summaryrefslogtreecommitdiff
path: root/fs/nfs/nfs4state.c
blob: 591ad1d51880aa08f71ee5cc203d9546347125ee (plain)
1
2
3
4
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
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
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
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
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
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
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
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
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
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
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
747
748
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
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
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
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
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
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
/*
 *  fs/nfs/nfs4state.c
 *
 *  Client-side XDR for NFSv4.
 *
 *  Copyright (c) 2002 The Regents of the University of Michigan.
 *  All rights reserved.
 *
 *  Kendrick Smith <kmsmith@umich.edu>
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *
 *  1. Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *  2. Redistributions in binary form must reproduce the above copyright
 *     notice, this list of conditions and the following disclaimer in the
 *     documentation and/or other materials provided with the distribution.
 *  3. Neither the name of the University nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
 *  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 *  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 *  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 *  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 *  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
 *  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
 *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
 *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 * Implementation of the NFSv4 state model.  For the time being,
 * this is minimal, but will be made much more complex in a
 * subsequent patch.
 */

#include <linux/config.h>
#include <linux/slab.h>
#include <linux/smp_lock.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_idmap.h>
#include <linux/workqueue.h>
#include <linux/bitops.h>

#include "nfs4_fs.h"
#include "callback.h"
#include "delegation.h"

#define OPENOWNER_POOL_SIZE	8

const nfs4_stateid zero_stateid;

static DEFINE_SPINLOCK(state_spinlock);
static LIST_HEAD(nfs4_clientid_list);

static void nfs4_recover_state(void *);

void
init_nfsv4_state(struct nfs_server *server)
{
	server->nfs4_state = NULL;
	INIT_LIST_HEAD(&server->nfs4_siblings);
}

void
destroy_nfsv4_state(struct nfs_server *server)
{
	if (server->mnt_path) {
		kfree(server->mnt_path);
		server->mnt_path = NULL;
	}
	if (server->nfs4_state) {
		nfs4_put_client(server->nfs4_state);
		server->nfs4_state = NULL;
	}
}

/*
 * nfs4_get_client(): returns an empty client structure
 * nfs4_put_client(): drops reference to client structure
 *
 * Since these are allocated/deallocated very rarely, we don't
 * bother putting them in a slab cache...
 */
static struct nfs4_client *
nfs4_alloc_client(struct in_addr *addr)
{
	struct nfs4_client *clp;

	if (nfs_callback_up() < 0)
		return NULL;
	if ((clp = kmalloc(sizeof(*clp), GFP_KERNEL)) == NULL) {
		nfs_callback_down();
		return NULL;
	}
	memset(clp, 0, sizeof(*clp));
	memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr));
	init_rwsem(&clp->cl_sem);
	INIT_LIST_HEAD(&clp->cl_delegations);
	INIT_LIST_HEAD(&clp->cl_state_owners);
	INIT_LIST_HEAD(&clp->cl_unused);
	spin_lock_init(&clp->cl_lock);
	atomic_set(&clp->cl_count, 1);
	INIT_WORK(&clp->cl_recoverd, nfs4_recover_state, clp);
	INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp);
	INIT_LIST_HEAD(&clp->cl_superblocks);
	init_waitqueue_head(&clp->cl_waitq);
	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS4 client");
	clp->cl_rpcclient = ERR_PTR(-EINVAL);
	clp->cl_boot_time = CURRENT_TIME;
	clp->cl_state = 1 << NFS4CLNT_OK;
	return clp;
}

static void
nfs4_free_client(struct nfs4_client *clp)
{
	struct nfs4_state_owner *sp;

	while (!list_empty(&clp->cl_unused)) {
		sp = list_entry(clp->cl_unused.next,
				struct nfs4_state_owner,
				so_list);
		list_del(&sp->so_list);
		kfree(sp);
	}
	BUG_ON(!list_empty(&clp->cl_state_owners));
	if (clp->cl_cred)
		put_rpccred(clp->cl_cred);
	nfs_idmap_delete(clp);
	if (!IS_ERR(clp->cl_rpcclient))
		rpc_shutdown_client(clp->cl_rpcclient);
	kfree(clp);
	nfs_callback_down();
}

