aboutsummaryrefslogtreecommitdiff
path: root/sys/fs/ext2fs/ext2_alloc.c
blob: 6ffb91069078089b775931eec10d41e52b3c20a4 (plain) (blame)
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
/*-
 *  modified for Lites 1.1
 *
 *  Aug 1995, Godmar Back (gback@cs.utah.edu)
 *  University of Utah, Department of Computer Science
 */
/*-
 * 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. 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
 * $FreeBSD$
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/conf.h>
#include <sys/vnode.h>
#include <sys/stat.h>
#include <sys/mount.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/buf.h>
#include <sys/endian.h>

#include <fs/ext2fs/fs.h>
#include <fs/ext2fs/inode.h>
#include <fs/ext2fs/ext2_mount.h>
#include <fs/ext2fs/ext2fs.h>
#include <fs/ext2fs/ext2_extern.h>

static daddr_t	ext2_alloccg(struct inode *, int, daddr_t, int);
static daddr_t	ext2_clusteralloc(struct inode *, int, daddr_t, int);
static u_long	ext2_dirpref(struct inode *);
static u_long	ext2_hashalloc(struct inode *, int, long, int,
				daddr_t (*)(struct inode *, int, daddr_t, 
						int));
static daddr_t	ext2_nodealloccg(struct inode *, int, daddr_t, int);
static daddr_t  ext2_mapsearch(struct m_ext2fs *, char *, daddr_t);

/*
 * Allocate a block in the filesystem.
 *
 * 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 hierarchy 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
ext2_alloc(struct inode *ip, daddr_t lbn, e4fs_daddr_t bpref, int size,
    struct ucred *cred, e4fs_daddr_t *bnp)
{
	struct m_ext2fs *fs;
	struct ext2mount *ump;
	int32_t bno;
	int cg;

	*bnp = 0;
	fs = ip->i_e2fs;
	ump = ip->i_ump;
	mtx_assert(EXT2_MTX(ump), MA_OWNED);
#ifdef INVARIANTS
	if ((u_int)size > fs->e2fs_bsize || blkoff(fs, size) != 0) {
		vn_printf(ip->i_devvp, "bsize = %lu, size = %d, fs = %s\n",
		    (long unsigned int)fs->e2fs_bsize, size, fs->e2fs_fsmnt);
		panic("ext2_alloc: bad size");
	}
	if (cred == NOCRED)
		panic("ext2_alloc: missing credential");
#endif		/* INVARIANTS */
	if (size == fs->e2fs_bsize && fs->e2fs->e2fs_fbcount == 0)
		goto nospace;
	if (cred->cr_uid != 0 &&
	    fs->e2fs->e2fs_fbcount < fs->e2fs->e2fs_rbcount)
		goto nospace;
	if (bpref >= fs->e2fs->e2fs_bcount)
		bpref = 0;
	if (bpref == 0)
		cg = ino_to_cg(fs, ip->i_number);
	else
		cg = dtog(fs, bpref);
	bno = (daddr_t)ext2_hashalloc(ip, cg, bpref, fs->e2fs_bsize,
	    ext2_alloccg);
	if (bno > 0) {
		/* set next_alloc fields as done in block_getblk */
		ip->i_next_alloc_block = lbn;
		ip->i_next_alloc_goal = bno;

		ip->i_blocks += btodb(fs->e2fs_bsize);
		ip->i_flag |= IN_CHANGE | IN_UPDATE;
		*bnp = bno;
		return (0);
	}
nospace:
	EXT2_UNLOCK(ump);
	ext2_fserr(fs, cred->cr_uid, "filesystem full");
	uprintf("\n%s: write failed, filesystem is full\n", fs->e2fs_fsmnt);
	return (ENOSPC);
}

/*
 * Allocate EA's block for inode.
 */
daddr_t
ext2_alloc_meta(struct inode *ip)
{
	struct m_ext2fs *fs;
	daddr_t blk;

	fs = ip->i_e2fs;

	EXT2_LOCK(ip->i_ump);
	blk = ext2_hashalloc(ip, ino_to_cg(fs, ip->i_number), 0, fs->e2fs_bsize,
	    ext2_alloccg);
	if (0 == blk)
		EXT2_UNLOCK(ip->i_ump);

	return (blk);
}

/*
 * 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 preceding 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.
 */

static SYSCTL_NODE(_vfs, OID_AUTO, ext2fs, CTLFLAG_RW, 0, "EXT2FS filesystem");

static int doasyncfree = 1;

SYSCTL_INT(_vfs_ext2fs, OID_AUTO, doasyncfree, CTLFLAG_RW, &doasyncfree, 0,
    "Use asychronous writes to update block pointers when freeing blocks");

static int doreallocblks = 1;

