summaryrefslogtreecommitdiff
path: root/sys/kern/subr_disk.c
blob: 2bf6f11b7e049fae740affe3e14f12dbdffc50b3 (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
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
/*	$OpenBSD: subr_disk.c,v 1.67 2007/12/16 20:57:17 otto Exp $	*/
/*	$NetBSD: subr_disk.c,v 1.17 1996/03/16 23:17:08 christos Exp $	*/

/*
 * Copyright (c) 1995 Jason R. Thorpe.  All rights reserved.
 * Copyright (c) 1982, 1986, 1988, 1993
 *	The Regents of the University of California.  All rights reserved.
 * (c) UNIX System Laboratories, Inc.
 * All or some portions of this file are derived from material licensed
 * to the University of California by American Telephone and Telegraph
 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
 * the permission of UNIX System Laboratories, Inc.
 *
 * 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.
 *
 *	@(#)ufs_disksubr.c	8.5 (Berkeley) 1/21/94
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/fcntl.h>
#include <sys/buf.h>
#include <sys/stat.h>
#include <sys/syslog.h>
#include <sys/device.h>
#include <sys/time.h>
#include <sys/disklabel.h>
#include <sys/conf.h>
#include <sys/lock.h>
#include <sys/disk.h>
#include <sys/reboot.h>
#include <sys/dkio.h>
#include <sys/dkstat.h>		/* XXX */
#include <sys/proc.h>
#include <uvm/uvm_extern.h>

#include <dev/rndvar.h>
#include <dev/cons.h>

/*
 * A global list of all disks attached to the system.  May grow or
 * shrink over time.
 */
struct	disklist_head disklist;	/* TAILQ_HEAD */
int	disk_count;		/* number of drives in global disklist */
int	disk_change;		/* set if a disk has been attached/detached
				 * since last we looked at this variable. This
				 * is reset by hw_sysctl()
				 */

/*
 * Seek sort for disks.  We depend on the driver which calls us using b_resid
 * as the current cylinder number.
 *
 * The argument ap structure holds a b_actf activity chain pointer on which we
 * keep two queues, sorted in ascending cylinder order.  The first queue holds
 * those requests which are positioned after the current cylinder (in the first
 * request); the second holds requests which came in after their cylinder number
 * was passed.  Thus we implement a one way scan, retracting after reaching the
 * end of the drive to the first request on the second queue, at which time it
 * becomes the first queue.
 *
 * A one-way scan is natural because of the way UNIX read-ahead blocks are
 * allocated.
 */

void
disksort(struct buf *ap, struct buf *bp)
{
	struct buf *bq;

	/* If the queue is empty, then it's easy. */
	if (ap->b_actf == NULL) {
		bp->b_actf = NULL;
		ap->b_actf = bp;
		return;
	}

	/*
	 * If we lie after the first (currently active) request, then we
	 * must locate the second request list and add ourselves to it.
	 */
	bq = ap->b_actf;
	if (bp->b_cylinder < bq->b_cylinder) {
		while (bq->b_actf) {
			/*
			 * Check for an ``inversion'' in the normally ascending
			 * cylinder numbers, indicating the start of the second
			 * request list.
			 */
			if (bq->b_actf->b_cylinder < bq->b_cylinder) {
				/*
				 * Search the second request list for the first
				 * request at a larger cylinder number.  We go
				 * before that; if there is no such request, we
				 * go at end.
				 */
				do {
					if (bp->b_cylinder <
					    bq->b_actf->b_cylinder)
						goto insert;
					if (bp->b_cylinder ==
					    bq->b_actf->b_cylinder &&
					    bp->b_blkno < bq->b_actf->b_blkno)
						goto insert;
					bq = bq->b_actf;
				} while (bq->b_actf);
				goto insert;		/* after last */
			}
			bq = bq->b_actf;
		}
		/*
		 * No inversions... we will go after the last, and
		 * be the first request in the second request list.
		 */
		goto insert;
	}
	/*
	 * Request is at/after the current request...
	 * sort in the first request list.
	 */
	while (bq->b_actf) {
		/*
		 * We want to go after the current request if there is an
		 * inversion after it (i.e. it is the end of the first
		 * request list), or if the next request is a larger cylinder
		 * than our request.
		 */
		if (bq->b_actf->b_cylinder < bq->b_cylinder ||
		    bp->b_cylinder < bq->b_actf->b_cylinder ||
		    (bp->b_cylinder == bq->b_actf->b_cylinder &&
		    bp->b_blkno < bq->b_actf->b_blkno))
			goto insert;
		bq = bq->b_actf;
	}
	/*
	 * Neither a second list nor a larger request... we go at the end of
	 * the first list, which is the same as the end of the whole schebang.
	 */
insert:	bp->b_actf = bq->b_actf;
	bq->b_actf = bp;
}

/*
 * Compute checksum for disk label.
 */
u_int
dkcksum(struct disklabel *lp)
{
	u_int16_t *start, *end;
	u_int16_t sum = 0;

	start = (u_int16_t *)lp;
	end = (u_int16_t *)&lp->d_partitions[lp->d_npartitions];
	while (start < end)
		sum ^= *start++;
	return (sum);
}

char *
initdisklabel(struct disklabel *lp)
{
	int i;

