_lkb *lkb, *safe;

	list_for_each_entry_safe(lkb, safe, &r->res_lookup, lkb_rsb_lookup) {
1925
		list_del_init(&lkb->lkb_rsb_lookup);
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947
		_request_lock(r, lkb);
		schedule();
	}
}

/* confirm_master -- confirm (or deny) an rsb's master nodeid */

static void confirm_master(struct dlm_rsb *r, int error)
{
	struct dlm_lkb *lkb;

	if (!r->res_first_lkid)
		return;

	switch (error) {
	case 0:
	case -EINPROGRESS:
		r->res_first_lkid = 0;
		process_lookup_list(r);
		break;

	case -EAGAIN:
1948 1949 1950 1951 1952
	case -EBADR:
	case -ENOTBLK:
		/* the remote request failed and won't be retried (it was
		   a NOQUEUE, or has been canceled/unlocked); make a waiting
		   lkb the first_lkid */
1953 1954 1955 1956 1957 1958

		r->res_first_lkid = 0;

		if (!list_empty(&r->res_lookup)) {
			lkb = list_entry(r->res_lookup.next, struct dlm_lkb,
					 lkb_rsb_lookup);
1959
			list_del_init(&lkb->lkb_rsb_lookup);
1960 1961
			r->res_first_lkid = lkb->lkb_id;
			_request_lock(r, lkb);
1962
		}
1963 1964 1965 1966 1967 1968 1969 1970
		break;

	default:
		log_error(r->res_ls, "confirm_master unknown error %d", error);
	}
}

static int set_lock_args(int mode, struct dlm_lksb *lksb, uint32_t flags,
1971 1972 1973 1974 1975
			 int namelen, unsigned long timeout_cs,
			 void (*ast) (void *astparam),
			 void *astparam,
			 void (*bast) (void *astparam, int mode),
			 struct dlm_args *args)
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
{
	int rv = -EINVAL;

	/* check for invalid arg usage */

	if (mode < 0 || mode > DLM_LOCK_EX)
		goto out;

	if (!(flags & DLM_LKF_CONVERT) && (namelen > DLM_RESNAME_MAXLEN))
		goto out;

	if (flags & DLM_LKF_CANCEL)
		goto out;

	if (flags & DLM_LKF_QUECVT && !(flags & DLM_LKF_CONVERT))
		goto out;

	if (flags & DLM_LKF_CONVDEADLK && !(flags & DLM_LKF_CONVERT))
		goto out;

	if (flags & DLM_LKF_CONVDEADLK && flags & DLM_LKF_NOQUEUE)
		goto out;

	if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_CONVERT)
		goto out;

	if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_QUECVT)
		goto out;

	if (flags & DLM_LKF_EXPEDITE && flags & DLM_LKF_NOQUEUE)
		goto out;

	if (flags & DLM_LKF_EXPEDITE && mode != DLM_LOCK_NL)
		goto out;

	if (!ast || !lksb)
		goto out;

	if (flags & DLM_LKF_VALBLK && !lksb->sb_lvbptr)
		goto out;

	if (flags & DLM_LKF_CONVERT && !lksb->sb_lkid)
		goto out;

	/* these args will be copied to the lkb in validate_lock_args,
	   it cannot be done now because when converting locks, fields in
	   an active lkb cannot be modified before locking the rsb */

	args->flags = flags;
2025 2026 2027
	args->astfn = ast;
	args->astparam = astparam;
	args->bastfn = bast;
2028
	args->timeout = timeout_cs;
2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041
	args->mode = mode;
	args->lksb = lksb;
	rv = 0;
 out:
	return rv;
}

static int set_unlock_args(uint32_t flags, void *astarg, struct dlm_args *args)
{
	if (flags & ~(DLM_LKF_CANCEL | DLM_LKF_VALBLK | DLM_LKF_IVVALBLK |
 		      DLM_LKF_FORCEUNLOCK))
		return -EINVAL;

2042 2043 2044
	if (flags & DLM_LKF_CANCEL && flags & DLM_LKF_FORCEUNLOCK)
		return -EINVAL;

2045
	args->flags = flags;
2046
	args->astparam = astarg;
2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068
	return 0;
}

static int validate_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
			      struct dlm_args *args)
{
	int rv = -EINVAL;

	if (args->flags & DLM_LKF_CONVERT) {
		if (lkb->lkb_flags & DLM_IFL_MSTCPY)
			goto out;

		if (args->flags & DLM_LKF_QUECVT &&
		    !__quecvt_compat_matrix[lkb->lkb_grmode+1][args->mode+1])
			goto out;

		rv = -EBUSY;
		if (lkb->lkb_status != DLM_LKSTS_GRANTED)
			goto out;

		if (lkb->lkb_wait_type)
			goto out;
2069 2070 2071

		if (is_overlap(lkb))
			goto out;
2072 2073 2074 2075
	}

	lkb->lkb_exflags = args->flags;
	lkb->lkb_sbflags = 0;
2076
	lkb->lkb_astfn = args->astfn;
2077
	lkb->lkb_astparam = args->astparam;
2078
	lkb->lkb_bastfn = args->bastfn;
2079 2080 2081 2082
	lkb->lkb_rqmode = args->mode;
	lkb->lkb_lksb = args->lksb;
	lkb->lkb_lvbptr = args->lksb->sb_lvbptr;
	lkb->lkb_ownpid = (int) current->pid;
2083
	lkb->lkb_timeout_cs = args->timeout;
2084 2085 2086 2087 2088
	rv = 0;
 out:
	return rv;
}

2089 2090 2091 2092 2093 2094 2095
/* when dlm_unlock() sees -EBUSY with CANCEL/FORCEUNLOCK it returns 0
   for success */

/* note: it's valid for lkb_nodeid/res_nodeid to be -1 when we get here
   because there may be a lookup in progress and it's valid to do
   cancel/unlockf on it */

2096 2097
static int validate_unlock_args(struct dlm_lkb *lkb, struct dlm_args *args)
{
2098
	struct dlm_ls *ls = lkb->lkb_resource->res_ls;
2099 2100
	int rv = -EINVAL;

2101 2102 2103
	if (lkb->lkb_flags & DLM_IFL_MSTCPY) {
		log_error(ls, "unlock on MSTCPY %x", lkb->lkb_id);
		dlm_print_lkb(lkb);
2104
		goto out;
2105
	}
2106

2107 2108 2109
	/* an lkb may still exist even though the lock is EOL'ed due to a
	   cancel, unlock or failed noqueue request; an app can't use these
	   locks; return same error as if the lkid had not been found at all */
2110

2111 2112 2113
	if (lkb->lkb_flags & DLM_IFL_ENDOFLIFE) {
		log_debug(ls, "unlock on ENDOFLIFE %x", lkb->lkb_id);
		rv = -ENOENT;
2114
		goto out;
2115
	}
2116

2117 2118 2119
	/* an lkb may be waiting for an rsb lookup to complete where the
	   lookup was initiated by another lock */

2120 2121
	if (!list_empty(&lkb->lkb_rsb_lookup)) {
		if (args->flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)) {
2122 2123 2124 2125 2126 2127 2128
			log_debug(ls, "unlock on rsb_lookup %x", lkb->lkb_id);
			list_del_init(&lkb->lkb_rsb_lookup);
			queue_cast(lkb->lkb_resource, lkb,
				   args->flags & DLM_LKF_CANCEL ?
				   -DLM_ECANCEL : -DLM_EUNLOCK);
			unhold_lkb(lkb); /* undoes create_lkb() */
		}
2129 2130 2131
		/* caller changes -EBUSY to 0 for CANCEL and FORCEUNLOCK */
		rv = -EBUSY;
		goto out;
2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142
	}

	/* cancel not allowed with another cancel/unlock in progress */

	if (args->flags & DLM_LKF_CANCEL) {
		if (lkb->lkb_exflags & DLM_LKF_CANCEL)
			goto out;

		if (is_overlap(lkb))
			goto out;

2143 2144 2145
		/* don't let scand try to do a cancel */
		del_timeout(lkb);

2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175
		if (lkb->lkb_flags & DLM_IFL_RESEND) {
			lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
			rv = -EBUSY;
			goto out;
		}

		switch (lkb->lkb_wait_type) {
		case DLM_MSG_LOOKUP:
		case DLM_MSG_REQUEST:
			lkb->lkb_flags |= DLM_IFL_OVERLAP_CANCEL;
			rv = -EBUSY;
			goto out;
		case DLM_MSG_UNLOCK:
		case DLM_MSG_CANCEL:
			goto out;
		}
		/* add_to_waiters() will set OVERLAP_CANCEL */
		goto out_ok;
	}

	/* do we need to allow a force-unlock if there's a normal unlock
	   already in progress?  in what conditions could the normal unlock
	   fail such that we'd want to send a force-unlock to be sure? */

	if (args->flags & DLM_LKF_FORCEUNLOCK) {
		if (lkb->lkb_exflags & DLM_LKF_FORCEUNLOCK)
			goto out;

		if (is_overlap_unlock(lkb))
			goto out;
2176

2177 2178 2179
		/* don't let scand try to do a cancel */
		del_timeout(lkb);

2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199
		if (lkb->lkb_flags & DLM_IFL_RESEND) {
			lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
			rv = -EBUSY;
			goto out;
		}

		switch (lkb->lkb_wait_type) {
		case DLM_MSG_LOOKUP:
		case DLM_MSG_REQUEST:
			lkb->lkb_flags |= DLM_IFL_OVERLAP_UNLOCK;
			rv = -EBUSY;
			goto out;
		case DLM_MSG_UNLOCK:
			goto out;
		}
		/* add_to_waiters() will set OVERLAP_UNLOCK */
		goto out_ok;
	}

	/* normal unlock not allowed if there's any op in progress */
2200
	rv = -EBUSY;
2201
	if (lkb->lkb_wait_type || lkb->lkb_wait_count)
2202 2203 2204
		goto out;

 out_ok:
2205 2206
	/* an overlapping op shouldn't blow away exflags from other op */
	lkb->lkb_exflags |= args->flags;
2207 2208 2209 2210
	lkb->lkb_sbflags = 0;
	lkb->lkb_astparam = args->astparam;
	rv = 0;
 out:
2211 2212 2213 2214 2215
	if (rv)
		log_debug(ls, "validate_unlock_args %d %x %x %x %x %d %s", rv,
			  lkb->lkb_id, lkb->lkb_flags, lkb->lkb_exflags,
			  args->flags, lkb->lkb_wait_type,
			  lkb->lkb_resource->res_name);
2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229
	return rv;
}

/*
 * Four stage 4 varieties:
 * do_request(), do_convert(), do_unlock(), do_cancel()
 * These are called on the master node for the given lock and
 * from the central locking logic.
 */

static int do_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error = 0;

2230
	if (can_be_granted(r, lkb, 1, NULL)) {
2231 2232 2233 2234 2235 2236 2237 2238 2239
		grant_lock(r, lkb);
		queue_cast(r, lkb, 0);
		goto out;
	}

	if (can_be_queued(lkb)) {
		error = -EINPROGRESS;
		add_lkb(r, lkb, DLM_LKSTS_WAITING);
		send_blocking_asts(r, lkb);
2240
		add_timeout(lkb);
2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255
		goto out;
	}

	error = -EAGAIN;
	if (force_blocking_asts(lkb))
		send_blocking_asts_all(r, lkb);
	queue_cast(r, lkb, -EAGAIN);

 out:
	return error;
}

static int do_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error = 0;
2256
	int deadlk = 0;
2257 2258 2259

	/* changing an existing lock may allow others to be granted */

2260
	if (can_be_granted(r, lkb, 1, &deadlk)) {
2261 2262 2263 2264 2265 2266
		grant_lock(r, lkb);
		queue_cast(r, lkb, 0);
		grant_pending_locks(r);
		goto out;
	}