SYSCTL_INT(_vfs_ext2fs, OID_AUTO, doreallocblks, CTLFLAG_RW, &doreallocblks, 0, "");

int
ext2_reallocblks(struct vop_reallocblks_args *ap)
{
	struct m_ext2fs *fs;
	struct inode *ip;
	struct vnode *vp;
	struct buf *sbp, *ebp;
	uint32_t *bap, *sbap, *ebap;
	struct ext2mount *ump;
	struct cluster_save *buflist;
	struct indir start_ap[EXT2_NIADDR + 1], end_ap[EXT2_NIADDR + 1], *idp;
	e2fs_lbn_t start_lbn, end_lbn;
	int soff;
	e2fs_daddr_t newblk, blkno;
	int i, len, start_lvl, end_lvl, pref, ssize;

	if (doreallocblks == 0)
		return (ENOSPC);

	vp = ap->a_vp;
	ip = VTOI(vp);
	fs = ip->i_e2fs;
	ump = ip->i_ump;

	if (fs->e2fs_contigsumsize <= 0 || ip->i_flag & IN_E4EXTENTS)
		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 INVARIANTS
	for (i = 1; i < len; i++)
		if (buflist->bs_children[i]->b_lblkno != start_lbn + i)
			panic("ext2_reallocblks: non-cluster");
#endif
	/*
	 * If the cluster crosses the boundary for the first indirect
	 * block, leave space for the indirect block. Indirect blocks
	 * are initially laid out in a position after the last direct
	 * block. Block reallocation would usually destroy locality by
	 * moving the indirect block out of the way to make room for
	 * data blocks if we didn't compensate here. We should also do
	 * this for other indirect block boundaries, but it is only
	 * important for the first one.
	 */
	if (start_lbn < EXT2_NDADDR && end_lbn >= EXT2_NDADDR)
		return (ENOSPC);
	/*
	 * 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 (ext2_getlbns(vp, start_lbn, start_ap, &start_lvl) ||
	    ext2_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->e2fs_bsize, NOCRED, &sbp)) {
			brelse(sbp);
			return (ENOSPC);
		}
		sbap = (u_int *)sbp->b_data;
		soff = idp->in_off;
	}
	/*
	 * If the block range spans two block maps, get the second map.
	 */
	ebap = NULL;
	if (end_lvl == 0 || (idp = &end_ap[end_lvl - 1])->in_off + 1 >= len) {
		ssize = len;
	} else {
#ifdef INVARIANTS
		if (start_ap[start_lvl - 1].in_lbn == idp->in_lbn)
			panic("ext2_reallocblks: start == end");
#endif
		ssize = len - (idp->in_off + 1);
		if (bread(vp, idp->in_lbn, (int)fs->e2fs_bsize, NOCRED, &ebp))
			goto fail;
		ebap = (u_int *)ebp->b_data;
	}
	/*
	 * Find the preferred location for the cluster.
	 */
	EXT2_LOCK(ump);
	pref = ext2_blkpref(ip, start_lbn, soff, sbap, 0);
	/*
	 * Search the block map looking for an allocation of the desired size.
	 */
	if ((newblk = (e2fs_daddr_t)ext2_hashalloc(ip, dtog(fs, pref), pref,
	    len, ext2_clusteralloc)) == 0) {
		EXT2_UNLOCK(ump);
		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.
	 */
#ifdef DEBUG
	printf("realloc: ino %ju, lbns %jd-%jd\n\told:",
	    (uintmax_t)ip->i_number, (intmax_t)start_lbn, (intmax_t)end_lbn);
#endif	/* DEBUG */
	blkno = newblk;
	for (bap = &sbap[soff], i = 0; i < len; i++, blkno += fs->e2fs_fpb) {
		if (i == ssize) {
			bap = ebap;
			soff = -i;
		}
#ifdef INVARIANTS
		if (buflist->bs_children[i]->b_blkno != fsbtodb(fs, *bap))
			panic("ext2_reallocblks: alloc mismatch");
#endif
#ifdef DEBUG
		printf(" %d,", *bap);
#endif	/* DEBUG */
		*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)
			ext2_update(vp, 1);
	}
	if (ssize < len) {
		if (doasyncfree)
			bdwrite(ebp);
		else
			bwrite(ebp);
	}
	/*
	 * Last, free the old blocks and assign the new blocks to the buffers.
	 */
#ifdef DEBUG
	printf("\n\tnew:");
#endif	/* DEBUG */
	for (blkno = newblk, i = 0; i < len; i++, blkno += fs->e2fs_fpb) {
		ext2_blkfree(ip, dbtofsb(fs, buflist->bs_children[i]->b_blkno),
		    fs->e2fs_bsize);
		buflist->bs_children[i]->b_blkno = fsbtodb(fs, blkno);
#ifdef DEBUG
		printf(" %d,", blkno);
#endif	/* DEBUG */
	}
#ifdef DEBUG
	printf("\n");
#endif	/* DEBUG */
	return (0);

fail:
	if (ssize < len)
		brelse(ebp);
	if (sbap != &ip->i_db[0])
		brelse(sbp);
	return (ENOSPC);
}

/*
 * Allocate an inode in the filesystem.
 *
 */
int
ext2_valloc(struct vnode *pvp, int mode, struct ucred *cred, struct vnode **vpp)
{
	struct timespec ts;
	struct inode *pip;
	struct m_ext2fs *fs;
	struct inode *ip;
	struct ext2mount *ump;
	ino_t ino, ipref;
	int error, cg;

	*vpp = NULL;
	pip = VTOI(pvp);
	fs = pip->i_e2fs;
	ump = pip->i_ump;

