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
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
|
/*
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)ffs_alloc.c 8.8 (Berkeley) 2/21/94
* $Id: ffs_alloc.c,v 1.19 1995/11/19 19:55:26 dyson Exp $
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/vnode.h>
#include <sys/mount.h>
#include <sys/kernel.h>
#include <sys/syslog.h>
#include <vm/vm.h>
#include <ufs/ufs/quota.h>
#include <ufs/ufs/inode.h>
#include <ufs/ufs/ufs_extern.h>
#include <ufs/ffs/fs.h>
#include <ufs/ffs/ffs_extern.h>
extern u_long nextgennumber;
typedef long allocfcn_t __P((struct inode *ip, int cg, daddr_t bpref,
int size));
static daddr_t ffs_alloccg __P((struct inode *, int, daddr_t, int));
static daddr_t ffs_alloccgblk __P((struct fs *, struct cg *, daddr_t));
static daddr_t ffs_clusteralloc __P((struct inode *, int, daddr_t, int));
static ino_t ffs_dirpref __P((struct fs *));
static daddr_t ffs_fragextend __P((struct inode *, int, long, int, int));
static void ffs_fserr __P((struct fs *, u_int, char *));
static u_long ffs_hashalloc
__P((struct inode *, int, long, int, allocfcn_t *));
static ino_t ffs_nodealloccg __P((struct inode *, int, daddr_t, int));
static daddr_t ffs_mapsearch __P((struct fs *, struct cg *, daddr_t, int));
void ffs_clusteracct __P((struct fs *, struct cg *, daddr_t, int));
/*
* Allocate a block in the file system.
*
* The size of the requested block is given, which must be some
* multiple of fs_fsize and <= fs_bsize.
* A preference may be optionally specified. If a preference is given
* the following hierarchy is used to allocate a block:
* 1) allocate the requested block.
* 2) allocate a rotationally optimal block in the same cylinder.
* 3) allocate a block in the same cylinder group.
* 4) quadradically rehash into other cylinder groups, until an
* available block is located.
* If no block preference is given the following heirarchy is used
* to allocate a block:
* 1) allocate a block in the cylinder group that contains the
* inode for the file.
* 2) quadradically rehash into other cylinder groups, until an
* available block is located.
*/
int
ffs_alloc(ip, lbn, bpref, size, cred, bnp)
register struct inode *ip;
daddr_t lbn, bpref;
int size;
struct ucred *cred;
daddr_t *bnp;
{
register struct fs *fs;
daddr_t bno;
int cg;
#ifdef QUOTA
int error;
#endif
*bnp = 0;
fs = ip->i_fs;
#ifdef DIAGNOSTIC
if ((u_int)size > fs->fs_bsize || fragoff(fs, size) != 0) {
printf("dev = 0x%lx, bsize = %ld, size = %d, fs = %s\n",
(u_long)ip->i_dev, fs->fs_bsize, size, fs->fs_fsmnt);
panic("ffs_alloc: bad size");
}
if (cred == NOCRED)
panic("ffs_alloc: missing credential");
#endif /* DIAGNOSTIC */
if (size == fs->fs_bsize && fs->fs_cstotal.cs_nbfree == 0)
goto nospace;
if (cred->cr_uid != 0 && freespace(fs, fs->fs_minfree) <= 0)
goto nospace;
#ifdef QUOTA
error = chkdq(ip, (long)btodb(size), cred, 0);
if (error)
return (error);
#endif
if (bpref >= fs->fs_size)
bpref = 0;
if (bpref == 0)
cg = ino_to_cg(fs, ip->i_number);
else
cg = dtog(fs, bpref);
bno = (daddr_t)ffs_hashalloc(ip, cg, (long)bpref, size, ffs_alloccg);
if (bno > 0) {
ip->i_blocks += btodb(size);
ip->i_flag |= IN_CHANGE | IN_UPDATE;
*bnp = bno;
return (0);
}
#ifdef QUOTA
/*
* Restore user's disk quota because allocation failed.
*/
(void) chkdq(ip, (long)-btodb(size), cred, FORCE);
#endif
nospace:
ffs_fserr(fs, cred->cr_uid, "file system full");
uprintf("\n%s: write failed, file system is full\n", fs->fs_fsmnt);
return (ENOSPC);
}
/*
* Reallocate a fragment to a bigger size
*
* The number and size of the old block is given, and a preference
* and new size is also specified. The allocator attempts to extend
* the original block. Failing that, the regular block allocator is
* invoked to get an appropriate block.
*/
int
ffs_realloccg(ip, lbprev, bpref, osize, nsize, cred, bpp)
register struct inode *ip;
daddr_t lbprev;
daddr_t bpref;
int osize, nsize;
struct ucred *cred;
struct buf **bpp;
{
register struct fs *fs;
struct buf *bp;
int cg, request, error;
daddr_t bprev, bno;
*bpp = 0;
fs = ip->i_fs;
#ifdef DIAGNOSTIC
if ((u_int)osize > fs->fs_bsize || fragoff(fs, osize) != 0 ||
(u_int)nsize > fs->fs_bsize || fragoff(fs, nsize) != 0) {
printf(
"dev = 0x%lx, bsize = %ld, osize = %d, "
"nsize = %d, fs = %s\n",
(u_long)ip->i_dev, fs->fs_bsize, osize,
nsize, fs->fs_fsmnt);
panic("ffs_realloccg: bad size");
}
if (cred == NOCRED)
panic("ffs_realloccg: missing credential");
#endif /* DIAGNOSTIC */
if (cred->cr_uid != 0 && freespace(fs, fs->fs_minfree) <= 0)
goto nospace;
if ((bprev = ip->i_db[lbprev]) == 0) {
printf("dev = 0x%lx, bsize = %ld, bprev = %ld, fs = %s\n",
(u_long) ip->i_dev, fs->fs_bsize, bprev, fs->fs_fsmnt);
panic("ffs_realloccg: bad bprev");
}
/*
* Allocate the extra space in the buffer.