	/* minimal requirements for archetypal disk label */
	if (lp->d_secsize < DEV_BSIZE)
		lp->d_secsize = DEV_BSIZE;
	if (DL_GETDSIZE(lp) == 0)
		DL_SETDSIZE(lp, MAXDISKSIZE);
	if (lp->d_secpercyl == 0)
		return ("invalid geometry");
	lp->d_npartitions = RAW_PART + 1;
	for (i = 0; i < RAW_PART; i++) {
		DL_SETPSIZE(&lp->d_partitions[i], 0);
		DL_SETPOFFSET(&lp->d_partitions[i], 0);
	}
	if (DL_GETPSIZE(&lp->d_partitions[RAW_PART]) == 0)
		DL_SETPSIZE(&lp->d_partitions[RAW_PART], DL_GETDSIZE(lp));
	DL_SETPOFFSET(&lp->d_partitions[RAW_PART], 0);
	lp->d_version = 1;
	lp->d_bbsize = 8192;
	lp->d_sbsize = 64*1024;			/* XXX ? */
	return (NULL);
}

/*
 * Check an incoming block to make sure it is a disklabel, convert it to
 * a newer version if needed, etc etc.
 */
char *
checkdisklabel(void *rlp, struct disklabel *lp)
{
	struct disklabel *dlp = rlp;
	struct __partitionv0 *v0pp;
	struct partition *pp;
	daddr64_t disksize;
	char *msg = NULL;
	int i;

	if (dlp->d_magic != DISKMAGIC || dlp->d_magic2 != DISKMAGIC)
		msg = "no disk label";
	else if (dlp->d_npartitions > MAXPARTITIONS)
		msg = "unreasonable partition count";
	else if (dkcksum(dlp) != 0)
		msg = "disk label corrupted";

	if (msg) {
		u_int16_t *start, *end, sum = 0;

		/* If it is byte-swapped, attempt to convert it */
		if (swap32(dlp->d_magic) != DISKMAGIC ||
		    swap32(dlp->d_magic2) != DISKMAGIC ||
		    swap16(dlp->d_npartitions) > MAXPARTITIONS)
			return (msg);

		/*
		 * Need a byte-swap aware dkcksum varient
		 * inlined, because dkcksum uses a sub-field
		 */
		start = (u_int16_t *)dlp;
		end = (u_int16_t *)&dlp->d_partitions[
		    swap16(dlp->d_npartitions)];
		while (start < end)
			sum ^= *start++;
		if (sum != 0)
			return (msg);

		dlp->d_magic = swap32(dlp->d_magic);
		dlp->d_type = swap16(dlp->d_type);
		dlp->d_subtype = swap16(dlp->d_subtype);

		/* d_typename and d_packname are strings */

		dlp->d_secsize = swap32(dlp->d_secsize);
		dlp->d_nsectors = swap32(dlp->d_nsectors);
		dlp->d_ntracks = swap32(dlp->d_ntracks);
		dlp->d_ncylinders = swap32(dlp->d_ncylinders);
		dlp->d_secpercyl = swap32(dlp->d_secpercyl);
		dlp->d_secperunit = swap32(dlp->d_secperunit);

		dlp->d_sparespertrack = swap16(dlp->d_sparespertrack);
		dlp->d_sparespercyl = swap16(dlp->d_sparespercyl);

		dlp->d_acylinders = swap32(dlp->d_acylinders);

		dlp->d_rpm = swap16(dlp->d_rpm);
		dlp->d_interleave = swap16(dlp->d_interleave);
		dlp->d_trackskew = swap16(dlp->d_trackskew);
		dlp->d_cylskew = swap16(dlp->d_cylskew);
		dlp->d_headswitch = swap32(dlp->d_headswitch);
		dlp->d_trkseek = swap32(dlp->d_trkseek);
		dlp->d_flags = swap32(dlp->d_flags);

		for (i = 0; i < NDDATA; i++)
			dlp->d_drivedata[i] = swap32(dlp->d_drivedata[i]);

		dlp->d_secperunith = swap16(dlp->d_secperunith);
		dlp->d_version = swap16(dlp->d_version);

		for (i = 0; i < NSPARE; i++)
			dlp->d_spare[i] = swap32(dlp->d_spare[i]);

		dlp->d_magic2 = swap32(dlp->d_magic2);
		dlp->d_checksum = swap16(dlp->d_checksum);

		dlp->d_npartitions = swap16(dlp->d_npartitions);
		dlp->d_bbsize = swap32(dlp->d_bbsize);
		dlp->d_sbsize = swap32(dlp->d_sbsize);

		for (i = 0; i < MAXPARTITIONS; i++) {
			pp = &dlp->d_partitions[i];
			pp->p_size = swap32(pp->p_size);
			pp->p_offset = swap32(pp->p_offset);
			if (dlp->d_version == 0) {
				v0pp = (struct __partitionv0 *)pp;
				v0pp->p_fsize = swap32(v0pp->p_fsize);
			} else {
				pp->p_offseth = swap16(pp->p_offseth);
				pp->p_sizeh = swap16(pp->p_sizeh);
			}
			pp->p_cpg = swap16(pp->p_cpg);
		}

		dlp->d_checksum = 0;
		dlp->d_checksum = dkcksum(dlp);
		msg = NULL;
	}

	/* XXX should verify lots of other fields and whine a lot */

	if (msg)
		return (msg);