	EXT2_LOCK(ump);
	if (fs->e2fs->e2fs_ficount == 0)
		goto noinodes;
	/*
	 * If it is a directory then obtain a cylinder group based on
	 * ext2_dirpref else obtain it using ino_to_cg. The preferred inode is
	 * always the next inode.
	 */
	if ((mode & IFMT) == IFDIR) {
		cg = ext2_dirpref(pip);
		if (fs->e2fs_contigdirs[cg] < 255)
			fs->e2fs_contigdirs[cg]++;
	} else {
		cg = ino_to_cg(fs, pip->i_number);
		if (fs->e2fs_contigdirs[cg] > 0)
			fs->e2fs_contigdirs[cg]--;
	}
	ipref = cg * fs->e2fs->e2fs_ipg + 1;
	ino = (ino_t)ext2_hashalloc(pip, cg, (long)ipref, mode, ext2_nodealloccg);

	if (ino == 0)
		goto noinodes;
	error = VFS_VGET(pvp->v_mount, ino, LK_EXCLUSIVE, vpp);
	if (error) {
		ext2_vfree(pvp, ino, mode);
		return (error);
	}
	ip = VTOI(*vpp);

	/*
	 * The question is whether using VGET was such good idea at all:
	 * Linux doesn't read the old inode in when it is allocating a
	 * new one. I will set at least i_size and i_blocks to zero.
	 */
	ip->i_flag = 0;
	ip->i_size = 0;
	ip->i_blocks = 0;
	ip->i_mode = 0;
	ip->i_flags = 0;
	if (EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_EXTENTS)
	    && (S_ISREG(mode) || S_ISDIR(mode)))
		ext4_ext_tree_init(ip);
	else
		memset(ip->i_data, 0, sizeof(ip->i_data));
	

	/*
	 * Set up a new generation number for this inode.
	 * Avoid zero values.
	 */
	do {
		ip->i_gen = arc4random();
	} while (ip->i_gen == 0);

	vfs_timestamp(&ts);
	ip->i_birthtime = ts.tv_sec;
	ip->i_birthnsec = ts.tv_nsec;

/*
printf("ext2_valloc: allocated inode %d\n", ino);
*/
	return (0);
noinodes:
	EXT2_UNLOCK(ump);
	ext2_fserr(fs, cred->cr_uid, "out of inodes");
	uprintf("\n%s: create/symlink failed, no inodes free\n", fs->e2fs_fsmnt);
	return (ENOSPC);
}

/*
 * Find a cylinder to place a directory.
 *
 * The policy implemented by this algorithm is to allocate a
 * directory inode in the same cylinder group as its parent
 * directory, but also to reserve space for its files inodes
 * and data. Restrict the number of directories which may be
 * allocated one after another in the same cylinder group
 * without intervening allocation of files.
 *
 * If we allocate a first level directory then force allocation
 * in another cylinder group.
 *
 */
static u_long
ext2_dirpref(struct inode *pip)
{
	struct m_ext2fs *fs;
	int cg, prefcg, cgsize;
	u_int avgifree, avgbfree, avgndir, curdirsize;
	u_int minifree, minbfree, maxndir;
	u_int mincg, minndir;
	u_int dirsize, maxcontigdirs;

	mtx_assert(EXT2_MTX(pip->i_ump), MA_OWNED);
	fs = pip->i_e2fs;

	avgifree = fs->e2fs->e2fs_ficount / fs->e2fs_gcount;
	avgbfree = fs->e2fs->e2fs_fbcount / fs->e2fs_gcount;
	avgndir = fs->e2fs_total_dir / fs->e2fs_gcount;

	/*
	 * Force allocation in another cg if creating a first level dir.
	 */
	ASSERT_VOP_LOCKED(ITOV(pip), "ext2fs_dirpref");
	if (ITOV(pip)->v_vflag & VV_ROOT) {
		prefcg = arc4random() % fs->e2fs_gcount;
		mincg = prefcg;
		minndir = fs->e2fs_ipg;
		for (cg = prefcg; cg < fs->e2fs_gcount; cg++)
			if (fs->e2fs_gd[cg].ext2bgd_ndirs < minndir &&
			    fs->e2fs_gd[cg].ext2bgd_nifree >= avgifree &&
			    fs->e2fs_gd[cg].ext2bgd_nbfree >= avgbfree) {
				mincg = cg;
				minndir = fs->e2fs_gd[cg].ext2bgd_ndirs;
			}
		for (cg = 0; cg < prefcg; cg++)
			if (fs->e2fs_gd[cg].ext2bgd_ndirs < minndir &&
			    fs->e2fs_gd[cg].ext2bgd_nifree >= avgifree &&
			    fs->e2fs_gd[cg].ext2bgd_nbfree >= avgbfree) {
				mincg = cg;
				minndir = fs->e2fs_gd[cg].ext2bgd_ndirs;
			}
		return (mincg);
	}
	/*
	 * Count various limits which used for
	 * optimal allocation of a directory inode.
	 */
	maxndir = min(avgndir + fs->e2fs_ipg / 16, fs->e2fs_ipg);
	minifree = avgifree - avgifree / 4;
	if (minifree < 1)
		minifree = 1;
	minbfree = avgbfree - avgbfree / 4;
	if (minbfree < 1)
		minbfree = 1;
	cgsize = fs->e2fs_fsize * fs->e2fs_fpg;
	dirsize = AVGDIRSIZE;
	curdirsize = avgndir ? (cgsize - avgbfree * fs->e2fs_bsize) / avgndir : 0;
	if (dirsize < curdirsize)
		dirsize = curdirsize;
	maxcontigdirs = min((avgbfree * fs->e2fs_bsize) / dirsize, 255);
	maxcontigdirs = min(maxcontigdirs, fs->e2fs_ipg / AFPDIR);
	if (maxcontigdirs == 0)
		maxcontigdirs = 1;

