forked from torvalds/linux
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathxfs_dquot.c
1167 lines (1024 loc) · 29.3 KB
/
xfs_dquot.c
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
/*
* Copyright (c) 2000-2003 Silicon Graphics, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_shared.h"
#include "xfs_trans_resv.h"
#include "xfs_bit.h"
#include "xfs_mount.h"
#include "xfs_defer.h"
#include "xfs_inode.h"
#include "xfs_bmap.h"
#include "xfs_bmap_util.h"
#include "xfs_alloc.h"
#include "xfs_quota.h"
#include "xfs_error.h"
#include "xfs_trans.h"
#include "xfs_buf_item.h"
#include "xfs_trans_space.h"
#include "xfs_trans_priv.h"
#include "xfs_qm.h"
#include "xfs_cksum.h"
#include "xfs_trace.h"
#include "xfs_log.h"
#include "xfs_bmap_btree.h"
/*
* Lock order:
*
* ip->i_lock
* qi->qi_tree_lock
* dquot->q_qlock (xfs_dqlock() and friends)
* dquot->q_flush (xfs_dqflock() and friends)
* qi->qi_lru_lock
*
* If two dquots need to be locked the order is user before group/project,
* otherwise by the lowest id first, see xfs_dqlock2.
*/
struct kmem_zone *xfs_qm_dqtrxzone;
static struct kmem_zone *xfs_qm_dqzone;
static struct lock_class_key xfs_dquot_group_class;
static struct lock_class_key xfs_dquot_project_class;
/*
* This is called to free all the memory associated with a dquot
*/
void
xfs_qm_dqdestroy(
xfs_dquot_t *dqp)
{
ASSERT(list_empty(&dqp->q_lru));
kmem_free(dqp->q_logitem.qli_item.li_lv_shadow);
mutex_destroy(&dqp->q_qlock);
XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot);
kmem_zone_free(xfs_qm_dqzone, dqp);
}
/*
* If default limits are in force, push them into the dquot now.
* We overwrite the dquot limits only if they are zero and this
* is not the root dquot.
*/
void
xfs_qm_adjust_dqlimits(
struct xfs_mount *mp,
struct xfs_dquot *dq)
{
struct xfs_quotainfo *q = mp->m_quotainfo;
struct xfs_disk_dquot *d = &dq->q_core;
struct xfs_def_quota *defq;
int prealloc = 0;
ASSERT(d->d_id);
defq = xfs_get_defquota(dq, q);
if (defq->bsoftlimit && !d->d_blk_softlimit) {
d->d_blk_softlimit = cpu_to_be64(defq->bsoftlimit);
prealloc = 1;
}
if (defq->bhardlimit && !d->d_blk_hardlimit) {
d->d_blk_hardlimit = cpu_to_be64(defq->bhardlimit);
prealloc = 1;
}
if (defq->isoftlimit && !d->d_ino_softlimit)
d->d_ino_softlimit = cpu_to_be64(defq->isoftlimit);
if (defq->ihardlimit && !d->d_ino_hardlimit)
d->d_ino_hardlimit = cpu_to_be64(defq->ihardlimit);
if (defq->rtbsoftlimit && !d->d_rtb_softlimit)
d->d_rtb_softlimit = cpu_to_be64(defq->rtbsoftlimit);
if (defq->rtbhardlimit && !d->d_rtb_hardlimit)
d->d_rtb_hardlimit = cpu_to_be64(defq->rtbhardlimit);
if (prealloc)
xfs_dquot_set_prealloc_limits(dq);
}
/*
* Check the limits and timers of a dquot and start or reset timers
* if necessary.
* This gets called even when quota enforcement is OFF, which makes our
* life a little less complicated. (We just don't reject any quota
* reservations in that case, when enforcement is off).
* We also return 0 as the values of the timers in Q_GETQUOTA calls, when
* enforcement's off.
* In contrast, warnings are a little different in that they don't
* 'automatically' get started when limits get exceeded. They do
* get reset to zero, however, when we find the count to be under
* the soft limit (they are only ever set non-zero via userspace).