static struct nfs4_client *__nfs4_find_client(struct in_addr *addr)
{
	struct nfs4_client *clp;
	list_for_each_entry(clp, &nfs4_clientid_list, cl_servers) {
		if (memcmp(&clp->cl_addr, addr, sizeof(clp->cl_addr)) == 0) {
			atomic_inc(&clp->cl_count);
			return clp;
		}
	}
	return NULL;
}

struct nfs4_client *nfs4_find_client(struct in_addr *addr)
{
	struct nfs4_client *clp;
	spin_lock(&state_spinlock);
	clp = __nfs4_find_client(addr);
	spin_unlock(&state_spinlock);
	return clp;
}

struct nfs4_client *
nfs4_get_client(struct in_addr *addr)
{
	struct nfs4_client *clp, *new = NULL;

	spin_lock(&state_spinlock);
	for (;;) {
		clp = __nfs4_find_client(addr);
		if (clp != NULL)
			break;
		clp = new;
		if (clp != NULL) {
			list_add(&clp->cl_servers, &nfs4_clientid_list);
			new = NULL;
			break;
		}
		spin_unlock(&state_spinlock);
		new = nfs4_alloc_client(addr);
		spin_lock(&state_spinlock);
		if (new == NULL)
			break;
	}
	spin_unlock(&state_spinlock);
	if (new)
		nfs4_free_client(new);
	return clp;
}

void
nfs4_put_client(struct nfs4_client *clp)
{
	if (!atomic_dec_and_lock(&clp->cl_count, &state_spinlock))
		return;
	list_del(&clp->cl_servers);
	spin_unlock(&state_spinlock);
	BUG_ON(!list_empty(&clp->cl_superblocks));
	wake_up_all(&clp->cl_waitq);
	rpc_wake_up(&clp->cl_rpcwaitq);
	nfs4_kill_renewd(clp);
	nfs4_free_client(clp);
}

static int __nfs4_init_client(struct nfs4_client *clp)
{
	int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, nfs_callback_tcpport);
	if (status == 0)
		status = nfs4_proc_setclientid_confirm(clp);
	if (status == 0)
		nfs4_schedule_state_renewal(clp);
	return status;
}

int nfs4_init_client(struct nfs4_client *clp)
{
	return nfs4_map_errors(__nfs4_init_client(clp));
}

u32
nfs4_alloc_lockowner_id(struct nfs4_client *clp)
{
	return clp->cl_lockowner_id ++;
}

static struct nfs4_state_owner *
nfs4_client_grab_unused(struct nfs4_client *clp, struct rpc_cred *cred)
{
	struct nfs4_state_owner *sp = NULL;

	if (!list_empty(&clp->cl_unused)) {
		sp = list_entry(clp->cl_unused.next, struct nfs4_state_owner, so_list);
		atomic_inc(&sp->so_count);
		sp->so_cred = cred;
		list_move(&sp->so_list, &clp->cl_state_owners);
		clp->cl_nunused--;
	}
	return sp;
}

static struct nfs4_state_owner *
nfs4_find_state_owner(struct nfs4_client *clp, struct rpc_cred *cred)
{
	struct nfs4_state_owner *sp, *res = NULL;

	list_for_each_entry(sp, &clp->cl_state_owners, so_list) {
		if (sp->so_cred != cred)
			continue;
		atomic_inc(&sp->so_count);
		/* Move to the head of the list */
		list_move(&sp->so_list, &clp->cl_state_owners);
		res = sp;
		break;
	}
	return res;
}

