forked from torvalds/linux
-
Notifications
You must be signed in to change notification settings - Fork 0
/
jfs_imap.c
3155 lines (2778 loc) · 83.1 KB
/
jfs_imap.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
// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Copyright (C) International Business Machines Corp., 2000-2004
*/
/*
* jfs_imap.c: inode allocation map manager
*
* Serialization:
* Each AG has a simple lock which is used to control the serialization of
* the AG level lists. This lock should be taken first whenever an AG
* level list will be modified or accessed.
*
* Each IAG is locked by obtaining the buffer for the IAG page.
*
* There is also a inode lock for the inode map inode. A read lock needs to
* be taken whenever an IAG is read from the map or the global level
* information is read. A write lock needs to be taken whenever the global
* level information is modified or an atomic operation needs to be used.
*
* If more than one IAG is read at one time, the read lock may not
* be given up until all of the IAG's are read. Otherwise, a deadlock
* may occur when trying to obtain the read lock while another thread
* holding the read lock is waiting on the IAG already being held.
*
* The control page of the inode map is read into memory by diMount().
* Thereafter it should only be modified in memory and then it will be
* written out when the filesystem is unmounted by diUnmount().
*/
#include <linux/fs.h>
#include <linux/buffer_head.h>
#include <linux/pagemap.h>
#include <linux/quotaops.h>
#include <linux/slab.h>
#include "jfs_incore.h"
#include "jfs_inode.h"
#include "jfs_filsys.h"
#include "jfs_dinode.h"
#include "jfs_dmap.h"
#include "jfs_imap.h"
#include "jfs_metapage.h"
#include "jfs_superblock.h"
#include "jfs_debug.h"
/*
* imap locks
*/
/* iag free list lock */
#define IAGFREE_LOCK_INIT(imap) mutex_init(&imap->im_freelock)
#define IAGFREE_LOCK(imap) mutex_lock(&imap->im_freelock)
#define IAGFREE_UNLOCK(imap) mutex_unlock(&imap->im_freelock)
/* per ag iag list locks */
#define AG_LOCK_INIT(imap,index) mutex_init(&(imap->im_aglock[index]))
#define AG_LOCK(imap,agno) mutex_lock(&imap->im_aglock[agno])
#define AG_UNLOCK(imap,agno) mutex_unlock(&imap->im_aglock[agno])
/*
* forward references
*/
static int diAllocAG(struct inomap *, int, bool, struct inode *);
static int diAllocAny(struct inomap *, int, bool, struct inode *);
static int diAllocBit(struct inomap *, struct iag *, int);
static int diAllocExt(struct inomap *, int, struct inode *);
static int diAllocIno(struct inomap *, int, struct inode *);
static int diFindFree(u32, int);
static int diNewExt(struct inomap *, struct iag *, int);
static int diNewIAG(struct inomap *, int *, int, struct metapage **);
static void duplicateIXtree(struct super_block *, s64, int, s64 *);
static int diIAGRead(struct inomap * imap, int, struct metapage **);
static int copy_from_dinode(struct dinode *, struct inode *);
static void copy_to_dinode(struct dinode *, struct inode *);
/*
* NAME: diMount()
*
* FUNCTION: initialize the incore inode map control structures for
* a fileset or aggregate init time.
*
* the inode map's control structure (dinomap) is
* brought in from disk and placed in virtual memory.
*
* PARAMETERS:
* ipimap - pointer to inode map inode for the aggregate or fileset.
*
* RETURN VALUES:
* 0 - success
* -ENOMEM - insufficient free virtual memory.
* -EIO - i/o error.
