summaryrefslogtreecommitdiffstats
path: root/net/sunrpc/cache.c
blob: 50b1a8b441fed54ad49bb830358843868cd26f4b (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
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
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
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
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
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
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
1173
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
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
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
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
/*
 * net/sunrpc/cache.c
 *
 * Generic code for various authentication-related caches
 * used by sunrpc clients and servers.
 *
 * Copyright (C) 2002 Neil Brown <neilb@cse.unsw.edu.au>
 *
 * Released under terms in GPL version 2.  See COPYING.
 *
 */

#include <linux/types.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/slab.h>
#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/kmod.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/ctype.h>
#include <asm/uaccess.h>
#include <linux/poll.h>
#include <linux/seq_file.h>
#include <linux/proc_fs.h>
#include <linux/net.h>
#include <linux/workqueue.h>
#include <linux/mutex.h>
#include <asm/ioctls.h>
#include <linux/sunrpc/types.h>
#include <linux/sunrpc/cache.h>
#include <linux/sunrpc/stats.h>

#define	 RPCDBG_FACILITY RPCDBG_CACHE

static int cache_defer_req(struct cache_req *req, struct cache_head *item);
static void cache_revisit_request(struct cache_head *item);

static void cache_init(struct cache_head *h)
{
	time_t now = get_seconds();
	h->next = NULL;
	h->flags = 0;
	kref_init(&h->ref);
	h->expiry_time = now + CACHE_NEW_EXPIRY;
	h->last_refresh = now;
}

struct cache_head *sunrpc_cache_lookup(struct cache_detail *detail,
				       struct cache_head *key, int hash)
{
	struct cache_head **head,  **hp;
	struct cache_head *new = NULL;

	head = &detail->hash_table[hash];

	read_lock(&detail->hash_lock);

	for (hp=head; *hp != NULL ; hp = &(*hp)->next) {
		struct cache_head *tmp = *hp;
		if (detail->match(tmp, key)) {
			cache_get(tmp);
			read_unlock(&detail->hash_lock);
			return tmp;
		}
	}
	read_unlock(&detail->hash_lock);
	/* Didn't find anything, insert an empty entry */

	new = detail->alloc();
	if (!new)
		return NULL;
	/* must fully initialise 'new', else
	 * we might get lose if we need to
	 * cache_put it soon.
	 */
	cache_init(new);
	detail->init(new, key);

	write_lock(&detail->hash_lock);

	/* check if entry appeared while we slept */
	for (hp=head; *hp != NULL ; hp = &(*hp)->next) {
		struct cache_head *tmp = *hp;
		if (detail->match(tmp, key)) {
			cache_get(tmp);
			write_unlock(&detail->hash_lock);
			cache_put(new, detail);
			return tmp;
		}
	}
	new->next = *head;
	*head = new;
	detail->entries++;
	cache_get(new);
	write_unlock(&detail->hash_lock);

	return new;
}
EXPORT_SYMBOL(sunrpc_cache_lookup);


static void queue_loose(struct cache_detail *detail, struct cache_head *ch);

static int cache_fresh_locked(struct cache_head *head, time_t expiry)
{
	head->expiry_time = expiry;
	head->last_refresh = get_seconds();
	return !test_and_set_bit(CACHE_VALID, &head->flags);
}

static void cache_fresh_unlocked(struct cache_head *head,
			struct cache_detail *detail, int new)
{
	if (new)
		cache_revisit_request(head);
	if (test_and_clear_bit(CACHE_PENDING, &head->flags)) {
		cache_revisit_request(head);
		queue_loose(detail, head);
	}
}

struct cache_head *sunrpc_cache_update(struct cache_detail *detail,
				       struct cache_head *new, struct cache_head *old, int hash)
{
	/* The 'old' entry is to be replaced by 'new'.
	 * If 'old' is not VALID, we update it directly,
	 * otherwise we need to replace it
	 */
	struct cache_head **head;
	struct cache_head *tmp;
	int is_new;

	if (!test_bit(CACHE_VALID, &old->flags)) {
		write_lock(&detail->hash_lock);
		if (!test_bit(CACHE_VALID, &old->flags)) {
			if (test_bit(CACHE_NEGATIVE, &new->flags))
				set_bit(CACHE_NEGATIVE, &old->flags);
			else
				detail->update(old, new);
			is_new = cache_fresh_locked(old, new->expiry_time);
			write_unlock(&detail->hash_lock);
			cache_fresh_unlocked(old, detail, is_new);
			return old;
		}
		write_unlock(&detail->hash_lock);
	}
	/* We need to insert a new entry */
	tmp = detail->alloc();
	if (!tmp) {
		cache_put(old, detail);
		return NULL;
	}
	cache_init(tmp);
	detail->init(tmp, old);
	head = &detail->hash_table[hash];