/*
 * nfs4_alloc_state_owner(): this is called on the OPEN or CREATE path to
 * create a new state_owner.
 *
 */
static struct nfs4_state_owner *
nfs4_alloc_state_owner(void)
{
	struct nfs4_state_owner *sp;

	sp = kmalloc(sizeof(*sp),GFP_KERNEL);
	if (!sp)
		return NULL;
	init_MUTEX(&sp->so_sema);
	sp->so_seqid = 0;                 /* arbitrary */
	INIT_LIST_HEAD(&sp->so_states);
	INIT_LIST_HEAD(&sp->so_delegations);
	atomic_set(&sp->so_count, 1);
	return sp;
}

void
nfs4_drop_state_owner(struct nfs4_state_owner *sp)
{
	struct nfs4_client *clp = sp->so_client;
	spin_lock(&clp->cl_lock);
	list_del_init(&sp->so_list);
	spin_unlock(&clp->cl_lock);
}

/*
 * Note: must be called with clp->cl_sem held in order to prevent races
 *       with reboot recovery!
 */
struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct rpc_cred *cred)
{
	struct nfs4_client *clp = server->nfs4_state;
	struct nfs4_state_owner *sp, *new;

	get_rpccred(cred);
	new = nfs4_alloc_state_owner();
	spin_lock(&clp->cl_lock);
	sp = nfs4_find_state_owner(clp, cred);
	if (sp == NULL)
		sp = nfs4_client_grab_unused(clp, cred);
	if (sp == NULL && new != NULL) {
		list_add(&new->so_list, &clp->cl_state_owners);
		new->so_client = clp;
		new->so_id = nfs4_alloc_lockowner_id(clp);
		new->so_cred = cred;
		sp = new;
		new = NULL;
	}
	spin_unlock(&clp->cl_lock);
	if (new)
		kfree(new);
	if (sp != NULL)
		return sp;
	put_rpccred(cred);
	return NULL;
}

/*
 * Must be called with clp->cl_sem held in order to avoid races
 * with state recovery...
 */
void nfs4_put_state_owner(struct nfs4_state_owner *sp)
{
	struct nfs4_client *clp = sp->so_client;
	struct rpc_cred *cred = sp->so_cred;

	if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock))
		return;
	if (clp->cl_nunused >= OPENOWNER_POOL_SIZE)
		goto out_free;
	if (list_empty(&sp->so_list))
		goto out_free;
	list_move(&sp->so_list, &clp->cl_unused);
	clp->cl_nunused++;
	spin_unlock(&clp->cl_lock);
	put_rpccred(cred);
	cred = NULL;
	return;
out_free:
	list_del(&sp->so_list);
	spin_unlock(&clp->cl_lock);
	put_rpccred(cred);
	kfree(sp);
}

static struct nfs4_state *
nfs4_alloc_open_state(void)
{
	struct nfs4_state *state;

	state = kmalloc(sizeof(*state), GFP_KERNEL);
	if (!state)
		return NULL;
	state->state = 0;
	state->nreaders = 0;
	state->nwriters = 0;
	state->flags = 0;
	memset(state->stateid.data, 0, sizeof(state->stateid.data));
	atomic_set(&state->count, 1);
	INIT_LIST_HEAD(&state->lock_states);
	init_MUTEX(&state->lock_sema);
	rwlock_init(&state->state_lock);
	return state;
}

static struct nfs4_state *
__nfs4_find_state(struct inode *inode, struct rpc_cred *cred, mode_t mode)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs4_state *state;

	mode &= (FMODE_READ|FMODE_WRITE);
	list_for_each_entry(state, &nfsi->open_states, inode_states) {
		if (state->owner->so_cred != cred)
			continue;
		if ((mode & FMODE_READ) != 0 && state->nreaders == 0)
			continue;
		if ((mode & FMODE_WRITE) != 0 && state->nwriters == 0)
			continue;
		if ((state->state & mode) != mode)
			continue;
		atomic_inc(&state->count);
		if (mode & FMODE_READ)
			state->nreaders++;
		if (mode & FMODE_WRITE)
			state->nwriters++;
		return state;
	}
	return NULL;
}

static struct nfs4_state *
__nfs4_find_state_byowner(struct inode *inode, struct nfs4_state_owner *owner)
{
	struct nfs_inode *nfsi = NFS_I(inode);
	struct nfs4_state *state;

