summaryrefslogtreecommitdiff
path: root/serverboot/ext2_file_io.c
blob: cb915c4a26b68cc1d4dd996108a7cd666ed4e36e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
/*
 * Mach Operating System
 * Copyright (c) 1991,1990 Carnegie Mellon University
 * All Rights Reserved.
 *
 * Permission to use, copy, modify and distribute this software and its
 * documentation is hereby granted, provided that both the copyright
 * notice and this permission notice appear in all copies of the
 * software, derivative works or modified versions, and any portions
 * thereof, and that both notices appear in supporting documentation.
 *
 * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
 * CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
 * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
 *
 * Carnegie Mellon requests users of this software to return to
 *
 *  Software Distribution Coordinator  or  Software.Distribution@CS.CMU.EDU
 *  School of Computer Science
 *  Carnegie Mellon University
 *  Pittsburgh PA 15213-3890
 *
 * any improvements or extensions that they make and grant Carnegie Mellon
 * the rights to redistribute these changes.
 */
/*
 * Stand-alone file reading package.
 */

#include <device/device_types.h>
#include <device/device.h>

#include <mach/mach_traps.h>
#include <mach/mach_interface.h>

#include "file_io.h"
#include "ffs_compat.h"

void	ext2_close_file();	/* forward */

/*
 * Free file buffers, but don't close file.
 */
static void
free_file_buffers(fp)
	register struct file	*fp;
{
	register int level;

	/*
	 * Free the indirect blocks
	 */
	for (level = 0; level < NIADDR; level++) {
	    if (fp->f_blk[level] != 0) {
		(void) vm_deallocate(mach_task_self(),
				     fp->f_blk[level],
				     fp->f_blksize[level]);
		fp->f_blk[level] = 0;
	    }
	    fp->f_blkno[level] = -1;
	}

	/*
	 * Free the data block
	 */
	if (fp->f_buf != 0) {
	    (void) vm_deallocate(mach_task_self(),
				 fp->f_buf,
				 fp->f_buf_size);
	    fp->f_buf = 0;
	}
	fp->f_buf_blkno = -1;
}

/*
 * Read a new inode into a file structure.
 */
static int
read_inode(inumber, fp)
	ino_t			inumber;
	register struct file	*fp;
{
	vm_offset_t		buf;
	mach_msg_type_number_t	buf_size;
	register
	struct ext2_super_block	*fs;
	daddr_t			disk_block;
	kern_return_t		rc;

	fs = fp->f_fs;
	disk_block = ino2blk(fs, fp->f_gd, inumber);

	rc = device_read(fp->f_dev,
			 0,
			 (recnum_t) fsbtodb(fp->f_fs, disk_block),
			 (int) EXT2_BLOCK_SIZE(fs),
			 (char **)&buf,
			 &buf_size);
	if (rc != KERN_SUCCESS)
	    return (rc);

	{
	    register struct ext2_inode *dp;

	    dp = (struct ext2_inode *)buf;
	    dp += itoo(fs, inumber);
	    fp->i_ic = *dp;
	    fp->f_size = dp->i_size;
	}

	(void) vm_deallocate(mach_task_self(), buf, buf_size);

	/*
	 * Clear out the old buffers
	 */
	free_file_buffers(fp);

	return (0);
}

/*
 * Given an offset in a file, find the disk block number that
 * contains that block.
 */
static int
block_map(fp, file_block, disk_block_p)
	struct file	*fp;
	daddr_t		file_block;
	daddr_t		*disk_block_p;	/* out */
{
	int		level;
	int		idx;
	daddr_t		ind_block_num;
	kern_return_t	rc;

	vm_offset_t	olddata[NIADDR+1];
	vm_size_t	oldsize[NIADDR+1];

	/*
	 * Index structure of an inode:
	 *
	 * i_db[0..NDADDR-1]	hold block numbers for blocks
	 *			0..NDADDR-1
	 *
	 * i_ib[0]		index block 0 is the single indirect
	 *			block
	 *			holds block numbers for blocks
	 *			NDADDR .. NDADDR + NINDIR(fs)-1
	 *
	 * i_ib[1]		index block 1 is the double indirect
	 *			block
	 *			holds block numbers for INDEX blocks
	 *			for blocks
	 *			NDADDR + NINDIR(fs) ..
	 *			NDADDR + NINDIR(fs) + NINDIR(fs)**2 - 1
	 *
	 * i_ib[2]		index block 2 is the triple indirect
	 *			block
	 *			holds block numbers for double-indirect
	 *			blocks for blocks
	 *			NDADDR + NINDIR(fs) + NINDIR(fs)**2 ..
	 *			NDADDR + NINDIR(fs) + NINDIR(fs)**2
	 *				+ NINDIR(fs)**3 - 1
	 */

	mutex_lock(&fp->f_lock);

	if (file_block < NDADDR) {
	    /* Direct block. */
	    *disk_block_p = fp->i_ic.i_block[file_block];
	    mutex_unlock(&fp->f_lock);
	    return (0);
	}

	file_block -= NDADDR;