*/
error = bread(ITOV(ip), lbprev, osize, NOCRED, &bp);
if (error) {
brelse(bp);
return (error);
}
if( bp->b_blkno == bp->b_lblkno) {
if( lbprev >= NDADDR)
panic("ffs_realloccg: lbprev out of range");
bp->b_blkno = fsbtodb(fs, bprev);
}
#ifdef QUOTA
error = chkdq(ip, (long)btodb(nsize - osize), cred, 0);
if (error) {
brelse(bp);
return (error);
}
#endif
/*
* Check for extension in the existing location.
*/
cg = dtog(fs, bprev);
bno = ffs_fragextend(ip, cg, (long)bprev, osize, nsize);
if (bno) {
if (bp->b_blkno != fsbtodb(fs, bno))
panic("bad blockno");
ip->i_blocks += btodb(nsize - osize);
ip->i_flag |= IN_CHANGE | IN_UPDATE;
allocbuf(bp, nsize);
bp->b_flags |= B_DONE;
bzero((char *)bp->b_data + osize, (u_int)nsize - osize);
*bpp = bp;
return (0);
}
/*
* Allocate a new disk location.
*/
if (bpref >= fs->fs_size)
bpref = 0;
switch ((int)fs->fs_optim) {
case FS_OPTSPACE:
/*
* Allocate an exact sized fragment. Although this makes
* best use of space, we will waste time relocating it if
* the file continues to grow. If the fragmentation is
* less than half of the minimum free reserve, we choose
* to begin optimizing for time.
*/
request = nsize;
if (fs->fs_minfree <= 5 ||
fs->fs_cstotal.cs_nffree >
fs->fs_dsize * fs->fs_minfree / (2 * 100))
break;
log(LOG_NOTICE, "%s: optimization changed from SPACE to TIME\n",
fs->fs_fsmnt);
fs->fs_optim = FS_OPTTIME;
break;
case FS_OPTTIME:
/*
* At this point we have discovered a file that is trying to
* grow a small fragment to a larger fragment. To save time,
* we allocate a full sized block, then free the unused portion.
* If the file continues to grow, the `ffs_fragextend' call
* above will be able to grow it in place without further
* copying. If aberrant programs cause disk fragmentation to
* grow within 2% of the free reserve, we choose to begin
* optimizing for space.
*/
request = fs->fs_bsize;
if (fs->fs_cstotal.cs_nffree <
fs->fs_dsize * (fs->fs_minfree - 2) / 100)
break;
log(LOG_NOTICE, "%s: optimization changed from TIME to SPACE\n",
fs->fs_fsmnt);
fs->fs_optim = FS_OPTSPACE;
break;
default:
printf("dev = 0x%lx, optim = %ld, fs = %s\n",
(u_long)ip->i_dev, fs->fs_optim, fs->fs_fsmnt);
panic("ffs_realloccg: bad optim");
/* NOTREACHED */
}
bno = (daddr_t)ffs_hashalloc(ip, cg, (long)bpref, request, ffs_alloccg);
if (bno > 0) {
bp->b_blkno = fsbtodb(fs, bno);
ffs_blkfree(ip, bprev, (long)osize);
if (nsize < request)
ffs_blkfree(ip, bno + numfrags(fs, nsize),
(long)(request - nsize));
ip->i_blocks += btodb(nsize - osize);
ip->i_flag |= IN_CHANGE | IN_UPDATE;
allocbuf(bp, nsize);
bp->b_flags |= B_DONE;
bzero((char *)bp->b_data + osize, (u_int)nsize - osize);
*bpp = bp;
return (0);
}
#ifdef QUOTA
/*
* Restore user's disk quota because allocation failed.
*/
(void) chkdq(ip, (long)-btodb(nsize - osize), cred, FORCE);
#endif
brelse(bp);
nospace:
/*
* no space available
*/
ffs_fserr(fs, cred->cr_uid, "file system full");
uprintf("\n%s: write failed, file system is full\n", fs->fs_fsmnt);
return (ENOSPC);
}
/*
* Reallocate a sequence of blocks into a contiguous sequence of blocks.
*
* The vnode and an array of buffer pointers for a range of sequential
* logical blocks to be made contiguous is given. The allocator attempts
* to find a range of sequential blocks starting as close as possible to
* an fs_rotdelay offset from the end of the allocation for the logical
* block immediately preceeding the current range. If successful, the
* physical block numbers in the buffer pointers and in the inode are
* changed to reflect the new allocation. If unsuccessful, the allocation
* is left unchanged. The success in doing the reallocation is returned.
* Note that the error return is not reflected back to the user. Rather
* the previous block allocation will be used.
*/
#include <sys/sysctl.h>
int doasyncfree = 1;
#ifdef DEBUG
SYSCTL_INT(_debug, 14, doasyncfree, CTLFLAG_RW, &doasyncfree, 0, "");
#endif
int
ffs_reallocblks(ap)
struct vop_reallocblks_args /* {
struct vnode *a_vp;
struct cluster_save *a_buflist;
} */ *ap;
{
#if 1
return (ENOSPC);
#else
struct fs *fs;
struct inode *ip;
struct vnode *vp;
struct buf *sbp, *ebp;
daddr_t *bap, *sbap, *ebap = 0;
struct cluster_save *buflist;
daddr_t start_lbn, end_lbn, soff, newblk, blkno;
struct indir start_ap[NIADDR + 1], end_ap[NIADDR + 1], *idp;
int i, len, start_lvl, end_lvl, pref, ssize;
struct timeval tv;
vp = ap->a_vp;
ip = VTOI(vp);
fs = ip->i_fs;
if (fs->fs_contigsumsize <= 0)
return (ENOSPC);
buflist = ap->a_buflist;
len = buflist->bs_nchildren;
start_lbn = buflist->bs_children[0]->b_lblkno;
end_lbn = start_lbn + len - 1;
#ifdef DIAGNOSTIC
for (i = 1; i < len; i++)
if (buflist->bs_children[i]->b_lblkno != start_lbn + i)
panic("ffs_reallocblks: non-cluster");
#endif
/*
* If the latest allocation is in a new cylinder group, assume that
* the filesystem has decided to move and do not force it back to
* the previous cylinder group.