	/*
	 * Limit number of dirs in one cg and reserve space for
	 * regular files, but only if we have no deficit in
	 * inodes or space.
	 */
	prefcg = ino_to_cg(fs, pip->i_number);
	for (cg = prefcg; cg < fs->e2fs_gcount; cg++)
		if (fs->e2fs_gd[cg].ext2bgd_ndirs < maxndir &&
		    fs->e2fs_gd[cg].ext2bgd_nifree >= minifree &&
		    fs->e2fs_gd[cg].ext2bgd_nbfree >= minbfree) {
			if (fs->e2fs_contigdirs[cg] < maxcontigdirs)
				return (cg);
		}
	for (cg = 0; cg < prefcg; cg++)
		if (fs->e2fs_gd[cg].ext2bgd_ndirs < maxndir &&
		    fs->e2fs_gd[cg].ext2bgd_nifree >= minifree &&
		    fs->e2fs_gd[cg].ext2bgd_nbfree >= minbfree) {
			if (fs->e2fs_contigdirs[cg] < maxcontigdirs)
				return (cg);
		}
	/*
	 * This is a backstop when we have deficit in space.
	 */
	for (cg = prefcg; cg < fs->e2fs_gcount; cg++)
		if (fs->e2fs_gd[cg].ext2bgd_nifree >= avgifree)
			return (cg);
	for (cg = 0; cg < prefcg; cg++)
		if (fs->e2fs_gd[cg].ext2bgd_nifree >= avgifree)
			break;
	return (cg);
}

/*
 * Select the desired position for the next block in a file.
 *
 * we try to mimic what Remy does in inode_getblk/block_getblk
 *
 * we note: blocknr == 0 means that we're about to allocate either
 * a direct block or a pointer block at the first level of indirection
 * (In other words, stuff that will go in i_db[] or i_ib[])
 *
 * blocknr != 0 means that we're allocating a block that is none
 * of the above. Then, blocknr tells us the number of the block
 * that will hold the pointer
 */
e4fs_daddr_t
ext2_blkpref(struct inode *ip, e2fs_lbn_t lbn, int indx, e2fs_daddr_t *bap,
    e2fs_daddr_t blocknr)
{
	struct m_ext2fs *fs;
	int tmp;

	fs = ip->i_e2fs;

	mtx_assert(EXT2_MTX(ip->i_ump), MA_OWNED);

	/*
	 * If the next block is actually what we thought it is, then set the
	 * goal to what we thought it should be.
	 */
	if (ip->i_next_alloc_block == lbn && ip->i_next_alloc_goal != 0)
		return ip->i_next_alloc_goal;

	/*
	 * Now check whether we were provided with an array that basically
	 * tells us previous blocks to which we want to stay close.
	 */
	if (bap)
		for (tmp = indx - 1; tmp >= 0; tmp--)
			if (bap[tmp])
				return bap[tmp];

	/*
	 * Else lets fall back to the blocknr or, if there is none, follow
	 * the rule that a block should be allocated near its inode.
	 */
	return (blocknr ? blocknr :
	    (e2fs_daddr_t)(ip->i_block_group *
	    EXT2_BLOCKS_PER_GROUP(fs)) + fs->e2fs->e2fs_first_dblock);
}

/*
 * 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.
 */
static u_long
ext2_hashalloc(struct inode *ip, int cg, long pref, int size,
    daddr_t (*allocator) (struct inode *, int, daddr_t, int))
{
	struct m_ext2fs *fs;
	ino_t result;
	int i, icg = cg;

	mtx_assert(EXT2_MTX(ip->i_ump), MA_OWNED);
	fs = ip->i_e2fs;
	/*
	 * 1: preferred cylinder group
	 */
	result = (*allocator)(ip, cg, pref, size);
	if (result)
		return (result);
	/*
	 * 2: quadratic rehash
	 */
	for (i = 1; i < fs->e2fs_gcount; i *= 2) {
		cg += i;
		if (cg >= fs->e2fs_gcount)
			cg -= fs->e2fs_gcount;
		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->e2fs_gcount;
	for (i = 2; i < fs->e2fs_gcount; i++) {
		result = (*allocator)(ip, cg, 0, size);
		if (result)
			return (result);
		cg++;
		if (cg == fs->e2fs_gcount)
			cg = 0;
	}
	return (0);
}

static unsigned long
ext2_cg_num_gdb(struct m_ext2fs *fs, int cg)
{
	int gd_per_block, metagroup, first, last;

	gd_per_block = fs->e2fs_bsize / sizeof(struct ext2_gd);
	metagroup = cg / gd_per_block;
	first = metagroup * gd_per_block;
	last = first + gd_per_block - 1;

	if (!EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_META_BG) ||
	    metagroup < fs->e2fs->e3fs_first_meta_bg) {
		if (!ext2_cg_has_sb(fs, cg))
			return (0);
		if (EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_META_BG))
			return (fs->e2fs->e3fs_first_meta_bg);
		return (fs->e2fs_gdbcount);
	}

	if (cg == first || cg == first + 1 || cg == last)
		return (1);
	return (0);