2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281
	/* can_be_granted() detected that this lock would block in a conversion
	   deadlock, so we leave it on the granted queue and return EDEADLK in
	   the ast for the convert. */

	if (deadlk) {
		/* it's left on the granted queue */
		log_debug(r->res_ls, "deadlock %x node %d sts%d g%d r%d %s",
			  lkb->lkb_id, lkb->lkb_nodeid, lkb->lkb_status,
			  lkb->lkb_grmode, lkb->lkb_rqmode, r->res_name);
		revert_lock(r, lkb);
		queue_cast(r, lkb, -EDEADLK);
		error = -EDEADLK;
		goto out;
	}

David Teigland's avatar
David Teigland committed
2282 2283 2284 2285 2286 2287 2288
	/* is_demoted() means the can_be_granted() above set the grmode
	   to NL, and left us on the granted queue.  This auto-demotion
	   (due to CONVDEADLK) might mean other locks, and/or this lock, are
	   now grantable.  We have to try to grant other converting locks
	   before we try again to grant this one. */

	if (is_demoted(lkb)) {
2289
		grant_pending_convert(r, DLM_LOCK_IV, NULL);
David Teigland's avatar
David Teigland committed
2290 2291 2292
		if (_can_be_granted(r, lkb, 1)) {
			grant_lock(r, lkb);
			queue_cast(r, lkb, 0);
2293
			grant_pending_locks(r);
David Teigland's avatar
David Teigland committed
2294 2295 2296 2297 2298 2299
			goto out;
		}
		/* else fall through and move to convert queue */
	}

	if (can_be_queued(lkb)) {
2300 2301 2302 2303
		error = -EINPROGRESS;
		del_lkb(r, lkb);
		add_lkb(r, lkb, DLM_LKSTS_CONVERT);
		send_blocking_asts(r, lkb);
2304
		add_timeout(lkb);
2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324
		goto out;
	}

	error = -EAGAIN;
	if (force_blocking_asts(lkb))
		send_blocking_asts_all(r, lkb);
	queue_cast(r, lkb, -EAGAIN);

 out:
	return error;
}

static int do_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	remove_lock(r, lkb);
	queue_cast(r, lkb, -DLM_EUNLOCK);
	grant_pending_locks(r);
	return -DLM_EUNLOCK;
}

2325
/* returns: 0 did nothing, -DLM_ECANCEL canceled lock */
2326
 
2327 2328
static int do_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
2329 2330 2331 2332 2333 2334 2335 2336 2337
	int error;

	error = revert_lock(r, lkb);
	if (error) {
		queue_cast(r, lkb, -DLM_ECANCEL);
		grant_pending_locks(r);
		return -DLM_ECANCEL;
	}
	return 0;
2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369
}

/*
 * Four stage 3 varieties:
 * _request_lock(), _convert_lock(), _unlock_lock(), _cancel_lock()
 */

/* add a new lkb to a possibly new rsb, called by requesting process */

static int _request_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error;

	/* set_master: sets lkb nodeid from r */

	error = set_master(r, lkb);
	if (error < 0)
		goto out;
	if (error) {
		error = 0;
		goto out;
	}

	if (is_remote(r))
		/* receive_request() calls do_request() on remote node */
		error = send_request(r, lkb);
	else
		error = do_request(r, lkb);
 out:
	return error;
}

2370
/* change some property of an existing lkb, e.g. mode */
2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526

static int _convert_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error;

	if (is_remote(r))
		/* receive_convert() calls do_convert() on remote node */
		error = send_convert(r, lkb);
	else
		error = do_convert(r, lkb);

	return error;
}

/* remove an existing lkb from the granted queue */

static int _unlock_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error;

	if (is_remote(r))
		/* receive_unlock() calls do_unlock() on remote node */
		error = send_unlock(r, lkb);
	else
		error = do_unlock(r, lkb);

	return error;
}

/* remove an existing lkb from the convert or wait queue */

static int _cancel_lock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error;

	if (is_remote(r))
		/* receive_cancel() calls do_cancel() on remote node */
		error = send_cancel(r, lkb);
	else
		error = do_cancel(r, lkb);

	return error;
}

/*
 * Four stage 2 varieties:
 * request_lock(), convert_lock(), unlock_lock(), cancel_lock()
 */

static int request_lock(struct dlm_ls *ls, struct dlm_lkb *lkb, char *name,
			int len, struct dlm_args *args)
{
	struct dlm_rsb *r;
	int error;

	error = validate_lock_args(ls, lkb, args);
	if (error)
		goto out;

	error = find_rsb(ls, name, len, R_CREATE, &r);
	if (error)
		goto out;

	lock_rsb(r);

	attach_lkb(r, lkb);
	lkb->lkb_lksb->sb_lkid = lkb->lkb_id;

	error = _request_lock(r, lkb);

	unlock_rsb(r);
	put_rsb(r);

 out:
	return error;
}

static int convert_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
			struct dlm_args *args)
{
	struct dlm_rsb *r;
	int error;

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

	error = validate_lock_args(ls, lkb, args);
	if (error)
		goto out;

	error = _convert_lock(r, lkb);
 out:
	unlock_rsb(r);
	put_rsb(r);
	return error;
}

static int unlock_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
		       struct dlm_args *args)
{
	struct dlm_rsb *r;
	int error;

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

	error = validate_unlock_args(lkb, args);
	if (error)
		goto out;

	error = _unlock_lock(r, lkb);
 out:
	unlock_rsb(r);
	put_rsb(r);
	return error;
}

static int cancel_lock(struct dlm_ls *ls, struct dlm_lkb *lkb,
		       struct dlm_args *args)
{
	struct dlm_rsb *r;
	int error;

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

	error = validate_unlock_args(lkb, args);
	if (error)
		goto out;

	error = _cancel_lock(r, lkb);
 out:
	unlock_rsb(r);
	put_rsb(r);
	return error;
}

/*
 * Two stage 1 varieties:  dlm_lock() and dlm_unlock()
 */

int dlm_lock(dlm_lockspace_t *lockspace,
	     int mode,
	     struct dlm_lksb *lksb,
	     uint32_t flags,
	     void *name,
	     unsigned int namelen,
	     uint32_t parent_lkid,
	     void (*ast) (void *astarg),
	     void *astarg,
2527
	     void (*bast) (void *astarg, int mode))
2528 2529 2530 2531 2532 2533 2534 2535 2536 2537
{
	struct dlm_ls *ls;
	struct dlm_lkb *lkb;
	struct dlm_args args;
	int error, convert = flags & DLM_LKF_CONVERT;

	ls = dlm_find_lockspace_local(lockspace);
	if (!ls)
		return -EINVAL;

2538
	dlm_lock_recovery(ls);
2539 2540 2541 2542 2543 2544 2545 2546 2547

	if (convert)
		error = find_lkb(ls, lksb->sb_lkid, &lkb);
	else
		error = create_lkb(ls, &lkb);

	if (error)
		goto out;

2548
	error = set_lock_args(mode, lksb, flags, namelen, 0, ast,
2549
			      astarg, bast, &args);
2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561
	if (error)
		goto out_put;

	if (convert)
		error = convert_lock(ls, lkb, &args);
	else
		error = request_lock(ls, lkb, name, namelen, &args);

	if (error == -EINPROGRESS)
		error = 0;
 out_put:
	if (convert || error)
2562
		__put_lkb(ls, lkb);
2563
	if (error == -EAGAIN || error == -EDEADLK)
2564 2565
		error = 0;
 out:
2566
	dlm_unlock_recovery(ls);
2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585
	dlm_put_lockspace(ls);
	return error;
}

int dlm_unlock(dlm_lockspace_t *lockspace,
	       uint32_t lkid,
	       uint32_t flags,
	       struct dlm_lksb *lksb,
	       void *astarg)
{
	struct dlm_ls *ls;
	struct dlm_lkb *lkb;
	struct dlm_args args;
	int error;

	ls = dlm_find_lockspace_local(lockspace);
	if (!ls)
		return -EINVAL;

2586
	dlm_lock_recovery(ls);
2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602

	error = find_lkb(ls, lkid, &lkb);
	if (error)
		goto out;

	error = set_unlock_args(flags, astarg, &args);
	if (error)
		goto out_put;

	if (flags & DLM_LKF_CANCEL)
		error = cancel_lock(ls, lkb, &args);
	else
		error = unlock_lock(ls, lkb, &args);

	if (error == -DLM_EUNLOCK || error == -DLM_ECANCEL)
		error = 0;
2603 2604
	if (error == -EBUSY && (flags & (DLM_LKF_CANCEL | DLM_LKF_FORCEUNLOCK)))
		error = 0;
2605
 out_put:
2606
	dlm_put_lkb(lkb);
2607
 out:
2608
	dlm_unlock_recovery(ls);
2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634
	dlm_put_lockspace(ls);
	return error;
}

/*
 * send/receive routines for remote operations and replies
 *
 * send_args
 * send_common
 * send_request			receive_request
 * send_convert			receive_convert
 * send_unlock			receive_unlock
 * send_cancel			receive_cancel
 * send_grant			receive_grant
 * send_bast			receive_bast
 * send_lookup			receive_lookup
 * send_remove			receive_remove
 *
 * 				send_common_reply
 * receive_request_reply	send_request_reply
 * receive_convert_reply	send_convert_reply
 * receive_unlock_reply		send_unlock_reply
 * receive_cancel_reply		send_cancel_reply
 * receive_lookup_reply		send_lookup_reply
 */

2635 2636 2637 2638
static int _create_message(struct dlm_ls *ls, int mb_len,
			   int to_nodeid, int mstype,
			   struct dlm_message **ms_ret,
			   struct dlm_mhandle **mh_ret)
2639 2640 2641 2642 2643 2644 2645 2646 2647
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	char *mb;

	/* get_buffer gives us a message handle (mh) that we need to
	   pass into lowcomms_commit and a message buffer (mb) that we
	   write our data into */

2648
	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, ls->ls_allocation, &mb);
2649 2650 2651 2652 2653 2654 2655 2656
	if (!mh)
		return -ENOBUFS;

	memset(mb, 0, mb_len);

	ms = (struct dlm_message *) mb;

	ms->m_header.h_version = (DLM_HEADER_MAJOR | DLM_HEADER_MINOR);
2657
	ms->m_header.h_lockspace = ls->ls_global_id;
2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668
	ms->m_header.h_nodeid = dlm_our_nodeid();
	ms->m_header.h_length = mb_len;
	ms->m_header.h_cmd = DLM_MSG;

	ms->m_type = mstype;

	*mh_ret = mh;
	*ms_ret = ms;
	return 0;
}

2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695
static int create_message(struct dlm_rsb *r, struct dlm_lkb *lkb,
			  int to_nodeid, int mstype,
			  struct dlm_message **ms_ret,
			  struct dlm_mhandle **mh_ret)
{
	int mb_len = sizeof(struct dlm_message);

	switch (mstype) {
	case DLM_MSG_REQUEST:
	case DLM_MSG_LOOKUP:
	case DLM_MSG_REMOVE:
		mb_len += r->res_length;
		break;
	case DLM_MSG_CONVERT:
	case DLM_MSG_UNLOCK:
	case DLM_MSG_REQUEST_REPLY:
	case DLM_MSG_CONVERT_REPLY:
	case DLM_MSG_GRANT:
		if (lkb && lkb->lkb_lvbptr)
			mb_len += r->res_ls->ls_lvblen;
		break;
	}

	return _create_message(r->res_ls, mb_len, to_nodeid, mstype,
			       ms_ret, mh_ret);
}

2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724
/* further lowcomms enhancements or alternate implementations may make
   the return value from this function useful at some point */

static int send_message(struct dlm_mhandle *mh, struct dlm_message *ms)
{
	dlm_message_out(ms);
	dlm_lowcomms_commit_buffer(mh);
	return 0;
}

static void send_args(struct dlm_rsb *r, struct dlm_lkb *lkb,
		      struct dlm_message *ms)
{
	ms->m_nodeid   = lkb->lkb_nodeid;
	ms->m_pid      = lkb->lkb_ownpid;
	ms->m_lkid     = lkb->lkb_id;
	ms->m_remid    = lkb->lkb_remid;
	ms->m_exflags  = lkb->lkb_exflags;
	ms->m_sbflags  = lkb->lkb_sbflags;
	ms->m_flags    = lkb->lkb_flags;
	ms->m_lvbseq   = lkb->lkb_lvbseq;
	ms->m_status   = lkb->lkb_status;
	ms->m_grmode   = lkb->lkb_grmode;
	ms->m_rqmode   = lkb->lkb_rqmode;
	ms->m_hash     = r->res_hash;