*/
void
xfs_qm_adjust_dqtimers(
xfs_mount_t *mp,
xfs_disk_dquot_t *d)
{
ASSERT(d->d_id);
#ifdef DEBUG
if (d->d_blk_hardlimit)
ASSERT(be64_to_cpu(d->d_blk_softlimit) <=
be64_to_cpu(d->d_blk_hardlimit));
if (d->d_ino_hardlimit)
ASSERT(be64_to_cpu(d->d_ino_softlimit) <=
be64_to_cpu(d->d_ino_hardlimit));
if (d->d_rtb_hardlimit)
ASSERT(be64_to_cpu(d->d_rtb_softlimit) <=
be64_to_cpu(d->d_rtb_hardlimit));
#endif
if (!d->d_btimer) {
if ((d->d_blk_softlimit &&
(be64_to_cpu(d->d_bcount) >
be64_to_cpu(d->d_blk_softlimit))) ||
(d->d_blk_hardlimit &&
(be64_to_cpu(d->d_bcount) >
be64_to_cpu(d->d_blk_hardlimit)))) {
d->d_btimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_btimelimit);
} else {
d->d_bwarns = 0;
}
} else {
if ((!d->d_blk_softlimit ||
(be64_to_cpu(d->d_bcount) <=
be64_to_cpu(d->d_blk_softlimit))) &&
(!d->d_blk_hardlimit ||
(be64_to_cpu(d->d_bcount) <=
be64_to_cpu(d->d_blk_hardlimit)))) {
d->d_btimer = 0;
}
}
if (!d->d_itimer) {
if ((d->d_ino_softlimit &&
(be64_to_cpu(d->d_icount) >
be64_to_cpu(d->d_ino_softlimit))) ||
(d->d_ino_hardlimit &&
(be64_to_cpu(d->d_icount) >
be64_to_cpu(d->d_ino_hardlimit)))) {
d->d_itimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_itimelimit);
} else {
d->d_iwarns = 0;
}
} else {
if ((!d->d_ino_softlimit ||
(be64_to_cpu(d->d_icount) <=
be64_to_cpu(d->d_ino_softlimit))) &&
(!d->d_ino_hardlimit ||
(be64_to_cpu(d->d_icount) <=
be64_to_cpu(d->d_ino_hardlimit)))) {
d->d_itimer = 0;
}
}
if (!d->d_rtbtimer) {
if ((d->d_rtb_softlimit &&
(be64_to_cpu(d->d_rtbcount) >
be64_to_cpu(d->d_rtb_softlimit))) ||
(d->d_rtb_hardlimit &&
(be64_to_cpu(d->d_rtbcount) >
be64_to_cpu(d->d_rtb_hardlimit)))) {
d->d_rtbtimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_rtbtimelimit);
} else {
d->d_rtbwarns = 0;
}
} else {
if ((!d->d_rtb_softlimit ||
(be64_to_cpu(d->d_rtbcount) <=
be64_to_cpu(d->d_rtb_softlimit))) &&
(!d->d_rtb_hardlimit ||
(be64_to_cpu(d->d_rtbcount) <=
be64_to_cpu(d->d_rtb_hardlimit)))) {
d->d_rtbtimer = 0;
}
}
}
/*
* initialize a buffer full of dquots and log the whole thing
*/
STATIC void
xfs_qm_init_dquot_blk(
xfs_trans_t *tp,
xfs_mount_t *mp,
xfs_dqid_t id,
uint type,
xfs_buf_t *bp)
{
struct xfs_quotainfo *q = mp->m_quotainfo;
xfs_dqblk_t *d;
xfs_dqid_t curid;
int i;
ASSERT(tp);
ASSERT(xfs_buf_islocked(bp));
d = bp->b_addr;
/*
* ID of the first dquot in the block - id's are zero based.