*/
if (dtog(fs, dbtofsb(fs, buflist->bs_children[0]->b_blkno)) !=
dtog(fs, dbtofsb(fs, buflist->bs_children[len - 1]->b_blkno)))
return (ENOSPC);
if (ufs_getlbns(vp, start_lbn, start_ap, &start_lvl) ||
ufs_getlbns(vp, end_lbn, end_ap, &end_lvl))
return (ENOSPC);
/*
* Get the starting offset and block map for the first block.
*/
if (start_lvl == 0) {
sbap = &ip->i_db[0];
soff = start_lbn;
} else {
idp = &start_ap[start_lvl - 1];
if (bread(vp, idp->in_lbn, (int)fs->fs_bsize, NOCRED, &sbp)) {
brelse(sbp);
return (ENOSPC);
}
sbap = (daddr_t *)sbp->b_data;
soff = idp->in_off;
}
/*
* Find the preferred location for the cluster.
*/
pref = ffs_blkpref(ip, start_lbn, soff, sbap);
/*
* If the block range spans two block maps, get the second map.
*/
if (end_lvl == 0 || (idp = &end_ap[end_lvl - 1])->in_off + 1 >= len) {
ssize = len;
} else {
#ifdef DIAGNOSTIC
if (start_ap[start_lvl-1].in_lbn == idp->in_lbn)
panic("ffs_reallocblk: start == end");
#endif
ssize = len - (idp->in_off + 1);
if (bread(vp, idp->in_lbn, (int)fs->fs_bsize, NOCRED, &ebp))
goto fail;
ebap = (daddr_t *)ebp->b_data;
}
/*
* Search the block map looking for an allocation of the desired size.
*/
if ((newblk = (daddr_t)ffs_hashalloc(ip, dtog(fs, pref), (long)pref,
len, ffs_clusteralloc)) == 0)
goto fail;
/*
* We have found a new contiguous block.
*
* First we have to replace the old block pointers with the new
* block pointers in the inode and indirect blocks associated
* with the file.
*/
blkno = newblk;
for (bap = &sbap[soff], i = 0; i < len; i++, blkno += fs->fs_frag) {
if (i == ssize)
bap = ebap;
#ifdef DIAGNOSTIC
if (buflist->bs_children[i]->b_blkno != fsbtodb(fs, *bap))
panic("ffs_reallocblks: alloc mismatch");
#endif
*bap++ = blkno;
}
/*
* Next we must write out the modified inode and indirect blocks.
* For strict correctness, the writes should be synchronous since
* the old block values may have been written to disk. In practise
* they are almost never written, but if we are concerned about
* strict correctness, the `doasyncfree' flag should be set to zero.
*
* The test on `doasyncfree' should be changed to test a flag
* that shows whether the associated buffers and inodes have
* been written. The flag should be set when the cluster is
* started and cleared whenever the buffer or inode is flushed.
* We can then check below to see if it is set, and do the
* synchronous write only when it has been cleared.
*/
if (sbap != &ip->i_db[0]) {
if (doasyncfree)
bdwrite(sbp);
else
bwrite(sbp);
} else {
ip->i_flag |= IN_CHANGE | IN_UPDATE;
if (!doasyncfree) {
tv = time;
VOP_UPDATE(vp, &tv, &tv, 1);
}
}
if (ssize < len)
if (doasyncfree)
bdwrite(ebp);
else
bwrite(ebp);
/*
* Last, free the old blocks and assign the new blocks to the buffers.
*/
for (blkno = newblk, i = 0; i < len; i++, blkno += fs->fs_frag) {
ffs_blkfree(ip, dbtofsb(fs, buflist->bs_children[i]->b_blkno),
fs->fs_bsize);
buflist->bs_children[i]->b_blkno = fsbtodb(fs, blkno);
}
return (0);
fail:
if (ssize < len)
brelse(ebp);
if (sbap != &ip->i_db[0])
brelse(sbp);
return (ENOSPC);
#endif
}
/*
* Allocate an inode in the file system.
*
* If allocating a directory, use ffs_dirpref to select the inode.
* If allocating in a directory, the following hierarchy is followed:
* 1) allocate the preferred inode.
* 2) allocate an inode in the same cylinder group.
* 3) quadradically rehash into other cylinder groups, until an
* available inode is located.
* If no inode preference is given the following heirarchy is used
* to allocate an inode:
* 1) allocate an inode in cylinder group 0.
* 2) quadradically rehash into other cylinder groups, until an
* available inode is located.
*/
int
ffs_valloc(ap)
struct vop_valloc_args /* {
struct vnode *a_pvp;
int a_mode;
struct ucred *a_cred;
struct vnode **a_vpp;
} */ *ap;
{
register struct vnode *pvp = ap->a_pvp;
register struct inode *pip;
register struct fs *fs;
register struct inode *ip;
mode_t mode = ap->a_mode;
ino_t ino, ipref;
int cg, error;
*ap->a_vpp = NULL;
pip = VTOI(pvp);
fs = pip->i_fs;
if (fs->fs_cstotal.cs_nifree == 0)
goto noinodes;
if ((mode & IFMT) == IFDIR)
ipref = ffs_dirpref(fs);
else
ipref = pip->i_number;
if (ipref >= fs->fs_ncg * fs->fs_ipg)
ipref = 0;
cg = ino_to_cg(fs, ipref);
ino = (ino_t)ffs_hashalloc(pip, cg, (long)ipref, mode, ffs_nodealloccg);
if (ino == 0)
goto noinodes;
error = VFS_VGET(pvp->v_mount, ino, ap->a_vpp);
if (error) {
VOP_VFREE(pvp, ino, mode);
return (error);
}
ip = VTOI(*ap->a_vpp);
if (ip->i_mode) {
printf("mode = 0%o, inum = %ld, fs = %s\n",
ip->i_mode, ip->i_number, fs->fs_fsmnt);
panic("ffs_valloc: dup alloc");
}
if (ip->i_blocks) { /* XXX */
printf("free inode %s/%ld had %ld blocks\n",
fs->fs_fsmnt, ino, ip->i_blocks);
ip->i_blocks = 0;
}
ip->i_flags = 0;
/*
* Set up a new generation number for this inode.