	write_lock(&detail->hash_lock);
	if (test_bit(CACHE_NEGATIVE, &new->flags))
		set_bit(CACHE_NEGATIVE, &tmp->flags);
	else
		detail->update(tmp, new);
	tmp->next = *head;
	*head = tmp;
	detail->entries++;
	cache_get(tmp);
	is_new = cache_fresh_locked(tmp, new->expiry_time);
	cache_fresh_locked(old, 0);
	write_unlock(&detail->hash_lock);
	cache_fresh_unlocked(tmp, detail, is_new);
	cache_fresh_unlocked(old, detail, 0);
	cache_put(old, detail);
	return tmp;
}
EXPORT_SYMBOL(sunrpc_cache_update);

static int cache_make_upcall(struct cache_detail *detail, struct cache_head *h);
/*
 * This is the generic cache management routine for all
 * the authentication caches.
 * It checks the currency of a cache item and will (later)
 * initiate an upcall to fill it if needed.
 *
 *
 * Returns 0 if the cache_head can be used, or cache_puts it and returns
 * -EAGAIN if upcall is pending,
 * -ETIMEDOUT if upcall failed and should be retried,
 * -ENOENT if cache entry was negative
 */
int cache_check(struct cache_detail *detail,
		    struct cache_head *h, struct cache_req *rqstp)
{
	int rv;
	long refresh_age, age;

	/* First decide return status as best we can */
	if (!test_bit(CACHE_VALID, &h->flags) ||
	    h->expiry_time < get_seconds())
		rv = -EAGAIN;
	else if (detail->flush_time > h->last_refresh)
		rv = -EAGAIN;
	else {
		/* entry is valid */
		if (test_bit(CACHE_NEGATIVE, &h->flags))
			rv = -ENOENT;
		else rv = 0;
	}

	/* now see if we want to start an upcall */
	refresh_age = (h->expiry_time - h->last_refresh);
	age = get_seconds() - h->last_refresh;

	if (rqstp == NULL) {
		if (rv == -EAGAIN)
			rv = -ENOENT;
	} else if (rv == -EAGAIN || age > refresh_age/2) {
		dprintk("RPC:       Want update, refage=%ld, age=%ld\n",
				refresh_age, age);
		if (!test_and_set_bit(CACHE_PENDING, &h->flags)) {
			switch (cache_make_upcall(detail, h)) {
			case -EINVAL:
				clear_bit(CACHE_PENDING, &h->flags);
				if (rv == -EAGAIN) {
					set_bit(CACHE_NEGATIVE, &h->flags);
					cache_fresh_unlocked(h, detail,
					     cache_fresh_locked(h, get_seconds()+CACHE_NEW_EXPIRY));
					rv = -ENOENT;
				}
				break;

			case -EAGAIN:
				clear_bit(CACHE_PENDING, &h->flags);
				cache_revisit_request(h);
				break;
			}
		}
	}

	if (rv == -EAGAIN)
		if (cache_defer_req(rqstp, h) != 0)
			rv = -ETIMEDOUT;

	if (rv)
		cache_put(h, detail);
	return rv;
}

/*
 * caches need to be periodically cleaned.
 * For this we maintain a list of cache_detail and
 * a current pointer into that list and into the table
 * for that entry.
 *
 * Each time clean_cache is called it finds the next non-empty entry
 * in the current table and walks the list in that entry
 * looking for entries that can be removed.
 *
 * An entry gets removed if:
 * - The expiry is before current time
 * - The last_refresh time is before the flush_time for that cache
 *
 * later we might drop old entries with non-NEVER expiry if that table
 * is getting 'full' for some definition of 'full'
 *
 * The question of "how often to scan a table" is an interesting one
 * and is answered in part by the use of the "nextcheck" field in the
 * cache_detail.
 * When a scan of a table begins, the nextcheck field is set to a time
 * that is well into the future.
 * While scanning, if an expiry time is found that is earlier than the
 * current nextcheck time, nextcheck is set to that expiry time.
 * If the flush_time is ever set to a time earlier than the nextcheck
 * time, the nextcheck time is then set to that flush_time.
 *
 * A table is then only scanned if the current time is at least
 * the nextcheck time.
 *
 */

static LIST_HEAD(cache_list);
static DEFINE_SPINLOCK(cache_list_lock);
static struct cache_detail *current_detail;
static int current_index;

static const struct file_operations cache_file_operations;
static const struct file_operations content_file_operations;
static const struct file_operations cache_flush_operations;

static void do_cache_clean(struct work_struct *work);
static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean);

static void remove_cache_proc_entries(struct cache_detail *cd)
{
	if (cd->proc_ent == NULL)
		return;
	if (cd->flush_ent)
		remove_proc_entry("flush", cd->proc_ent);
	if (cd->channel_ent)
		remove_proc_entry("channel", cd->proc_ent);
	if (cd->content_ent)
		remove_proc_entry("content", cd->proc_ent);
	cd->proc_ent = NULL;
	remove_proc_entry(cd->name, proc_net_rpc);
}