*/
curid = id - (id % q->qi_dqperchunk);
memset(d, 0, BBTOB(q->qi_dqchunklen));
for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) {
d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
d->dd_diskdq.d_id = cpu_to_be32(curid);
d->dd_diskdq.d_flags = type;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid);
xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
XFS_DQUOT_CRC_OFF);
}
}
xfs_trans_dquot_buf(tp, bp,
(type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF :
((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF :
XFS_BLF_GDQUOT_BUF)));
xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1);
}
/*
* Initialize the dynamic speculative preallocation thresholds. The lo/hi
* watermarks correspond to the soft and hard limits by default. If a soft limit
* is not specified, we use 95% of the hard limit.
*/
void
xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp)
{
uint64_t space;
dqp->q_prealloc_hi_wmark = be64_to_cpu(dqp->q_core.d_blk_hardlimit);
dqp->q_prealloc_lo_wmark = be64_to_cpu(dqp->q_core.d_blk_softlimit);
if (!dqp->q_prealloc_lo_wmark) {
dqp->q_prealloc_lo_wmark = dqp->q_prealloc_hi_wmark;
do_div(dqp->q_prealloc_lo_wmark, 100);
dqp->q_prealloc_lo_wmark *= 95;
}
space = dqp->q_prealloc_hi_wmark;
do_div(space, 100);
dqp->q_low_space[XFS_QLOWSP_1_PCNT] = space;
dqp->q_low_space[XFS_QLOWSP_3_PCNT] = space * 3;
dqp->q_low_space[XFS_QLOWSP_5_PCNT] = space * 5;
}
/*
* Allocate a block and fill it with dquots.
* This is called when the bmapi finds a hole.
*/
STATIC int
xfs_qm_dqalloc(
xfs_trans_t **tpp,
xfs_mount_t *mp,
xfs_dquot_t *dqp,
xfs_inode_t *quotip,
xfs_fileoff_t offset_fsb,
xfs_buf_t **O_bpp)
{
xfs_fsblock_t firstblock;
struct xfs_defer_ops dfops;
xfs_bmbt_irec_t map;
int nmaps, error;
xfs_buf_t *bp;
xfs_trans_t *tp = *tpp;
ASSERT(tp != NULL);
trace_xfs_dqalloc(dqp);
/*
* Initialize the bmap freelist prior to calling bmapi code.
*/
xfs_defer_init(&dfops, &firstblock);
xfs_ilock(quotip, XFS_ILOCK_EXCL);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return -ESRCH;
}
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
error = xfs_bmapi_write(tp, quotip, offset_fsb,
XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA,
&firstblock, XFS_QM_DQALLOC_SPACE_RES(mp),
&map, &nmaps, &dfops);
if (error)
goto error0;
ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
ASSERT(nmaps == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
/*
* Keep track of the blkno to save a lookup later
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0);
if (!bp) {
error = -ENOMEM;
goto error1;
}
bp->b_ops = &xfs_dquot_buf_ops;
/*
* Make a chunk of dquots out of this buffer and log
* the entire thing.
*/
xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id),
dqp->dq_flags & XFS_DQ_ALLTYPES, bp);
/*
* xfs_defer_finish() may commit the current transaction and
* start a second transaction if the freelist is not empty.
*
* Since we still want to modify this buffer, we need to
* ensure that the buffer is not released on commit of
* the first transaction and ensure the buffer is added to the
* second transaction.
*
* If there is only one transaction then don't stop the buffer
* from being released when it commits later on.
*/
xfs_trans_bhold(tp, bp);
error = xfs_defer_finish(tpp, &dfops);
if (error)
goto error1;
/* Transaction was committed? */
if (*tpp != tp) {
tp = *tpp;
xfs_trans_bjoin(tp, bp);
} else {
xfs_trans_bhold_release(tp, bp);
}
*O_bpp = bp;
return 0;
error1:
xfs_defer_cancel(&dfops);
error0:
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return error;
}
STATIC int
xfs_qm_dqrepair(
struct xfs_mount *mp,
struct xfs_trans *tp,
struct xfs_dquot *dqp,
xfs_dqid_t firstid,
struct xfs_buf **bpp)
{
int error;
struct xfs_disk_dquot *ddq;
struct xfs_dqblk *d;
int i;
/*
* Read the buffer without verification so we get the corrupted
* buffer returned to us. make sure we verify it on write, though.