*/
int diMount(struct inode *ipimap)
{
struct inomap *imap;
struct metapage *mp;
int index;
struct dinomap_disk *dinom_le;
/*
* allocate/initialize the in-memory inode map control structure
*/
/* allocate the in-memory inode map control structure. */
imap = kmalloc(sizeof(struct inomap), GFP_KERNEL);
if (imap == NULL)
return -ENOMEM;
/* read the on-disk inode map control structure. */
mp = read_metapage(ipimap,
IMAPBLKNO << JFS_SBI(ipimap->i_sb)->l2nbperpage,
PSIZE, 0);
if (mp == NULL) {
kfree(imap);
return -EIO;
}
/* copy the on-disk version to the in-memory version. */
dinom_le = (struct dinomap_disk *) mp->data;
imap->im_freeiag = le32_to_cpu(dinom_le->in_freeiag);
imap->im_nextiag = le32_to_cpu(dinom_le->in_nextiag);
atomic_set(&imap->im_numinos, le32_to_cpu(dinom_le->in_numinos));
atomic_set(&imap->im_numfree, le32_to_cpu(dinom_le->in_numfree));
imap->im_nbperiext = le32_to_cpu(dinom_le->in_nbperiext);
imap->im_l2nbperiext = le32_to_cpu(dinom_le->in_l2nbperiext);
for (index = 0; index < MAXAG; index++) {
imap->im_agctl[index].inofree =
le32_to_cpu(dinom_le->in_agctl[index].inofree);
imap->im_agctl[index].extfree =
le32_to_cpu(dinom_le->in_agctl[index].extfree);
imap->im_agctl[index].numinos =
le32_to_cpu(dinom_le->in_agctl[index].numinos);
imap->im_agctl[index].numfree =
le32_to_cpu(dinom_le->in_agctl[index].numfree);
}
/* release the buffer. */
release_metapage(mp);
/*
* allocate/initialize inode allocation map locks
*/
/* allocate and init iag free list lock */
IAGFREE_LOCK_INIT(imap);
/* allocate and init ag list locks */
for (index = 0; index < MAXAG; index++) {
AG_LOCK_INIT(imap, index);
}
/* bind the inode map inode and inode map control structure
* to each other.
*/
imap->im_ipimap = ipimap;
JFS_IP(ipimap)->i_imap = imap;
return (0);
}
/*
* NAME: diUnmount()
*
* FUNCTION: write to disk the incore inode map control structures for
* a fileset or aggregate at unmount time.
*
* PARAMETERS:
* ipimap - pointer to inode map inode for the aggregate or fileset.
*
* RETURN VALUES:
* 0 - success
* -ENOMEM - insufficient free virtual memory.
* -EIO - i/o error.
*/
int diUnmount(struct inode *ipimap, int mounterror)
{
struct inomap *imap = JFS_IP(ipimap)->i_imap;
/*
* update the on-disk inode map control structure
*/
if (!(mounterror || isReadOnly(ipimap)))
diSync(ipimap);
/*
* Invalidate the page cache buffers
*/
truncate_inode_pages(ipimap->i_mapping, 0);
/*
* free in-memory control structure
*/
kfree(imap);
return (0);
}
/*
* diSync()
*/
int diSync(struct inode *ipimap)
{
struct dinomap_disk *dinom_le;
struct inomap *imp = JFS_IP(ipimap)->i_imap;
struct metapage *mp;
int index;
/*
* write imap global conrol page
*/
/* read the on-disk inode map control structure */
mp = get_metapage(ipimap,
IMAPBLKNO << JFS_SBI(ipimap->i_sb)->l2nbperpage,
PSIZE, 0);
if (mp == NULL) {
jfs_err("diSync: get_metapage failed!");
return -EIO;
}
/* copy the in-memory version to the on-disk version */
dinom_le = (struct dinomap_disk *) mp->data;
dinom_le->in_freeiag = cpu_to_le32(imp->im_freeiag);
dinom_le->in_nextiag = cpu_to_le32(imp->im_nextiag);
dinom_le->in_numinos = cpu_to_le32(atomic_read(&imp->im_numinos));
dinom_le->in_numfree = cpu_to_le32(atomic_read(&imp->im_numfree));
dinom_le->in_nbperiext = cpu_to_le32(imp->im_nbperiext);
dinom_le->in_l2nbperiext = cpu_to_le32(imp->im_l2nbperiext);
for (index = 0; index < MAXAG; index++) {
dinom_le->in_agctl[index].inofree =
cpu_to_le32(imp->im_agctl[index].inofree);
dinom_le->in_agctl[index].extfree =
cpu_to_le32(imp->im_agctl[index].extfree);
dinom_le->in_agctl[index].numinos =
cpu_to_le32(imp->im_agctl[index].numinos);
dinom_le->in_agctl[index].numfree =
cpu_to_le32(imp->im_agctl[index].numfree);
}
/* write out the control structure */
write_metapage(mp);
/*
* write out dirty pages of imap
*/
filemap_write_and_wait(ipimap->i_mapping);
diWriteSpecial(ipimap, 0);
return (0);
}
/*
* NAME: diRead()
*
* FUNCTION: initialize an incore inode from disk.