*/
if (++nextgennumber < (u_long)time.tv_sec)
nextgennumber = time.tv_sec;
ip->i_gen = nextgennumber;
return (0);
noinodes:
ffs_fserr(fs, ap->a_cred->cr_uid, "out of inodes");
uprintf("\n%s: create/symlink failed, no inodes free\n", fs->fs_fsmnt);
return (ENOSPC);
}
/*
* Find a cylinder to place a directory.
*
* The policy implemented by this algorithm is to select from
* among those cylinder groups with above the average number of
* free inodes, the one with the smallest number of directories.
*/
static ino_t
ffs_dirpref(fs)
register struct fs *fs;
{
int cg, minndir, mincg, avgifree;
avgifree = fs->fs_cstotal.cs_nifree / fs->fs_ncg;
minndir = fs->fs_ipg;
mincg = 0;
for (cg = 0; cg < fs->fs_ncg; cg++)
if (fs->fs_cs(fs, cg).cs_ndir < minndir &&
fs->fs_cs(fs, cg).cs_nifree >= avgifree) {
mincg = cg;
minndir = fs->fs_cs(fs, cg).cs_ndir;
}
return ((ino_t)(fs->fs_ipg * mincg));
}
/*
* Select the desired position for the next block in a file. The file is
* logically divided into sections. The first section is composed of the
* direct blocks. Each additional section contains fs_maxbpg blocks.
*
* If no blocks have been allocated in the first section, the policy is to
* request a block in the same cylinder group as the inode that describes
* the file. If no blocks have been allocated in any other section, the
* policy is to place the section in a cylinder group with a greater than
* average number of free blocks. An appropriate cylinder group is found
* by using a rotor that sweeps the cylinder groups. When a new group of
* blocks is needed, the sweep begins in the cylinder group following the
* cylinder group from which the previous allocation was made. The sweep
* continues until a cylinder group with greater than the average number
* of free blocks is found. If the allocation is for the first block in an
* indirect block, the information on the previous allocation is unavailable;
* here a best guess is made based upon the logical block number being
* allocated.
*
* If a section is already partially allocated, the policy is to
* contiguously allocate fs_maxcontig blocks. The end of one of these
* contiguous blocks and the beginning of the next is physically separated
* so that the disk head will be in transit between them for at least
* fs_rotdelay milliseconds. This is to allow time for the processor to
* schedule another I/O transfer.
*/
daddr_t
ffs_blkpref(ip, lbn, indx, bap)
struct inode *ip;
daddr_t lbn;
int indx;
daddr_t *bap;
{
register struct fs *fs;
register int cg;
int avgbfree, startcg;
daddr_t nextblk;
fs = ip->i_fs;
if (indx % fs->fs_maxbpg == 0 || bap[indx - 1] == 0) {
if (lbn < NDADDR) {
cg = ino_to_cg(fs, ip->i_number);
return (fs->fs_fpg * cg + fs->fs_frag);
}
/*
* Find a cylinder with greater than average number of
* unused data blocks.
*/
if (indx == 0 || bap[indx - 1] == 0)
startcg =
ino_to_cg(fs, ip->i_number) + lbn / fs->fs_maxbpg;
else
startcg = dtog(fs, bap[indx - 1]) + 1;
startcg %= fs->fs_ncg;
avgbfree = fs->fs_cstotal.cs_nbfree / fs->fs_ncg;
for (cg = startcg; cg < fs->fs_ncg; cg++)
if (fs->fs_cs(fs, cg).cs_nbfree >= avgbfree) {
fs->fs_cgrotor = cg;
return (fs->fs_fpg * cg + fs->fs_frag);
}
for (cg = 0; cg <= startcg; cg++)
if (fs->fs_cs(fs, cg).cs_nbfree >= avgbfree) {
fs->fs_cgrotor = cg;
return (fs->fs_fpg * cg + fs->fs_frag);
}
return (NULL);
}
/*
* One or more previous blocks have been laid out. If less
* than fs_maxcontig previous blocks are contiguous, the
* next block is requested contiguously, otherwise it is
* requested rotationally delayed by fs_rotdelay milliseconds.
*/
nextblk = bap[indx - 1] + fs->fs_frag;
if (fs->fs_rotdelay == 0 || indx < fs->fs_maxcontig ||
bap[indx - fs->fs_maxcontig] +
blkstofrags(fs, fs->fs_maxcontig) != nextblk)
return (nextblk);
/*
* Here we convert ms of delay to frags as:
* (frags) = (ms) * (rev/sec) * (sect/rev) /
* ((sect/frag) * (ms/sec))
* then round up to the next block.
*/
nextblk += roundup(fs->fs_rotdelay * fs->fs_rps * fs->fs_nsect /
(NSPF(fs) * 1000), fs->fs_frag);
return (nextblk);
}
/*
* Implement the cylinder overflow algorithm.
*
* The policy implemented by this algorithm is:
* 1) allocate the block in its requested cylinder group.
* 2) quadradically rehash on the cylinder group number.
* 3) brute force search for a free block.
*/
/*VARARGS5*/
static u_long
ffs_hashalloc(ip, cg, pref, size, allocator)
struct inode *ip;
int cg;
long pref;
int size; /* size for data blocks, mode for inodes */
allocfcn_t *allocator;
{
register struct fs *fs;
long result; /* XXX why not same type as we return? */
int i, icg = cg;
fs = ip->i_fs;
/*
* 1: preferred cylinder group
*/
result = (*allocator)(ip, cg, pref, size);
if (result)
return (result);
/*
* 2: quadratic rehash
*/
for (i = 1; i < fs->fs_ncg; i *= 2) {
cg += i;
if (cg >= fs->fs_ncg)
cg -= fs->fs_ncg;
result = (*allocator)(ip, cg, 0, size);
if (result)
return (result);
}
/*
* 3: brute force search
* Note that we start at i == 2, since 0 was checked initially,
* and 1 is always checked in the quadratic rehash.
*/
cg = (icg + 2) % fs->fs_ncg;
for (i = 2; i < fs->fs_ncg; i++) {
result = (*allocator)(ip, cg, 0, size);
if (result)
return (result);
cg++;
if (cg == fs->fs_ncg)
cg = 0;
}
return (0);
}
/*
* Determine whether a fragment can be extended.
*
* Check to see if the necessary fragments are available, and
* if they are, allocate them.