	list_for_each_entry(state, &nfsi->open_states, inode_states) {
		/* Is this in the process of being freed? */
		if (state->nreaders == 0 && state->nwriters == 0)
			continue;
		if (state->owner == owner) {
			atomic_inc(&state->count);
			return state;
		}
	}
	return NULL;
}

struct nfs4_state *
nfs4_find_state(struct inode *inode, struct rpc_cred *cred, mode_t mode)
{
	struct nfs4_state *state;

	spin_lock(&inode->i_lock);
	state = __nfs4_find_state(inode, cred, mode);
	spin_unlock(&inode->i_lock);
	return state;
}

static void
nfs4_free_open_state(struct nfs4_state *state)
{
	kfree(state);
}

struct nfs4_state *
nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner)
{
	struct nfs4_state *state, *new;
	struct nfs_inode *nfsi = NFS_I(inode);

	spin_lock(&inode->i_lock);
	state = __nfs4_find_state_byowner(inode, owner);
	spin_unlock(&inode->i_lock);
	if (state)
		goto out;
	new = nfs4_alloc_open_state();
	spin_lock(&inode->i_lock);
	state = __nfs4_find_state_byowner(inode, owner);
	if (state == NULL && new != NULL) {
		state = new;
		/* Caller *must* be holding owner->so_sem */
		/* Note: The reclaim code dictates that we add stateless
		 * and read-only stateids to the end of the list */
		list_add_tail(&state->open_states, &owner->so_states);
		state->owner = owner;
		atomic_inc(&owner->so_count);
		list_add(&state->inode_states, &nfsi->open_states);
		state->inode = igrab(inode);
		spin_unlock(&inode->i_lock);
	} else {
		spin_unlock(&inode->i_lock);
		if (new)
			nfs4_free_open_state(new);
	}
out:
	return state;
}

/*
 * Beware! Caller must be holding exactly one
 * reference to clp->cl_sem and owner->so_sema!
 */
void nfs4_put_open_state(struct nfs4_state *state)
{
	struct inode *inode = state->inode;
	struct nfs4_state_owner *owner = state->owner;

	if (!atomic_dec_and_lock(&state->count, &inode->i_lock))
		return;
	if (!list_empty(&state->inode_states))
		list_del(&state->inode_states);
	spin_unlock(&inode->i_lock);
	list_del(&state->open_states);
	iput(inode);
	BUG_ON (state->state != 0);
	nfs4_free_open_state(state);
	nfs4_put_state_owner(owner);
}

/*
 * Beware! Caller must be holding no references to clp->cl_sem!
 * of owner->so_sema!
 */
void nfs4_close_state(struct nfs4_state *state, mode_t mode)
{
	struct inode *inode = state->inode;
	struct nfs4_state_owner *owner = state->owner;
	struct nfs4_client *clp = owner->so_client;
	int newstate;

	atomic_inc(&owner->so_count);
	down_read(&clp->cl_sem);
	down(&owner->so_sema);
	/* Protect against nfs4_find_state() */
	spin_lock(&inode->i_lock);
	if (mode & FMODE_READ)
		state->nreaders--;
	if (mode & FMODE_WRITE)
		state->nwriters--;
	if (state->nwriters == 0) {
		if (state->nreaders == 0)
			list_del_init(&state->inode_states);
		/* See reclaim code */
		list_move_tail(&state->open_states, &owner->so_states);
	}
	spin_unlock(&inode->i_lock);
	newstate = 0;
	if (state->state != 0) {
		if (state->nreaders)
			newstate |= FMODE_READ;
		if (state->nwriters)
			newstate |= FMODE_WRITE;
		if (state->state == newstate)
			goto out;
		if (nfs4_do_close(inode, state, newstate) == -EINPROGRESS)
			return;
	}
out:
	nfs4_put_open_state(state);
	up(&owner->so_sema);
	nfs4_put_state_owner(owner);
	up_read(&clp->cl_sem);
}