}

static int
ext2_num_base_meta_blocks(struct m_ext2fs *fs, int cg)
{
	int num, gd_per_block;

	gd_per_block = fs->e2fs_bsize / sizeof(struct ext2_gd);
	num = ext2_cg_has_sb(fs, cg);

	if (!EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_META_BG) ||
	    cg < fs->e2fs->e3fs_first_meta_bg * gd_per_block) {
		if (num) {
			num += ext2_cg_num_gdb(fs, cg);
			num += fs->e2fs->e2fs_reserved_ngdb;
		}
	} else {
		num += ext2_cg_num_gdb(fs, cg);
	}
	
	return (num);
}

static int
ext2_get_cg_number(struct m_ext2fs *fs, daddr_t blk)
{
	int cg;

	if (fs->e2fs->e2fs_bpg == fs->e2fs_bsize * 8)
		cg = (blk - fs->e2fs->e2fs_first_dblock) / (fs->e2fs_bsize * 8);
	else
		cg = blk - fs->e2fs->e2fs_first_dblock;

	return (cg);
}

static void
ext2_mark_bitmap_end(int start_bit, int end_bit, char *bitmap)
{
	int i;

	if (start_bit >= end_bit)
		return;

	for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++)
		setbit(bitmap, i);
	if (i < end_bit)
		memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3);
}

static int
ext2_cg_block_bitmap_init(struct m_ext2fs *fs, int cg, struct buf *bp)
{
	int bit, bit_max, inodes_per_block;
	uint32_t start, tmp;

	if (!EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_GDT_CSUM) ||
	    !(fs->e2fs_gd[cg].ext4bgd_flags & EXT2_BG_BLOCK_UNINIT))
		return (0);

	memset(bp->b_data, 0, fs->e2fs_bsize);

	bit_max = ext2_num_base_meta_blocks(fs, cg);
	if ((bit_max >> 3) >= fs->e2fs_bsize)
		return (EINVAL);

	for (bit = 0; bit < bit_max; bit++)
		setbit(bp->b_data, bit);

	start = cg * fs->e2fs->e2fs_bpg + fs->e2fs->e2fs_first_dblock;

	/* Set bits for block and inode bitmaps, and inode table */
	tmp = fs->e2fs_gd[cg].ext2bgd_b_bitmap;
	if (!EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_FLEX_BG) ||
	    tmp == ext2_get_cg_number(fs, cg))
		setbit(bp->b_data, tmp - start);

	tmp = fs->e2fs_gd[cg].ext2bgd_i_bitmap;
	if (!EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_FLEX_BG) ||
	    tmp == ext2_get_cg_number(fs, cg))
		setbit(bp->b_data, tmp - start);

	tmp = fs->e2fs_gd[cg].ext2bgd_i_tables;
	inodes_per_block = fs->e2fs_bsize/EXT2_INODE_SIZE(fs);
	while( tmp < fs->e2fs_gd[cg].ext2bgd_i_tables +
	    fs->e2fs->e2fs_ipg / inodes_per_block ) {
		if (!EXT2_HAS_INCOMPAT_FEATURE(fs, EXT2F_INCOMPAT_FLEX_BG) ||
		    tmp == ext2_get_cg_number(fs, cg))
			setbit(bp->b_data, tmp - start);
		tmp++;
	}

	/*
	 * Also if the number of blocks within the group is less than
	 * the blocksize * 8 ( which is the size of bitmap ), set rest
	 * of the block bitmap to 1
	 */
	ext2_mark_bitmap_end(fs->e2fs->e2fs_bpg, fs->e2fs_bsize * 8,
	    bp->b_data);

	/* Clean the flag */
	fs->e2fs_gd[cg].ext4bgd_flags &= ~EXT2_BG_BLOCK_UNINIT;

	return (0);
}

/*
 * 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
ext2_alloccg(struct inode *ip, int cg, daddr_t bpref, int size)
{
	struct m_ext2fs *fs;
	struct buf *bp;
	struct ext2mount *ump;
	daddr_t bno, runstart, runlen;
	int bit, loc, end, error, start;
	char *bbp;
	/* XXX ondisk32 */
	fs = ip->i_e2fs;
	ump = ip->i_ump;
	if (fs->e2fs_gd[cg].ext2bgd_nbfree == 0)
		return (0);
	EXT2_UNLOCK(ump);
	error = bread(ip->i_devvp, fsbtodb(fs,
	    fs->e2fs_gd[cg].ext2bgd_b_bitmap),
	    (int)fs->e2fs_bsize, NOCRED, &bp);
	if (error) {
		brelse(bp);
		EXT2_LOCK(ump);
		return (0);
	}
	if (EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_GDT_CSUM)) {
		error = ext2_cg_block_bitmap_init(fs, cg, bp);
		if (error) {
			brelse(bp);
			EXT2_LOCK(ump);
			return (0);
		}
	}
	if (fs->e2fs_gd[cg].ext2bgd_nbfree == 0) {
		/*
		 * Another thread allocated the last block in this
		 * group while we were waiting for the buffer.
		 */
		brelse(bp);
		EXT2_LOCK(ump);
		return (0);
	}
	bbp = (char *)bp->b_data;

	if (dtog(fs, bpref) != cg)
		bpref = 0;
	if (bpref != 0) {
		bpref = dtogd(fs, bpref);
		/*
		 * if the requested block is available, use it
		 */
		if (isclr(bbp, bpref)) {
			bno = bpref;
			goto gotit;
		}
	}
	/*
	 * no blocks in the requested cylinder, so take next
	 * available one in this cylinder group.
	 * first try to get 8 contigous blocks, then fall back to a single
	 * block.
	 */
	if (bpref)
		start = dtogd(fs, bpref) / NBBY;
	else
		start = 0;
	end = howmany(fs->e2fs->e2fs_fpg, NBBY) - start;
retry:
	runlen = 0;
	runstart = 0;
	for (loc = start; loc < end; loc++) {
		if (bbp[loc] == (char)0xff) {
			runlen = 0;
			continue;
		}