*/
static daddr_t
ffs_fragextend(ip, cg, bprev, osize, nsize)
struct inode *ip;
int cg;
long bprev;
int osize, nsize;
{
register struct fs *fs;
register struct cg *cgp;
struct buf *bp;
long bno;
int frags, bbase;
int i, error;
fs = ip->i_fs;
if (fs->fs_cs(fs, cg).cs_nffree < numfrags(fs, nsize - osize))
return (NULL);
frags = numfrags(fs, nsize);
bbase = fragnum(fs, bprev);
if (bbase > fragnum(fs, (bprev + frags - 1))) {
/* cannot extend across a block boundary */
return (NULL);
}
error = bread(ip->i_devvp, fsbtodb(fs, cgtod(fs, cg)),
(int)fs->fs_cgsize, NOCRED, &bp);
if (error) {
brelse(bp);
return (NULL);
}
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp)) {
brelse(bp);
return (NULL);
}
cgp->cg_time = time.tv_sec;
bno = dtogd(fs, bprev);
for (i = numfrags(fs, osize); i < frags; i++)
if (isclr(cg_blksfree(cgp), bno + i)) {
brelse(bp);
return (NULL);
}
/*
* the current fragment can be extended
* deduct the count on fragment being extended into
* increase the count on the remaining fragment (if any)
* allocate the extended piece
*/
for (i = frags; i < fs->fs_frag - bbase; i++)
if (isclr(cg_blksfree(cgp), bno + i))
break;
cgp->cg_frsum[i - numfrags(fs, osize)]--;
if (i != frags)
cgp->cg_frsum[i - frags]++;
for (i = numfrags(fs, osize); i < frags; i++) {
clrbit(cg_blksfree(cgp), bno + i);
cgp->cg_cs.cs_nffree--;
fs->fs_cstotal.cs_nffree--;
fs->fs_cs(fs, cg).cs_nffree--;
}
fs->fs_fmod = 1;
bdwrite(bp);
return (bprev);
}
/*
* Determine whether a block can be allocated.
*
* Check to see if a block of the appropriate size is available,
* and if it is, allocate it.
*/
static daddr_t
ffs_alloccg(ip, cg, bpref, size)
struct inode *ip;
int cg;
daddr_t bpref;
int size;
{
register struct fs *fs;
register struct cg *cgp;
struct buf *bp;
register int i;
int error, bno, frags, allocsiz;
fs = ip->i_fs;
if (fs->fs_cs(fs, cg).cs_nbfree == 0 && size == fs->fs_bsize)
return (NULL);
error = bread(ip->i_devvp, fsbtodb(fs, cgtod(fs, cg)),
(int)fs->fs_cgsize, NOCRED, &bp);
if (error) {
brelse(bp);
return (NULL);
}
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp) ||
(cgp->cg_cs.cs_nbfree == 0 && size == fs->fs_bsize)) {
brelse(bp);
return (NULL);
}
cgp->cg_time = time.tv_sec;
if (size == fs->fs_bsize) {
bno = ffs_alloccgblk(fs, cgp, bpref);
bdwrite(bp);
return (bno);
}
/*
* check to see if any fragments are already available
* allocsiz is the size which will be allocated, hacking
* it down to a smaller size if necessary
*/
frags = numfrags(fs, size);
for (allocsiz = frags; allocsiz < fs->fs_frag; allocsiz++)
if (cgp->cg_frsum[allocsiz] != 0)
break;
if (allocsiz == fs->fs_frag) {
/*
* no fragments were available, so a block will be
* allocated, and hacked up
*/
if (cgp->cg_cs.cs_nbfree == 0) {
brelse(bp);
return (NULL);
}
bno = ffs_alloccgblk(fs, cgp, bpref);
bpref = dtogd(fs, bno);
for (i = frags; i < fs->fs_frag; i++)
setbit(cg_blksfree(cgp), bpref + i);
i = fs->fs_frag - frags;
cgp->cg_cs.cs_nffree += i;
fs->fs_cstotal.cs_nffree += i;
fs->fs_cs(fs, cg).cs_nffree += i;
fs->fs_fmod = 1;
cgp->cg_frsum[i]++;
bdwrite(bp);
return (bno);
}
bno = ffs_mapsearch(fs, cgp, bpref, allocsiz);
if (bno < 0) {
brelse(bp);
return (NULL);
}
for (i = 0; i < frags; i++)
clrbit(cg_blksfree(cgp), bno + i);
cgp->cg_cs.cs_nffree -= frags;
fs->fs_cstotal.cs_nffree -= frags;
fs->fs_cs(fs, cg).cs_nffree -= frags;
fs->fs_fmod = 1;
cgp->cg_frsum[allocsiz]--;
if (frags != allocsiz)
cgp->cg_frsum[allocsiz - frags]++;
bdwrite(bp);
return (cg * fs->fs_fpg + bno);
}
/*
* Allocate a block in a cylinder group.
*
* This algorithm implements the following policy:
* 1) allocate the requested block.
* 2) allocate a rotationally optimal block in the same cylinder.
* 3) allocate the next available block on the block rotor for the
* specified cylinder group.
* Note that this routine only allocates fs_bsize blocks; these
* blocks may be fragmented by the routine that allocates them.
*/
static daddr_t
ffs_alloccgblk(fs, cgp, bpref)
register struct fs *fs;
register struct cg *cgp;
daddr_t bpref;
{
daddr_t bno, blkno;
int cylno, pos, delta;
short *cylbp;
register int i;
if (bpref == 0 || dtog(fs, bpref) != cgp->cg_cgx) {
bpref = cgp->cg_rotor;
goto norot;
}
bpref = blknum(fs, bpref);
bpref = dtogd(fs, bpref);
/*
* if the requested block is available, use it
*/
if (ffs_isblock(fs, cg_blksfree(cgp), fragstoblks(fs, bpref))) {
bno = bpref;
goto gotit;
}
if (fs->fs_nrpos <= 1 || fs->fs_cpc == 0) {
/*
* Block layout information is not available.
* Leaving bpref unchanged means we take the
* next available free block following the one
* we just allocated. Hopefully this will at
* least hit a track cache on drives of unknown
* geometry (e.g. SCSI).