	/*
	 * nindir[0] = NINDIR
	 * nindir[1] = NINDIR**2
	 * nindir[2] = NINDIR**3
	 *	etc
	 */
	for (level = 0; level < NIADDR; level++) {
	    if (file_block < fp->f_nindir[level])
		break;
	    file_block -= fp->f_nindir[level];
	}
	if (level == NIADDR) {
	    /* Block number too high */
	    mutex_unlock(&fp->f_lock);
	    return (FS_NOT_IN_FILE);
	}

	ind_block_num = fp->i_ic.i_block[level + NDADDR];

	/*
	 * Initialize array of blocks to free.
	 */
	for (idx = 0; idx < NIADDR; idx++)
	    oldsize[idx] = 0;

	for (; level >= 0; level--) {

	    vm_offset_t	data;
	    mach_msg_type_number_t	size;

	    if (ind_block_num == 0)
		break;

	    if (fp->f_blkno[level] == ind_block_num) {
		/*
		 *	Cache hit.  Just pick up the data.
		 */

		data = fp->f_blk[level];
	    }
	    else {
		/*
		 *	Drop our lock while doing the read.
		 *	(The f_dev and f_fs fields don`t change.)
		 */
		mutex_unlock(&fp->f_lock);

		rc = device_read(fp->f_dev,
				0,
				(recnum_t) fsbtodb(fp->f_fs, ind_block_num),
				EXT2_BLOCK_SIZE(fp->f_fs),
				(char **)&data,
				&size);
		if (rc != KERN_SUCCESS)
		    return (rc);

		/*
		 *	See if we can cache the data.  Need a write lock to
		 *	do this.  While we hold the write lock, we can`t do
		 *	*anything* which might block for memory.  Otherwise
		 *	a non-privileged thread might deadlock with the
		 *	privileged threads.  We can`t block while taking the
		 *	write lock.  Otherwise a non-privileged thread
		 *	blocked in the vm_deallocate (while holding a read
		 *	lock) will block a privileged thread.  For the same
		 *	reason, we can`t take a read lock and then use
		 *	lock_read_to_write.
		 */

		mutex_lock(&fp->f_lock);

		olddata[level] = fp->f_blk[level];
		oldsize[level] = fp->f_blksize[level];

		fp->f_blkno[level] = ind_block_num;
		fp->f_blk[level] = data;
		fp->f_blksize[level] = size;

		/*
		 *	Return to holding a read lock, and
		 *	dispose of old data.
		 */

	    }

	    if (level > 0) {
		idx = file_block / fp->f_nindir[level-1];
		file_block %= fp->f_nindir[level-1];
	    }
	    else
		idx = file_block;

	    ind_block_num = ((daddr_t *)data)[idx];
	}

	mutex_unlock(&fp->f_lock);

	/*
	 * After unlocking the file, free any blocks that
	 * we need to free.
	 */
	for (idx = 0; idx < NIADDR; idx++)
	    if (oldsize[idx] != 0)
		(void) vm_deallocate(mach_task_self(),
				     olddata[idx],
				     oldsize[idx]);

	*disk_block_p = ind_block_num;
	return (0);
}

/*
 * Read a portion of a file into an internal buffer.  Return
 * the location in the buffer and the amount in the buffer.
 */
static int
buf_read_file(fp, offset, buf_p, size_p)
	register struct file	*fp;
	vm_offset_t		offset;
	vm_offset_t		*buf_p;		/* out */
	vm_size_t		*size_p;	/* out */
{
	register
	struct ext2_super_block	*fs;
	vm_offset_t		off;
	register daddr_t	file_block;
	daddr_t			disk_block;
	int			rc;
	vm_offset_t		block_size;

	if (offset >= fp->i_ic.i_size)
	    return (FS_NOT_IN_FILE);

	fs = fp->f_fs;