	/* Initial passed in lp contains the real disk size. */
	disksize = DL_GETDSIZE(lp);

	if (lp != dlp)
		*lp = *dlp;

	if (lp->d_version == 0) {
		lp->d_version = 1;
		lp->d_secperunith = 0;

		v0pp = (struct __partitionv0 *)lp->d_partitions;
		pp = lp->d_partitions;
		for (i = 0; i < lp->d_npartitions; i++, pp++, v0pp++) {
			pp->p_fragblock = DISKLABELV1_FFS_FRAGBLOCK(v0pp->
			    p_fsize, v0pp->p_frag);
			pp->p_offseth = 0;
			pp->p_sizeh = 0;
		}
	}

#ifdef DEBUG
	if (DL_GETDSIZE(lp) != disksize)
		printf("on-disk disklabel has incorrect disksize (%lld)\n",
		    DL_GETDSIZE(lp));
	if (DL_GETPSIZE(&lp->d_partitions[RAW_PART]) != disksize)
		printf("on-disk disklabel RAW_PART has incorrect size (%lld)\n",
		    DL_GETPSIZE(&lp->d_partitions[RAW_PART]));
	if (DL_GETPOFFSET(&lp->d_partitions[RAW_PART]) != 0)
		printf("on-disk disklabel RAW_PART offset != 0 (%lld)\n",
		    DL_GETPOFFSET(&lp->d_partitions[RAW_PART]));
#endif
	DL_SETDSIZE(lp, disksize);
	DL_SETPSIZE(&lp->d_partitions[RAW_PART], disksize);
	DL_SETPOFFSET(&lp->d_partitions[RAW_PART], 0);

	lp->d_checksum = 0;
	lp->d_checksum = dkcksum(lp);
	return (msg);
}

/*
 * If dos partition table requested, attempt to load it and
 * find disklabel inside a DOS partition. Return buffer
 * for use in signalling errors if requested.
 *
 * We would like to check if each MBR has a valid BOOT_MAGIC, but
 * we cannot because it doesn't always exist. So.. we assume the
 * MBR is valid.
 */
char *
readdoslabel(struct buf *bp, void (*strat)(struct buf *),
    struct disklabel *lp, int *partoffp, int spoofonly)
{
	struct dos_partition dp[NDOSPART], *dp2;
	u_int32_t extoff = 0;
	daddr64_t part_blkno = DOSBBSECTOR;
	int dospartoff = 0, i, ourpart = -1;
	int wander = 1, n = 0, loop = 0;

	if (lp->d_secpercyl == 0)
		return ("invalid label, d_secpercyl == 0");
	if (lp->d_secsize == 0)
		return ("invalid label, d_secsize == 0");

	/* do DOS partitions in the process of getting disklabel? */

	/*
	 * Read dos partition table, follow extended partitions.
	 * Map the partitions to disklabel entries i-p
	 */
	while (wander && n < 8 && loop < 8) {
		loop++;
		wander = 0;
		if (part_blkno < extoff)
			part_blkno = extoff;

		/* read boot record */
		bp->b_blkno = part_blkno;
		bp->b_bcount = lp->d_secsize;
		bp->b_flags = B_BUSY | B_READ;
		(*strat)(bp);
		if (biowait(bp)) {
/*wrong*/		if (partoffp)
/*wrong*/			*partoffp = -1;
			return ("dos partition I/O error");
		}

		bcopy(bp->b_data + DOSPARTOFF, dp, sizeof(dp));

		if (ourpart == -1 && part_blkno == DOSBBSECTOR) {
			/* Search for our MBR partition */
			for (dp2=dp, i=0; i < NDOSPART && ourpart == -1;
			    i++, dp2++)
				if (letoh32(dp2->dp_size) &&
				    dp2->dp_typ == DOSPTYP_OPENBSD)
					ourpart = i;
			if (ourpart == -1)
				goto donot;
			/*
			 * This is our MBR partition. need sector
			 * address for SCSI/IDE, cylinder for
			 * ESDI/ST506/RLL
			 */
			dp2 = &dp[ourpart];
			dospartoff = letoh32(dp2->dp_start) + part_blkno;

			/* found our OpenBSD partition, finish up */
			if (partoffp)
				goto notfat;

			if (lp->d_ntracks == 0)
				lp->d_ntracks = dp2->dp_ehd + 1;
			if (lp->d_nsectors == 0)
				lp->d_nsectors = DPSECT(dp2->dp_esect);
			if (lp->d_secpercyl == 0)
				lp->d_secpercyl = lp->d_ntracks *
				    lp->d_nsectors;
		}
donot:
		/*
		 * In case the disklabel read below fails, we want to
		 * provide a fake label in i-p.
		 */
		for (dp2=dp, i=0; i < NDOSPART && n < 8; i++, dp2++) {
			struct partition *pp = &lp->d_partitions[8+n];

			if (dp2->dp_typ == DOSPTYP_OPENBSD)
				continue;
			if (letoh32(dp2->dp_size) > DL_GETDSIZE(lp))
				continue;
			if (letoh32(dp2->dp_start) > DL_GETDSIZE(lp))
				continue;
			if (letoh32(dp2->dp_size) == 0)
				continue;
			if (letoh32(dp2->dp_start))
				DL_SETPOFFSET(pp,
				    letoh32(dp2->dp_start) + part_blkno);