*
* on entry, the specifed incore inode should itself
* specify the disk inode number corresponding to the
* incore inode (i.e. i_number should be initialized).
*
* this routine handles incore inode initialization for
* both "special" and "regular" inodes. special inodes
* are those required early in the mount process and
* require special handling since much of the file system
* is not yet initialized. these "special" inodes are
* identified by a NULL inode map inode pointer and are
* actually initialized by a call to diReadSpecial().
*
* for regular inodes, the iag describing the disk inode
* is read from disk to determine the inode extent address
* for the disk inode. with the inode extent address in
* hand, the page of the extent that contains the disk
* inode is read and the disk inode is copied to the
* incore inode.
*
* PARAMETERS:
* ip - pointer to incore inode to be initialized from disk.
*
* RETURN VALUES:
* 0 - success
* -EIO - i/o error.
* -ENOMEM - insufficient memory
*
*/
int diRead(struct inode *ip)
{
struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb);
int iagno, ino, extno, rc;
struct inode *ipimap;
struct dinode *dp;
struct iag *iagp;
struct metapage *mp;
s64 blkno, agstart;
struct inomap *imap;
int block_offset;
int inodes_left;
unsigned long pageno;
int rel_inode;
jfs_info("diRead: ino = %ld", ip->i_ino);
ipimap = sbi->ipimap;
JFS_IP(ip)->ipimap = ipimap;
/* determine the iag number for this inode (number) */
iagno = INOTOIAG(ip->i_ino);
/* read the iag */
IREAD_LOCK(ipimap, RDWRLOCK_IMAP);
imap = JFS_IP(ipimap)->i_imap;
rc = diIAGRead(imap, iagno, &mp);
IREAD_UNLOCK(ipimap);
if (rc) {
jfs_err("diRead: diIAGRead returned %d", rc);
return (rc);
}
iagp = (struct iag *) mp->data;
/* determine inode extent that holds the disk inode */
ino = ip->i_ino & (INOSPERIAG - 1);
extno = ino >> L2INOSPEREXT;
if ((lengthPXD(&iagp->inoext[extno]) != imap->im_nbperiext) ||
(addressPXD(&iagp->inoext[extno]) == 0)) {
release_metapage(mp);
return -ESTALE;
}
/* get disk block number of the page within the inode extent
* that holds the disk inode.