#ifdef CONFIG_PROC_FS
static int create_cache_proc_entries(struct cache_detail *cd)
{
	struct proc_dir_entry *p;

	cd->proc_ent = proc_mkdir(cd->name, proc_net_rpc);
	if (cd->proc_ent == NULL)
		goto out_nomem;
	cd->proc_ent->owner = cd->owner;
	cd->channel_ent = cd->content_ent = NULL;

	p = create_proc_entry("flush", S_IFREG|S_IRUSR|S_IWUSR, cd->proc_ent);
	cd->flush_ent = p;
	if (p == NULL)
		goto out_nomem;
	p->proc_fops = &cache_flush_operations;
	p->owner = cd->owner;
	p->data = cd;

	if (cd->cache_request || cd->cache_parse) {
		p = create_proc_entry("channel", S_IFREG|S_IRUSR|S_IWUSR,
				      cd->proc_ent);
		cd->channel_ent = p;
		if (p == NULL)
			goto out_nomem;
		p->proc_fops = &cache_file_operations;
		p->owner = cd->owner;
		p->data = cd;
	}
	if (cd->cache_show) {
		p = create_proc_entry("content", S_IFREG|S_IRUSR|S_IWUSR,
				      cd->proc_ent);
		cd->content_ent = p;
		if (p == NULL)
			goto out_nomem;
		p->proc_fops = &content_file_operations;
		p->owner = cd->owner;
		p->data = cd;
	}
	return 0;
out_nomem:
	remove_cache_proc_entries(cd);
	return -ENOMEM;
}
#else /* CONFIG_PROC_FS */
static int create_cache_proc_entries(struct cache_detail *cd)
{
	return 0;
}
#endif

int cache_register(struct cache_detail *cd)
{
	int ret;

	ret = create_cache_proc_entries(cd);
	if (ret)
		return ret;
	rwlock_init(&cd->hash_lock);
	INIT_LIST_HEAD(&cd->queue);
	spin_lock(&cache_list_lock);
	cd->nextcheck = 0;
	cd->entries = 0;
	atomic_set(&cd->readers, 0);
	cd->last_close = 0;
	cd->last_warn = -1;
	list_add(&cd->others, &cache_list);
	spin_unlock(&cache_list_lock);

	/* start the cleaning process */
	schedule_delayed_work(&cache_cleaner, 0);
	return 0;
}

void cache_unregister(struct cache_detail *cd)
{
	cache_purge(cd);
	spin_lock(&cache_list_lock);
	write_lock(&cd->hash_lock);
	if (cd->entries || atomic_read(&cd->inuse)) {
		write_unlock(&cd->hash_lock);
		spin_unlock(&cache_list_lock);
		goto out;
	}
	if (current_detail == cd)
		current_detail = NULL;
	list_del_init(&cd->others);
	write_unlock(&cd->hash_lock);
	spin_unlock(&cache_list_lock);
	remove_cache_proc_entries(cd);
	if (list_empty(&cache_list)) {
		/* module must be being unloaded so its safe to kill the worker */
		cancel_delayed_work_sync(&cache_cleaner);
	}
	return;
out:
	printk(KERN_ERR "nfsd: failed to unregister %s cache\n", cd->name);
}

/* clean cache tries to find something to clean
 * and cleans it.
 * It returns 1 if it cleaned something,
 *            0 if it didn't find anything this time
 *           -1 if it fell off the end of the list.
 */
static int cache_clean(void)
{
	int rv = 0;
	struct list_head *next;

	spin_lock(&cache_list_lock);

	/* find a suitable table if we don't already have one */
	while (current_detail == NULL ||
	    current_index >= current_detail->hash_size) {
		if (current_detail)
			next = current_detail->others.next;
		else
			next = cache_list.next;
		if (next == &cache_list) {
			current_detail = NULL;
			spin_unlock(&cache_list_lock);
			return -1;
		}
		current_detail = list_entry(next, struct cache_detail, others);
		if (current_detail->nextcheck > get_seconds())
			current_index = current_detail->hash_size;
		else {
			current_index = 0;
			current_detail->nextcheck = get_seconds()+30*60;
		}
	}

	/* find a non-empty bucket in the table */
	while (current_detail &&
	       current_index < current_detail->hash_size &&
	       current_detail->hash_table[current_index] == NULL)
		current_index++;

	/* find a cleanable entry in the bucket and clean it, or set to next bucket */

	if (current_detail && current_index < current_detail->hash_size) {
		struct cache_head *ch, **cp;
		struct cache_detail *d;

		write_lock(&current_detail->hash_lock);