/*
 * Search the state->lock_states for an existing lock_owner
 * that is compatible with current->files
 */
static struct nfs4_lock_state *
__nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
{
	struct nfs4_lock_state *pos;
	list_for_each_entry(pos, &state->lock_states, ls_locks) {
		if (pos->ls_owner != fl_owner)
			continue;
		atomic_inc(&pos->ls_count);
		return pos;
	}
	return NULL;
}

struct nfs4_lock_state *
nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
{
	struct nfs4_lock_state *lsp;
	read_lock(&state->state_lock);
	lsp = __nfs4_find_lock_state(state, fl_owner);
	read_unlock(&state->state_lock);
	return lsp;
}

/*
 * Return a compatible lock_state. If no initialized lock_state structure
 * exists, return an uninitialized one.
 *
 * The caller must be holding state->lock_sema
 */
static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner)
{
	struct nfs4_lock_state *lsp;
	struct nfs4_client *clp = state->owner->so_client;

	lsp = kmalloc(sizeof(*lsp), GFP_KERNEL);
	if (lsp == NULL)
		return NULL;
	lsp->ls_flags = 0;
	lsp->ls_seqid = 0;	/* arbitrary */
	lsp->ls_id = -1; 
	memset(lsp->ls_stateid.data, 0, sizeof(lsp->ls_stateid.data));
	atomic_set(&lsp->ls_count, 1);
	lsp->ls_owner = fl_owner;
	INIT_LIST_HEAD(&lsp->ls_locks);
	spin_lock(&clp->cl_lock);
	lsp->ls_id = nfs4_alloc_lockowner_id(clp);
	spin_unlock(&clp->cl_lock);
	return lsp;
}

/*
 * Return a compatible lock_state. If no initialized lock_state structure
 * exists, return an uninitialized one.
 *
 * The caller must be holding state->lock_sema and clp->cl_sem
 */
struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner)
{
	struct nfs4_lock_state * lsp;
	
	lsp = nfs4_find_lock_state(state, owner);
	if (lsp == NULL)
		lsp = nfs4_alloc_lock_state(state, owner);
	return lsp;
}

/*
 * Byte-range lock aware utility to initialize the stateid of read/write
 * requests.
 */
void
nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t fl_owner)
{
	if (test_bit(LK_STATE_IN_USE, &state->flags)) {
		struct nfs4_lock_state *lsp;

		lsp = nfs4_find_lock_state(state, fl_owner);
		if (lsp) {
			memcpy(dst, &lsp->ls_stateid, sizeof(*dst));
			nfs4_put_lock_state(lsp);
			return;
		}
	}
	memcpy(dst, &state->stateid, sizeof(*dst));
}

/*
* Called with state->lock_sema and clp->cl_sem held.
*/
void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp)
{
	if (status == NFS_OK || seqid_mutating_err(-status))
		lsp->ls_seqid++;
}

/* 
* Check to see if the request lock (type FL_UNLK) effects the fl lock.
*
* fl and request must have the same posix owner
*
* return: 
* 0 -> fl not effected by request
* 1 -> fl consumed by request
*/

static int
nfs4_check_unlock(struct file_lock *fl, struct file_lock *request)
{
	if (fl->fl_start >= request->fl_start && fl->fl_end <= request->fl_end)
		return 1;
	return 0;
}

/*
 * Post an initialized lock_state on the state->lock_states list.
 */
void nfs4_notify_setlk(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp)
{
	if (!list_empty(&lsp->ls_locks))
		return;
	atomic_inc(&lsp->ls_count);
	write_lock(&state->state_lock);
	list_add(&lsp->ls_locks, &state->lock_states);
	set_bit(LK_STATE_IN_USE, &state->flags);
	write_unlock(&state->state_lock);
}

/* 
 * to decide to 'reap' lock state:
 * 1) search i_flock for file_locks with fl.lock_state = to ls.
 * 2) determine if unlock will consume found lock. 
 * 	if so, reap
 *
 * 	else, don't reap.
 *
 */
void
nfs4_notify_unlck(struct nfs4_state *state, struct file_lock *request, struct nfs4_lock_state *lsp)
{
	struct inode *inode = state->inode;
	struct file_lock *fl;