*/
goto norot;
}
/*
* check for a block available on the same cylinder
*/
cylno = cbtocylno(fs, bpref);
if (cg_blktot(cgp)[cylno] == 0)
goto norot;
/*
* check the summary information to see if a block is
* available in the requested cylinder starting at the
* requested rotational position and proceeding around.
*/
cylbp = cg_blks(fs, cgp, cylno);
pos = cbtorpos(fs, bpref);
for (i = pos; i < fs->fs_nrpos; i++)
if (cylbp[i] > 0)
break;
if (i == fs->fs_nrpos)
for (i = 0; i < pos; i++)
if (cylbp[i] > 0)
break;
if (cylbp[i] > 0) {
/*
* found a rotational position, now find the actual
* block. A panic if none is actually there.
*/
pos = cylno % fs->fs_cpc;
bno = (cylno - pos) * fs->fs_spc / NSPB(fs);
if (fs_postbl(fs, pos)[i] == -1) {
printf("pos = %d, i = %d, fs = %s\n",
pos, i, fs->fs_fsmnt);
panic("ffs_alloccgblk: cyl groups corrupted");
}
for (i = fs_postbl(fs, pos)[i];; ) {
if (ffs_isblock(fs, cg_blksfree(cgp), bno + i)) {
bno = blkstofrags(fs, (bno + i));
goto gotit;
}
delta = fs_rotbl(fs)[i];
if (delta <= 0 ||
delta + i > fragstoblks(fs, fs->fs_fpg))
break;
i += delta;
}
printf("pos = %d, i = %d, fs = %s\n", pos, i, fs->fs_fsmnt);
panic("ffs_alloccgblk: can't find blk in cyl");
}
norot:
/*
* no blocks in the requested cylinder, so take next
* available one in this cylinder group.
*/
bno = ffs_mapsearch(fs, cgp, bpref, (int)fs->fs_frag);
if (bno < 0)
return (NULL);
cgp->cg_rotor = bno;
gotit:
blkno = fragstoblks(fs, bno);
ffs_clrblock(fs, cg_blksfree(cgp), (long)blkno);
ffs_clusteracct(fs, cgp, blkno, -1);
cgp->cg_cs.cs_nbfree--;
fs->fs_cstotal.cs_nbfree--;
fs->fs_cs(fs, cgp->cg_cgx).cs_nbfree--;
cylno = cbtocylno(fs, bno);
cg_blks(fs, cgp, cylno)[cbtorpos(fs, bno)]--;
cg_blktot(cgp)[cylno]--;
fs->fs_fmod = 1;
return (cgp->cg_cgx * fs->fs_fpg + bno);
}
/*
* Determine whether a cluster can be allocated.
*
* We do not currently check for optimal rotational layout if there
* are multiple choices in the same cylinder group. Instead we just
* take the first one that we find following bpref.
*/
static daddr_t
ffs_clusteralloc(ip, cg, bpref, len)
struct inode *ip;
int cg;
daddr_t bpref;
int len;
{
register struct fs *fs;
register struct cg *cgp;
struct buf *bp;
int i, run, bno, bit, map;
u_char *mapp;
fs = ip->i_fs;
if (fs->fs_cs(fs, cg).cs_nbfree < len)
return (NULL);
if (bread(ip->i_devvp, fsbtodb(fs, cgtod(fs, cg)), (int)fs->fs_cgsize,
NOCRED, &bp))
goto fail;
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp))
goto fail;
/*
* Check to see if a cluster of the needed size (or bigger) is
* available in this cylinder group.
*/
for (i = len; i <= fs->fs_contigsumsize; i++)
if (cg_clustersum(cgp)[i] > 0)
break;
if (i > fs->fs_contigsumsize)
goto fail;
/*
* Search the cluster map to find a big enough cluster.
* We take the first one that we find, even if it is larger
* than we need as we prefer to get one close to the previous
* block allocation. We do not search before the current
* preference point as we do not want to allocate a block
* that is allocated before the previous one (as we will
* then have to wait for another pass of the elevator
* algorithm before it will be read). We prefer to fail and
* be recalled to try an allocation in the next cylinder group.
*/
if (dtog(fs, bpref) != cg)
bpref = 0;
else
bpref = fragstoblks(fs, dtogd(fs, blknum(fs, bpref)));
mapp = &cg_clustersfree(cgp)[bpref / NBBY];
map = *mapp++;
bit = 1 << (bpref % NBBY);
for (run = 0, i = bpref; i < cgp->cg_nclusterblks; i++) {
if ((map & bit) == 0) {
run = 0;
} else {
run++;
if (run == len)
break;
}
if ((i & (NBBY - 1)) != (NBBY - 1)) {
bit <<= 1;
} else {
map = *mapp++;
bit = 1;
}
}
if (i == cgp->cg_nclusterblks)
goto fail;
/*
* Allocate the cluster that we have found.
*/
bno = cg * fs->fs_fpg + blkstofrags(fs, i - run + 1);
len = blkstofrags(fs, len);
for (i = 0; i < len; i += fs->fs_frag)
if (ffs_alloccgblk(fs, cgp, bno + i) != bno + i)
panic("ffs_clusteralloc: lost block");
bdwrite(bp);
return (bno);
fail:
brelse(bp);
return (0);
}
/*
* Determine whether an inode can be allocated.
*
* Check to see if an inode is available, and if it is,
* allocate it using the following policy:
* 1) allocate the requested inode.
* 2) allocate the next available inode after the requested
* inode in the specified cylinder group.