*/
blkno = INOPBLK(&iagp->inoext[extno], ino, sbi->l2nbperpage);
/* get the ag for the iag */
agstart = le64_to_cpu(iagp->agstart);
release_metapage(mp);
rel_inode = (ino & (INOSPERPAGE - 1));
pageno = blkno >> sbi->l2nbperpage;
if ((block_offset = ((u32) blkno & (sbi->nbperpage - 1)))) {
/*
* OS/2 didn't always align inode extents on page boundaries
*/
inodes_left =
(sbi->nbperpage - block_offset) << sbi->l2niperblk;
if (rel_inode < inodes_left)
rel_inode += block_offset << sbi->l2niperblk;
else {
pageno += 1;
rel_inode -= inodes_left;
}
}
/* read the page of disk inode */
mp = read_metapage(ipimap, pageno << sbi->l2nbperpage, PSIZE, 1);
if (!mp) {
jfs_err("diRead: read_metapage failed");
return -EIO;
}
/* locate the disk inode requested */
dp = (struct dinode *) mp->data;
dp += rel_inode;
if (ip->i_ino != le32_to_cpu(dp->di_number)) {
jfs_error(ip->i_sb, "i_ino != di_number\n");
rc = -EIO;
} else if (le32_to_cpu(dp->di_nlink) == 0)
rc = -ESTALE;
else
/* copy the disk inode to the in-memory inode */
rc = copy_from_dinode(dp, ip);
release_metapage(mp);
/* set the ag for the inode */
JFS_IP(ip)->agstart = agstart;
JFS_IP(ip)->active_ag = -1;
return (rc);
}
/*
* NAME: diReadSpecial()
*
* FUNCTION: initialize a 'special' inode from disk.
*
* this routines handles aggregate level inodes. The
* inode cache cannot differentiate between the
* aggregate inodes and the filesystem inodes, so we
* handle these here. We don't actually use the aggregate
* inode map, since these inodes are at a fixed location
* and in some cases the aggregate inode map isn't initialized
* yet.
*
* PARAMETERS:
* sb - filesystem superblock
* inum - aggregate inode number
* secondary - 1 if secondary aggregate inode table
*
* RETURN VALUES:
* new inode - success
* NULL - i/o error.
*/
struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary)
{
struct jfs_sb_info *sbi = JFS_SBI(sb);
uint address;
struct dinode *dp;
struct inode *ip;
struct metapage *mp;
ip = new_inode(sb);
if (ip == NULL) {
jfs_err("diReadSpecial: new_inode returned NULL!");
return ip;
}
if (secondary) {
address = addressPXD(&sbi->ait2) >> sbi->l2nbperpage;
JFS_IP(ip)->ipimap = sbi->ipaimap2;
} else {
address = AITBL_OFF >> L2PSIZE;
JFS_IP(ip)->ipimap = sbi->ipaimap;
}
ASSERT(inum < INOSPEREXT);
ip->i_ino = inum;
address += inum >> 3; /* 8 inodes per 4K page */
/* read the page of fixed disk inode (AIT) in raw mode */
mp = read_metapage(ip, address << sbi->l2nbperpage, PSIZE, 1);
if (mp == NULL) {
set_nlink(ip, 1); /* Don't want iput() deleting it */
iput(ip);
return (NULL);
}
/* get the pointer to the disk inode of interest */
dp = (struct dinode *) (mp->data);
dp += inum % 8; /* 8 inodes per 4K page */
/* copy on-disk inode to in-memory inode */
if ((copy_from_dinode(dp, ip)) != 0) {
/* handle bad return by returning NULL for ip */
set_nlink(ip, 1); /* Don't want iput() deleting it */
iput(ip);
/* release the page */
release_metapage(mp);
return (NULL);
}
ip->i_mapping->a_ops = &jfs_metapage_aops;
mapping_set_gfp_mask(ip->i_mapping, GFP_NOFS);
/* Allocations to metadata inodes should not affect quotas */
ip->i_flags |= S_NOQUOTA;
if ((inum == FILESYSTEM_I) && (JFS_IP(ip)->ipimap == sbi->ipaimap)) {
sbi->gengen = le32_to_cpu(dp->di_gengen);