*/
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0, bpp, NULL);
if (error) {
ASSERT(*bpp == NULL);
return error;
}
(*bpp)->b_ops = &xfs_dquot_buf_ops;
ASSERT(xfs_buf_islocked(*bpp));
d = (struct xfs_dqblk *)(*bpp)->b_addr;
/* Do the actual repair of dquots in this buffer */
for (i = 0; i < mp->m_quotainfo->qi_dqperchunk; i++) {
ddq = &d[i].dd_diskdq;
error = xfs_dqcheck(mp, ddq, firstid + i,
dqp->dq_flags & XFS_DQ_ALLTYPES,
XFS_QMOPT_DQREPAIR, "xfs_qm_dqrepair");
if (error) {
/* repair failed, we're screwed */
xfs_trans_brelse(tp, *bpp);
return -EIO;
}
}
return 0;
}
/*
* Maps a dquot to the buffer containing its on-disk version.
* This returns a ptr to the buffer containing the on-disk dquot
* in the bpp param, and a ptr to the on-disk dquot within that buffer
*/
STATIC int
xfs_qm_dqtobp(
xfs_trans_t **tpp,
xfs_dquot_t *dqp,
xfs_disk_dquot_t **O_ddpp,
xfs_buf_t **O_bpp,
uint flags)
{
struct xfs_bmbt_irec map;
int nmaps = 1, error;
struct xfs_buf *bp;
struct xfs_inode *quotip;
struct xfs_mount *mp = dqp->q_mount;
xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id);
struct xfs_trans *tp = (tpp ? *tpp : NULL);
uint lock_mode;
quotip = xfs_quota_inode(dqp->q_mount, dqp->dq_flags);
dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
lock_mode = xfs_ilock_data_map_shared(quotip);
if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
/*
* Return if this type of quotas is turned off while we
* didn't have the quota inode lock.
*/
xfs_iunlock(quotip, lock_mode);
return -ESRCH;
}
/*
* Find the block map; no allocations yet
*/
error = xfs_bmapi_read(quotip, dqp->q_fileoffset,
XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0);
xfs_iunlock(quotip, lock_mode);
if (error)
return error;
ASSERT(nmaps == 1);
ASSERT(map.br_blockcount == 1);
/*
* Offset of dquot in the (fixed sized) dquot chunk.
*/
dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
sizeof(xfs_dqblk_t);
ASSERT(map.br_startblock != DELAYSTARTBLOCK);
if (map.br_startblock == HOLESTARTBLOCK) {
/*
* We don't allocate unless we're asked to
*/
if (!(flags & XFS_QMOPT_DQALLOC))
return -ENOENT;
ASSERT(tp);
error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
dqp->q_fileoffset, &bp);
if (error)
return error;
tp = *tpp;
} else {
trace_xfs_dqtobp_read(dqp);
/*
* store the blkno etc so that we don't have to do the
* mapping all the time
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0, &bp, &xfs_dquot_buf_ops);
if (error == -EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
xfs_dqid_t firstid = (xfs_dqid_t)map.br_startoff *
mp->m_quotainfo->qi_dqperchunk;
ASSERT(bp == NULL);
error = xfs_qm_dqrepair(mp, tp, dqp, firstid, &bp);
}
if (error) {
ASSERT(bp == NULL);
return error;
}
}
ASSERT(xfs_buf_islocked(bp));
*O_bpp = bp;
*O_ddpp = bp->b_addr + dqp->q_bufoffset;
return 0;
}
/*
* Read in the ondisk dquot using dqtobp() then copy it to an incore version,
* and release the buffer immediately.
*
* If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed.