	/* m_result and m_bastmode are set from function args,
	   not from lkb fields */

2725
	if (lkb->lkb_bastfn)
2726
		ms->m_asts |= AST_BAST;
2727
	if (lkb->lkb_astfn)
2728 2729
		ms->m_asts |= AST_COMP;

2730 2731
	/* compare with switch in create_message; send_remove() doesn't
	   use send_args() */
2732

2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744
	switch (ms->m_type) {
	case DLM_MSG_REQUEST:
	case DLM_MSG_LOOKUP:
		memcpy(ms->m_extra, r->res_name, r->res_length);
		break;
	case DLM_MSG_CONVERT:
	case DLM_MSG_UNLOCK:
	case DLM_MSG_REQUEST_REPLY:
	case DLM_MSG_CONVERT_REPLY:
	case DLM_MSG_GRANT:
		if (!lkb->lkb_lvbptr)
			break;
2745
		memcpy(ms->m_extra, lkb->lkb_lvbptr, r->res_ls->ls_lvblen);
2746 2747
		break;
	}
2748 2749 2750 2751 2752 2753 2754 2755
}

static int send_common(struct dlm_rsb *r, struct dlm_lkb *lkb, int mstype)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

2756 2757 2758
	error = add_to_waiters(lkb, mstype);
	if (error)
		return error;
2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773

	to_nodeid = r->res_nodeid;

	error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh);
	if (error)
		goto fail;

	send_args(r, lkb, ms);

	error = send_message(mh, ms);
	if (error)
		goto fail;
	return 0;

 fail:
2774
	remove_from_waiters(lkb, msg_reply_type(mstype));
2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790
	return error;
}

static int send_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	return send_common(r, lkb, DLM_MSG_REQUEST);
}

static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	int error;

	error = send_common(r, lkb, DLM_MSG_CONVERT);

	/* down conversions go without a reply from the master */
	if (!error && down_conversion(lkb)) {
2791 2792
		remove_from_waiters(lkb, DLM_MSG_CONVERT_REPLY);
		r->res_ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY;
2793
		r->res_ls->ls_stub_ms.m_result = 0;
2794
		r->res_ls->ls_stub_ms.m_flags = lkb->lkb_flags;
2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862
		__receive_convert_reply(r, lkb, &r->res_ls->ls_stub_ms);
	}

	return error;
}

/* FIXME: if this lkb is the only lock we hold on the rsb, then set
   MASTER_UNCERTAIN to force the next request on the rsb to confirm
   that the master is still correct. */

static int send_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	return send_common(r, lkb, DLM_MSG_UNLOCK);
}

static int send_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	return send_common(r, lkb, DLM_MSG_CANCEL);
}

static int send_grant(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

	to_nodeid = lkb->lkb_nodeid;

	error = create_message(r, lkb, to_nodeid, DLM_MSG_GRANT, &ms, &mh);
	if (error)
		goto out;

	send_args(r, lkb, ms);

	ms->m_result = 0;

	error = send_message(mh, ms);
 out:
	return error;
}

static int send_bast(struct dlm_rsb *r, struct dlm_lkb *lkb, int mode)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

	to_nodeid = lkb->lkb_nodeid;

	error = create_message(r, NULL, to_nodeid, DLM_MSG_BAST, &ms, &mh);
	if (error)
		goto out;

	send_args(r, lkb, ms);

	ms->m_bastmode = mode;

	error = send_message(mh, ms);
 out:
	return error;
}

static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

2863 2864 2865
	error = add_to_waiters(lkb, DLM_MSG_LOOKUP);
	if (error)
		return error;
2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880

	to_nodeid = dlm_dir_nodeid(r);

	error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh);
	if (error)
		goto fail;

	send_args(r, lkb, ms);

	error = send_message(mh, ms);
	if (error)
		goto fail;
	return 0;

 fail:
2881
	remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974
	return error;
}

static int send_remove(struct dlm_rsb *r)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

	to_nodeid = dlm_dir_nodeid(r);

	error = create_message(r, NULL, to_nodeid, DLM_MSG_REMOVE, &ms, &mh);
	if (error)
		goto out;

	memcpy(ms->m_extra, r->res_name, r->res_length);
	ms->m_hash = r->res_hash;

	error = send_message(mh, ms);
 out:
	return error;
}

static int send_common_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
			     int mstype, int rv)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int to_nodeid, error;

	to_nodeid = lkb->lkb_nodeid;

	error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh);
	if (error)
		goto out;

	send_args(r, lkb, ms);

	ms->m_result = rv;

	error = send_message(mh, ms);
 out:
	return error;
}

static int send_request_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
	return send_common_reply(r, lkb, DLM_MSG_REQUEST_REPLY, rv);
}

static int send_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
	return send_common_reply(r, lkb, DLM_MSG_CONVERT_REPLY, rv);
}

static int send_unlock_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
	return send_common_reply(r, lkb, DLM_MSG_UNLOCK_REPLY, rv);
}

static int send_cancel_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, int rv)
{
	return send_common_reply(r, lkb, DLM_MSG_CANCEL_REPLY, rv);
}

static int send_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms_in,
			     int ret_nodeid, int rv)
{
	struct dlm_rsb *r = &ls->ls_stub_rsb;
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int error, nodeid = ms_in->m_header.h_nodeid;

	error = create_message(r, NULL, nodeid, DLM_MSG_LOOKUP_REPLY, &ms, &mh);
	if (error)
		goto out;

	ms->m_lkid = ms_in->m_lkid;
	ms->m_result = rv;
	ms->m_nodeid = ret_nodeid;

	error = send_message(mh, ms);
 out:
	return error;
}

/* which args we save from a received message depends heavily on the type
   of message, unlike the send side where we can safely send everything about
   the lkb for any type of message */

static void receive_flags(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	lkb->lkb_exflags = ms->m_exflags;
2975
	lkb->lkb_sbflags = ms->m_sbflags;
2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998
	lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) |
		         (ms->m_flags & 0x0000FFFF);
}

static void receive_flags_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	lkb->lkb_sbflags = ms->m_sbflags;
	lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) |
		         (ms->m_flags & 0x0000FFFF);
}

static int receive_extralen(struct dlm_message *ms)
{
	return (ms->m_header.h_length - sizeof(struct dlm_message));
}

static int receive_lvb(struct dlm_ls *ls, struct dlm_lkb *lkb,
		       struct dlm_message *ms)
{
	int len;

	if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
		if (!lkb->lkb_lvbptr)
2999
			lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3000 3001 3002
		if (!lkb->lkb_lvbptr)
			return -ENOMEM;
		len = receive_extralen(ms);
3003 3004
		if (len > DLM_RESNAME_MAXLEN)
			len = DLM_RESNAME_MAXLEN;
3005 3006 3007 3008 3009
		memcpy(lkb->lkb_lvbptr, ms->m_extra, len);
	}
	return 0;
}

3010 3011 3012 3013 3014 3015 3016 3017 3018 3019
static void fake_bastfn(void *astparam, int mode)
{
	log_print("fake_bastfn should not be called");
}

static void fake_astfn(void *astparam)
{
	log_print("fake_astfn should not be called");
}

3020 3021 3022 3023 3024 3025 3026 3027
static int receive_request_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
				struct dlm_message *ms)
{
	lkb->lkb_nodeid = ms->m_header.h_nodeid;
	lkb->lkb_ownpid = ms->m_pid;
	lkb->lkb_remid = ms->m_lkid;
	lkb->lkb_grmode = DLM_LOCK_IV;
	lkb->lkb_rqmode = ms->m_rqmode;
3028 3029 3030

	lkb->lkb_bastfn = (ms->m_asts & AST_BAST) ? &fake_bastfn : NULL;
	lkb->lkb_astfn = (ms->m_asts & AST_COMP) ? &fake_astfn : NULL;
3031

3032 3033
	if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
		/* lkb was just created so there won't be an lvb yet */
3034
		lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
3035 3036 3037
		if (!lkb->lkb_lvbptr)
			return -ENOMEM;
	}
3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074

	return 0;
}

static int receive_convert_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
				struct dlm_message *ms)
{
	if (lkb->lkb_status != DLM_LKSTS_GRANTED)
		return -EBUSY;

	if (receive_lvb(ls, lkb, ms))
		return -ENOMEM;

	lkb->lkb_rqmode = ms->m_rqmode;
	lkb->lkb_lvbseq = ms->m_lvbseq;

	return 0;
}

static int receive_unlock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
			       struct dlm_message *ms)
{
	if (receive_lvb(ls, lkb, ms))
		return -ENOMEM;
	return 0;
}

/* We fill in the stub-lkb fields with the info that send_xxxx_reply()
   uses to send a reply and that the remote end uses to process the reply. */

static void setup_stub_lkb(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb = &ls->ls_stub_lkb;
	lkb->lkb_nodeid = ms->m_header.h_nodeid;
	lkb->lkb_remid = ms->m_lkid;
}

3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118
/* This is called after the rsb is locked so that we can safely inspect
   fields in the lkb. */

static int validate_message(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	int from = ms->m_header.h_nodeid;
	int error = 0;

	switch (ms->m_type) {
	case DLM_MSG_CONVERT:
	case DLM_MSG_UNLOCK:
	case DLM_MSG_CANCEL:
		if (!is_master_copy(lkb) || lkb->lkb_nodeid != from)
			error = -EINVAL;
		break;

	case DLM_MSG_CONVERT_REPLY:
	case DLM_MSG_UNLOCK_REPLY:
	case DLM_MSG_CANCEL_REPLY:
	case DLM_MSG_GRANT:
	case DLM_MSG_BAST:
		if (!is_process_copy(lkb) || lkb->lkb_nodeid != from)
			error = -EINVAL;
		break;

	case DLM_MSG_REQUEST_REPLY:
		if (!is_process_copy(lkb))
			error = -EINVAL;
		else if (lkb->lkb_nodeid != -1 && lkb->lkb_nodeid != from)
			error = -EINVAL;
		break;

	default:
		error = -EINVAL;
	}

	if (error)
		log_error(lkb->lkb_resource->res_ls,
			  "ignore invalid message %d from %d %x %x %x %d",
			  ms->m_type, from, lkb->lkb_id, lkb->lkb_remid,
			  lkb->lkb_flags, lkb->lkb_nodeid);
	return error;
}

3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132
static void receive_request(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error, namelen;

	error = create_lkb(ls, &lkb);
	if (error)
		goto fail;

	receive_flags(lkb, ms);
	lkb->lkb_flags |= DLM_IFL_MSTCPY;
	error = receive_request_args(ls, lkb, ms);
	if (error) {
3133
		__put_lkb(ls, lkb);
3134 3135 3136 3137 3138 3139 3140
		goto fail;
	}

	namelen = receive_extralen(ms);

	error = find_rsb(ls, ms->m_extra, namelen, R_MASTER, &r);
	if (error) {
3141
		__put_lkb(ls, lkb);
3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156
		goto fail;
	}

	lock_rsb(r);

	attach_lkb(r, lkb);
	error = do_request(r, lkb);
	send_request_reply(r, lkb, error);

	unlock_rsb(r);
	put_rsb(r);

	if (error == -EINPROGRESS)
		error = 0;
	if (error)
3157
		dlm_put_lkb(lkb);
3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168
	return;

 fail:
	setup_stub_lkb(ls, ms);
	send_request_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
}

static void receive_convert(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
3169
	int error, reply = 1;
3170 3171 3172 3173 3174 3175 3176 3177 3178 3179

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error)
		goto fail;