			DL_SETPSIZE(pp, letoh32(dp2->dp_size));

			switch (dp2->dp_typ) {
			case DOSPTYP_UNUSED:
				pp->p_fstype = FS_UNUSED;
				n++;
				break;

			case DOSPTYP_LINUX:
				pp->p_fstype = FS_EXT2FS;
				n++;
				break;

			case DOSPTYP_NTFS:
				pp->p_fstype = FS_NTFS;
				n++;
				break;

			case DOSPTYP_FAT12:
			case DOSPTYP_FAT16S:
			case DOSPTYP_FAT16B:
			case DOSPTYP_FAT16L:
			case DOSPTYP_FAT32:
			case DOSPTYP_FAT32L:
				pp->p_fstype = FS_MSDOS;
				n++;
				break;
			case DOSPTYP_EXTEND:
			case DOSPTYP_EXTENDL:
				part_blkno = letoh32(dp2->dp_start) + extoff;
				if (!extoff) {
					extoff = letoh32(dp2->dp_start);
					part_blkno = 0;
				}
				wander = 1;
				break;
			default:
				pp->p_fstype = FS_OTHER;
				n++;
				break;
			}
		}
	}
	lp->d_npartitions = MAXPARTITIONS;

	if (n == 0 && part_blkno == DOSBBSECTOR) {
		u_int16_t fattest;

		/* Check for a short jump instruction. */
		fattest = ((bp->b_data[0] << 8) & 0xff00) |
		    (bp->b_data[2] & 0xff);
		if (fattest != 0xeb90 && fattest != 0xe900)
			goto notfat;

		/* Check for a valid bytes per sector value. */
		fattest = ((bp->b_data[12] << 8) & 0xff00) |
		    (bp->b_data[11] & 0xff);
		if (fattest < 512 || fattest > 4096 || (fattest % 512 != 0))
			goto notfat;

		/* Check the end of sector marker. */
		fattest = ((bp->b_data[510] << 8) & 0xff00) |
		    (bp->b_data[511] & 0xff);
		if (fattest != 0x55aa)
			goto notfat;

		/* Looks like a FAT filesystem. Spoof 'i'. */
		DL_SETPSIZE(&lp->d_partitions['i' - 'a'],
		    DL_GETPSIZE(&lp->d_partitions[RAW_PART]));
		DL_SETPOFFSET(&lp->d_partitions['i' - 'a'], 0);
		lp->d_partitions['i' - 'a'].p_fstype = FS_MSDOS;
	}
notfat:

	/* record the OpenBSD partition's placement for the caller */
	if (partoffp)
		*partoffp = dospartoff;

	/* don't read the on-disk label if we are in spoofed-only mode */
	if (spoofonly)
		return (NULL);

	bp->b_blkno = dospartoff + DOS_LABELSECTOR;
	bp->b_bcount = lp->d_secsize;
	bp->b_flags = B_BUSY | B_READ;
	(*strat)(bp);
	if (biowait(bp))
		return ("disk label I/O error");

	/* sub-MBR disklabels are always at a LABELOFFSET of 0 */
	return checkdisklabel(bp->b_data, lp);
}

/*
 * Check new disk label for sensibility
 * before setting it.
 */
int
setdisklabel(struct disklabel *olp, struct disklabel *nlp, u_int openmask)
{
	int i;
	struct partition *opp, *npp;

	/* sanity clause */
	if (nlp->d_secpercyl == 0 || nlp->d_secsize == 0 ||
	    (nlp->d_secsize % DEV_BSIZE) != 0)
		return (EINVAL);

	/* special case to allow disklabel to be invalidated */
	if (nlp->d_magic == 0xffffffff) {
		*olp = *nlp;
		return (0);
	}

	if (nlp->d_magic != DISKMAGIC || nlp->d_magic2 != DISKMAGIC ||
	    dkcksum(nlp) != 0)
		return (EINVAL);

	/* XXX missing check if other dos partitions will be overwritten */

	while (openmask != 0) {
		i = ffs(openmask) - 1;
		openmask &= ~(1 << i);
		if (nlp->d_npartitions <= i)
			return (EBUSY);
		opp = &olp->d_partitions[i];
		npp = &nlp->d_partitions[i];
		if (DL_GETPOFFSET(npp) != DL_GETPOFFSET(opp) ||
		    DL_GETPSIZE(npp) < DL_GETPSIZE(opp))
			return (EBUSY);
		/*
		 * Copy internally-set partition information
		 * if new label doesn't include it.		XXX
		 */
		if (npp->p_fstype == FS_UNUSED && opp->p_fstype != FS_UNUSED) {
			npp->p_fstype = opp->p_fstype;
			npp->p_fragblock = opp->p_fragblock;
			npp->p_cpg = opp->p_cpg;
		}
	}
	nlp->d_checksum = 0;
	nlp->d_checksum = dkcksum(nlp);
	*olp = *nlp;
	return (0);
}