*/
int
xfs_qm_dqread(
struct xfs_mount *mp,
xfs_dqid_t id,
uint type,
uint flags,
struct xfs_dquot **O_dqpp)
{
struct xfs_dquot *dqp;
struct xfs_disk_dquot *ddqp;
struct xfs_buf *bp;
struct xfs_trans *tp = NULL;
int error;
dqp = kmem_zone_zalloc(xfs_qm_dqzone, KM_SLEEP);
dqp->dq_flags = type;
dqp->q_core.d_id = cpu_to_be32(id);
dqp->q_mount = mp;
INIT_LIST_HEAD(&dqp->q_lru);
mutex_init(&dqp->q_qlock);
init_waitqueue_head(&dqp->q_pinwait);
/*
* Because we want to use a counting completion, complete
* the flush completion once to allow a single access to
* the flush completion without blocking.
*/
init_completion(&dqp->q_flush);
complete(&dqp->q_flush);
/*
* Make sure group quotas have a different lock class than user
* quotas.
*/
switch (type) {
case XFS_DQ_USER:
/* uses the default lock class */
break;
case XFS_DQ_GROUP:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_group_class);
break;
case XFS_DQ_PROJ:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_project_class);
break;
default:
ASSERT(0);
break;
}
XFS_STATS_INC(mp, xs_qm_dquot);
trace_xfs_dqread(dqp);
if (flags & XFS_QMOPT_DQALLOC) {
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc,
XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp);
if (error)
goto error0;
}
/*
* get a pointer to the on-disk dquot and the buffer containing it
* dqp already knows its own type (GROUP/USER).
*/
error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags);
if (error) {
/*
* This can happen if quotas got turned off (ESRCH),
* or if the dquot didn't exist on disk and we ask to
* allocate (ENOENT).
*/
trace_xfs_dqread_fail(dqp);
goto error1;
}
/* copy everything from disk dquot to the incore dquot */
memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
xfs_qm_dquot_logitem_init(dqp);
/*
* Reservation counters are defined as reservation plus current usage
* to avoid having to add every time.
*/
dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount);
dqp->q_res_icount = be64_to_cpu(ddqp->d_icount);
dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount);
/* initialize the dquot speculative prealloc thresholds */
xfs_dquot_set_prealloc_limits(dqp);
/* Mark the buf so that this will stay incore a little longer */
xfs_buf_set_ref(bp, XFS_DQUOT_REF);
/*
* We got the buffer with a xfs_trans_read_buf() (in dqtobp())
* So we need to release with xfs_trans_brelse().
* The strategy here is identical to that of inodes; we lock
* the dquot in xfs_qm_dqget() before making it accessible to
* others. This is because dquots, like inodes, need a good level of
* concurrency, and we don't want to take locks on the entire buffers
* for dquot accesses.
* Note also that the dquot buffer may even be dirty at this point, if
* this particular dquot was repaired. We still aren't afraid to
* brelse it because we have the changes incore.
*/
ASSERT(xfs_buf_islocked(bp));
xfs_trans_brelse(tp, bp);
if (tp) {
error = xfs_trans_commit(tp);
if (error)
goto error0;
}
*O_dqpp = dqp;
return error;
error1:
if (tp)
xfs_trans_cancel(tp);
error0:
xfs_qm_dqdestroy(dqp);
*O_dqpp = NULL;
return error;
}
/*
* Advance to the next id in the current chunk, or if at the
* end of the chunk, skip ahead to first id in next allocated chunk
* using the SEEK_DATA interface.
*/
static int
xfs_dq_get_next_id(
struct xfs_mount *mp,
uint type,
xfs_dqid_t *id)
{
struct xfs_inode *quotip = xfs_quota_inode(mp, type);
xfs_dqid_t next_id = *id + 1; /* simple advance */
uint lock_flags;
struct xfs_bmbt_irec got;
struct xfs_iext_cursor cur;
xfs_fsblock_t start;
int error = 0;
/* If we'd wrap past the max ID, stop */
if (next_id < *id)
return -ENOENT;
/* If new ID is within the current chunk, advancing it sufficed */
if (next_id % mp->m_quotainfo->qi_dqperchunk) {
*id = next_id;
return 0;
}
/* Nope, next_id is now past the current chunk, so find the next one */
start = (xfs_fsblock_t)next_id / mp->m_quotainfo->qi_dqperchunk;
lock_flags = xfs_ilock_data_map_shared(quotip);
if (!(quotip->i_df.if_flags & XFS_IFEXTENTS)) {
error = xfs_iread_extents(NULL, quotip, XFS_DATA_FORK);
if (error)
return error;
}
if (xfs_iext_lookup_extent(quotip, "ip->i_df, start, &cur, &got)) {
/* contiguous chunk, bump startoff for the id calculation */
if (got.br_startoff < start)
got.br_startoff = start;
*id = got.br_startoff * mp->m_quotainfo->qi_dqperchunk;
} else {
error = -ENOENT;
}
xfs_iunlock(quotip, lock_flags);
return error;
}
/*
* Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a
* a locked dquot, doing an allocation (if requested) as needed.