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

3180 3181 3182 3183
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3184 3185 3186
	receive_flags(lkb, ms);
	error = receive_convert_args(ls, lkb, ms);
	if (error)
3187
		goto out_reply;
3188 3189 3190
	reply = !down_conversion(lkb);

	error = do_convert(r, lkb);
3191
 out_reply:
3192 3193
	if (reply)
		send_convert_reply(r, lkb, error);
3194
 out:
3195 3196
	unlock_rsb(r);
	put_rsb(r);
3197
	dlm_put_lkb(lkb);
3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219
	return;

 fail:
	setup_stub_lkb(ls, ms);
	send_convert_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
}

static void receive_unlock(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error)
		goto fail;

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

3220 3221 3222 3223
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3224 3225 3226
	receive_flags(lkb, ms);
	error = receive_unlock_args(ls, lkb, ms);
	if (error)
3227
		goto out_reply;
3228 3229

	error = do_unlock(r, lkb);
3230
 out_reply:
3231
	send_unlock_reply(r, lkb, error);
3232
 out:
3233 3234
	unlock_rsb(r);
	put_rsb(r);
3235
	dlm_put_lkb(lkb);
3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259
	return;

 fail:
	setup_stub_lkb(ls, ms);
	send_unlock_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
}

static void receive_cancel(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error)
		goto fail;

	receive_flags(lkb, ms);

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

3260 3261 3262 3263
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3264 3265
	error = do_cancel(r, lkb);
	send_cancel_reply(r, lkb, error);
3266
 out:
3267 3268
	unlock_rsb(r);
	put_rsb(r);
3269
	dlm_put_lkb(lkb);
3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284
	return;

 fail:
	setup_stub_lkb(ls, ms);
	send_cancel_reply(&ls->ls_stub_rsb, &ls->ls_stub_lkb, error);
}

static void receive_grant(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3285 3286
		log_debug(ls, "receive_grant from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3287 3288 3289 3290 3291 3292 3293 3294
		return;
	}

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

3295 3296 3297 3298
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3299
	receive_flags_reply(lkb, ms);
David Teigland's avatar
David Teigland committed
3300 3301
	if (is_altmode(lkb))
		munge_altmode(lkb, ms);
3302 3303
	grant_lock_pc(r, lkb, ms);
	queue_cast(r, lkb, 0);
3304
 out:
3305 3306
	unlock_rsb(r);
	put_rsb(r);
3307
	dlm_put_lkb(lkb);
3308 3309 3310 3311 3312 3313 3314 3315 3316 3317
}

static void receive_bast(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3318 3319
		log_debug(ls, "receive_bast from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3320 3321 3322 3323 3324 3325 3326 3327
		return;
	}

	r = lkb->lkb_resource;

	hold_rsb(r);
	lock_rsb(r);

3328 3329 3330
	error = validate_message(lkb, ms);
	if (error)
		goto out;
3331

3332 3333
	queue_bast(r, lkb, ms->m_bastmode);
 out:
3334 3335
	unlock_rsb(r);
	put_rsb(r);
3336
	dlm_put_lkb(lkb);
3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385
}

static void receive_lookup(struct dlm_ls *ls, struct dlm_message *ms)
{
	int len, error, ret_nodeid, dir_nodeid, from_nodeid, our_nodeid;

	from_nodeid = ms->m_header.h_nodeid;
	our_nodeid = dlm_our_nodeid();

	len = receive_extralen(ms);

	dir_nodeid = dlm_hash2nodeid(ls, ms->m_hash);
	if (dir_nodeid != our_nodeid) {
		log_error(ls, "lookup dir_nodeid %d from %d",
			  dir_nodeid, from_nodeid);
		error = -EINVAL;
		ret_nodeid = -1;
		goto out;
	}

	error = dlm_dir_lookup(ls, from_nodeid, ms->m_extra, len, &ret_nodeid);

	/* Optimization: we're master so treat lookup as a request */
	if (!error && ret_nodeid == our_nodeid) {
		receive_request(ls, ms);
		return;
	}
 out:
	send_lookup_reply(ls, ms, ret_nodeid, error);
}

static void receive_remove(struct dlm_ls *ls, struct dlm_message *ms)
{
	int len, dir_nodeid, from_nodeid;

	from_nodeid = ms->m_header.h_nodeid;

	len = receive_extralen(ms);

	dir_nodeid = dlm_hash2nodeid(ls, ms->m_hash);
	if (dir_nodeid != dlm_our_nodeid()) {
		log_error(ls, "remove dir entry dir_nodeid %d from %d",
			  dir_nodeid, from_nodeid);
		return;
	}

	dlm_dir_remove_entry(ls, from_nodeid, ms->m_extra, len);
}

3386 3387 3388 3389 3390
static void receive_purge(struct dlm_ls *ls, struct dlm_message *ms)
{
	do_purge(ls, ms->m_nodeid, ms->m_pid);
}

3391 3392 3393 3394
static void receive_request_reply(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
3395
	int error, mstype, result;
3396 3397 3398

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3399 3400
		log_debug(ls, "receive_request_reply from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3401 3402 3403 3404 3405 3406 3407
		return;
	}

	r = lkb->lkb_resource;
	hold_rsb(r);
	lock_rsb(r);

3408 3409 3410 3411
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3412 3413 3414 3415 3416
	mstype = lkb->lkb_wait_type;
	error = remove_from_waiters(lkb, DLM_MSG_REQUEST_REPLY);
	if (error)
		goto out;

3417 3418 3419 3420 3421 3422 3423
	/* Optimization: the dir node was also the master, so it took our
	   lookup as a request and sent request reply instead of lookup reply */
	if (mstype == DLM_MSG_LOOKUP) {
		r->res_nodeid = ms->m_header.h_nodeid;
		lkb->lkb_nodeid = r->res_nodeid;
	}

3424 3425 3426 3427
	/* this is the value returned from do_request() on the master */
	result = ms->m_result;

	switch (result) {
3428
	case -EAGAIN:
3429
		/* request would block (be queued) on remote master */
3430 3431
		queue_cast(r, lkb, -EAGAIN);
		confirm_master(r, -EAGAIN);
3432
		unhold_lkb(lkb); /* undoes create_lkb() */
3433 3434 3435 3436 3437 3438 3439
		break;

	case -EINPROGRESS:
	case 0:
		/* request was queued or granted on remote master */
		receive_flags_reply(lkb, ms);
		lkb->lkb_remid = ms->m_lkid;
David Teigland's avatar
David Teigland committed
3440 3441
		if (is_altmode(lkb))
			munge_altmode(lkb, ms);
3442
		if (result) {
3443
			add_lkb(r, lkb, DLM_LKSTS_WAITING);
3444 3445
			add_timeout(lkb);
		} else {
3446 3447 3448
			grant_lock_pc(r, lkb, ms);
			queue_cast(r, lkb, 0);
		}
3449
		confirm_master(r, result);
3450 3451
		break;

David Teigland's avatar
David Teigland committed
3452
	case -EBADR:
3453 3454
	case -ENOTBLK:
		/* find_rsb failed to find rsb or rsb wasn't master */
3455 3456
		log_debug(ls, "receive_request_reply %x %x master diff %d %d",
			  lkb->lkb_id, lkb->lkb_flags, r->res_nodeid, result);
3457 3458
		r->res_nodeid = -1;
		lkb->lkb_nodeid = -1;
3459 3460 3461 3462

		if (is_overlap(lkb)) {
			/* we'll ignore error in cancel/unlock reply */
			queue_cast_overlap(r, lkb);
3463
			confirm_master(r, result);
3464 3465 3466
			unhold_lkb(lkb); /* undoes create_lkb() */
		} else
			_request_lock(r, lkb);
3467 3468 3469
		break;

	default:
3470 3471
		log_error(ls, "receive_request_reply %x error %d",
			  lkb->lkb_id, result);
3472 3473
	}

3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489
	if (is_overlap_unlock(lkb) && (result == 0 || result == -EINPROGRESS)) {
		log_debug(ls, "receive_request_reply %x result %d unlock",
			  lkb->lkb_id, result);
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
		send_unlock(r, lkb);
	} else if (is_overlap_cancel(lkb) && (result == -EINPROGRESS)) {
		log_debug(ls, "receive_request_reply %x cancel", lkb->lkb_id);
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
		send_cancel(r, lkb);
	} else {
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
	}
 out:
3490 3491
	unlock_rsb(r);
	put_rsb(r);
3492
	dlm_put_lkb(lkb);
3493 3494 3495 3496 3497 3498
}

static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
				    struct dlm_message *ms)
{
	/* this is the value returned from do_convert() on the master */
3499
	switch (ms->m_result) {
3500 3501 3502 3503 3504
	case -EAGAIN:
		/* convert would block (be queued) on remote master */
		queue_cast(r, lkb, -EAGAIN);
		break;

3505 3506 3507 3508 3509 3510
	case -EDEADLK:
		receive_flags_reply(lkb, ms);
		revert_lock_pc(r, lkb);
		queue_cast(r, lkb, -EDEADLK);
		break;

3511 3512
	case -EINPROGRESS:
		/* convert was queued on remote master */
David Teigland's avatar
David Teigland committed
3513 3514 3515
		receive_flags_reply(lkb, ms);
		if (is_demoted(lkb))
			munge_demoted(lkb, ms);
3516 3517
		del_lkb(r, lkb);
		add_lkb(r, lkb, DLM_LKSTS_CONVERT);
3518
		add_timeout(lkb);
3519 3520 3521 3522 3523
		break;

	case 0:
		/* convert was granted on remote master */
		receive_flags_reply(lkb, ms);
David Teigland's avatar
David Teigland committed
3524 3525
		if (is_demoted(lkb))
			munge_demoted(lkb, ms);
3526 3527 3528 3529 3530
		grant_lock_pc(r, lkb, ms);
		queue_cast(r, lkb, 0);
		break;

	default:
3531 3532
		log_error(r->res_ls, "receive_convert_reply %x error %d",
			  lkb->lkb_id, ms->m_result);
3533 3534 3535 3536 3537 3538
	}
}

static void _receive_convert_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	struct dlm_rsb *r = lkb->lkb_resource;
3539
	int error;
3540 3541 3542 3543

	hold_rsb(r);
	lock_rsb(r);

3544 3545 3546 3547
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3548 3549 3550 3551
	/* stub reply can happen with waiters_mutex held */
	error = remove_from_waiters_ms(lkb, ms);
	if (error)
		goto out;
3552

3553 3554
	__receive_convert_reply(r, lkb, ms);
 out:
3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565
	unlock_rsb(r);
	put_rsb(r);
}

static void receive_convert_reply(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3566 3567
		log_debug(ls, "receive_convert_reply from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3568 3569 3570 3571
		return;
	}

	_receive_convert_reply(lkb, ms);
3572
	dlm_put_lkb(lkb);
3573 3574 3575 3576 3577
}

static void _receive_unlock_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	struct dlm_rsb *r = lkb->lkb_resource;
3578
	int error;
3579 3580 3581 3582

	hold_rsb(r);
	lock_rsb(r);

3583 3584 3585 3586
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3587 3588 3589 3590 3591
	/* stub reply can happen with waiters_mutex held */
	error = remove_from_waiters_ms(lkb, ms);
	if (error)
		goto out;

3592 3593
	/* this is the value returned from do_unlock() on the master */

3594
	switch (ms->m_result) {
3595 3596 3597 3598 3599
	case -DLM_EUNLOCK:
		receive_flags_reply(lkb, ms);
		remove_lock_pc(r, lkb);
		queue_cast(r, lkb, -DLM_EUNLOCK);
		break;
3600 3601
	case -ENOENT:
		break;
3602
	default:
3603 3604
		log_error(r->res_ls, "receive_unlock_reply %x error %d",
			  lkb->lkb_id, ms->m_result);
3605
	}
3606
 out:
3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617
	unlock_rsb(r);
	put_rsb(r);
}

static void receive_unlock_reply(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3618 3619
		log_debug(ls, "receive_unlock_reply from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3620 3621 3622 3623
		return;
	}