		/* Ok, now to clean this strand */

		cp = & current_detail->hash_table[current_index];
		ch = *cp;
		for (; ch; cp= & ch->next, ch= *cp) {
			if (current_detail->nextcheck > ch->expiry_time)
				current_detail->nextcheck = ch->expiry_time+1;
			if (ch->expiry_time >= get_seconds()
			    && ch->last_refresh >= current_detail->flush_time
				)
				continue;
			if (test_and_clear_bit(CACHE_PENDING, &ch->flags))
				queue_loose(current_detail, ch);

			if (atomic_read(&ch->ref.refcount) == 1)
				break;
		}
		if (ch) {
			*cp = ch->next;
			ch->next = NULL;
			current_detail->entries--;
			rv = 1;
		}
		write_unlock(&current_detail->hash_lock);
		d = current_detail;
		if (!ch)
			current_index ++;
		spin_unlock(&cache_list_lock);
		if (ch)
			cache_put(ch, d);
	} else
		spin_unlock(&cache_list_lock);

	return rv;
}

/*
 * We want to regularly clean the cache, so we need to schedule some work ...
 */
static void do_cache_clean(struct work_struct *work)
{
	int delay = 5;
	if (cache_clean() == -1)
		delay = 30*HZ;

	if (list_empty(&cache_list))
		delay = 0;

	if (delay)
		schedule_delayed_work(&cache_cleaner, delay);
}


/*
 * Clean all caches promptly.  This just calls cache_clean
 * repeatedly until we are sure that every cache has had a chance to
 * be fully cleaned
 */
void cache_flush(void)
{
	while (cache_clean() != -1)
		cond_resched();
	while (cache_clean() != -1)
		cond_resched();
}

void cache_purge(struct cache_detail *detail)
{
	detail->flush_time = LONG_MAX;
	detail->nextcheck = get_seconds();
	cache_flush();
	detail->flush_time = 1;
}



/*
 * Deferral and Revisiting of Requests.
 *
 * If a cache lookup finds a pending entry, we
 * need to defer the request and revisit it later.
 * All deferred requests are stored in a hash table,
 * indexed by "struct cache_head *".
 * As it may be wasteful to store a whole request
 * structure, we allow the request to provide a
 * deferred form, which must contain a
 * 'struct cache_deferred_req'
 * This cache_deferred_req contains a method to allow
 * it to be revisited when cache info is available
 */

#define	DFR_HASHSIZE	(PAGE_SIZE/sizeof(struct list_head))
#define	DFR_HASH(item)	((((long)item)>>4 ^ (((long)item)>>13)) % DFR_HASHSIZE)

#define	DFR_MAX	300	/* ??? */

static DEFINE_SPINLOCK(cache_defer_lock);
static LIST_HEAD(cache_defer_list);
static struct list_head cache_defer_hash[DFR_HASHSIZE];
static int cache_defer_cnt;

static int cache_defer_req(struct cache_req *req, struct cache_head *item)
{
	struct cache_deferred_req *dreq;
	int hash = DFR_HASH(item);

	if (cache_defer_cnt >= DFR_MAX) {
		/* too much in the cache, randomly drop this one,
		 * or continue and drop the oldest below
		 */
		if (net_random()&1)
			return -ETIMEDOUT;
	}
	dreq = req->defer(req);
	if (dreq == NULL)
		return -ETIMEDOUT;

	dreq->item = item;
	dreq->recv_time = get_seconds();

	spin_lock(&cache_defer_lock);

	list_add(&dreq->recent, &cache_defer_list);

	if (cache_defer_hash[hash].next == NULL)
		INIT_LIST_HEAD(&cache_defer_hash[hash]);
	list_add(&dreq->hash, &cache_defer_hash[hash]);

	/* it is in, now maybe clean up */
	dreq = NULL;
	if (++cache_defer_cnt > DFR_MAX) {
		dreq = list_entry(cache_defer_list.prev,
				  struct cache_deferred_req, recent);
		list_del(&dreq->recent);
		list_del(&dreq->hash);
		cache_defer_cnt--;
	}
	spin_unlock(&cache_defer_lock);

	if (dreq) {
		/* there was one too many */
		dreq->revisit(dreq, 1);
	}
	if (!test_bit(CACHE_PENDING, &item->flags)) {
		/* must have just been validated... */
		cache_revisit_request(item);
	}
	return 0;
}

static void cache_revisit_request(struct cache_head *item)
{
	struct cache_deferred_req *dreq;
	struct list_head pending;

	struct list_head *lp;
	int hash = DFR_HASH(item);