*/
static ino_t
ffs_nodealloccg(ip, cg, ipref, mode)
struct inode *ip;
int cg;
daddr_t ipref;
int mode;
{
register struct fs *fs;
register struct cg *cgp;
struct buf *bp;
int error, start, len, loc, map, i;
fs = ip->i_fs;
if (fs->fs_cs(fs, cg).cs_nifree == 0)
return (NULL);
error = bread(ip->i_devvp, fsbtodb(fs, cgtod(fs, cg)),
(int)fs->fs_cgsize, NOCRED, &bp);
if (error) {
brelse(bp);
return (NULL);
}
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp) || cgp->cg_cs.cs_nifree == 0) {
brelse(bp);
return (NULL);
}
cgp->cg_time = time.tv_sec;
if (ipref) {
ipref %= fs->fs_ipg;
if (isclr(cg_inosused(cgp), ipref))
goto gotit;
}
start = cgp->cg_irotor / NBBY;
len = howmany(fs->fs_ipg - cgp->cg_irotor, NBBY);
loc = skpc(0xff, len, &cg_inosused(cgp)[start]);
if (loc == 0) {
len = start + 1;
start = 0;
loc = skpc(0xff, len, &cg_inosused(cgp)[0]);
if (loc == 0) {
printf("cg = %d, irotor = %ld, fs = %s\n",
cg, cgp->cg_irotor, fs->fs_fsmnt);
panic("ffs_nodealloccg: map corrupted");
/* NOTREACHED */
}
}
i = start + len - loc;
map = cg_inosused(cgp)[i];
ipref = i * NBBY;
for (i = 1; i < (1 << NBBY); i <<= 1, ipref++) {
if ((map & i) == 0) {
cgp->cg_irotor = ipref;
goto gotit;
}
}
printf("fs = %s\n", fs->fs_fsmnt);
panic("ffs_nodealloccg: block not in map");
/* NOTREACHED */
gotit:
setbit(cg_inosused(cgp), ipref);
cgp->cg_cs.cs_nifree--;
fs->fs_cstotal.cs_nifree--;
fs->fs_cs(fs, cg).cs_nifree--;
fs->fs_fmod = 1;
if ((mode & IFMT) == IFDIR) {
cgp->cg_cs.cs_ndir++;
fs->fs_cstotal.cs_ndir++;
fs->fs_cs(fs, cg).cs_ndir++;
}
bdwrite(bp);
return (cg * fs->fs_ipg + ipref);
}
/*
* Free a block or fragment.
*
* The specified block or fragment is placed back in the
* free map. If a fragment is deallocated, a possible
* block reassembly is checked.
*/
void
ffs_blkfree(ip, bno, size)
register struct inode *ip;
daddr_t bno;
long size;
{
register struct fs *fs;
register struct cg *cgp;
struct buf *bp;
daddr_t blkno;
int i, error, cg, blk, frags, bbase;
fs = ip->i_fs;
if ((u_int)size > fs->fs_bsize || fragoff(fs, size) != 0) {
printf("dev = 0x%lx, bsize = %ld, size = %ld, fs = %s\n",
(u_long)ip->i_dev, fs->fs_bsize, size, fs->fs_fsmnt);
panic("blkfree: bad size");
}
cg = dtog(fs, bno);
if ((u_int)bno >= fs->fs_size) {
printf("bad block %ld, ino %ld\n", bno, ip->i_number);
ffs_fserr(fs, ip->i_uid, "bad block");
return;
}
error = bread(ip->i_devvp, fsbtodb(fs, cgtod(fs, cg)),
(int)fs->fs_cgsize, NOCRED, &bp);
if (error) {
brelse(bp);
return;
}
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp)) {
brelse(bp);
return;
}
cgp->cg_time = time.tv_sec;
bno = dtogd(fs, bno);
if (size == fs->fs_bsize) {
blkno = fragstoblks(fs, bno);
if (ffs_isblock(fs, cg_blksfree(cgp), blkno)) {
printf("dev = 0x%lx, block = %ld, fs = %s\n",
(u_long) ip->i_dev, bno, fs->fs_fsmnt);
panic("blkfree: freeing free block");
}
ffs_setblock(fs, cg_blksfree(cgp), blkno);
ffs_clusteracct(fs, cgp, blkno, 1);
cgp->cg_cs.cs_nbfree++;
fs->fs_cstotal.cs_nbfree++;
fs->fs_cs(fs, cg).cs_nbfree++;
i = cbtocylno(fs, bno);
cg_blks(fs, cgp, i)[cbtorpos(fs, bno)]++;
cg_blktot(cgp)[i]++;
} else {
bbase = bno - fragnum(fs, bno);
/*
* decrement the counts associated with the old frags
*/
blk = blkmap(fs, cg_blksfree(cgp), bbase);
ffs_fragacct(fs, blk, cgp->cg_frsum, -1);
/*
* deallocate the fragment
*/
frags = numfrags(fs, size);
for (i = 0; i < frags; i++) {
if (isset(cg_blksfree(cgp), bno + i)) {
printf("dev = 0x%lx, block = %ld, fs = %s\n",
(u_long) ip->i_dev, bno + i, fs->fs_fsmnt);
panic("blkfree: freeing free frag");
}
setbit(cg_blksfree(cgp), bno + i);
}
cgp->cg_cs.cs_nffree += i;
fs->fs_cstotal.cs_nffree += i;
fs->fs_cs(fs, cg).cs_nffree += i;
/*
* add back in counts associated with the new frags
*/
blk = blkmap(fs, cg_blksfree(cgp), bbase);
ffs_fragacct(fs, blk, cgp->cg_frsum, 1);
/*
* if a complete block has been reassembled, account for it
*/
blkno = fragstoblks(fs, bbase);
if (ffs_isblock(fs, cg_blksfree(cgp), blkno)) {
cgp->cg_cs.cs_nffree -= fs->fs_frag;
fs->fs_cstotal.cs_nffree -= fs->fs_frag;
fs->fs_cs(fs, cg).cs_nffree -= fs->fs_frag;
ffs_clusteracct(fs, cgp, blkno, 1);
cgp->cg_cs.cs_nbfree++;
fs->fs_cstotal.cs_nbfree++;
fs->fs_cs(fs, cg).cs_nbfree++;
i = cbtocylno(fs, bbase);
cg_blks(fs, cgp, i)[cbtorpos(fs, bbase)]++;
cg_blktot(cgp)[i]++;
}
}
fs->fs_fmod = 1;
bdwrite(bp);
}
/*
* Free an inode.
*
* The specified inode is placed back in the free map.