/*
 * Determine the size of the transfer, and make sure it is within the
 * boundaries of the partition. Adjust transfer if needed, and signal errors or
 * early completion.
 */
int
bounds_check_with_label(struct buf *bp, struct disklabel *lp, int wlabel)
{
#define blockpersec(count, lp) ((count) * (((lp)->d_secsize) / DEV_BSIZE))
	struct partition *p = &lp->d_partitions[DISKPART(bp->b_dev)];
	daddr64_t sz = howmany(bp->b_bcount, DEV_BSIZE);

	/* avoid division by zero */
	if (lp->d_secpercyl == 0)
		goto bad;

	if (bp->b_blkno < 0 || sz < 0)
		panic("bounds_check_with_label %lld %lld\n", bp->b_blkno, sz);

	/* beyond partition? */
	if (bp->b_blkno + sz > blockpersec(DL_GETPSIZE(p), lp)) {
		sz = blockpersec(DL_GETPSIZE(p), lp) - bp->b_blkno;
		if (sz == 0) {
			/* If exactly at end of disk, return EOF. */
			bp->b_resid = bp->b_bcount;
			return (-1);
		}
		if (sz < 0)
			/* If past end of disk, return EINVAL. */
			goto bad;

		/* Otherwise, truncate request. */
		bp->b_bcount = sz << DEV_BSHIFT;
	}

	/* calculate cylinder for disksort to order transfers with */
	bp->b_cylinder = (bp->b_blkno + blockpersec(DL_GETPOFFSET(p), lp)) /
	    blockpersec(lp->d_secpercyl, lp);
	return (1);

bad:
	bp->b_error = EINVAL;
	bp->b_flags |= B_ERROR;
	return (-1);
}

/*
 * Disk error is the preface to plaintive error messages
 * about failing disk transfers.  It prints messages of the form

hp0g: hard error reading fsbn 12345 of 12344-12347 (hp0 bn %d cn %d tn %d sn %d)

 * if the offset of the error in the transfer and a disk label
 * are both available.  blkdone should be -1 if the position of the error
 * is unknown; the disklabel pointer may be null from drivers that have not
 * been converted to use them.  The message is printed with printf
 * if pri is LOG_PRINTF, otherwise it uses log at the specified priority.
 * The message should be completed (with at least a newline) with printf
 * or addlog, respectively.  There is no trailing space.
 */
void
diskerr(struct buf *bp, char *dname, char *what, int pri, int blkdone,
    struct disklabel *lp)
{
	int unit = DISKUNIT(bp->b_dev), part = DISKPART(bp->b_dev);
	int (*pr)(const char *, ...);
	char partname = 'a' + part;
	daddr64_t sn;

	if (pri != LOG_PRINTF) {
		static const char fmt[] = "";
		log(pri, fmt);
		pr = addlog;
	} else
		pr = printf;
	(*pr)("%s%d%c: %s %sing fsbn ", dname, unit, partname, what,
	    bp->b_flags & B_READ ? "read" : "writ");
	sn = bp->b_blkno;
	if (bp->b_bcount <= DEV_BSIZE)
		(*pr)("%lld", sn);
	else {
		if (blkdone >= 0) {
			sn += blkdone;
			(*pr)("%lld of ", sn);
		}
		(*pr)("%lld-%lld", bp->b_blkno,
		    bp->b_blkno + (bp->b_bcount - 1) / DEV_BSIZE);
	}
	if (lp && (blkdone >= 0 || bp->b_bcount <= lp->d_secsize)) {
		sn += DL_GETPOFFSET(&lp->d_partitions[part]);
		(*pr)(" (%s%d bn %lld; cn %lld", dname, unit, sn,
		    sn / lp->d_secpercyl);
		sn %= lp->d_secpercyl;
		(*pr)(" tn %lld sn %lld)", sn / lp->d_nsectors,
		    sn % lp->d_nsectors);
	}
}

/*
 * Initialize the disklist.  Called by main() before autoconfiguration.
 */
void
disk_init(void)
{

	TAILQ_INIT(&disklist);
	disk_count = disk_change = 0;
}

int
disk_construct(struct disk *diskp, char *lockname)
{
	rw_init(&diskp->dk_lock, lockname);
	
	diskp->dk_flags |= DKF_CONSTRUCTED;
	    
	return (0);
}

/*
 * Attach a disk.
 */
void
disk_attach(struct disk *diskp)
{

	if (!ISSET(diskp->dk_flags, DKF_CONSTRUCTED))
		disk_construct(diskp, diskp->dk_name);