* When both an inode and an id are given, the inode's id takes precedence.
* That is, if the id changes while we don't hold the ilock inside this
* function, the new dquot is returned, not necessarily the one requested
* in the id argument.
*/
int
xfs_qm_dqget(
xfs_mount_t *mp,
xfs_inode_t *ip, /* locked inode (optional) */
xfs_dqid_t id, /* uid/projid/gid depending on type */
uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */
uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */
xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */
{
struct xfs_quotainfo *qi = mp->m_quotainfo;
struct radix_tree_root *tree = xfs_dquot_tree(qi, type);
struct xfs_dquot *dqp;
int error;
ASSERT(XFS_IS_QUOTA_RUNNING(mp));
if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
(! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) ||
(! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
return -ESRCH;
}
ASSERT(type == XFS_DQ_USER ||
type == XFS_DQ_PROJ ||
type == XFS_DQ_GROUP);
if (ip) {
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
ASSERT(xfs_inode_dquot(ip, type) == NULL);
}
restart:
mutex_lock(&qi->qi_tree_lock);
dqp = radix_tree_lookup(tree, id);
if (dqp) {
xfs_dqlock(dqp);
if (dqp->dq_flags & XFS_DQ_FREEING) {
xfs_dqunlock(dqp);
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_freeing(dqp);
delay(1);
goto restart;
}
/* uninit / unused quota found in radix tree, keep looking */
if (flags & XFS_QMOPT_DQNEXT) {
if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
xfs_dqunlock(dqp);
mutex_unlock(&qi->qi_tree_lock);
error = xfs_dq_get_next_id(mp, type, &id);
if (error)
return error;
goto restart;
}
}
dqp->q_nrefs++;
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_hit(dqp);
XFS_STATS_INC(mp, xs_qm_dqcachehits);
*O_dqpp = dqp;
return 0;
}
mutex_unlock(&qi->qi_tree_lock);
XFS_STATS_INC(mp, xs_qm_dqcachemisses);
/*
* Dquot cache miss. We don't want to keep the inode lock across
* a (potential) disk read. Also we don't want to deal with the lock
* ordering between quotainode and this inode. OTOH, dropping the inode
* lock here means dealing with a chown that can happen before
* we re-acquire the lock.
*/
if (ip)
xfs_iunlock(ip, XFS_ILOCK_EXCL);
error = xfs_qm_dqread(mp, id, type, flags, &dqp);
if (ip)
xfs_ilock(ip, XFS_ILOCK_EXCL);
/* If we are asked to find next active id, keep looking */
if (error == -ENOENT && (flags & XFS_QMOPT_DQNEXT)) {
error = xfs_dq_get_next_id(mp, type, &id);
if (!error)
goto restart;
}
if (error)
return error;
if (ip) {
/*
* A dquot could be attached to this inode by now, since
* we had dropped the ilock.
*/
if (xfs_this_quota_on(mp, type)) {
struct xfs_dquot *dqp1;
dqp1 = xfs_inode_dquot(ip, type);
if (dqp1) {
xfs_qm_dqdestroy(dqp);
dqp = dqp1;
xfs_dqlock(dqp);
goto dqret;
}
} else {
/* inode stays locked on return */
xfs_qm_dqdestroy(dqp);
return -ESRCH;
}
}
mutex_lock(&qi->qi_tree_lock);
error = radix_tree_insert(tree, id, dqp);
if (unlikely(error)) {
WARN_ON(error != -EEXIST);
/*
* Duplicate found. Just throw away the new dquot and start
* over.