	_receive_unlock_reply(lkb, ms);
3624
	dlm_put_lkb(lkb);
3625 3626 3627 3628 3629
}

static void _receive_cancel_reply(struct dlm_lkb *lkb, struct dlm_message *ms)
{
	struct dlm_rsb *r = lkb->lkb_resource;
3630
	int error;
3631 3632 3633 3634

	hold_rsb(r);
	lock_rsb(r);

3635 3636 3637 3638
	error = validate_message(lkb, ms);
	if (error)
		goto out;

3639 3640 3641 3642 3643
	/* stub reply can happen with waiters_mutex held */
	error = remove_from_waiters_ms(lkb, ms);
	if (error)
		goto out;

3644 3645
	/* this is the value returned from do_cancel() on the master */

3646
	switch (ms->m_result) {
3647 3648 3649
	case -DLM_ECANCEL:
		receive_flags_reply(lkb, ms);
		revert_lock_pc(r, lkb);
David Teigland's avatar
David Teigland committed
3650
		queue_cast(r, lkb, -DLM_ECANCEL);
3651 3652
		break;
	case 0:
3653 3654
		break;
	default:
3655 3656
		log_error(r->res_ls, "receive_cancel_reply %x error %d",
			  lkb->lkb_id, ms->m_result);
3657
	}
3658
 out:
3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669
	unlock_rsb(r);
	put_rsb(r);
}

static void receive_cancel_reply(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	int error;

	error = find_lkb(ls, ms->m_remid, &lkb);
	if (error) {
3670 3671
		log_debug(ls, "receive_cancel_reply from %d no lkb %x",
			  ms->m_header.h_nodeid, ms->m_remid);
3672 3673 3674 3675
		return;
	}

	_receive_cancel_reply(lkb, ms);
3676
	dlm_put_lkb(lkb);
3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690
}

static void receive_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
	int error, ret_nodeid;

	error = find_lkb(ls, ms->m_lkid, &lkb);
	if (error) {
		log_error(ls, "receive_lookup_reply no lkb");
		return;
	}

3691
	/* ms->m_result is the value returned by dlm_dir_lookup on dir node
3692 3693 3694 3695 3696 3697
	   FIXME: will a non-zero error ever be returned? */

	r = lkb->lkb_resource;
	hold_rsb(r);
	lock_rsb(r);

3698 3699 3700 3701
	error = remove_from_waiters(lkb, DLM_MSG_LOOKUP_REPLY);
	if (error)
		goto out;

3702 3703 3704 3705 3706 3707 3708 3709 3710 3711
	ret_nodeid = ms->m_nodeid;
	if (ret_nodeid == dlm_our_nodeid()) {
		r->res_nodeid = 0;
		ret_nodeid = 0;
		r->res_first_lkid = 0;
	} else {
		/* set_master() will copy res_nodeid to lkb_nodeid */
		r->res_nodeid = ret_nodeid;
	}

3712 3713 3714 3715 3716 3717 3718 3719
	if (is_overlap(lkb)) {
		log_debug(ls, "receive_lookup_reply %x unlock %x",
			  lkb->lkb_id, lkb->lkb_flags);
		queue_cast_overlap(r, lkb);
		unhold_lkb(lkb); /* undoes create_lkb() */
		goto out_list;
	}

3720 3721
	_request_lock(r, lkb);

3722
 out_list:
3723 3724
	if (!ret_nodeid)
		process_lookup_list(r);
3725
 out:
3726 3727
	unlock_rsb(r);
	put_rsb(r);
3728
	dlm_put_lkb(lkb);
3729 3730
}

3731
static void _receive_message(struct dlm_ls *ls, struct dlm_message *ms)
3732
{
3733 3734 3735 3736 3737 3738 3739
	if (!dlm_is_member(ls, ms->m_header.h_nodeid)) {
		log_debug(ls, "ignore non-member message %d from %d %x %x %d",
			  ms->m_type, ms->m_header.h_nodeid, ms->m_lkid,
			  ms->m_remid, ms->m_result);
		return;
	}

3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803
	switch (ms->m_type) {

	/* messages sent to a master node */

	case DLM_MSG_REQUEST:
		receive_request(ls, ms);
		break;

	case DLM_MSG_CONVERT:
		receive_convert(ls, ms);
		break;

	case DLM_MSG_UNLOCK:
		receive_unlock(ls, ms);
		break;

	case DLM_MSG_CANCEL:
		receive_cancel(ls, ms);
		break;

	/* messages sent from a master node (replies to above) */

	case DLM_MSG_REQUEST_REPLY:
		receive_request_reply(ls, ms);
		break;

	case DLM_MSG_CONVERT_REPLY:
		receive_convert_reply(ls, ms);
		break;

	case DLM_MSG_UNLOCK_REPLY:
		receive_unlock_reply(ls, ms);
		break;

	case DLM_MSG_CANCEL_REPLY:
		receive_cancel_reply(ls, ms);
		break;

	/* messages sent from a master node (only two types of async msg) */

	case DLM_MSG_GRANT:
		receive_grant(ls, ms);
		break;

	case DLM_MSG_BAST:
		receive_bast(ls, ms);
		break;

	/* messages sent to a dir node */

	case DLM_MSG_LOOKUP:
		receive_lookup(ls, ms);
		break;

	case DLM_MSG_REMOVE:
		receive_remove(ls, ms);
		break;

	/* messages sent from a dir node (remove has no reply) */

	case DLM_MSG_LOOKUP_REPLY:
		receive_lookup_reply(ls, ms);
		break;

3804 3805 3806 3807 3808 3809
	/* other messages */

	case DLM_MSG_PURGE:
		receive_purge(ls, ms);
		break;

3810 3811 3812 3813 3814 3815 3816
	default:
		log_error(ls, "unknown message type %d", ms->m_type);
	}

	dlm_astd_wake();
}

3817 3818 3819 3820 3821 3822 3823
/* If the lockspace is in recovery mode (locking stopped), then normal
   messages are saved on the requestqueue for processing after recovery is
   done.  When not in recovery mode, we wait for dlm_recoverd to drain saved
   messages off the requestqueue before we process new ones. This occurs right
   after recovery completes when we transition from saving all messages on
   requestqueue, to processing all the saved messages, to processing new
   messages as they arrive. */
3824

3825 3826 3827 3828
static void dlm_receive_message(struct dlm_ls *ls, struct dlm_message *ms,
				int nodeid)
{
	if (dlm_locking_stopped(ls)) {
3829
		dlm_add_requestqueue(ls, nodeid, ms);
3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848
	} else {
		dlm_wait_requestqueue(ls);
		_receive_message(ls, ms);
	}
}

/* This is called by dlm_recoverd to process messages that were saved on
   the requestqueue. */

void dlm_receive_message_saved(struct dlm_ls *ls, struct dlm_message *ms)
{
	_receive_message(ls, ms);
}

/* This is called by the midcomms layer when something is received for
   the lockspace.  It could be either a MSG (normal message sent as part of
   standard locking activity) or an RCOM (recovery message sent as part of
   lockspace recovery). */

3849
void dlm_receive_buffer(union dlm_packet *p, int nodeid)
3850
{
3851
	struct dlm_header *hd = &p->header;
3852 3853 3854 3855 3856
	struct dlm_ls *ls;
	int type = 0;

	switch (hd->h_cmd) {
	case DLM_MSG:
3857 3858
		dlm_message_in(&p->message);
		type = p->message.m_type;
3859 3860
		break;
	case DLM_RCOM:
3861 3862
		dlm_rcom_in(&p->rcom);
		type = p->rcom.rc_type;
3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876
		break;
	default:
		log_print("invalid h_cmd %d from %u", hd->h_cmd, nodeid);
		return;
	}

	if (hd->h_nodeid != nodeid) {
		log_print("invalid h_nodeid %d from %d lockspace %x",
			  hd->h_nodeid, nodeid, hd->h_lockspace);
		return;
	}

	ls = dlm_find_lockspace_global(hd->h_lockspace);
	if (!ls) {
3877 3878 3879
		if (dlm_config.ci_log_debug)
			log_print("invalid lockspace %x from %d cmd %d type %d",
				  hd->h_lockspace, nodeid, hd->h_cmd, type);
3880 3881

		if (hd->h_cmd == DLM_RCOM && type == DLM_RCOM_STATUS)
3882
			dlm_send_ls_not_ready(nodeid, &p->rcom);
3883 3884 3885 3886 3887 3888 3889 3890
		return;
	}

	/* this rwsem allows dlm_ls_stop() to wait for all dlm_recv threads to
	   be inactive (in this ls) before transitioning to recovery mode */

	down_read(&ls->ls_recv_active);
	if (hd->h_cmd == DLM_MSG)
3891
		dlm_receive_message(ls, &p->message, nodeid);
3892
	else
3893
		dlm_receive_rcom(ls, &p->rcom, nodeid);
3894 3895 3896 3897
	up_read(&ls->ls_recv_active);

	dlm_put_lockspace(ls);
}
3898 3899 3900 3901 3902

static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
	if (middle_conversion(lkb)) {
		hold_lkb(lkb);
3903
		ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY;
3904
		ls->ls_stub_ms.m_result = -EINPROGRESS;
3905
		ls->ls_stub_ms.m_flags = lkb->lkb_flags;
3906
		ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid;
3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947
		_receive_convert_reply(lkb, &ls->ls_stub_ms);

		/* Same special case as in receive_rcom_lock_args() */
		lkb->lkb_grmode = DLM_LOCK_IV;
		rsb_set_flag(lkb->lkb_resource, RSB_RECOVER_CONVERT);
		unhold_lkb(lkb);

	} else if (lkb->lkb_rqmode >= lkb->lkb_grmode) {
		lkb->lkb_flags |= DLM_IFL_RESEND;
	}

	/* lkb->lkb_rqmode < lkb->lkb_grmode shouldn't happen since down
	   conversions are async; there's no reply from the remote master */
}

/* A waiting lkb needs recovery if the master node has failed, or
   the master node is changing (only when no directory is used) */

static int waiter_needs_recovery(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
	if (dlm_is_removed(ls, lkb->lkb_nodeid))
		return 1;

	if (!dlm_no_directory(ls))
		return 0;

	if (dlm_dir_nodeid(lkb->lkb_resource) != lkb->lkb_nodeid)
		return 1;

	return 0;
}

/* Recovery for locks that are waiting for replies from nodes that are now
   gone.  We can just complete unlocks and cancels by faking a reply from the
   dead node.  Requests and up-conversions we flag to be resent after
   recovery.  Down-conversions can just be completed with a fake reply like
   unlocks.  Conversions between PR and CW need special attention. */

void dlm_recover_waiters_pre(struct dlm_ls *ls)
{
	struct dlm_lkb *lkb, *safe;
3948
	int wait_type, stub_unlock_result, stub_cancel_result;
3949

3950
	mutex_lock(&ls->ls_waiters_mutex);
3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966

	list_for_each_entry_safe(lkb, safe, &ls->ls_waiters, lkb_wait_reply) {
		log_debug(ls, "pre recover waiter lkid %x type %d flags %x",
			  lkb->lkb_id, lkb->lkb_wait_type, lkb->lkb_flags);

		/* all outstanding lookups, regardless of destination  will be
		   resent after recovery is done */

		if (lkb->lkb_wait_type == DLM_MSG_LOOKUP) {
			lkb->lkb_flags |= DLM_IFL_RESEND;
			continue;
		}

		if (!waiter_needs_recovery(ls, lkb))
			continue;

3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993
		wait_type = lkb->lkb_wait_type;
		stub_unlock_result = -DLM_EUNLOCK;
		stub_cancel_result = -DLM_ECANCEL;