	/*
	 * Allocate and initialize the disklabel structures.  Note that
	 * it's not safe to sleep here, since we're probably going to be
	 * called during autoconfiguration.
	 */
	diskp->dk_label = malloc(sizeof(struct disklabel), M_DEVBUF,
	    M_NOWAIT|M_ZERO);
	if (diskp->dk_label == NULL)
		panic("disk_attach: can't allocate storage for disklabel");

	/*
	 * Set the attached timestamp.
	 */
	microuptime(&diskp->dk_attachtime);

	/*
	 * Link into the disklist.
	 */
	TAILQ_INSERT_TAIL(&disklist, diskp, dk_link);
	++disk_count;
	disk_change = 1;
}

/*
 * Detach a disk.
 */
void
disk_detach(struct disk *diskp)
{

	/*
	 * Free the space used by the disklabel structures.
	 */
	free(diskp->dk_label, M_DEVBUF);

	/*
	 * Remove from the disklist.
	 */
	TAILQ_REMOVE(&disklist, diskp, dk_link);
	disk_change = 1;
	if (--disk_count < 0)
		panic("disk_detach: disk_count < 0");
}

/*
 * Increment a disk's busy counter.  If the counter is going from
 * 0 to 1, set the timestamp.
 */
void
disk_busy(struct disk *diskp)
{

	/*
	 * XXX We'd like to use something as accurate as microtime(),
	 * but that doesn't depend on the system TOD clock.
	 */
	if (diskp->dk_busy++ == 0) {
		microuptime(&diskp->dk_timestamp);
	}
}

/*
 * Decrement a disk's busy counter, increment the byte count, total busy
 * time, and reset the timestamp.
 */
void
disk_unbusy(struct disk *diskp, long bcount, int read)
{
	struct timeval dv_time, diff_time;

	if (diskp->dk_busy-- == 0)
		printf("disk_unbusy: %s: dk_busy < 0\n", diskp->dk_name);

	microuptime(&dv_time);

	timersub(&dv_time, &diskp->dk_timestamp, &diff_time);
	timeradd(&diskp->dk_time, &diff_time, &diskp->dk_time);

	diskp->dk_timestamp = dv_time;
	if (bcount > 0) {
		if (read) {
			diskp->dk_rbytes += bcount;
			diskp->dk_rxfer++;
		} else {
			diskp->dk_wbytes += bcount;
			diskp->dk_wxfer++;
		}
	} else
		diskp->dk_seek++;

	add_disk_randomness(bcount ^ diff_time.tv_usec);
}

int
disk_lock(struct disk *dk)
{
	int error;

	error = rw_enter(&dk->dk_lock, RW_WRITE|RW_INTR);

	return (error);
}

void
disk_unlock(struct disk *dk)
{
	rw_exit(&dk->dk_lock);
}

int
dk_mountroot(void)
{
	dev_t rawdev, rrootdev;
	int part = DISKPART(rootdev);
	int (*mountrootfn)(void);
	struct disklabel dl;
	int error;

	rrootdev = blktochr(rootdev);
	rawdev = MAKEDISKDEV(major(rrootdev), DISKUNIT(rootdev), RAW_PART);
#ifdef DEBUG
	printf("rootdev=0x%x rrootdev=0x%x rawdev=0x%x\n", rootdev,
	    rrootdev, rawdev);
#endif

	/*
	 * open device, ioctl for the disklabel, and close it.
	 */
	error = (cdevsw[major(rrootdev)].d_open)(rawdev, FREAD,
	    S_IFCHR, curproc);
	if (error)
		panic("cannot open disk, 0x%x/0x%x, error %d",
		    rootdev, rrootdev, error);
	error = (cdevsw[major(rrootdev)].d_ioctl)(rawdev, DIOCGDINFO,
	    (caddr_t)&dl, FREAD, curproc);
	if (error)
		panic("cannot read disk label, 0x%x/0x%x, error %d",
		    rootdev, rrootdev, error);
	(void) (cdevsw[major(rrootdev)].d_close)(rawdev, FREAD,
	    S_IFCHR, curproc);

	if (DL_GETPSIZE(&dl.d_partitions[part]) == 0)
		panic("root filesystem has size 0");
	switch (dl.d_partitions[part].p_fstype) {
#ifdef EXT2FS
	case FS_EXT2FS:
		{
		extern int ext2fs_mountroot(void);
		mountrootfn = ext2fs_mountroot;
		}
		break;
#endif
#ifdef FFS
	case FS_BSDFFS:
		{
		extern int ffs_mountroot(void);
		mountrootfn = ffs_mountroot;
		}
		break;
#endif
#ifdef CD9660
	case FS_ISO9660:
		{
		extern int cd9660_mountroot(void);
		mountrootfn = cd9660_mountroot;
		}
		break;
#endif
	default:
#ifdef FFS
		{ 
		extern int ffs_mountroot(void);

		printf("filesystem type %d not known.. assuming ffs\n",
		    dl.d_partitions[part].p_fstype);
		mountrootfn = ffs_mountroot;
		}
#else
		panic("disk 0x%x/0x%x filesystem type %d not known", 
		    rootdev, rrootdev, dl.d_partitions[part].p_fstype);
#endif
	}
	return (*mountrootfn)();
}

struct bufq *
bufq_default_alloc(void)
{
	struct bufq_default *bq;