	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
		if (!(fl->fl_flags & FL_POSIX))
			continue;
		if (fl->fl_owner != lsp->ls_owner)
			continue;
		/* Exit if we find at least one lock which is not consumed */
		if (nfs4_check_unlock(fl,request) == 0)
			return;
	}

	write_lock(&state->state_lock);
	list_del_init(&lsp->ls_locks);
	if (list_empty(&state->lock_states))
		clear_bit(LK_STATE_IN_USE, &state->flags);
	write_unlock(&state->state_lock);
	nfs4_put_lock_state(lsp);
}

/*
 * Release reference to lock_state, and free it if we see that
 * it is no longer in use
 */
void
nfs4_put_lock_state(struct nfs4_lock_state *lsp)
{
	if (!atomic_dec_and_test(&lsp->ls_count))
		return;
	BUG_ON (!list_empty(&lsp->ls_locks));
	kfree(lsp);
}

/*
* Called with sp->so_sema and clp->cl_sem held.
*
* Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or
* failed with a seqid incrementing error -
* see comments nfs_fs.h:seqid_mutating_error()
*/
void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp)
{
	if (status == NFS_OK || seqid_mutating_err(-status))
		sp->so_seqid++;
	/* If the server returns BAD_SEQID, unhash state_owner here */
	if (status == -NFS4ERR_BAD_SEQID)
		nfs4_drop_state_owner(sp);
}

static int reclaimer(void *);
struct reclaimer_args {
	struct nfs4_client *clp;
	struct completion complete;
};

/*
 * State recovery routine
 */
void
nfs4_recover_state(void *data)
{
	struct nfs4_client *clp = (struct nfs4_client *)data;
	struct reclaimer_args args = {
		.clp = clp,
	};
	might_sleep();

	init_completion(&args.complete);

	if (kernel_thread(reclaimer, &args, CLONE_KERNEL) < 0)
		goto out_failed_clear;
	wait_for_completion(&args.complete);
	return;
out_failed_clear:
	set_bit(NFS4CLNT_OK, &clp->cl_state);
	wake_up_all(&clp->cl_waitq);
	rpc_wake_up(&clp->cl_rpcwaitq);
}

/*
 * Schedule a state recovery attempt
 */
void
nfs4_schedule_state_recovery(struct nfs4_client *clp)
{
	if (!clp)
		return;
	if (test_and_clear_bit(NFS4CLNT_OK, &clp->cl_state))
		schedule_work(&clp->cl_recoverd);
}

static int nfs4_reclaim_locks(struct nfs4_state_recovery_ops *ops, struct nfs4_state *state)
{
	struct inode *inode = state->inode;
	struct file_lock *fl;
	int status = 0;

	for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) {
		if (!(fl->fl_flags & FL_POSIX))
			continue;
		if (((struct nfs_open_context *)fl->fl_file->private_data)->state != state)
			continue;
		status = ops->recover_lock(state, fl);
		if (status >= 0)
			continue;
		switch (status) {
			default:
				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
						__FUNCTION__, status);
			case -NFS4ERR_EXPIRED:
			case -NFS4ERR_NO_GRACE:
			case -NFS4ERR_RECLAIM_BAD:
			case -NFS4ERR_RECLAIM_CONFLICT:
				/* kill_proc(fl->fl_owner, SIGLOST, 1); */
				break;
			case -NFS4ERR_STALE_CLIENTID:
				goto out_err;
		}
	}
	return 0;
out_err:
	return status;
}

static int nfs4_reclaim_open_state(struct nfs4_state_recovery_ops *ops, struct nfs4_state_owner *sp)
{
	struct nfs4_state *state;
	struct nfs4_lock_state *lock;
	int status = 0;