sbi->inostamp = le32_to_cpu(dp->di_inostamp);
}
/* release the page */
release_metapage(mp);
inode_fake_hash(ip);
return (ip);
}
/*
* NAME: diWriteSpecial()
*
* FUNCTION: Write the special inode to disk
*
* PARAMETERS:
* ip - special inode
* secondary - 1 if secondary aggregate inode table
*
* RETURN VALUES: none
*/
void diWriteSpecial(struct inode *ip, int secondary)
{
struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb);
uint address;
struct dinode *dp;
ino_t inum = ip->i_ino;
struct metapage *mp;
if (secondary)
address = addressPXD(&sbi->ait2) >> sbi->l2nbperpage;
else
address = AITBL_OFF >> L2PSIZE;
ASSERT(inum < INOSPEREXT);
address += inum >> 3; /* 8 inodes per 4K page */
/* read the page of fixed disk inode (AIT) in raw mode */
mp = read_metapage(ip, address << sbi->l2nbperpage, PSIZE, 1);
if (mp == NULL) {
jfs_err("diWriteSpecial: failed to read aggregate inode extent!");
return;
}
/* get the pointer to the disk inode of interest */
dp = (struct dinode *) (mp->data);
dp += inum % 8; /* 8 inodes per 4K page */
/* copy on-disk inode to in-memory inode */
copy_to_dinode(dp, ip);
memcpy(&dp->di_xtroot, &JFS_IP(ip)->i_xtroot, 288);
if (inum == FILESYSTEM_I)
dp->di_gengen = cpu_to_le32(sbi->gengen);
/* write the page */
write_metapage(mp);
}
/*
* NAME: diFreeSpecial()
*
* FUNCTION: Free allocated space for special inode
*/
void diFreeSpecial(struct inode *ip)
{
if (ip == NULL) {
jfs_err("diFreeSpecial called with NULL ip!");
return;
}
filemap_write_and_wait(ip->i_mapping);
truncate_inode_pages(ip->i_mapping, 0);
iput(ip);
}
/*
* NAME: diWrite()
*
* FUNCTION: write the on-disk inode portion of the in-memory inode
* to its corresponding on-disk inode.
*
* on entry, the specifed incore inode should itself
* specify the disk inode number corresponding to the
* incore inode (i.e. i_number should be initialized).
*
* the inode contains the inode extent address for the disk
* inode. with the inode extent address in hand, the
* page of the extent that contains the disk inode is
* read and the disk inode portion of the incore inode
* is copied to the disk inode.
*
* PARAMETERS:
* tid - transacation id
* ip - pointer to incore inode to be written to the inode extent.
*
* RETURN VALUES:
* 0 - success
* -EIO - i/o error.
*/
int diWrite(tid_t tid, struct inode *ip)
{
struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb);
struct jfs_inode_info *jfs_ip = JFS_IP(ip);
int rc = 0;
s32 ino;
struct dinode *dp;
s64 blkno;
int block_offset;
int inodes_left;
struct metapage *mp;
unsigned long pageno;
int rel_inode;
int dioffset;
struct inode *ipimap;
uint type;
lid_t lid;
struct tlock *ditlck, *tlck;
struct linelock *dilinelock, *ilinelock;
struct lv *lv;
int n;
ipimap = jfs_ip->ipimap;
ino = ip->i_ino & (INOSPERIAG - 1);
if (!addressPXD(&(jfs_ip->ixpxd)) ||
(lengthPXD(&(jfs_ip->ixpxd)) !=
JFS_IP(ipimap)->i_imap->im_nbperiext)) {
jfs_error(ip->i_sb, "ixpxd invalid\n");
return -EIO;
}
/*
* read the page of disk inode containing the specified inode:
*/
/* compute the block address of the page */
blkno = INOPBLK(&(jfs_ip->ixpxd), ino, sbi->l2nbperpage);
rel_inode = (ino & (INOSPERPAGE - 1));
pageno = blkno >> sbi->l2nbperpage;
if ((block_offset = ((u32) blkno & (sbi->nbperpage - 1)))) {