	bq = malloc(sizeof(*bq), M_DEVBUF, M_NOWAIT|M_ZERO);
	if (bq == NULL)
		panic("bufq_default_alloc: no memory");

	bq->bufq.bufq_free = bufq_default_free;
	bq->bufq.bufq_add = bufq_default_add;
	bq->bufq.bufq_get = bufq_default_get;

	return ((struct bufq *)bq);
}

void
bufq_default_free(struct bufq *bq)
{
	free(bq, M_DEVBUF);
}

void
bufq_default_add(struct bufq *bq, struct buf *bp)
{
	struct bufq_default *bufq = (struct bufq_default *)bq;
	struct proc *p = bp->b_proc;
	struct buf *head;

	if (p == NULL || p->p_nice < NZERO)
		head = &bufq->bufq_head[0];
	else if (p->p_nice == NZERO)
		head = &bufq->bufq_head[1];
	else
		head = &bufq->bufq_head[2];

	disksort(head, bp);
}

struct buf *
bufq_default_get(struct bufq *bq)
{
	struct bufq_default *bufq = (struct bufq_default *)bq;
	struct buf *bp, *head;
	int i;

	for (i = 0; i < 3; i++) {
		head = &bufq->bufq_head[i];
		if ((bp = head->b_actf))
			break;
	}
	if (bp == NULL)
		return (NULL);
	head->b_actf = bp->b_actf;
	return (bp);
}

#ifdef RAMDISK_HOOKS
static struct device fakerdrootdev = { DV_DISK, {}, NULL, 0, "rd0", NULL };
#endif

struct device *
getdisk(char *str, int len, int defpart, dev_t *devp)
{
	struct device *dv;

	if ((dv = parsedisk(str, len, defpart, devp)) == NULL) {
		printf("use one of: exit");
#ifdef RAMDISK_HOOKS
		printf(" %s[a-p]", fakerdrootdev.dv_xname);
#endif
		TAILQ_FOREACH(dv, &alldevs, dv_list) {
			if (dv->dv_class == DV_DISK)
				printf(" %s[a-p]", dv->dv_xname);
#if defined(NFSCLIENT)
			if (dv->dv_class == DV_IFNET)
				printf(" %s", dv->dv_xname);
#endif
		}
		printf("\n");
	}
	return (dv);
}

struct device *
parsedisk(char *str, int len, int defpart, dev_t *devp)
{
	struct device *dv;
	char c;
	int majdev, part;

	if (len == 0)
		return (NULL);
	c = str[len-1];
	if (c >= 'a' && (c - 'a') < MAXPARTITIONS) {
		part = c - 'a';
		len -= 1;
	} else
		part = defpart;

#ifdef RAMDISK_HOOKS
	if (strcmp(str, fakerdrootdev.dv_xname) == 0) {
		dv = &fakerdrootdev;
		goto gotdisk;
	}
#endif

	TAILQ_FOREACH(dv, &alldevs, dv_list) {
		if (dv->dv_class == DV_DISK &&
		    strncmp(str, dv->dv_xname, len) == 0 &&
		    dv->dv_xname[len] == '\0') {
#ifdef RAMDISK_HOOKS
gotdisk:
#endif
			majdev = findblkmajor(dv);
			if (majdev < 0)
				panic("parsedisk");
			*devp = MAKEDISKDEV(majdev, dv->dv_unit, part);
			break;
		}
#if defined(NFSCLIENT)
		if (dv->dv_class == DV_IFNET &&
		    strncmp(str, dv->dv_xname, len) == 0 &&
		    dv->dv_xname[len] == '\0') {
			*devp = NODEV;
			break;
		}
#endif
	}

	return (dv);
}

void
setroot(struct device *bootdv, int part, int exitflags)
{
	int majdev, unit, len, s;
	struct swdevt *swp;
	struct device *rootdv, *dv;
	dev_t nrootdev, nswapdev = NODEV, temp = NODEV;
	char buf[128];
#if defined(NFSCLIENT)
	extern char *nfsbootdevname;
#endif

	if (boothowto & RB_DFLTROOT)
		return;

#ifdef RAMDISK_HOOKS
	bootdv = &fakerdrootdev;
	mountroot = NULL;
	part = 0;
#endif

	/*
	 * If `swap generic' and we couldn't determine boot device,
	 * ask the user.
	 */
	if (mountroot == NULL && bootdv == NULL)
		boothowto |= RB_ASKNAME;
	if (boothowto & RB_ASKNAME) {
		while (1) {
			printf("root device");
			if (bootdv != NULL) {
				printf(" (default %s", bootdv->dv_xname);
				if (bootdv->dv_class == DV_DISK)
					printf("%c", 'a' + part);
				printf(")");
			}
			printf(": ");
			s = splhigh();
			cnpollc(TRUE);
			len = getsn(buf, sizeof(buf));
			cnpollc(FALSE);
			splx(s);
			if (strcmp(buf, "exit") == 0)
				boot(exitflags);
			if (len == 0 && bootdv != NULL) {
				strlcpy(buf, bootdv->dv_xname, sizeof buf);
				len = strlen(buf);
			}
			if (len > 0 && buf[len - 1] == '*') {
				buf[--len] = '\0';
				dv = getdisk(buf, len, part, &nrootdev);
				if (dv != NULL) {
					rootdv = dv;
					nswapdev = nrootdev;
					goto gotswap;
				}
			}
			dv = getdisk(buf, len, part, &nrootdev);
			if (dv != NULL) {
				rootdv = dv;
				break;
			}
		}

		if (rootdv->dv_class == DV_IFNET)
			goto gotswap;