	off = blkoff(fs, offset);
	file_block = lblkno(fs, offset);
	block_size = blksize(fs, fp, file_block);

	if (file_block != fp->f_buf_blkno) {
	    rc = block_map(fp, file_block, &disk_block);
	    if (rc != 0)
		return (rc);

	    if (fp->f_buf)
		(void)vm_deallocate(mach_task_self(),
				    fp->f_buf,
				    fp->f_buf_size);

	    if (disk_block == 0) {
		(void)vm_allocate(mach_task_self(),
				  &fp->f_buf,
				  block_size,
				  TRUE);
		fp->f_buf_size = block_size;
	    }
	    else {
		rc = device_read(fp->f_dev,
				0,
				(recnum_t) fsbtodb(fs, disk_block),
				(int) block_size,
				(char **) &fp->f_buf,
				(mach_msg_type_number_t *)&fp->f_buf_size);
	    }
	    if (rc)
		return (rc);

	    fp->f_buf_blkno = file_block;
	}

	/*
	 * Return address of byte in buffer corresponding to
	 * offset, and size of remainder of buffer after that
	 * byte.
	 */
	*buf_p = fp->f_buf + off;
	*size_p = block_size - off;

	/*
	 * But truncate buffer at end of file.
	 */
	if (*size_p > fp->i_ic.i_size - offset)
	    *size_p = fp->i_ic.i_size - offset;

	return (0);
}

/*
 * Search a directory for a name and return its
 * i_number.
 */
static int
search_directory(name, fp, inumber_p)
	char *		name;
	register struct file *fp;
	ino_t		*inumber_p;	/* out */
{
	vm_offset_t	buf;
	vm_size_t	buf_size;
	vm_offset_t	offset;
	struct ext2_dir_entry_2 *dp;
	int		length;
	kern_return_t	rc;
	char		tmp_name[256];

	length = strlen(name);

	offset = 0;
	while (offset < fp->i_ic.i_size) {
	    rc = buf_read_file(fp, offset, &buf, &buf_size);
	    if (rc != KERN_SUCCESS)
		return (rc);

	    dp = (struct ext2_dir_entry_2 *)buf;
	    if (dp->inode != 0) {
		strncpy (tmp_name, dp->name, dp->name_len);
		tmp_name[dp->name_len] = '\0';
		if (dp->name_len == length &&
		    !strcmp(name, tmp_name))
	    	{
		    /* found entry */
		    *inumber_p = dp->inode;
		    return (0);
		}
	    }
	    offset += dp->rec_len;
	}
	return (FS_NO_ENTRY);
}

static int
read_fs(dev, fsp, gdp, gd_size_p)
	mach_port_t dev;
	struct ext2_super_block **fsp;
	struct ext2_group_desc  **gdp;
	vm_size_t		*gd_size_p;
{
	register
	struct ext2_super_block *fs;
	vm_offset_t		buf;
	vm_offset_t		buf2;
	mach_msg_type_number_t	buf_size;
	mach_msg_type_number_t	buf2_size;
	int			error;
	int			gd_count;
	int			gd_blocks;
	int			gd_size;
	int			gd_location;
	int			gd_sector;

	/*
	 * Read the super block
	 */
	error = device_read(dev, 0, (recnum_t) SBLOCK, SBSIZE,
			    (char **) &buf, &buf_size);
	if (error)
	    return (error);

	/*
	 * Check the superblock
	 */
	fs = (struct ext2_super_block *)buf;
	if (fs->s_magic != EXT2_SUPER_MAGIC) {
		(void) vm_deallocate(mach_task_self(), buf, buf_size);
		return (FS_INVALID_FS);
	}

	*fsp = fs;

	/*
	 * Compute the groups informations
	 */
	gd_count = (fs->s_blocks_count - fs->s_first_data_block +
		    fs->s_blocks_per_group - 1) / fs->s_blocks_per_group;
	gd_blocks = (gd_count + EXT2_DESC_PER_BLOCK(fs) - 1) /
		    EXT2_DESC_PER_BLOCK(fs);
	gd_size = gd_blocks * EXT2_BLOCK_SIZE(fs);
	gd_location = fs->s_first_data_block + 1;
	gd_sector = (gd_location * EXT2_BLOCK_SIZE(fs)) / DEV_BSIZE;

	/*
	 * Read the groups descriptors
	 */
	error = device_read(dev, 0, (recnum_t) gd_sector, gd_size,
			    (char **) &buf2, &buf2_size);
	if (error) {
		(void) vm_deallocate(mach_task_self(), buf, buf_size);
		return error;
	}