	/* Note: we rely on the sp->so_states list being ordered 
	 * so that we always reclaim open(O_RDWR) and/or open(O_WRITE)
	 * states first.
	 * This is needed to ensure that the server won't give us any
	 * read delegations that we have to return if, say, we are
	 * recovering after a network partition or a reboot from a
	 * server that doesn't support a grace period.
	 */
	list_for_each_entry(state, &sp->so_states, open_states) {
		if (state->state == 0)
			continue;
		status = ops->recover_open(sp, state);
		list_for_each_entry(lock, &state->lock_states, ls_locks)
			lock->ls_flags &= ~NFS_LOCK_INITIALIZED;
		if (status >= 0) {
			status = nfs4_reclaim_locks(ops, state);
			if (status < 0)
				goto out_err;
			list_for_each_entry(lock, &state->lock_states, ls_locks) {
				if (!(lock->ls_flags & NFS_LOCK_INITIALIZED))
					printk("%s: Lock reclaim failed!\n",
							__FUNCTION__);
			}
			continue;
		}
		switch (status) {
			default:
				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
						__FUNCTION__, status);
			case -ENOENT:
			case -NFS4ERR_RECLAIM_BAD:
			case -NFS4ERR_RECLAIM_CONFLICT:
				/*
				 * Open state on this file cannot be recovered
				 * All we can do is revert to using the zero stateid.
				 */
				memset(state->stateid.data, 0,
					sizeof(state->stateid.data));
				/* Mark the file as being 'closed' */
				state->state = 0;
				break;
			case -NFS4ERR_EXPIRED:
			case -NFS4ERR_NO_GRACE:
			case -NFS4ERR_STALE_CLIENTID:
				goto out_err;
		}
	}
	return 0;
out_err:
	return status;
}

static int reclaimer(void *ptr)
{
	struct reclaimer_args *args = (struct reclaimer_args *)ptr;
	struct nfs4_client *clp = args->clp;
	struct nfs4_state_owner *sp;
	struct nfs4_state_recovery_ops *ops;
	int status = 0;

	daemonize("%u.%u.%u.%u-reclaim", NIPQUAD(clp->cl_addr));
	allow_signal(SIGKILL);

	atomic_inc(&clp->cl_count);
	complete(&args->complete);

	/* Ensure exclusive access to NFSv4 state */
	lock_kernel();
	down_write(&clp->cl_sem);
	/* Are there any NFS mounts out there? */
	if (list_empty(&clp->cl_superblocks))
		goto out;
restart_loop:
	status = nfs4_proc_renew(clp);
	switch (status) {
		case 0:
		case -NFS4ERR_CB_PATH_DOWN:
			goto out;
		case -NFS4ERR_STALE_CLIENTID:
		case -NFS4ERR_LEASE_MOVED:
			ops = &nfs4_reboot_recovery_ops;
			break;
		default:
			ops = &nfs4_network_partition_recovery_ops;
	};
	status = __nfs4_init_client(clp);
	if (status)
		goto out_error;
	/* Mark all delegations for reclaim */
	nfs_delegation_mark_reclaim(clp);
	/* Note: list is protected by exclusive lock on cl->cl_sem */
	list_for_each_entry(sp, &clp->cl_state_owners, so_list) {
		status = nfs4_reclaim_open_state(ops, sp);
		if (status < 0) {
			if (status == -NFS4ERR_NO_GRACE) {
				ops = &nfs4_network_partition_recovery_ops;
				status = nfs4_reclaim_open_state(ops, sp);
			}
			if (status == -NFS4ERR_STALE_CLIENTID)
				goto restart_loop;
			if (status == -NFS4ERR_EXPIRED)
				goto restart_loop;
		}
	}
	nfs_delegation_reap_unclaimed(clp);
out:
	set_bit(NFS4CLNT_OK, &clp->cl_state);
	up_write(&clp->cl_sem);
	unlock_kernel();
	wake_up_all(&clp->cl_waitq);
	rpc_wake_up(&clp->cl_rpcwaitq);
	if (status == -NFS4ERR_CB_PATH_DOWN)
		nfs_handle_cb_pathdown(clp);
	nfs4_put_client(clp);
	return 0;
out_error:
	printk(KERN_WARNING "Error: state recovery failed on NFSv4 server %u.%u.%u.%u with error %d\n",
				NIPQUAD(clp->cl_addr.s_addr), -status);
	goto out;
}

/*
 * Local variables:
 *  c-basic-offset: 8
 * End:
 */