		/* Main reply may have been received leaving a zero wait_type,
		   but a reply for the overlapping op may not have been
		   received.  In that case we need to fake the appropriate
		   reply for the overlap op. */

		if (!wait_type) {
			if (is_overlap_cancel(lkb)) {
				wait_type = DLM_MSG_CANCEL;
				if (lkb->lkb_grmode == DLM_LOCK_IV)
					stub_cancel_result = 0;
			}
			if (is_overlap_unlock(lkb)) {
				wait_type = DLM_MSG_UNLOCK;
				if (lkb->lkb_grmode == DLM_LOCK_IV)
					stub_unlock_result = -ENOENT;
			}

			log_debug(ls, "rwpre overlap %x %x %d %d %d",
				  lkb->lkb_id, lkb->lkb_flags, wait_type,
				  stub_cancel_result, stub_unlock_result);
		}

		switch (wait_type) {
3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004

		case DLM_MSG_REQUEST:
			lkb->lkb_flags |= DLM_IFL_RESEND;
			break;

		case DLM_MSG_CONVERT:
			recover_convert_waiter(ls, lkb);
			break;

		case DLM_MSG_UNLOCK:
			hold_lkb(lkb);
4005
			ls->ls_stub_ms.m_type = DLM_MSG_UNLOCK_REPLY;
4006
			ls->ls_stub_ms.m_result = stub_unlock_result;
4007
			ls->ls_stub_ms.m_flags = lkb->lkb_flags;
4008
			ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid;
4009
			_receive_unlock_reply(lkb, &ls->ls_stub_ms);
4010
			dlm_put_lkb(lkb);
4011 4012 4013 4014
			break;

		case DLM_MSG_CANCEL:
			hold_lkb(lkb);
4015
			ls->ls_stub_ms.m_type = DLM_MSG_CANCEL_REPLY;
4016
			ls->ls_stub_ms.m_result = stub_cancel_result;
4017
			ls->ls_stub_ms.m_flags = lkb->lkb_flags;
4018
			ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid;
4019
			_receive_cancel_reply(lkb, &ls->ls_stub_ms);
4020
			dlm_put_lkb(lkb);
4021 4022 4023
			break;

		default:
4024 4025
			log_error(ls, "invalid lkb wait_type %d %d",
				  lkb->lkb_wait_type, wait_type);
4026
		}
4027
		schedule();
4028
	}
4029
	mutex_unlock(&ls->ls_waiters_mutex);
4030 4031
}

4032
static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls)
4033 4034
{
	struct dlm_lkb *lkb;
4035
	int found = 0;
4036

4037
	mutex_lock(&ls->ls_waiters_mutex);
4038 4039
	list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) {
		if (lkb->lkb_flags & DLM_IFL_RESEND) {
4040 4041
			hold_lkb(lkb);
			found = 1;
4042 4043 4044
			break;
		}
	}
4045
	mutex_unlock(&ls->ls_waiters_mutex);
4046

4047
	if (!found)
4048
		lkb = NULL;
4049
	return lkb;
4050 4051 4052 4053 4054 4055
}

/* Deal with lookups and lkb's marked RESEND from _pre.  We may now be the
   master or dir-node for r.  Processing the lkb may result in it being placed
   back on waiters. */

4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067
/* We do this after normal locking has been enabled and any saved messages
   (in requestqueue) have been processed.  We should be confident that at
   this point we won't get or process a reply to any of these waiting
   operations.  But, new ops may be coming in on the rsbs/locks here from
   userspace or remotely. */

/* there may have been an overlap unlock/cancel prior to recovery or after
   recovery.  if before, the lkb may still have a pos wait_count; if after, the
   overlap flag would just have been set and nothing new sent.  we can be
   confident here than any replies to either the initial op or overlap ops
   prior to recovery have been received. */

4068 4069 4070 4071
int dlm_recover_waiters_post(struct dlm_ls *ls)
{
	struct dlm_lkb *lkb;
	struct dlm_rsb *r;
4072
	int error = 0, mstype, err, oc, ou;
4073 4074 4075 4076 4077 4078 4079 4080

	while (1) {
		if (dlm_locking_stopped(ls)) {
			log_debug(ls, "recover_waiters_post aborted");
			error = -EINTR;
			break;
		}

4081 4082
		lkb = find_resend_waiter(ls);
		if (!lkb)
4083 4084 4085
			break;

		r = lkb->lkb_resource;
4086 4087 4088 4089 4090 4091 4092
		hold_rsb(r);
		lock_rsb(r);

		mstype = lkb->lkb_wait_type;
		oc = is_overlap_cancel(lkb);
		ou = is_overlap_unlock(lkb);
		err = 0;
4093 4094 4095 4096

		log_debug(ls, "recover_waiters_post %x type %d flags %x %s",
			  lkb->lkb_id, mstype, lkb->lkb_flags, r->res_name);

4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144
		/* At this point we assume that we won't get a reply to any
		   previous op or overlap op on this lock.  First, do a big
		   remove_from_waiters() for all previous ops. */

		lkb->lkb_flags &= ~DLM_IFL_RESEND;
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_UNLOCK;
		lkb->lkb_flags &= ~DLM_IFL_OVERLAP_CANCEL;
		lkb->lkb_wait_type = 0;
		lkb->lkb_wait_count = 0;
		mutex_lock(&ls->ls_waiters_mutex);
		list_del_init(&lkb->lkb_wait_reply);
		mutex_unlock(&ls->ls_waiters_mutex);
		unhold_lkb(lkb); /* for waiters list */

		if (oc || ou) {
			/* do an unlock or cancel instead of resending */
			switch (mstype) {
			case DLM_MSG_LOOKUP:
			case DLM_MSG_REQUEST:
				queue_cast(r, lkb, ou ? -DLM_EUNLOCK :
							-DLM_ECANCEL);
				unhold_lkb(lkb); /* undoes create_lkb() */
				break;
			case DLM_MSG_CONVERT:
				if (oc) {
					queue_cast(r, lkb, -DLM_ECANCEL);
				} else {
					lkb->lkb_exflags |= DLM_LKF_FORCEUNLOCK;
					_unlock_lock(r, lkb);
				}
				break;
			default:
				err = 1;
			}
		} else {
			switch (mstype) {
			case DLM_MSG_LOOKUP:
			case DLM_MSG_REQUEST:
				_request_lock(r, lkb);
				if (is_master(r))
					confirm_master(r, 0);
				break;
			case DLM_MSG_CONVERT:
				_convert_lock(r, lkb);
				break;
			default:
				err = 1;
			}
4145
		}
4146 4147 4148 4149 4150 4151 4152

		if (err)
			log_error(ls, "recover_waiters_post %x %d %x %d %d",
			  	  lkb->lkb_id, mstype, lkb->lkb_flags, oc, ou);
		unlock_rsb(r);
		put_rsb(r);
		dlm_put_lkb(lkb);
4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165
	}

	return error;
}

static void purge_queue(struct dlm_rsb *r, struct list_head *queue,
			int (*test)(struct dlm_ls *ls, struct dlm_lkb *lkb))
{
	struct dlm_ls *ls = r->res_ls;
	struct dlm_lkb *lkb, *safe;

	list_for_each_entry_safe(lkb, safe, queue, lkb_statequeue) {
		if (test(ls, lkb)) {
4166
			rsb_set_flag(r, RSB_LOCKS_PURGED);
4167 4168
			del_lkb(r, lkb);
			/* this put should free the lkb */
4169
			if (!dlm_put_lkb(lkb))
4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222
				log_error(ls, "purged lkb not released");
		}
	}
}

static int purge_dead_test(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
	return (is_master_copy(lkb) && dlm_is_removed(ls, lkb->lkb_nodeid));
}

static int purge_mstcpy_test(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
	return is_master_copy(lkb);
}

static void purge_dead_locks(struct dlm_rsb *r)
{
	purge_queue(r, &r->res_grantqueue, &purge_dead_test);
	purge_queue(r, &r->res_convertqueue, &purge_dead_test);
	purge_queue(r, &r->res_waitqueue, &purge_dead_test);
}

void dlm_purge_mstcpy_locks(struct dlm_rsb *r)
{
	purge_queue(r, &r->res_grantqueue, &purge_mstcpy_test);
	purge_queue(r, &r->res_convertqueue, &purge_mstcpy_test);
	purge_queue(r, &r->res_waitqueue, &purge_mstcpy_test);
}

/* Get rid of locks held by nodes that are gone. */

int dlm_purge_locks(struct dlm_ls *ls)
{
	struct dlm_rsb *r;

	log_debug(ls, "dlm_purge_locks");

	down_write(&ls->ls_root_sem);
	list_for_each_entry(r, &ls->ls_root_list, res_root_list) {
		hold_rsb(r);
		lock_rsb(r);
		if (is_master(r))
			purge_dead_locks(r);
		unlock_rsb(r);
		unhold_rsb(r);

		schedule();
	}
	up_write(&ls->ls_root_sem);

	return 0;
}

4223 4224 4225 4226
static struct dlm_rsb *find_purged_rsb(struct dlm_ls *ls, int bucket)
{
	struct dlm_rsb *r, *r_ret = NULL;

4227
	spin_lock(&ls->ls_rsbtbl[bucket].lock);
4228 4229 4230 4231 4232 4233 4234 4235
	list_for_each_entry(r, &ls->ls_rsbtbl[bucket].list, res_hashchain) {
		if (!rsb_flag(r, RSB_LOCKS_PURGED))
			continue;
		hold_rsb(r);
		rsb_clear_flag(r, RSB_LOCKS_PURGED);
		r_ret = r;
		break;
	}
4236
	spin_unlock(&ls->ls_rsbtbl[bucket].lock);
4237 4238 4239 4240
	return r_ret;
}

void dlm_grant_after_purge(struct dlm_ls *ls)
4241 4242
{
	struct dlm_rsb *r;
4243
	int bucket = 0;
4244

4245 4246 4247 4248 4249 4250
	while (1) {
		r = find_purged_rsb(ls, bucket);
		if (!r) {
			if (bucket == ls->ls_rsbtbl_size - 1)
				break;
			bucket++;
4251
			continue;
4252
		}
4253 4254 4255 4256
		lock_rsb(r);
		if (is_master(r)) {
			grant_pending_locks(r);
			confirm_master(r, 0);
4257
		}
4258 4259
		unlock_rsb(r);
		put_rsb(r);
4260
		schedule();
4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292
	}
}

static struct dlm_lkb *search_remid_list(struct list_head *head, int nodeid,
					 uint32_t remid)
{
	struct dlm_lkb *lkb;

	list_for_each_entry(lkb, head, lkb_statequeue) {
		if (lkb->lkb_nodeid == nodeid && lkb->lkb_remid == remid)
			return lkb;
	}
	return NULL;
}

static struct dlm_lkb *search_remid(struct dlm_rsb *r, int nodeid,
				    uint32_t remid)
{
	struct dlm_lkb *lkb;

	lkb = search_remid_list(&r->res_grantqueue, nodeid, remid);
	if (lkb)
		return lkb;
	lkb = search_remid_list(&r->res_convertqueue, nodeid, remid);
	if (lkb)
		return lkb;
	lkb = search_remid_list(&r->res_waitqueue, nodeid, remid);
	if (lkb)
		return lkb;
	return NULL;
}

4293
/* needs at least dlm_rcom + rcom_lock */
4294 4295 4296 4297 4298 4299
static int receive_rcom_lock_args(struct dlm_ls *ls, struct dlm_lkb *lkb,
				  struct dlm_rsb *r, struct dlm_rcom *rc)
{
	struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;

	lkb->lkb_nodeid = rc->rc_header.h_nodeid;
Al Viro's avatar
Al Viro committed
4300 4301 4302 4303
	lkb->lkb_ownpid = le32_to_cpu(rl->rl_ownpid);
	lkb->lkb_remid = le32_to_cpu(rl->rl_lkid);
	lkb->lkb_exflags = le32_to_cpu(rl->rl_exflags);
	lkb->lkb_flags = le32_to_cpu(rl->rl_flags) & 0x0000FFFF;
4304
	lkb->lkb_flags |= DLM_IFL_MSTCPY;
Al Viro's avatar
Al Viro committed
4305
	lkb->lkb_lvbseq = le32_to_cpu(rl->rl_lvbseq);
4306 4307 4308 4309
	lkb->lkb_rqmode = rl->rl_rqmode;
	lkb->lkb_grmode = rl->rl_grmode;
	/* don't set lkb_status because add_lkb wants to itself */