/*
* OS/2 didn't always align inode extents on page boundaries
*/
inodes_left =
(sbi->nbperpage - block_offset) << sbi->l2niperblk;
if (rel_inode < inodes_left)
rel_inode += block_offset << sbi->l2niperblk;
else {
pageno += 1;
rel_inode -= inodes_left;
}
}
/* read the page of disk inode */
retry:
mp = read_metapage(ipimap, pageno << sbi->l2nbperpage, PSIZE, 1);
if (!mp)
return -EIO;
/* get the pointer to the disk inode */
dp = (struct dinode *) mp->data;
dp += rel_inode;
dioffset = (ino & (INOSPERPAGE - 1)) << L2DISIZE;
/*
* acquire transaction lock on the on-disk inode;
* N.B. tlock is acquired on ipimap not ip;
*/
if ((ditlck =
txLock(tid, ipimap, mp, tlckINODE | tlckENTRY)) == NULL)
goto retry;
dilinelock = (struct linelock *) & ditlck->lock;
/*
* copy btree root from in-memory inode to on-disk inode
*
* (tlock is taken from inline B+-tree root in in-memory
* inode when the B+-tree root is updated, which is pointed
* by jfs_ip->blid as well as being on tx tlock list)
*
* further processing of btree root is based on the copy
* in in-memory inode, where txLog() will log from, and,
* for xtree root, txUpdateMap() will update map and reset
* XAD_NEW bit;
*/
if (S_ISDIR(ip->i_mode) && (lid = jfs_ip->xtlid)) {
/*
* This is the special xtree inside the directory for storing
* the directory table
*/
xtpage_t *p, *xp;
xad_t *xad;
jfs_ip->xtlid = 0;
tlck = lid_to_tlock(lid);
assert(tlck->type & tlckXTREE);
tlck->type |= tlckBTROOT;
tlck->mp = mp;
ilinelock = (struct linelock *) & tlck->lock;
/*
* copy xtree root from inode to dinode:
*/
p = &jfs_ip->i_xtroot;
xp = (xtpage_t *) &dp->di_dirtable;
lv = ilinelock->lv;
for (n = 0; n < ilinelock->index; n++, lv++) {
memcpy(&xp->xad[lv->offset], &p->xad[lv->offset],
lv->length << L2XTSLOTSIZE);
}
/* reset on-disk (metadata page) xtree XAD_NEW bit */
xad = &xp->xad[XTENTRYSTART];
for (n = XTENTRYSTART;
n < le16_to_cpu(xp->header.nextindex); n++, xad++)
if (xad->flag & (XAD_NEW | XAD_EXTENDED))
xad->flag &= ~(XAD_NEW | XAD_EXTENDED);
}
if ((lid = jfs_ip->blid) == 0)
goto inlineData;
jfs_ip->blid = 0;
tlck = lid_to_tlock(lid);
type = tlck->type;
tlck->type |= tlckBTROOT;
tlck->mp = mp;
ilinelock = (struct linelock *) & tlck->lock;
/*
* regular file: 16 byte (XAD slot) granularity
*/
if (type & tlckXTREE) {
xtpage_t *p, *xp;
xad_t *xad;
/*
* copy xtree root from inode to dinode:
*/
p = &jfs_ip->i_xtroot;
xp = &dp->di_xtroot;
lv = ilinelock->lv;
for (n = 0; n < ilinelock->index; n++, lv++) {
memcpy(&xp->xad[lv->offset], &p->xad[lv->offset],
lv->length << L2XTSLOTSIZE);
}
/* reset on-disk (metadata page) xtree XAD_NEW bit */
xad = &xp->xad[XTENTRYSTART];
for (n = XTENTRYSTART;
n < le16_to_cpu(xp->header.nextindex); n++, xad++)
if (xad->flag & (XAD_NEW | XAD_EXTENDED))
xad->flag &= ~(XAD_NEW | XAD_EXTENDED);
}
/*
* directory: 32 byte (directory entry slot) granularity
*/
else if (type & tlckDTREE) {
dtpage_t *p, *xp;
/*
* copy dtree root from inode to dinode:
*/
p = (dtpage_t *) &jfs_ip->i_dtroot;
xp = (dtpage_t *) & dp->di_dtroot;
lv = ilinelock->lv;
for (n = 0; n < ilinelock->index; n++, lv++) {
memcpy(&xp->slot[lv->offset], &p->slot[lv->offset],
lv->length << L2DTSLOTSIZE);
}
} else {
jfs_err("diWrite: UFO tlock");