		/* Start of a run, find the number of high clear bits. */
		if (runlen == 0) {
			bit = fls(bbp[loc]);
			runlen = NBBY - bit;
			runstart = loc * NBBY + bit;
		} else if (bbp[loc] == 0) {
			/* Continue a run. */
			runlen += NBBY;
		} else {
			/*
			 * Finish the current run.  If it isn't long
			 * enough, start a new one.
			 */
			bit = ffs(bbp[loc]) - 1;
			runlen += bit;
			if (runlen >= 8) {
				bno = runstart;
				goto gotit;
			}

			/* Run was too short, start a new one. */
			bit = fls(bbp[loc]);
			runlen = NBBY - bit;
			runstart = loc * NBBY + bit;
		}

		/* If the current run is long enough, use it. */
		if (runlen >= 8) {
			bno = runstart;
			goto gotit;
		}
	}
	if (start != 0) {
		end = start;
		start = 0;
		goto retry;
	}
	bno = ext2_mapsearch(fs, bbp, bpref);
	if (bno < 0) {
		brelse(bp);
		EXT2_LOCK(ump);
		return (0);
	}
gotit:
#ifdef INVARIANTS
	if (isset(bbp, bno)) {
		printf("ext2fs_alloccgblk: cg=%d bno=%jd fs=%s\n",
		    cg, (intmax_t)bno, fs->e2fs_fsmnt);
		panic("ext2fs_alloccg: dup alloc");
	}
#endif
	setbit(bbp, bno);
	EXT2_LOCK(ump);
	ext2_clusteracct(fs, bbp, cg, bno, -1);
	fs->e2fs->e2fs_fbcount--;
	fs->e2fs_gd[cg].ext2bgd_nbfree--;
	fs->e2fs_fmod = 1;
	EXT2_UNLOCK(ump);
	bdwrite(bp);
	return (cg * fs->e2fs->e2fs_fpg + fs->e2fs->e2fs_first_dblock + bno);
}

/*
 * Determine whether a cluster can be allocated.
 */
static daddr_t
ext2_clusteralloc(struct inode *ip, int cg, daddr_t bpref, int len)
{
	struct m_ext2fs *fs;
	struct ext2mount *ump;
	struct buf *bp;
	char *bbp;
	int bit, error, got, i, loc, run;
	int32_t *lp;
	daddr_t bno;

	fs = ip->i_e2fs;
	ump = ip->i_ump;

	if (fs->e2fs_maxcluster[cg] < len)
		return (0);

	EXT2_UNLOCK(ump);
	error = bread(ip->i_devvp,
	    fsbtodb(fs, fs->e2fs_gd[cg].ext2bgd_b_bitmap),
	    (int)fs->e2fs_bsize, NOCRED, &bp);
	if (error)
		goto fail_lock;

	bbp = (char *)bp->b_data;
	EXT2_LOCK(ump);
	/*
	 * Check to see if a cluster of the needed size (or bigger) is
	 * available in this cylinder group.
	 */
	lp = &fs->e2fs_clustersum[cg].cs_sum[len];
	for (i = len; i <= fs->e2fs_contigsumsize; i++)
		if (*lp++ > 0)
			break;
	if (i > fs->e2fs_contigsumsize) {
		/*
		 * Update the cluster summary information to reflect
		 * the true maximum-sized cluster so that future cluster
		 * allocation requests can avoid reading the bitmap only
		 * to find no cluster.
		 */
		lp = &fs->e2fs_clustersum[cg].cs_sum[len - 1];
		for (i = len - 1; i > 0; i--)
			if (*lp-- > 0)
				break;
		fs->e2fs_maxcluster[cg] = i;
		goto fail;
	}
	EXT2_UNLOCK(ump);

	/* Search the bitmap to find a big enough cluster like in FFS. */
	if (dtog(fs, bpref) != cg)
		bpref = 0;
	if (bpref != 0)
		bpref = dtogd(fs, bpref);
	loc = bpref / NBBY;
	bit = 1 << (bpref % NBBY);
	for (run = 0, got = bpref; got < fs->e2fs->e2fs_fpg; got++) {
		if ((bbp[loc] & bit) != 0)
			run = 0;
		else {
			run++;
			if (run == len)
				break;
		}
		if ((got & (NBBY - 1)) != (NBBY - 1))
			bit <<= 1;
		else {
			loc++;
			bit = 1;
		}
	}

	if (got >= fs->e2fs->e2fs_fpg)
		goto fail_lock;

	/* Allocate the cluster that we found. */
	for (i = 1; i < len; i++)
		if (!isclr(bbp, got - run + i))
			panic("ext2_clusteralloc: map mismatch");

	bno = got - run + 1;
	if (bno >= fs->e2fs->e2fs_fpg)
		panic("ext2_clusteralloc: allocated out of group");