4310 4311
	lkb->lkb_bastfn = (rl->rl_asts & AST_BAST) ? &fake_bastfn : NULL;
	lkb->lkb_astfn = (rl->rl_asts & AST_COMP) ? &fake_astfn : NULL;
4312 4313

	if (lkb->lkb_exflags & DLM_LKF_VALBLK) {
4314 4315 4316 4317
		int lvblen = rc->rc_header.h_length - sizeof(struct dlm_rcom) -
			 sizeof(struct rcom_lock);
		if (lvblen > ls->ls_lvblen)
			return -EINVAL;
4318
		lkb->lkb_lvbptr = dlm_allocate_lvb(ls);
4319 4320 4321 4322 4323 4324 4325 4326 4327
		if (!lkb->lkb_lvbptr)
			return -ENOMEM;
		memcpy(lkb->lkb_lvbptr, rl->rl_lvb, lvblen);
	}

	/* Conversions between PR and CW (middle modes) need special handling.
	   The real granted mode of these converting locks cannot be determined
	   until all locks have been rebuilt on the rsb (recover_conversion) */

Al Viro's avatar
Al Viro committed
4328 4329
	if (rl->rl_wait_type == cpu_to_le16(DLM_MSG_CONVERT) &&
	    middle_conversion(lkb)) {
4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343
		rl->rl_status = DLM_LKSTS_CONVERT;
		lkb->lkb_grmode = DLM_LOCK_IV;
		rsb_set_flag(r, RSB_RECOVER_CONVERT);
	}

	return 0;
}

/* This lkb may have been recovered in a previous aborted recovery so we need
   to check if the rsb already has an lkb with the given remote nodeid/lkid.
   If so we just send back a standard reply.  If not, we create a new lkb with
   the given values and send back our lkid.  We send back our lkid by sending
   back the rcom_lock struct we got but with the remid field filled in. */

4344
/* needs at least dlm_rcom + rcom_lock */
4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356
int dlm_recover_master_copy(struct dlm_ls *ls, struct dlm_rcom *rc)
{
	struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
	struct dlm_rsb *r;
	struct dlm_lkb *lkb;
	int error;

	if (rl->rl_parent_lkid) {
		error = -EOPNOTSUPP;
		goto out;
	}

Al Viro's avatar
Al Viro committed
4357 4358
	error = find_rsb(ls, rl->rl_name, le16_to_cpu(rl->rl_namelen),
			 R_MASTER, &r);
4359 4360 4361 4362 4363
	if (error)
		goto out;

	lock_rsb(r);

Al Viro's avatar
Al Viro committed
4364
	lkb = search_remid(r, rc->rc_header.h_nodeid, le32_to_cpu(rl->rl_lkid));
4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375
	if (lkb) {
		error = -EEXIST;
		goto out_remid;
	}

	error = create_lkb(ls, &lkb);
	if (error)
		goto out_unlock;

	error = receive_rcom_lock_args(ls, lkb, r, rc);
	if (error) {
4376
		__put_lkb(ls, lkb);
4377 4378 4379 4380 4381 4382 4383 4384 4385 4386
		goto out_unlock;
	}

	attach_lkb(r, lkb);
	add_lkb(r, lkb, rl->rl_status);
	error = 0;

 out_remid:
	/* this is the new value returned to the lock holder for
	   saving in its process-copy lkb */
Al Viro's avatar
Al Viro committed
4387
	rl->rl_remid = cpu_to_le32(lkb->lkb_id);
4388 4389 4390 4391 4392 4393

 out_unlock:
	unlock_rsb(r);
	put_rsb(r);
 out:
	if (error)
Al Viro's avatar
Al Viro committed
4394 4395 4396
		log_debug(ls, "recover_master_copy %d %x", error,
			  le32_to_cpu(rl->rl_lkid));
	rl->rl_result = cpu_to_le32(error);
4397 4398 4399
	return error;
}

4400
/* needs at least dlm_rcom + rcom_lock */
4401 4402 4403 4404 4405 4406 4407
int dlm_recover_process_copy(struct dlm_ls *ls, struct dlm_rcom *rc)
{
	struct rcom_lock *rl = (struct rcom_lock *) rc->rc_buf;
	struct dlm_rsb *r;
	struct dlm_lkb *lkb;
	int error;

Al Viro's avatar
Al Viro committed
4408
	error = find_lkb(ls, le32_to_cpu(rl->rl_lkid), &lkb);
4409
	if (error) {
Al Viro's avatar
Al Viro committed
4410 4411
		log_error(ls, "recover_process_copy no lkid %x",
				le32_to_cpu(rl->rl_lkid));
4412 4413 4414 4415 4416
		return error;
	}

	DLM_ASSERT(is_process_copy(lkb), dlm_print_lkb(lkb););

Al Viro's avatar
Al Viro committed
4417
	error = le32_to_cpu(rl->rl_result);
4418 4419 4420 4421 4422 4423

	r = lkb->lkb_resource;
	hold_rsb(r);
	lock_rsb(r);

	switch (error) {
David Teigland's avatar
David Teigland committed
4424 4425 4426 4427 4428 4429 4430 4431
	case -EBADR:
		/* There's a chance the new master received our lock before
		   dlm_recover_master_reply(), this wouldn't happen if we did
		   a barrier between recover_masters and recover_locks. */
		log_debug(ls, "master copy not ready %x r %lx %s", lkb->lkb_id,
			  (unsigned long)r, r->res_name);
		dlm_send_rcom_lock(r, lkb);
		goto out;
4432 4433 4434 4435
	case -EEXIST:
		log_debug(ls, "master copy exists %x", lkb->lkb_id);
		/* fall through */
	case 0:
Al Viro's avatar
Al Viro committed
4436
		lkb->lkb_remid = le32_to_cpu(rl->rl_remid);
4437 4438 4439 4440 4441 4442 4443 4444 4445
		break;
	default:
		log_error(ls, "dlm_recover_process_copy unknown error %d %x",
			  error, lkb->lkb_id);
	}

	/* an ack for dlm_recover_locks() which waits for replies from
	   all the locks it sends to new masters */
	dlm_recovered_lock(r);
David Teigland's avatar
David Teigland committed
4446
 out:
4447 4448
	unlock_rsb(r);
	put_rsb(r);
4449
	dlm_put_lkb(lkb);
4450 4451 4452 4453

	return 0;
}

David Teigland's avatar
David Teigland committed
4454 4455
int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua,
		     int mode, uint32_t flags, void *name, unsigned int namelen,
4456
		     unsigned long timeout_cs)
David Teigland's avatar
David Teigland committed
4457 4458 4459 4460 4461
{
	struct dlm_lkb *lkb;
	struct dlm_args args;
	int error;

4462
	dlm_lock_recovery(ls);
David Teigland's avatar
David Teigland committed
4463 4464 4465 4466 4467 4468 4469 4470

	error = create_lkb(ls, &lkb);
	if (error) {
		kfree(ua);
		goto out;
	}

	if (flags & DLM_LKF_VALBLK) {
David Teigland's avatar
David Teigland committed
4471
		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
David Teigland's avatar
David Teigland committed
4472 4473 4474 4475 4476 4477 4478 4479
		if (!ua->lksb.sb_lvbptr) {
			kfree(ua);
			__put_lkb(ls, lkb);
			error = -ENOMEM;
			goto out;
		}
	}

4480
	/* After ua is attached to lkb it will be freed by dlm_free_lkb().
David Teigland's avatar
David Teigland committed
4481 4482 4483
	   When DLM_IFL_USER is set, the dlm knows that this is a userspace
	   lock and that lkb_astparam is the dlm_user_args structure. */

4484
	error = set_lock_args(mode, &ua->lksb, flags, namelen, timeout_cs,
4485
			      fake_astfn, ua, fake_bastfn, &args);
David Teigland's avatar
David Teigland committed
4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511
	lkb->lkb_flags |= DLM_IFL_USER;
	ua->old_mode = DLM_LOCK_IV;

	if (error) {
		__put_lkb(ls, lkb);
		goto out;
	}

	error = request_lock(ls, lkb, name, namelen, &args);

	switch (error) {
	case 0:
		break;
	case -EINPROGRESS:
		error = 0;
		break;
	case -EAGAIN:
		error = 0;
		/* fall through */
	default:
		__put_lkb(ls, lkb);
		goto out;
	}

	/* add this new lkb to the per-process list of locks */
	spin_lock(&ua->proc->locks_spin);
4512
	hold_lkb(lkb);
David Teigland's avatar
David Teigland committed
4513 4514 4515
	list_add_tail(&lkb->lkb_ownqueue, &ua->proc->locks);
	spin_unlock(&ua->proc->locks_spin);
 out:
4516
	dlm_unlock_recovery(ls);
David Teigland's avatar
David Teigland committed
4517 4518 4519 4520
	return error;
}

int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
4521 4522
		     int mode, uint32_t flags, uint32_t lkid, char *lvb_in,
		     unsigned long timeout_cs)
David Teigland's avatar
David Teigland committed
4523 4524 4525 4526 4527 4528
{
	struct dlm_lkb *lkb;
	struct dlm_args args;
	struct dlm_user_args *ua;
	int error;

4529
	dlm_lock_recovery(ls);
David Teigland's avatar
David Teigland committed
4530 4531 4532 4533 4534 4535 4536 4537

	error = find_lkb(ls, lkid, &lkb);
	if (error)
		goto out;

	/* user can change the params on its lock when it converts it, or
	   add an lvb that didn't exist before */

4538
	ua = lkb->lkb_ua;
David Teigland's avatar
David Teigland committed
4539 4540

	if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) {
David Teigland's avatar
David Teigland committed
4541
		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
David Teigland's avatar
David Teigland committed
4542 4543 4544 4545 4546 4547 4548 4549
		if (!ua->lksb.sb_lvbptr) {
			error = -ENOMEM;
			goto out_put;
		}
	}
	if (lvb_in && ua->lksb.sb_lvbptr)
		memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);

4550
	ua->xid = ua_tmp->xid;
David Teigland's avatar
David Teigland committed
4551 4552 4553 4554
	ua->castparam = ua_tmp->castparam;
	ua->castaddr = ua_tmp->castaddr;
	ua->bastparam = ua_tmp->bastparam;
	ua->bastaddr = ua_tmp->bastaddr;
4555
	ua->user_lksb = ua_tmp->user_lksb;
David Teigland's avatar
David Teigland committed
4556 4557
	ua->old_mode = lkb->lkb_grmode;

4558
	error = set_lock_args(mode, &ua->lksb, flags, 0, timeout_cs,
4559
			      fake_astfn, ua, fake_bastfn, &args);
David Teigland's avatar
David Teigland committed
4560 4561 4562 4563 4564
	if (error)
		goto out_put;

	error = convert_lock(ls, lkb, &args);

4565
	if (error == -EINPROGRESS || error == -EAGAIN || error == -EDEADLK)
David Teigland's avatar
David Teigland committed
4566 4567 4568 4569
		error = 0;
 out_put:
	dlm_put_lkb(lkb);
 out:
4570
	dlm_unlock_recovery(ls);
David Teigland's avatar
David Teigland committed
4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582
	kfree(ua_tmp);
	return error;
}

int dlm_user_unlock(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
		    uint32_t flags, uint32_t lkid, char *lvb_in)
{
	struct dlm_lkb *lkb;
	struct dlm_args args;
	struct dlm_user_args *ua;
	int error;

4583
	dlm_lock_recovery(ls);
David Teigland's avatar
David Teigland committed
4584 4585 4586 4587 4588

	error = find_lkb(ls, lkid, &lkb);
	if (error)
		goto out;