	*gdp = (struct ext2_group_desc *) buf2;
	*gd_size_p = gd_size;

	return 0;
}

static int
mount_fs(fp)
	register struct file	*fp;
{
	register struct ext2_super_block *fs;
	int error;

	error = read_fs(fp->f_dev, &fp->f_fs, &fp->f_gd, &fp->f_gd_size);
	if (error)
	    return (error);

	fs = fp->f_fs;

	/*
	 * Calculate indirect block levels.
	 */
	{
	    register int	mult;
	    register int	level;

	    mult = 1;
	    for (level = 0; level < NIADDR; level++) {
		mult *= NINDIR(fs);
		fp->f_nindir[level] = mult;
	    }
	}

	return (0);
}

static void
unmount_fs(fp)
	register struct file	*fp;
{
	if (file_is_structured(fp)) {
	    (void) vm_deallocate(mach_task_self(),
				 (vm_offset_t) fp->f_fs,
				 SBSIZE);
	    (void) vm_deallocate(mach_task_self(),
				 (vm_offset_t) fp->f_gd,
				 fp->f_gd_size);
	    fp->f_fs = 0;
	}
}

/*
 * Open a file.
 */
int
ext2_open_file(master_device_port, path, fp)
	mach_port_t	master_device_port;
	char *		path;
	struct file	*fp;
{
#define	RETURN(code)	{ rc = (code); goto exit; }

	register char	*cp, *component;
	register int	c;	/* char */
	register int	rc;
	ino_t		inumber, parent_inumber;
	int		nlinks = 0;

	char	namebuf[MAXPATHLEN+1];

	if (path == 0 || *path == '\0') {
	    return FS_NO_ENTRY;
	}

	/*
	 * Copy name into buffer to allow modifying it.
	 */
	strcpy(namebuf, path);

	/*
	 * Look for '/dev/xxx' at start of path, for
	 * root device.
	 */
	if (!strprefix(namebuf, "/dev/")) {
	    printf("no device name\n");
	    return FS_NO_ENTRY;
	}

	cp = namebuf + 5;	/* device */
	component = cp;
	while ((c = *cp) != '\0' && c != '/') {
	    cp++;
	}
	*cp = '\0';

	bzero (fp, sizeof (struct file));

	rc = device_open(master_device_port,
			D_READ|D_WRITE,
			component,
			&fp->f_dev);
	if (rc)
	    return rc;

	if (c == 0) {
	    fp->f_fs = 0;
	    goto out_ok;
	}

	*cp = c;

	rc = mount_fs(fp);
	if (rc)
	    return rc;

	inumber = (ino_t) ROOTINO;
	if ((rc = read_inode(inumber, fp)) != 0) {
	    printf("can't read root inode\n");
	    goto exit;
	}

	while (*cp) {

	    /*
	     * Check that current node is a directory.
	     */
	    if ((fp->i_ic.i_mode & IFMT) != IFDIR)
		RETURN (FS_NOT_DIRECTORY);

	    /*
	     * Remove extra separators
	     */
	    while (*cp == '/')
		cp++;

	    /*
	     * Get next component of path name.
	     */
	    component = cp;
	    {
		register int	len = 0;

		while ((c = *cp) != '\0' && c != '/') {
		    if (len++ > MAXNAMLEN)
			RETURN (FS_NAME_TOO_LONG);
		    if (c & 0200)
			RETURN (FS_INVALID_PARAMETER);
		    cp++;
		}
		*cp = 0;
	    }

	    /*
	     * Look up component in current directory.
	     * Save directory inumber in case we find a
	     * symbolic link.
	     */
	    parent_inumber = inumber;
	    rc = search_directory(component, fp, &inumber);
	    if (rc) {
		printf("%s: not found\n", path);
		goto exit;
	    }
	    *cp = c;

	    /*
	     * Open next component.
	     */
	    if ((rc = read_inode(inumber, fp)) != 0)
		goto exit;

	    /*
	     * Check for symbolic link.
	     */
	    if ((fp->i_ic.i_mode & IFMT) == IFLNK) {

		int	link_len = fp->i_ic.i_size;
		int	len;

		len = strlen(cp) + 1;

		if (link_len + len >= MAXPATHLEN - 1)
		    RETURN (FS_NAME_TOO_LONG);

		if (++nlinks > MAXSYMLINKS)
		    RETURN (FS_SYMLINK_LOOP);

		memmove(&namebuf[link_len], cp, len);