	INIT_LIST_HEAD(&pending);
	spin_lock(&cache_defer_lock);

	lp = cache_defer_hash[hash].next;
	if (lp) {
		while (lp != &cache_defer_hash[hash]) {
			dreq = list_entry(lp, struct cache_deferred_req, hash);
			lp = lp->next;
			if (dreq->item == item) {
				list_del(&dreq->hash);
				list_move(&dreq->recent, &pending);
				cache_defer_cnt--;
			}
		}
	}
	spin_unlock(&cache_defer_lock);

	while (!list_empty(&pending)) {
		dreq = list_entry(pending.next, struct cache_deferred_req, recent);
		list_del_init(&dreq->recent);
		dreq->revisit(dreq, 0);
	}
}

void cache_clean_deferred(void *owner)
{
	struct cache_deferred_req *dreq, *tmp;
	struct list_head pending;


	INIT_LIST_HEAD(&pending);
	spin_lock(&cache_defer_lock);

	list_for_each_entry_safe(dreq, tmp, &cache_defer_list, recent) {
		if (dreq->owner == owner) {
			list_del(&dreq->hash);
			list_move(&dreq->recent, &pending);
			cache_defer_cnt--;
		}
	}
	spin_unlock(&cache_defer_lock);

	while (!list_empty(&pending)) {
		dreq = list_entry(pending.next, struct cache_deferred_req, recent);
		list_del_init(&dreq->recent);
		dreq->revisit(dreq, 1);
	}
}

/*
 * communicate with user-space
 *
 * We have a magic /proc file - /proc/sunrpc/<cachename>/channel.
 * On read, you get a full request, or block.
 * On write, an update request is processed.
 * Poll works if anything to read, and always allows write.
 *
 * Implemented by linked list of requests.  Each open file has
 * a ->private that also exists in this list.  New requests are added
 * to the end and may wakeup and preceding readers.
 * New readers are added to the head.  If, on read, an item is found with
 * CACHE_UPCALLING clear, we free it from the list.
 *
 */

static DEFINE_SPINLOCK(queue_lock);
static DEFINE_MUTEX(queue_io_mutex);

struct cache_queue {
	struct list_head	list;
	int			reader;	/* if 0, then request */
};
struct cache_request {
	struct cache_queue	q;
	struct cache_head	*item;
	char			* buf;
	int			len;
	int			readers;
};
struct cache_reader {
	struct cache_queue	q;
	int			offset;	/* if non-0, we have a refcnt on next request */
};

static ssize_t
cache_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
{
	struct cache_reader *rp = filp->private_data;
	struct cache_request *rq;
	struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data;
	int err;

	if (count == 0)
		return 0;

	mutex_lock(&queue_io_mutex); /* protect against multiple concurrent
			      * readers on this file */
 again:
	spin_lock(&queue_lock);
	/* need to find next request */
	while (rp->q.list.next != &cd->queue &&
	       list_entry(rp->q.list.next, struct cache_queue, list)
	       ->reader) {
		struct list_head *next = rp->q.list.next;
		list_move(&rp->q.list, next);
	}
	if (rp->q.list.next == &cd->queue) {
		spin_unlock(&queue_lock);
		mutex_unlock(&queue_io_mutex);
		BUG_ON(rp->offset);
		return 0;
	}
	rq = container_of(rp->q.list.next, struct cache_request, q.list);
	BUG_ON(rq->q.reader);
	if (rp->offset == 0)
		rq->readers++;
	spin_unlock(&queue_lock);

	if (rp->offset == 0 && !test_bit(CACHE_PENDING, &rq->item->flags)) {
		err = -EAGAIN;
		spin_lock(&queue_lock);
		list_move(&rp->q.list, &rq->q.list);
		spin_unlock(&queue_lock);
	} else {
		if (rp->offset + count > rq->len)
			count = rq->len - rp->offset;
		err = -EFAULT;
		if (copy_to_user(buf, rq->buf + rp->offset, count))
			goto out;
		rp->offset += count;
		if (rp->offset >= rq->len) {
			rp->offset = 0;
			spin_lock(&queue_lock);
			list_move(&rp->q.list, &rq->q.list);
			spin_unlock(&queue_lock);
		}
		err = 0;
	}
 out:
	if (rp->offset == 0) {
		/* need to release rq */
		spin_lock(&queue_lock);
		rq->readers--;
		if (rq->readers == 0 &&
		    !test_bit(CACHE_PENDING, &rq->item->flags)) {
			list_del(&rq->q.list);
			spin_unlock(&queue_lock);
			cache_put(rq->item, cd);
			kfree(rq->buf);
			kfree(rq);
		} else
			spin_unlock(&queue_lock);
	}
	if (err == -EAGAIN)
		goto again;
	mutex_unlock(&queue_io_mutex);
	return err ? err :  count;
}

static char write_buf[8192]; /* protected by queue_io_mutex */

static ssize_t
cache_write(struct file *filp, const char __user *buf, size_t count,
	    loff_t *ppos)
{
	int err;
	struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data;

	if (count == 0)
		return 0;
	if (count >= sizeof(write_buf))
		return -EINVAL;

	mutex_lock(&queue_io_mutex);

	if (copy_from_user(write_buf, buf, count)) {
		mutex_unlock(&queue_io_mutex);
		return -EFAULT;
	}
	write_buf[count] = '\0';
	if (cd->cache_parse)
		err = cd->cache_parse(cd, write_buf, count);
	else
		err = -EINVAL;

	mutex_unlock(&queue_io_mutex);
	return err ? err : count;
}

static DECLARE_WAIT_QUEUE_HEAD(queue_wait);

static unsigned int
cache_poll(struct file *filp, poll_table *wait)
{
	unsigned int mask;
	struct cache_reader *rp = filp->private_data;
	struct cache_queue *cq;
	struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data;

	poll_wait(filp, &queue_wait, wait);