	EXT2_LOCK(ump);
	for (i = 0; i < len; i += fs->e2fs_fpb) {
		setbit(bbp, bno + i);
		ext2_clusteracct(fs, bbp, cg, bno + i, -1);
		fs->e2fs->e2fs_fbcount--;
		fs->e2fs_gd[cg].ext2bgd_nbfree--;
	}
	fs->e2fs_fmod = 1;
	EXT2_UNLOCK(ump);

	bdwrite(bp);
	return (cg * fs->e2fs->e2fs_fpg + fs->e2fs->e2fs_first_dblock + bno);

fail_lock:
	EXT2_LOCK(ump);
fail:
	brelse(bp);
	return (0);
}

static int
ext2_zero_inode_table(struct inode *ip, int cg)
{
	struct m_ext2fs *fs;
	struct buf *bp;
	int i, all_blks, used_blks;

	fs = ip->i_e2fs;

	if (fs->e2fs_gd[cg].ext4bgd_flags & EXT2_BG_INODE_ZEROED)
		return (0);

	all_blks = fs->e2fs->e2fs_inode_size * fs->e2fs->e2fs_ipg /
	    fs->e2fs_bsize;

	used_blks = howmany(fs->e2fs->e2fs_ipg -
	    fs->e2fs_gd[cg].ext4bgd_i_unused,
	    fs->e2fs_bsize / EXT2_INODE_SIZE(fs));

	for (i = 0; i < all_blks - used_blks; i++) {
		bp = getblk(ip->i_devvp, fsbtodb(fs,
		    fs->e2fs_gd[cg].ext2bgd_i_tables + used_blks + i),
		    fs->e2fs_bsize, 0, 0, 0);
		if (!bp)
			return (EIO);

		vfs_bio_bzero_buf(bp, 0, fs->e2fs_bsize);
		bawrite(bp);
	}

	fs->e2fs_gd[cg].ext4bgd_flags |= EXT2_BG_INODE_ZEROED;

	return (0);
}

/*
 * Determine whether an inode can be allocated.
 *
 * Check to see if an inode is available, and if it is,
 * allocate it using tode in the specified cylinder group.
 */
static daddr_t
ext2_nodealloccg(struct inode *ip, int cg, daddr_t ipref, int mode)
{
	struct m_ext2fs *fs;
	struct buf *bp;
	struct ext2mount *ump;
	int error, start, len;
	char *ibp, *loc;

	ipref--;	/* to avoid a lot of (ipref -1) */
	if (ipref == -1)
		ipref = 0;
	fs = ip->i_e2fs;
	ump = ip->i_ump;
	if (fs->e2fs_gd[cg].ext2bgd_nifree == 0)
		return (0);
	EXT2_UNLOCK(ump);
	error = bread(ip->i_devvp, fsbtodb(fs,
	    fs->e2fs_gd[cg].ext2bgd_i_bitmap),
	    (int)fs->e2fs_bsize, NOCRED, &bp);
	if (error) {
		brelse(bp);
		EXT2_LOCK(ump);
		return (0);
	}
	if (EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_GDT_CSUM)) {
		if (fs->e2fs_gd[cg].ext4bgd_flags & EXT2_BG_INODE_UNINIT) {
			memset(bp->b_data, 0, fs->e2fs_bsize);
			fs->e2fs_gd[cg].ext4bgd_flags &= ~EXT2_BG_INODE_UNINIT;
		}
		error = ext2_zero_inode_table(ip, cg);
		if (error) {
			brelse(bp);
			EXT2_LOCK(ump);
			return (0);
		}
	}
	if (fs->e2fs_gd[cg].ext2bgd_nifree == 0) {
		/*
		 * Another thread allocated the last i-node in this
		 * group while we were waiting for the buffer.
		 */
		brelse(bp);
		EXT2_LOCK(ump);
		return (0);
	}
	ibp = (char *)bp->b_data;
	if (ipref) {
		ipref %= fs->e2fs->e2fs_ipg;
		if (isclr(ibp, ipref))
			goto gotit;
	}
	start = ipref / NBBY;
	len = howmany(fs->e2fs->e2fs_ipg - ipref, NBBY);
	loc = memcchr(&ibp[start], 0xff, len);
	if (loc == NULL) {
		len = start + 1;
		start = 0;
		loc = memcchr(&ibp[start], 0xff, len);
		if (loc == NULL) {
			printf("cg = %d, ipref = %lld, fs = %s\n",
			    cg, (long long)ipref, fs->e2fs_fsmnt);
			panic("ext2fs_nodealloccg: map corrupted");
			/* NOTREACHED */
		}
	}
	ipref = (loc - ibp) * NBBY + ffs(~*loc) - 1;
gotit:
	setbit(ibp, ipref);
	EXT2_LOCK(ump);
	fs->e2fs_gd[cg].ext2bgd_nifree--;
	if (EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_GDT_CSUM))
		fs->e2fs_gd[cg].ext4bgd_i_unused--;
	fs->e2fs->e2fs_ficount--;
	fs->e2fs_fmod = 1;
	if ((mode & IFMT) == IFDIR) {
		fs->e2fs_gd[cg].ext2bgd_ndirs++;
		fs->e2fs_total_dir++;
	}
	EXT2_UNLOCK(ump);
	bdwrite(bp);
	return (cg * fs->e2fs->e2fs_ipg + ipref + 1);
}