#ifdef	IC_FASTLINK
		if (fp->i_ic.i_blocks == 0) {
		    bcopy(fp->i_ic.i_block, namebuf, (unsigned) link_len);
		}
		else
#endif	/* IC_FASTLINK */
		{
		    /*
		     * Read file for symbolic link
		     */
		    vm_offset_t	buf;
		    mach_msg_type_number_t	buf_size;
		    daddr_t	disk_block;
		    register struct ext2_super_block *fs = fp->f_fs;

		    (void) block_map(fp, (daddr_t)0, &disk_block);
		    rc = device_read(fp->f_dev,
				     0,
				     (recnum_t) fsbtodb(fs, disk_block),
				     (int) blksize(fs, fp, 0),
				     (char **) &buf,
				     &buf_size);
		    if (rc)
			goto exit;

		    bcopy((char *)buf, namebuf, (unsigned)link_len);
		    (void) vm_deallocate(mach_task_self(), buf, buf_size);
		}

		/*
		 * If relative pathname, restart at parent directory.
		 * If absolute pathname, restart at root.
		 * If pathname begins '/dev/<device>/',
		 *	restart at root of that device.
		 */
		cp = namebuf;
		if (*cp != '/') {
		    inumber = parent_inumber;
		}
		else if (!strprefix(cp, "/dev/")) {
		    inumber = (ino_t)ROOTINO;
		}
		else {
		    cp += 5;
		    component = cp;
		    while ((c = *cp) != '\0' && c != '/') {
			cp++;
		    }
		    *cp = '\0';

		    /*
		     * Unmount current file system and free buffers.
		     */
		    ext2_close_file(fp);

		    /*
		     * Open new root device.
		     */
		    rc = device_open(master_device_port,
				D_READ,
				component,
				&fp->f_dev);
		    if (rc)
			return (rc);

		    if (c == 0) {
			fp->f_fs = 0;
			goto out_ok;
		    }

		    *cp = c;

		    rc = mount_fs(fp);
		    if (rc)
			return (rc);

		    inumber = (ino_t)ROOTINO;
		}
		if ((rc = read_inode(inumber, fp)) != 0)
		    goto exit;
	    }
	}

	/*
	 * Found terminal component.
	 */
    out_ok:
	mutex_init(&fp->f_lock);
	return 0;

	/*
	 * At error exit, close file to free storage.
	 */
    exit:
	ext2_close_file(fp);
	return rc;
}

/*
 * Close file - free all storage used.
 */
void
ext2_close_file(fp)
	register struct file	*fp;
{
	register int	i;

	/*
	 * Free the disk super-block.
	 */
	unmount_fs(fp);

	/*
	 * Free the inode and data buffers.
	 */
	free_file_buffers(fp);
}

int
ext2_file_is_directory(struct file *fp)
{
	return (fp->i_ic.i_mode & IFMT) == IFDIR;
}

int
ext2_file_is_regular(struct file *fp)
{
	return (fp->i_ic.i_mode & IFMT) == IFREG;
}

/*
 * Copy a portion of a file into kernel memory.
 * Cross block boundaries when necessary.
 */
int
ext2_read_file(fp, offset, start, size, resid)
	register struct file	*fp;
	vm_offset_t		offset;
	vm_offset_t		start;
	vm_size_t		size;
	vm_size_t		*resid;	/* out */
{
	int			rc;
	register vm_size_t	csize;
	vm_offset_t		buf;
	vm_size_t		buf_size;

	while (size != 0) {
	    rc = buf_read_file(fp, offset, &buf, &buf_size);
	    if (rc)
		return (rc);

	    csize = size;
	    if (csize > buf_size)
		csize = buf_size;
	    if (csize == 0)
		break;

	    bcopy((char *)buf, (char *)start, csize);

	    offset += csize;
	    start  += csize;
	    size   -= csize;
	}
	if (resid)
	    *resid = size;

	return (0);
}

/* simple utility: only works for 2^n */
static int
log2(n)
	register unsigned int n;
{
	register int i = 0;

	while ((n & 1) == 0) {
		i++;
		n >>= 1;
	}
	return i;
}

/*
 * Make an empty file_direct for a device.
 */
int
ext2_open_file_direct(dev, fdp, is_structured)
	mach_port_t	dev;
	register struct file_direct *fdp;
	boolean_t	is_structured;
{
	struct ext2_super_block *fs;
	struct ext2_group_desc	*gd;
	vm_size_t		gd_size;
	int			rc;