}
inlineData:
/*
* copy inline symlink from in-memory inode to on-disk inode
*/
if (S_ISLNK(ip->i_mode) && ip->i_size < IDATASIZE) {
lv = & dilinelock->lv[dilinelock->index];
lv->offset = (dioffset + 2 * 128) >> L2INODESLOTSIZE;
lv->length = 2;
memcpy(&dp->di_inline_all, jfs_ip->i_inline_all, IDATASIZE);
dilinelock->index++;
}
/*
* copy inline data from in-memory inode to on-disk inode:
* 128 byte slot granularity
*/
if (test_cflag(COMMIT_Inlineea, ip)) {
lv = & dilinelock->lv[dilinelock->index];
lv->offset = (dioffset + 3 * 128) >> L2INODESLOTSIZE;
lv->length = 1;
memcpy(&dp->di_inlineea, jfs_ip->i_inline_ea, INODESLOTSIZE);
dilinelock->index++;
clear_cflag(COMMIT_Inlineea, ip);
}
/*
* lock/copy inode base: 128 byte slot granularity
*/
lv = & dilinelock->lv[dilinelock->index];
lv->offset = dioffset >> L2INODESLOTSIZE;
copy_to_dinode(dp, ip);
if (test_and_clear_cflag(COMMIT_Dirtable, ip)) {
lv->length = 2;
memcpy(&dp->di_dirtable, &jfs_ip->i_dirtable, 96);
} else
lv->length = 1;
dilinelock->index++;
/* release the buffer holding the updated on-disk inode.
* the buffer will be later written by commit processing.
*/
write_metapage(mp);
return (rc);
}
/*
* NAME: diFree(ip)
*
* FUNCTION: free a specified inode from the inode working map
* for a fileset or aggregate.
*
* if the inode to be freed represents the first (only)
* free inode within the iag, the iag will be placed on
* the ag free inode list.
*
* freeing the inode will cause the inode extent to be
* freed if the inode is the only allocated inode within
* the extent. in this case all the disk resource backing
* up the inode extent will be freed. in addition, the iag
* will be placed on the ag extent free list if the extent
* is the first free extent in the iag. if freeing the
* extent also means that no free inodes will exist for
* the iag, the iag will also be removed from the ag free
* inode list.
*
* the iag describing the inode will be freed if the extent
* is to be freed and it is the only backed extent within
* the iag. in this case, the iag will be removed from the
* ag free extent list and ag free inode list and placed on
* the inode map's free iag list.
*
* a careful update approach is used to provide consistency
* in the face of updates to multiple buffers. under this
* approach, all required buffers are obtained before making
* any updates and are held until all updates are complete.
*
* PARAMETERS:
* ip - inode to be freed.
*
* RETURN VALUES:
* 0 - success
* -EIO - i/o error.
*/
int diFree(struct inode *ip)
{
int rc;
ino_t inum = ip->i_ino;
struct iag *iagp, *aiagp, *biagp, *ciagp, *diagp;
struct metapage *mp, *amp, *bmp, *cmp, *dmp;
int iagno, ino, extno, bitno, sword, agno;
int back, fwd;
u32 bitmap, mask;
struct inode *ipimap = JFS_SBI(ip->i_sb)->ipimap;
struct inomap *imap = JFS_IP(ipimap)->i_imap;
pxd_t freepxd;
tid_t tid;
struct inode *iplist[3];
struct tlock *tlck;
struct pxd_lock *pxdlock;
/*
* This is just to suppress compiler warnings. The same logic that
* references these variables is used to initialize them.
*/
aiagp = biagp = ciagp = diagp = NULL;
/* get the iag number containing the inode.
*/
iagno = INOTOIAG(inum);
/* make sure that the iag is contained within
* the map.