	/* alway allow write */
	mask = POLL_OUT | POLLWRNORM;

	if (!rp)
		return mask;

	spin_lock(&queue_lock);

	for (cq= &rp->q; &cq->list != &cd->queue;
	     cq = list_entry(cq->list.next, struct cache_queue, list))
		if (!cq->reader) {
			mask |= POLLIN | POLLRDNORM;
			break;
		}
	spin_unlock(&queue_lock);
	return mask;
}

static int
cache_ioctl(struct inode *ino, struct file *filp,
	    unsigned int cmd, unsigned long arg)
{
	int len = 0;
	struct cache_reader *rp = filp->private_data;
	struct cache_queue *cq;
	struct cache_detail *cd = PDE(ino)->data;

	if (cmd != FIONREAD || !rp)
		return -EINVAL;

	spin_lock(&queue_lock);

	/* only find the length remaining in current request,
	 * or the length of the next request
	 */
	for (cq= &rp->q; &cq->list != &cd->queue;
	     cq = list_entry(cq->list.next, struct cache_queue, list))
		if (!cq->reader) {
			struct cache_request *cr =
				container_of(cq, struct cache_request, q);
			len = cr->len - rp->offset;
			break;
		}
	spin_unlock(&queue_lock);

	return put_user(len, (int __user *)arg);
}

static int
cache_open(struct inode *inode, struct file *filp)
{
	struct cache_reader *rp = NULL;

	nonseekable_open(inode, filp);
	if (filp->f_mode & FMODE_READ) {
		struct cache_detail *cd = PDE(inode)->data;

		rp = kmalloc(sizeof(*rp), GFP_KERNEL);
		if (!rp)
			return -ENOMEM;
		rp->offset = 0;
		rp->q.reader = 1;
		atomic_inc(&cd->readers);
		spin_lock(&queue_lock);
		list_add(&rp->q.list, &cd->queue);
		spin_unlock(&queue_lock);
	}
	filp->private_data = rp;
	return 0;
}

static int
cache_release(struct inode *inode, struct file *filp)
{
	struct cache_reader *rp = filp->private_data;
	struct cache_detail *cd = PDE(inode)->data;

	if (rp) {
		spin_lock(&queue_lock);
		if (rp->offset) {
			struct cache_queue *cq;
			for (cq= &rp->q; &cq->list != &cd->queue;
			     cq = list_entry(cq->list.next, struct cache_queue, list))
				if (!cq->reader) {
					container_of(cq, struct cache_request, q)
						->readers--;
					break;
				}
			rp->offset = 0;
		}
		list_del(&rp->q.list);
		spin_unlock(&queue_lock);

		filp->private_data = NULL;
		kfree(rp);

		cd->last_close = get_seconds();
		atomic_dec(&cd->readers);
	}
	return 0;
}



static const struct file_operations cache_file_operations = {
	.owner		= THIS_MODULE,
	.llseek		= no_llseek,
	.read		= cache_read,
	.write		= cache_write,
	.poll		= cache_poll,
	.ioctl		= cache_ioctl, /* for FIONREAD */
	.open		= cache_open,
	.release	= cache_release,
};


static void queue_loose(struct cache_detail *detail, struct cache_head *ch)
{
	struct cache_queue *cq;
	spin_lock(&queue_lock);
	list_for_each_entry(cq, &detail->queue, list)
		if (!cq->reader) {
			struct cache_request *cr = container_of(cq, struct cache_request, q);
			if (cr->item != ch)
				continue;
			if (cr->readers != 0)
				continue;
			list_del(&cr->q.list);
			spin_unlock(&queue_lock);
			cache_put(cr->item, detail);
			kfree(cr->buf);
			kfree(cr);
			return;
		}
	spin_unlock(&queue_lock);
}