/*
 * Free a block or fragment.
 *
 */
void
ext2_blkfree(struct inode *ip, e4fs_daddr_t bno, long size)
{
	struct m_ext2fs *fs;
	struct buf *bp;
	struct ext2mount *ump;
	int cg, error;
	char *bbp;

	fs = ip->i_e2fs;
	ump = ip->i_ump;
	cg = dtog(fs, bno);
	if ((u_int)bno >= fs->e2fs->e2fs_bcount) {
		printf("bad block %lld, ino %ju\n", (long long)bno,
		    (uintmax_t)ip->i_number);
		ext2_fserr(fs, ip->i_uid, "bad block");
		return;
	}
	error = bread(ip->i_devvp,
	    fsbtodb(fs, fs->e2fs_gd[cg].ext2bgd_b_bitmap),
	    (int)fs->e2fs_bsize, NOCRED, &bp);
	if (error) {
		brelse(bp);
		return;
	}
	bbp = (char *)bp->b_data;
	bno = dtogd(fs, bno);
	if (isclr(bbp, bno)) {
		printf("block = %lld, fs = %s\n",
		    (long long)bno, fs->e2fs_fsmnt);
		panic("ext2_blkfree: freeing free block");
	}
	clrbit(bbp, bno);
	EXT2_LOCK(ump);
	ext2_clusteracct(fs, bbp, cg, bno, 1);
	fs->e2fs->e2fs_fbcount++;
	fs->e2fs_gd[cg].ext2bgd_nbfree++;
	fs->e2fs_fmod = 1;
	EXT2_UNLOCK(ump);
	bdwrite(bp);
}

/*
 * Free an inode.
 *
 */
int
ext2_vfree(struct vnode *pvp, ino_t ino, int mode)
{
	struct m_ext2fs *fs;
	struct inode *pip;
	struct buf *bp;
	struct ext2mount *ump;
	int error, cg;
	char *ibp;

	pip = VTOI(pvp);
	fs = pip->i_e2fs;
	ump = pip->i_ump;
	if ((u_int)ino > fs->e2fs_ipg * fs->e2fs_gcount)
		panic("ext2_vfree: range: devvp = %p, ino = %ju, fs = %s",
		    pip->i_devvp, (uintmax_t)ino, fs->e2fs_fsmnt);

	cg = ino_to_cg(fs, ino);
	error = bread(pip->i_devvp,
	    fsbtodb(fs, fs->e2fs_gd[cg].ext2bgd_i_bitmap),
	    (int)fs->e2fs_bsize, NOCRED, &bp);
	if (error) {
		brelse(bp);
		return (0);
	}
	ibp = (char *)bp->b_data;
	ino = (ino - 1) % fs->e2fs->e2fs_ipg;
	if (isclr(ibp, ino)) {
		printf("ino = %llu, fs = %s\n",
		    (unsigned long long)ino, fs->e2fs_fsmnt);
		if (fs->e2fs_ronly == 0)
			panic("ext2_vfree: freeing free inode");
	}
	clrbit(ibp, ino);
	EXT2_LOCK(ump);
	fs->e2fs->e2fs_ficount++;
	fs->e2fs_gd[cg].ext2bgd_nifree++;
	if (EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_GDT_CSUM))
		fs->e2fs_gd[cg].ext4bgd_i_unused++;
	if ((mode & IFMT) == IFDIR) {
		fs->e2fs_gd[cg].ext2bgd_ndirs--;
		fs->e2fs_total_dir--;
	}
	fs->e2fs_fmod = 1;
	EXT2_UNLOCK(ump);
	bdwrite(bp);
	return (0);
}

/*
 * Find a block 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
ext2_mapsearch(struct m_ext2fs *fs, char *bbp, daddr_t bpref)
{
	char *loc;
	int start, len;

	/*
	 * find the fragment by searching through the free block
	 * map for an appropriate bit pattern
	 */
	if (bpref)
		start = dtogd(fs, bpref) / NBBY;
	else
		start = 0;
	len = howmany(fs->e2fs->e2fs_fpg, NBBY) - start;
	loc = memcchr(&bbp[start], 0xff, len);
	if (loc == NULL) {
		len = start + 1;
		start = 0;
		loc = memcchr(&bbp[start], 0xff, len);
		if (loc == NULL) {
			printf("start = %d, len = %d, fs = %s\n",
			    start, len, fs->e2fs_fsmnt);
			panic("ext2_mapsearch: map corrupted");
			/* NOTREACHED */
		}
	}
	return ((loc - bbp) * NBBY + ffs(~*loc) - 1);
}

/*
 * Fserr prints the name of a filesystem with an error diagnostic.
 *
 * The form of the error message is:
 *	fs: error message
 */
void
ext2_fserr(struct m_ext2fs *fs, uid_t uid, char *cp)
{

	log(LOG_ERR, "uid %u on %s: %s\n", uid, fs->e2fs_fsmnt, cp);
}

int
ext2_cg_has_sb(struct m_ext2fs *fs, int cg)
{
	int a3, a5, a7;

	if (cg == 0)
		return (1);

	if (EXT2_HAS_COMPAT_FEATURE(fs, EXT2F_COMPAT_SPARSESUPER2)) {
		if (cg == fs->e2fs->e4fs_backup_bgs[0] ||
		    cg == fs->e2fs->e4fs_backup_bgs[1])
			return (1);
		return (0);
	}

	if ((cg <= 1) ||
	    !EXT2_HAS_RO_COMPAT_FEATURE(fs, EXT2F_ROCOMPAT_SPARSESUPER))
		return (1);

	if (!(cg & 1))
		return (0);

	for (a3 = 3, a5 = 5, a7 = 7;
	    a3 <= cg || a5 <= cg || a7 <= cg;
	    a3 *= 3, a5 *= 5, a7 *= 7)
		if (cg == a3 || cg == a5 || cg == a7)
			return (1);
	return (0);
}