*/
if (iagno >= imap->im_nextiag) {
print_hex_dump(KERN_ERR, "imap: ", DUMP_PREFIX_ADDRESS, 16, 4,
imap, 32, 0);
jfs_error(ip->i_sb, "inum = %d, iagno = %d, nextiag = %d\n",
(uint) inum, iagno, imap->im_nextiag);
return -EIO;
}
/* get the allocation group for this ino.
*/
agno = BLKTOAG(JFS_IP(ip)->agstart, JFS_SBI(ip->i_sb));
/* Lock the AG specific inode map information
*/
AG_LOCK(imap, agno);
/* Obtain read lock in imap inode. Don't release it until we have
* read all of the IAG's that we are going to.
*/
IREAD_LOCK(ipimap, RDWRLOCK_IMAP);
/* read the iag.
*/
if ((rc = diIAGRead(imap, iagno, &mp))) {
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
return (rc);
}
iagp = (struct iag *) mp->data;
/* get the inode number and extent number of the inode within
* the iag and the inode number within the extent.
*/
ino = inum & (INOSPERIAG - 1);
extno = ino >> L2INOSPEREXT;
bitno = ino & (INOSPEREXT - 1);
mask = HIGHORDER >> bitno;
if (!(le32_to_cpu(iagp->wmap[extno]) & mask)) {
jfs_error(ip->i_sb, "wmap shows inode already free\n");
}
if (!addressPXD(&iagp->inoext[extno])) {
release_metapage(mp);
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
jfs_error(ip->i_sb, "invalid inoext\n");
return -EIO;
}
/* compute the bitmap for the extent reflecting the freed inode.
*/
bitmap = le32_to_cpu(iagp->wmap[extno]) & ~mask;
if (imap->im_agctl[agno].numfree > imap->im_agctl[agno].numinos) {
release_metapage(mp);
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
jfs_error(ip->i_sb, "numfree > numinos\n");
return -EIO;
}
/*
* inode extent still has some inodes or below low water mark:
* keep the inode extent;
*/
if (bitmap ||
imap->im_agctl[agno].numfree < 96 ||
(imap->im_agctl[agno].numfree < 288 &&
(((imap->im_agctl[agno].numfree * 100) /
imap->im_agctl[agno].numinos) <= 25))) {
/* if the iag currently has no free inodes (i.e.,
* the inode being freed is the first free inode of iag),
* insert the iag at head of the inode free list for the ag.
*/
if (iagp->nfreeinos == 0) {
/* check if there are any iags on the ag inode
* free list. if so, read the first one so that
* we can link the current iag onto the list at
* the head.
*/
if ((fwd = imap->im_agctl[agno].inofree) >= 0) {
/* read the iag that currently is the head
* of the list.
*/
if ((rc = diIAGRead(imap, fwd, &))) {
IREAD_UNLOCK(ipimap);
AG_UNLOCK(imap, agno);
release_metapage(mp);
return (rc);
}
aiagp = (struct iag *) amp->data;
/* make current head point back to the iag.
*/
aiagp->inofreeback = cpu_to_le32(iagno);
write_metapage(amp);
}
/* iag points forward to current head and iag
* becomes the new head of the list.
*/
iagp->inofreefwd =
cpu_to_le32(imap->im_agctl[agno].inofree);
iagp->inofreeback = cpu_to_le32(-1);
imap->im_agctl[agno].inofree = iagno;
}
IREAD_UNLOCK(ipimap);
/* update the free inode summary map for the extent if
* freeing the inode means the extent will now have free
* inodes (i.e., the inode being freed is the first free
* inode of extent),
*/
if (iagp->wmap[extno] == cpu_to_le32(ONES)) {
sword = extno >> L2EXTSPERSUM;
bitno = extno & (EXTSPERSUM - 1);
iagp->inosmap[sword] &=
cpu_to_le32(~(HIGHORDER >> bitno));
}
/* update the bitmap.
*/
iagp->wmap[extno] = cpu_to_le32(bitmap);
/* update the free inode counts at the iag, ag and
* map level.
*/
le32_add_cpu(&iagp->nfreeinos, 1);
imap->im_agctl[agno].numfree += 1;