	if (!is_structured) {
		fdp->fd_dev     = dev;
		fdp->fd_blocks  = (daddr_t *) 0;
		fdp->fd_bsize   = vm_page_size;
		fdp->fd_bshift  = log2(vm_page_size);
		fdp->fd_fsbtodb = 0;	/* later */
		fdp->fd_size    = 0;	/* later */
		return 0;
	}

	rc = read_fs(dev, &fs, &gd, &gd_size);
	if (rc)
		return rc;

	fdp->fd_dev = dev;
	fdp->fd_blocks = (daddr_t *) 0;
	fdp->fd_size = 0;
	fdp->fd_bsize = EXT2_BLOCK_SIZE(fs);
	fdp->fd_bshift = log2(fdp->fd_bsize);
	fdp->fd_fsbtodb = log2(fdp->fd_bsize / DEV_BSIZE);

	(void) vm_deallocate(mach_task_self(),
			     (vm_offset_t) fs,
			     SBSIZE);
	(void) vm_deallocate(mach_task_self(),
			     (vm_offset_t) gd,
			     gd_size);

	return 0;
}

/*
 * Add blocks from a file to a file_direct.
 */
int
ext2_add_file_direct(fdp, fp)
	register struct file_direct *fdp;
	register struct file *fp;
{
	register struct ext2_super_block *fs;
	long num_blocks, i;
	vm_offset_t buffer;
	vm_size_t size;
	int rc;

	/* the file must be on the same device */

	if (fdp->fd_dev != fp->f_dev)
		return FS_INVALID_FS;

	if (!file_is_structured(fp)) {
		int	result[DEV_GET_SIZE_COUNT];
		natural_t count;

		count = DEV_GET_SIZE_COUNT;
		rc = device_get_status(	fdp->fd_dev, DEV_GET_SIZE,
					result, &count);
		if (rc)
			return rc;
		fdp->fd_size = result[DEV_GET_SIZE_DEVICE_SIZE] >> fdp->fd_bshift;
		fdp->fd_fsbtodb = log2(fdp->fd_bsize/result[DEV_GET_SIZE_RECORD_SIZE]);
		return 0;
	}

	/* it must hold a file system */

	fs = fp->f_fs;
/*
	if (fdp->fd_bsize != fs->fs_bsize ||
	    fdp->fd_fsbtodb != fs->fs_fsbtodb)
*/
	if (fdp->fd_bsize != EXT2_BLOCK_SIZE(fs))
		return FS_INVALID_FS;

	/* calculate number of blocks in the file, ignoring fragments */

	num_blocks = lblkno(fs, fp->i_ic.i_size);

	/* allocate memory for a bigger array */

	size = (num_blocks + fdp->fd_size) * sizeof(daddr_t);
	rc = vm_allocate(mach_task_self(), &buffer, size, TRUE);
	if (rc != KERN_SUCCESS)
		return rc;

	/* lookup new block addresses */

	for (i = 0; i < num_blocks; i++) {
		daddr_t disk_block;

		rc = block_map(fp, (daddr_t) i, &disk_block);
		if (rc != 0) {
			(void) vm_deallocate(mach_task_self(), buffer, size);
			return rc;
		}

		((daddr_t *) buffer)[fdp->fd_size + i] = disk_block;
	}

	/* copy old addresses and install the new array */

	if (fdp->fd_blocks != 0) {
		bcopy((char *) fdp->fd_blocks, (char *) buffer,
		      fdp->fd_size * sizeof(daddr_t));

		(void) vm_deallocate(mach_task_self(),
				(vm_offset_t) fdp->fd_blocks,
				(vm_size_t) (fdp->fd_size * sizeof(daddr_t)));
	}
	fdp->fd_blocks = (daddr_t *) buffer;
	fdp->fd_size += num_blocks;

	/* deallocate cached blocks */

	free_file_buffers(fp);

	return 0;
}

int
ext2_remove_file_direct(fdp)
	struct file_direct	*fdp;
{
	if (fdp->fd_blocks)
	(void) vm_deallocate(mach_task_self(),
			     (vm_offset_t) fdp->fd_blocks,
			     (vm_size_t) (fdp->fd_size * sizeof(daddr_t)));
	fdp->fd_blocks = 0; /* sanity */
	/* xxx should lose a ref to fdp->fd_dev here (and elsewhere) xxx */
}