*/
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_dup(dqp);
xfs_qm_dqdestroy(dqp);
XFS_STATS_INC(mp, xs_qm_dquot_dups);
goto restart;
}
/*
* We return a locked dquot to the caller, with a reference taken
*/
xfs_dqlock(dqp);
dqp->q_nrefs = 1;
qi->qi_dquots++;
mutex_unlock(&qi->qi_tree_lock);
/* If we are asked to find next active id, keep looking */
if (flags & XFS_QMOPT_DQNEXT) {
if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
xfs_qm_dqput(dqp);
error = xfs_dq_get_next_id(mp, type, &id);
if (error)
return error;
goto restart;
}
}
dqret:
ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
trace_xfs_dqget_miss(dqp);
*O_dqpp = dqp;
return 0;
}
/*
* Release a reference to the dquot (decrement ref-count) and unlock it.
*
* If there is a group quota attached to this dquot, carefully release that
* too without tripping over deadlocks'n'stuff.
*/
void
xfs_qm_dqput(
struct xfs_dquot *dqp)
{
ASSERT(dqp->q_nrefs > 0);
ASSERT(XFS_DQ_IS_LOCKED(dqp));
trace_xfs_dqput(dqp);
if (--dqp->q_nrefs == 0) {
struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo;
trace_xfs_dqput_free(dqp);
if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
XFS_STATS_INC(dqp->q_mount, xs_qm_dquot_unused);
}
xfs_dqunlock(dqp);
}
/*
* Release a dquot. Flush it if dirty, then dqput() it.
* dquot must not be locked.
*/
void
xfs_qm_dqrele(
xfs_dquot_t *dqp)
{
if (!dqp)
return;
trace_xfs_dqrele(dqp);
xfs_dqlock(dqp);
/*
* We don't care to flush it if the dquot is dirty here.
* That will create stutters that we want to avoid.
* Instead we do a delayed write when we try to reclaim
* a dirty dquot. Also xfs_sync will take part of the burden...
*/
xfs_qm_dqput(dqp);
}
/*
* This is the dquot flushing I/O completion routine. It is called
* from interrupt level when the buffer containing the dquot is
* flushed to disk. It is responsible for removing the dquot logitem
* from the AIL if it has not been re-logged, and unlocking the dquot's
* flush lock. This behavior is very similar to that of inodes..
*/
STATIC void
xfs_qm_dqflush_done(
struct xfs_buf *bp,
struct xfs_log_item *lip)
{
xfs_dq_logitem_t *qip = (struct xfs_dq_logitem *)lip;
xfs_dquot_t *dqp = qip->qli_dquot;
struct xfs_ail *ailp = lip->li_ailp;
/*
* We only want to pull the item from the AIL if its
* location in the log has not changed since we started the flush.
* Thus, we only bother if the dquot's lsn has
* not changed. First we check the lsn outside the lock
* since it's cheaper, and then we recheck while
* holding the lock before removing the dquot from the AIL.
*/
if ((lip->li_flags & XFS_LI_IN_AIL) &&
lip->li_lsn == qip->qli_flush_lsn) {
/* xfs_trans_ail_delete() drops the AIL lock. */
spin_lock(&ailp->xa_lock);
if (lip->li_lsn == qip->qli_flush_lsn)
xfs_trans_ail_delete(ailp, lip, SHUTDOWN_CORRUPT_INCORE);
else
spin_unlock(&ailp->xa_lock);
}
/*
* Release the dq's flush lock since we're done with it.
*/
xfs_dqfunlock(dqp);
}
/*
* Write a modified dquot to disk.
* The dquot must be locked and the flush lock too taken by caller.
* The flush lock will not be unlocked until the dquot reaches the disk,
* but the dquot is free to be unlocked and modified by the caller
* in the interim. Dquot is still locked on return. This behavior is
* identical to that of inodes.
*/
int
xfs_qm_dqflush(
struct xfs_dquot *dqp,
struct xfs_buf **bpp)