4589
	ua = lkb->lkb_ua;
David Teigland's avatar
David Teigland committed
4590 4591 4592

	if (lvb_in && ua->lksb.sb_lvbptr)
		memcpy(ua->lksb.sb_lvbptr, lvb_in, DLM_USER_LVB_LEN);
4593 4594
	if (ua_tmp->castparam)
		ua->castparam = ua_tmp->castparam;
4595
	ua->user_lksb = ua_tmp->user_lksb;
David Teigland's avatar
David Teigland committed
4596 4597 4598 4599 4600 4601 4602 4603 4604

	error = set_unlock_args(flags, ua, &args);
	if (error)
		goto out_put;

	error = unlock_lock(ls, lkb, &args);

	if (error == -DLM_EUNLOCK)
		error = 0;
4605 4606 4607
	/* from validate_unlock_args() */
	if (error == -EBUSY && (flags & DLM_LKF_FORCEUNLOCK))
		error = 0;
David Teigland's avatar
David Teigland committed
4608 4609 4610 4611
	if (error)
		goto out_put;

	spin_lock(&ua->proc->locks_spin);
David Teigland's avatar
David Teigland committed
4612 4613 4614
	/* dlm_user_add_ast() may have already taken lkb off the proc list */
	if (!list_empty(&lkb->lkb_ownqueue))
		list_move(&lkb->lkb_ownqueue, &ua->proc->unlocking);
David Teigland's avatar
David Teigland committed
4615 4616 4617 4618
	spin_unlock(&ua->proc->locks_spin);
 out_put:
	dlm_put_lkb(lkb);
 out:
4619
	dlm_unlock_recovery(ls);
4620
	kfree(ua_tmp);
David Teigland's avatar
David Teigland committed
4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631
	return error;
}

int dlm_user_cancel(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
		    uint32_t flags, uint32_t lkid)
{
	struct dlm_lkb *lkb;
	struct dlm_args args;
	struct dlm_user_args *ua;
	int error;

4632
	dlm_lock_recovery(ls);
David Teigland's avatar
David Teigland committed
4633 4634 4635 4636 4637

	error = find_lkb(ls, lkid, &lkb);
	if (error)
		goto out;

4638
	ua = lkb->lkb_ua;
4639 4640
	if (ua_tmp->castparam)
		ua->castparam = ua_tmp->castparam;
4641
	ua->user_lksb = ua_tmp->user_lksb;
David Teigland's avatar
David Teigland committed
4642 4643 4644 4645 4646 4647 4648 4649 4650

	error = set_unlock_args(flags, ua, &args);
	if (error)
		goto out_put;

	error = cancel_lock(ls, lkb, &args);

	if (error == -DLM_ECANCEL)
		error = 0;
4651 4652 4653
	/* from validate_unlock_args() */
	if (error == -EBUSY)
		error = 0;
David Teigland's avatar
David Teigland committed
4654 4655 4656
 out_put:
	dlm_put_lkb(lkb);
 out:
4657
	dlm_unlock_recovery(ls);
4658
	kfree(ua_tmp);
David Teigland's avatar
David Teigland committed
4659 4660 4661
	return error;
}

4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675
int dlm_user_deadlock(struct dlm_ls *ls, uint32_t flags, uint32_t lkid)
{
	struct dlm_lkb *lkb;
	struct dlm_args args;
	struct dlm_user_args *ua;
	struct dlm_rsb *r;
	int error;

	dlm_lock_recovery(ls);

	error = find_lkb(ls, lkid, &lkb);
	if (error)
		goto out;

4676
	ua = lkb->lkb_ua;
4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709

	error = set_unlock_args(flags, ua, &args);
	if (error)
		goto out_put;

	/* same as cancel_lock(), but set DEADLOCK_CANCEL after lock_rsb */

	r = lkb->lkb_resource;
	hold_rsb(r);
	lock_rsb(r);

	error = validate_unlock_args(lkb, &args);
	if (error)
		goto out_r;
	lkb->lkb_flags |= DLM_IFL_DEADLOCK_CANCEL;

	error = _cancel_lock(r, lkb);
 out_r:
	unlock_rsb(r);
	put_rsb(r);

	if (error == -DLM_ECANCEL)
		error = 0;
	/* from validate_unlock_args() */
	if (error == -EBUSY)
		error = 0;
 out_put:
	dlm_put_lkb(lkb);
 out:
	dlm_unlock_recovery(ls);
	return error;
}

4710 4711 4712
/* lkb's that are removed from the waiters list by revert are just left on the
   orphans list with the granted orphan locks, to be freed by purge */

David Teigland's avatar
David Teigland committed
4713 4714
static int orphan_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
4715 4716
	struct dlm_args args;
	int error;
David Teigland's avatar
David Teigland committed
4717

4718 4719 4720 4721
	hold_lkb(lkb);
	mutex_lock(&ls->ls_orphans_mutex);
	list_add_tail(&lkb->lkb_ownqueue, &ls->ls_orphans);
	mutex_unlock(&ls->ls_orphans_mutex);
David Teigland's avatar
David Teigland committed
4722

4723
	set_unlock_args(0, lkb->lkb_ua, &args);
4724 4725 4726 4727 4728

	error = cancel_lock(ls, lkb, &args);
	if (error == -DLM_ECANCEL)
		error = 0;
	return error;
David Teigland's avatar
David Teigland committed
4729 4730 4731 4732 4733 4734 4735 4736 4737 4738
}

/* The force flag allows the unlock to go ahead even if the lkb isn't granted.
   Regardless of what rsb queue the lock is on, it's removed and freed. */

static int unlock_proc_lock(struct dlm_ls *ls, struct dlm_lkb *lkb)
{
	struct dlm_args args;
	int error;

4739
	set_unlock_args(DLM_LKF_FORCEUNLOCK, lkb->lkb_ua, &args);
David Teigland's avatar
David Teigland committed
4740 4741 4742 4743 4744 4745 4746

	error = unlock_lock(ls, lkb, &args);
	if (error == -DLM_EUNLOCK)
		error = 0;
	return error;
}

4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771
/* We have to release clear_proc_locks mutex before calling unlock_proc_lock()
   (which does lock_rsb) due to deadlock with receiving a message that does
   lock_rsb followed by dlm_user_add_ast() */

static struct dlm_lkb *del_proc_lock(struct dlm_ls *ls,
				     struct dlm_user_proc *proc)
{
	struct dlm_lkb *lkb = NULL;

	mutex_lock(&ls->ls_clear_proc_locks);
	if (list_empty(&proc->locks))
		goto out;

	lkb = list_entry(proc->locks.next, struct dlm_lkb, lkb_ownqueue);
	list_del_init(&lkb->lkb_ownqueue);

	if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
		lkb->lkb_flags |= DLM_IFL_ORPHAN;
	else
		lkb->lkb_flags |= DLM_IFL_DEAD;
 out:
	mutex_unlock(&ls->ls_clear_proc_locks);
	return lkb;
}

David Teigland's avatar
David Teigland committed
4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785
/* The ls_clear_proc_locks mutex protects against dlm_user_add_asts() which
   1) references lkb->ua which we free here and 2) adds lkbs to proc->asts,
   which we clear here. */

/* proc CLOSING flag is set so no more device_reads should look at proc->asts
   list, and no more device_writes should add lkb's to proc->locks list; so we
   shouldn't need to take asts_spin or locks_spin here.  this assumes that
   device reads/writes/closes are serialized -- FIXME: we may need to serialize
   them ourself. */

void dlm_clear_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
{
	struct dlm_lkb *lkb, *safe;

4786
	dlm_lock_recovery(ls);
David Teigland's avatar
David Teigland committed
4787

4788 4789 4790 4791
	while (1) {
		lkb = del_proc_lock(ls, proc);
		if (!lkb)
			break;
David Teigland's avatar
David Teigland committed
4792
		del_timeout(lkb);
4793
		if (lkb->lkb_exflags & DLM_LKF_PERSISTENT)
David Teigland's avatar
David Teigland committed
4794
			orphan_proc_lock(ls, lkb);
4795
		else
David Teigland's avatar
David Teigland committed
4796 4797 4798 4799 4800 4801 4802 4803
			unlock_proc_lock(ls, lkb);

		/* this removes the reference for the proc->locks list
		   added by dlm_user_request, it may result in the lkb
		   being freed */

		dlm_put_lkb(lkb);
	}
David Teigland's avatar
David Teigland committed
4804

4805 4806
	mutex_lock(&ls->ls_clear_proc_locks);

David Teigland's avatar
David Teigland committed
4807 4808 4809 4810 4811 4812 4813 4814
	/* in-progress unlocks */
	list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
		list_del_init(&lkb->lkb_ownqueue);
		lkb->lkb_flags |= DLM_IFL_DEAD;
		dlm_put_lkb(lkb);
	}

	list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_astqueue) {
4815
		lkb->lkb_ast_type = 0;
David Teigland's avatar
David Teigland committed
4816 4817 4818 4819
		list_del(&lkb->lkb_astqueue);
		dlm_put_lkb(lkb);
	}

David Teigland's avatar
David Teigland committed
4820
	mutex_unlock(&ls->ls_clear_proc_locks);
4821
	dlm_unlock_recovery(ls);
David Teigland's avatar
David Teigland committed
4822
}
David Teigland's avatar
David Teigland committed
4823

4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902
static void purge_proc_locks(struct dlm_ls *ls, struct dlm_user_proc *proc)
{
	struct dlm_lkb *lkb, *safe;

	while (1) {
		lkb = NULL;
		spin_lock(&proc->locks_spin);
		if (!list_empty(&proc->locks)) {
			lkb = list_entry(proc->locks.next, struct dlm_lkb,
					 lkb_ownqueue);
			list_del_init(&lkb->lkb_ownqueue);
		}
		spin_unlock(&proc->locks_spin);

		if (!lkb)
			break;

		lkb->lkb_flags |= DLM_IFL_DEAD;
		unlock_proc_lock(ls, lkb);
		dlm_put_lkb(lkb); /* ref from proc->locks list */
	}

	spin_lock(&proc->locks_spin);
	list_for_each_entry_safe(lkb, safe, &proc->unlocking, lkb_ownqueue) {
		list_del_init(&lkb->lkb_ownqueue);
		lkb->lkb_flags |= DLM_IFL_DEAD;
		dlm_put_lkb(lkb);
	}
	spin_unlock(&proc->locks_spin);

	spin_lock(&proc->asts_spin);
	list_for_each_entry_safe(lkb, safe, &proc->asts, lkb_astqueue) {
		list_del(&lkb->lkb_astqueue);
		dlm_put_lkb(lkb);
	}
	spin_unlock(&proc->asts_spin);
}

/* pid of 0 means purge all orphans */

static void do_purge(struct dlm_ls *ls, int nodeid, int pid)
{
	struct dlm_lkb *lkb, *safe;

	mutex_lock(&ls->ls_orphans_mutex);
	list_for_each_entry_safe(lkb, safe, &ls->ls_orphans, lkb_ownqueue) {
		if (pid && lkb->lkb_ownpid != pid)
			continue;
		unlock_proc_lock(ls, lkb);
		list_del_init(&lkb->lkb_ownqueue);
		dlm_put_lkb(lkb);
	}
	mutex_unlock(&ls->ls_orphans_mutex);
}

static int send_purge(struct dlm_ls *ls, int nodeid, int pid)
{
	struct dlm_message *ms;
	struct dlm_mhandle *mh;
	int error;

	error = _create_message(ls, sizeof(struct dlm_message), nodeid,
				DLM_MSG_PURGE, &ms, &mh);
	if (error)
		return error;
	ms->m_nodeid = nodeid;
	ms->m_pid = pid;

	return send_message(mh, ms);
}

int dlm_user_purge(struct dlm_ls *ls, struct dlm_user_proc *proc,
		   int nodeid, int pid)
{
	int error = 0;

	if (nodeid != dlm_our_nodeid()) {
		error = send_purge(ls, nodeid, pid);
	} else {
4903
		dlm_lock_recovery(ls);
4904 4905 4906 4907
		if (pid == current->pid)
			purge_proc_locks(ls, proc);
		else
			do_purge(ls, nodeid, pid);
4908
		dlm_unlock_recovery(ls);
4909 4910 4911 4912
	}
	return error;
}