*/
int
ffs_vfree(ap)
struct vop_vfree_args /* {
struct vnode *a_pvp;
ino_t a_ino;
int a_mode;
} */ *ap;
{
register struct fs *fs;
register struct cg *cgp;
register struct inode *pip;
ino_t ino = ap->a_ino;
struct buf *bp;
int error, cg;
pip = VTOI(ap->a_pvp);
fs = pip->i_fs;
if ((u_int)ino >= fs->fs_ipg * fs->fs_ncg)
panic("ifree: range: dev = 0x%x, ino = %d, fs = %s",
pip->i_dev, ino, fs->fs_fsmnt);
cg = ino_to_cg(fs, ino);
error = bread(pip->i_devvp, fsbtodb(fs, cgtod(fs, cg)),
(int)fs->fs_cgsize, NOCRED, &bp);
if (error) {
brelse(bp);
return (0);
}
cgp = (struct cg *)bp->b_data;
if (!cg_chkmagic(cgp)) {
brelse(bp);
return (0);
}
cgp->cg_time = time.tv_sec;
ino %= fs->fs_ipg;
if (isclr(cg_inosused(cgp), ino)) {
printf("dev = 0x%lx, ino = %ld, fs = %s\n",
(u_long)pip->i_dev, ino, fs->fs_fsmnt);
if (fs->fs_ronly == 0)
panic("ifree: freeing free inode");
}
clrbit(cg_inosused(cgp), ino);
if (ino < cgp->cg_irotor)
cgp->cg_irotor = ino;
cgp->cg_cs.cs_nifree++;
fs->fs_cstotal.cs_nifree++;
fs->fs_cs(fs, cg).cs_nifree++;
if ((ap->a_mode & IFMT) == IFDIR) {
cgp->cg_cs.cs_ndir--;
fs->fs_cstotal.cs_ndir--;
fs->fs_cs(fs, cg).cs_ndir--;
}
fs->fs_fmod = 1;
bdwrite(bp);
return (0);
}
/*
* Find a block of the specified size in the specified cylinder group.
*
* It is a panic if a request is made to find a block if none are
* available.
*/
static daddr_t
ffs_mapsearch(fs, cgp, bpref, allocsiz)
register struct fs *fs;
register struct cg *cgp;
daddr_t bpref;
int allocsiz;
{
daddr_t bno;
int start, len, loc, i;
int blk, field, subfield, pos;
/*
* find the fragment by searching through the free block
* map for an appropriate bit pattern
*/
if (bpref)
start = dtogd(fs, bpref) / NBBY;
else
start = cgp->cg_frotor / NBBY;
len = howmany(fs->fs_fpg, NBBY) - start;
loc = scanc((u_int)len, (u_char *)&cg_blksfree(cgp)[start],
(u_char *)fragtbl[fs->fs_frag],
(u_char)(1 << (allocsiz - 1 + (fs->fs_frag % NBBY))));
if (loc == 0) {
len = start + 1;
start = 0;
loc = scanc((u_int)len, (u_char *)&cg_blksfree(cgp)[0],
(u_char *)fragtbl[fs->fs_frag],
(u_char)(1 << (allocsiz - 1 + (fs->fs_frag % NBBY))));
if (loc == 0) {
printf("start = %d, len = %d, fs = %s\n",
start, len, fs->fs_fsmnt);
panic("ffs_alloccg: map corrupted");
/* NOTREACHED */
}
}
bno = (start + len - loc) * NBBY;
cgp->cg_frotor = bno;
/*
* found the byte in the map
* sift through the bits to find the selected frag
*/
for (i = bno + NBBY; bno < i; bno += fs->fs_frag) {
blk = blkmap(fs, cg_blksfree(cgp), bno);
blk <<= 1;
field = around[allocsiz];
subfield = inside[allocsiz];
for (pos = 0; pos <= fs->fs_frag - allocsiz; pos++) {
if ((blk & field) == subfield)
return (bno + pos);
field <<= 1;
subfield <<= 1;
}
}
printf("bno = %lu, fs = %s\n", (u_long)bno, fs->fs_fsmnt);
panic("ffs_alloccg: block not in map");
return (-1);
}
/*
* Update the cluster map because of an allocation or free.
*
* Cnt == 1 means free; cnt == -1 means allocating.
*/
void
ffs_clusteracct(fs, cgp, blkno, cnt)
struct fs *fs;
struct cg *cgp;
daddr_t blkno;
int cnt;
{
long *sump;
u_char *freemapp, *mapp;
int i, start, end, forw, back, map, bit;
if (fs->fs_contigsumsize <= 0)
return;
freemapp = cg_clustersfree(cgp);
sump = cg_clustersum(cgp);
/*
* Allocate or clear the actual block.
*/
if (cnt > 0)
setbit(freemapp, blkno);
else
clrbit(freemapp, blkno);
/*
* Find the size of the cluster going forward.
*/
start = blkno + 1;
end = start + fs->fs_contigsumsize;
if (end >= cgp->cg_nclusterblks)
end = cgp->cg_nclusterblks;
mapp = &freemapp[start / NBBY];
map = *mapp++;
bit = 1 << (start % NBBY);
for (i = start; i < end; i++) {
if ((map & bit) == 0)
break;
if ((i & (NBBY - 1)) != (NBBY - 1)) {
bit <<= 1;
} else {
map = *mapp++;
bit = 1;
}
}
forw = i - start;
/*
* Find the size of the cluster going backward.
*/
start = blkno - 1;
end = start - fs->fs_contigsumsize;
if (end < 0)
end = -1;
mapp = &freemapp[start / NBBY];
map = *mapp--;
bit = 1 << (start % NBBY);
for (i = start; i > end; i--) {
if ((map & bit) == 0)
break;
if ((i & (NBBY - 1)) != 0) {
bit >>= 1;
} else {
map = *mapp--;
bit = 1 << (NBBY - 1);
}
}
back = start - i;
/*
* Account for old cluster and the possibly new forward and
* back clusters.
*/
i = back + forw + 1;
if (i > fs->fs_contigsumsize)
i = fs->fs_contigsumsize;
sump[i] += cnt;
if (back > 0)
sump[back] -= cnt;
if (forw > 0)
sump[forw] -= cnt;
}
/*
* Fserr prints the name of a file system with an error diagnostic.
*
* The form of the error message is:
* fs: error message
*/
static void
ffs_fserr(fs, uid, cp)
struct fs *fs;
u_int uid;
char *cp;
{
log(LOG_ERR, "uid %d on %s: %s\n", uid, fs->fs_fsmnt, cp);
}
|