/*
 * Support routines for text-based upcalls.
 * Fields are separated by spaces.
 * Fields are either mangled to quote space tab newline slosh with slosh
 * or a hexified with a leading \x
 * Record is terminated with newline.
 *
 */

void qword_add(char **bpp, int *lp, char *str)
{
	char *bp = *bpp;
	int len = *lp;
	char c;

	if (len < 0) return;

	while ((c=*str++) && len)
		switch(c) {
		case ' ':
		case '\t':
		case '\n':
		case '\\':
			if (len >= 4) {
				*bp++ = '\\';
				*bp++ = '0' + ((c & 0300)>>6);
				*bp++ = '0' + ((c & 0070)>>3);
				*bp++ = '0' + ((c & 0007)>>0);
			}
			len -= 4;
			break;
		default:
			*bp++ = c;
			len--;
		}
	if (c || len <1) len = -1;
	else {
		*bp++ = ' ';
		len--;
	}
	*bpp = bp;
	*lp = len;
}

void qword_addhex(char **bpp, int *lp, char *buf, int blen)
{
	char *bp = *bpp;
	int len = *lp;

	if (len < 0) return;

	if (len > 2) {
		*bp++ = '\\';
		*bp++ = 'x';
		len -= 2;
		while (blen && len >= 2) {
			unsigned char c = *buf++;
			*bp++ = '0' + ((c&0xf0)>>4) + (c>=0xa0)*('a'-'9'-1);
			*bp++ = '0' + (c&0x0f) + ((c&0x0f)>=0x0a)*('a'-'9'-1);
			len -= 2;
			blen--;
		}
	}
	if (blen || len<1) len = -1;
	else {
		*bp++ = ' ';
		len--;
	}
	*bpp = bp;
	*lp = len;
}

static void warn_no_listener(struct cache_detail *detail)
{
	if (detail->last_warn != detail->last_close) {
		detail->last_warn = detail->last_close;
		if (detail->warn_no_listener)
			detail->warn_no_listener(detail);
	}
}

/*
 * register an upcall request to user-space.
 * Each request is at most one page long.
 */
static int cache_make_upcall(struct cache_detail *detail, struct cache_head *h)
{

	char *buf;
	struct cache_request *crq;
	char *bp;
	int len;

	if (detail->cache_request == NULL)
		return -EINVAL;

	if (atomic_read(&detail->readers) == 0 &&
	    detail->last_close < get_seconds() - 30) {
			warn_no_listener(detail);
			return -EINVAL;
	}

	buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
	if (!buf)
		return -EAGAIN;

	crq = kmalloc(sizeof (*crq), GFP_KERNEL);
	if (!crq) {
		kfree(buf);
		return -EAGAIN;
	}

	bp = buf; len = PAGE_SIZE;

	detail->cache_request(detail, h, &bp, &len);

	if (len < 0) {
		kfree(buf);
		kfree(crq);
		return -EAGAIN;
	}
	crq->q.reader = 0;
	crq->item = cache_get(h);
	crq->buf = buf;
	crq->len = PAGE_SIZE - len;
	crq->readers = 0;
	spin_lock(&queue_lock);
	list_add_tail(&crq->q.list, &detail->queue);
	spin_unlock(&queue_lock);
	wake_up(&queue_wait);
	return 0;
}

/*
 * parse a message from user-space and pass it
 * to an appropriate cache
 * Messages are, like requests, separated into fields by
 * spaces and dequotes as \xHEXSTRING or embedded \nnn octal
 *
 * Message is
 *   reply cachename expiry key ... content....
 *
 * key and content are both parsed by cache
 */

#define isodigit(c) (isdigit(c) && c <= '7')
int qword_get(char **bpp, char *dest, int bufsize)
{
	/* return bytes copied, or -1 on error */
	char *bp = *bpp;
	int len = 0;

	while (*bp == ' ') bp++;

	if (bp[0] == '\\' && bp[1] == 'x') {
		/* HEX STRING */
		bp += 2;
		while (isxdigit(bp[0]) && isxdigit(bp[1]) && len < bufsize) {
			int byte = isdigit(*bp) ? *bp-'0' : toupper(*bp)-'A'+10;
			bp++;
			byte <<= 4;
			byte |= isdigit(*bp) ? *bp-'0' : toupper(*bp)-'A'+10;
			*dest++ = byte;
			bp++;
			len++;
		}
	} else {
		/* text with \nnn octal quoting */
		while (*bp != ' ' && *bp != '\n' && *bp && len < bufsize-1) {
			if (*bp == '\\' &&
			    isodigit(bp[1]) && (bp[1] <= '3') &&
			    isodigit(bp[2]) &&
			    isodigit(bp[3])) {
				int byte = (*++bp -'0');
				bp++;
				byte = (byte << 3) | (*bp++ - '0');
				byte = (byte << 3) | (*bp++ - '0');
				*dest++ = byte;
				len++;
			} else {
				*dest++ = *bp++;
				len++;
			}
		}
	}

	if (*bp != ' ' && *bp != '\n' && *bp != '\0')
		return -1;
	while (*bp == ' ') bp++;
	*bpp = bp;
	*dest = '\0';
	return len;
}


/*
 * support /proc/sunrpc/cache/$CACHENAME/content
 * as a seqfile.
 * We call ->cache_show passing NULL for the item to
 * get a header, then pass each real item in the cache
 */

struct handle {
	struct cache_detail *cd;
};

static void *c_start(struct seq_file *m, loff_t *pos)
	__acquires(cd->hash_lock)
{
	loff_t n = *pos;
	unsigned hash, entry;
	struct cache_head *ch;
	struct cache_detail *cd = ((struct handle*)m->private)->cd;


	read_lock(&cd->hash_lock);
	if (!n--)
		return SEQ_START_TOKEN;
	hash = n >> 32;
	entry = n & ((1LL<<32) - 1);

	for (ch=cd->hash_table[hash]; ch; ch=ch->next)
		if (!entry--)
			return ch;
	n &= ~((1LL<<32) - 1);
	do {
		hash++;
		n += 1LL<<32;
	} while(hash < cd->hash_size &&
		cd->hash_table[hash]==NULL);
	if (hash >= cd->hash_size)
		return NULL;
	*pos = n+1;
	return cd->hash_table[hash];
}

static void *c_next(struct seq_file *m, void *p, loff_t *pos)
{
	struct cache_head *ch = p;
	int hash = (*pos >> 32);
	struct cache_detail *cd = ((struct handle*)m->private)->cd;

	if (p == SEQ_START_TOKEN)
		hash = 0;
	else if (ch->next == NULL) {
		hash++;
		*pos += 1LL<<32;
	} else {
		++*pos;
		return ch->next;
	}
	*pos &= ~((1LL<<32) - 1);
	while (hash < cd->hash_size &&
	       cd->hash_table[hash] == NULL) {
		hash++;
		*pos += 1LL<<32;
	}
	if (hash >= cd->hash_size)
		return NULL;
	++*pos;
	return cd->hash_table[hash];
}

static void c_stop(struct seq_file *m, void *p)
	__releases(cd->hash_lock)
{
	struct cache_detail *cd = ((struct handle*)m->private)->cd;
	read_unlock(&cd->hash_lock);
}

static int c_show(struct seq_file *m, void *p)
{
	struct cache_head *cp = p;
	struct cache_detail *cd = ((struct handle*)m->private)->cd;

	if (p == SEQ_START_TOKEN)
		return cd->cache_show(m, cd, NULL);

	ifdebug(CACHE)
		seq_printf(m, "# expiry=%ld refcnt=%d flags=%lx\n",
			   cp->expiry_time, atomic_read(&cp->ref.refcount), cp->flags);
	cache_get(cp);
	if (cache_check(cd, cp, NULL))
		/* cache_check does a cache_put on failure */
		seq_printf(m, "# ");
	else
		cache_put(cp, cd);

	return cd->cache_show(m, cd, cp);
}

static const struct seq_operations cache_content_op = {
	.start	= c_start,
	.next	= c_next,
	.stop	= c_stop,
	.show	= c_show,
};

static int content_open(struct inode *inode, struct file *file)
{
	struct handle *han;
	struct cache_detail *cd = PDE(inode)->data;

	han = __seq_open_private(file, &cache_content_op, sizeof(*han));
	if (han == NULL)
		return -ENOMEM;

	han->cd = cd;
	return 0;
}

static const struct file_operations content_file_operations = {
	.open		= content_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= seq_release_private,
};

static ssize_t read_flush(struct file *file, char __user *buf,
			    size_t count, loff_t *ppos)
{
	struct cache_detail *cd = PDE(file->f_path.dentry->d_inode)->data;
	char tbuf[20];
	unsigned long p = *ppos;
	size_t len;

	sprintf(tbuf, "%lu\n", cd->flush_time);
	len = strlen(tbuf);
	if (p >= len)
		return 0;
	len -= p;
	if (len > count)
		len = count;
	if (copy_to_user(buf, (void*)(tbuf+p), len))
		return -EFAULT;
	*ppos += len;
	return len;
}

static ssize_t write_flush(struct file * file, const char __user * buf,
			     size_t count, loff_t *ppos)
{
	struct cache_detail *cd = PDE(file->f_path.dentry->d_inode)->data;
	char tbuf[20];
	char *ep;
	long flushtime;
	if (*ppos || count > sizeof(tbuf)-1)
		return -EINVAL;
	if (copy_from_user(tbuf, buf, count))
		return -EFAULT;
	tbuf[count] = 0;
	flushtime = simple_strtoul(tbuf, &ep, 0);
	if (*ep && *ep != '\n')
		return -EINVAL;

	cd->flush_time = flushtime;
	cd->nextcheck = get_seconds();
	cache_flush();

	*ppos += count;
	return count;
}

static const struct file_operations cache_flush_operations = {
	.open		= nonseekable_open,
	.read		= read_flush,
	.write		= write_flush,
};