		/* try to build swap device out of new root device */
		while (1) {
			printf("swap device");
			if (rootdv != NULL)
				printf(" (default %s%s)", rootdv->dv_xname,
				    rootdv->dv_class == DV_DISK ? "b" : "");
			printf(": ");
			s = splhigh();
			cnpollc(TRUE);
			len = getsn(buf, sizeof(buf));
			cnpollc(FALSE);
			splx(s);
			if (strcmp(buf, "exit") == 0)
				boot(exitflags);
			if (len == 0 && rootdv != NULL) {
				switch (rootdv->dv_class) {
				case DV_IFNET:
					nswapdev = NODEV;
					break;
				case DV_DISK:
					nswapdev = MAKEDISKDEV(major(nrootdev),
					    DISKUNIT(nrootdev), 1);
					if (nswapdev == nrootdev)
						continue;
					break;
				default:
					break;
				}
				break;
			}
			dv = getdisk(buf, len, 1, &nswapdev);
			if (dv) {
				if (dv->dv_class == DV_IFNET)
					nswapdev = NODEV;
				if (nswapdev == nrootdev)
					continue;
				break;
			}
		}
gotswap:
		rootdev = nrootdev;
		dumpdev = nswapdev;
		swdevt[0].sw_dev = nswapdev;
		swdevt[1].sw_dev = NODEV;
#if defined(NFSCLIENT)
	} else if (mountroot == nfs_mountroot) {
		rootdv = bootdv;
		rootdev = dumpdev = swapdev = NODEV;
#endif
	} else if (mountroot == NULL) {
		/* `swap generic': Use the device the ROM told us to use */
		rootdv = bootdv;
		majdev = findblkmajor(rootdv);
		if (majdev >= 0) {
			/*
			 * Root and swap are on the disk.
			 * Assume swap is on partition b.
			 */
			rootdev = MAKEDISKDEV(majdev, rootdv->dv_unit, part);
			nswapdev = MAKEDISKDEV(majdev, rootdv->dv_unit, 1);
		} else {
			/*
			 * Root and swap are on a net.
			 */
			nswapdev = NODEV;
		}
		dumpdev = nswapdev;
		swdevt[0].sw_dev = nswapdev;
		/* swdevt[1].sw_dev = NODEV; */
	} else {
		/* Completely pre-configured, but we want rootdv .. */
		majdev = major(rootdev);
		if (findblkname(majdev) == NULL)
			return;
		unit = DISKUNIT(rootdev);
		part = DISKPART(rootdev);
		snprintf(buf, sizeof buf, "%s%d%c",
		    findblkname(majdev), unit, 'a' + part);
		rootdv = parsedisk(buf, strlen(buf), 0, &nrootdev);
	}

	switch (rootdv->dv_class) {
#if defined(NFSCLIENT)
	case DV_IFNET:
		mountroot = nfs_mountroot;
		nfsbootdevname = rootdv->dv_xname;
		return;
#endif
	case DV_DISK:
		mountroot = dk_mountroot;
		part = DISKPART(rootdev);
		break;
	default:
		printf("can't figure root, hope your kernel is right\n");
		return;
	}

	printf("root on %s%c", rootdv->dv_xname, 'a' + part);

	/*
	 * Make the swap partition on the root drive the primary swap.
	 */
	for (swp = swdevt; swp->sw_dev != NODEV; swp++) {
		if (major(rootdev) == major(swp->sw_dev) &&
		    DISKUNIT(rootdev) == DISKUNIT(swp->sw_dev)) {
			temp = swdevt[0].sw_dev;
			swdevt[0].sw_dev = swp->sw_dev;
			swp->sw_dev = temp;
			break;
		}
	}
	if (swp->sw_dev != NODEV) {
		/*
		 * If dumpdev was the same as the old primary swap device,
		 * move it to the new primary swap device.
		 */
		if (temp == dumpdev)
			dumpdev = swdevt[0].sw_dev;
	}
	if (swdevt[0].sw_dev != NODEV)
		printf(" swap on %s%d%c", findblkname(major(swdevt[0].sw_dev)),
		    DISKUNIT(swdevt[0].sw_dev),
		    'a' + DISKPART(swdevt[0].sw_dev));
	if (dumpdev != NODEV)
		printf(" dump on %s%d%c", findblkname(major(dumpdev)),
		    DISKUNIT(dumpdev), 'a' + DISKPART(dumpdev));
	printf("\n");
}

extern struct nam2blk nam2blk[];

int
findblkmajor(struct device *dv)
{
	char *name = dv->dv_xname;
	int i;

	for (i = 0; nam2blk[i].name; i++)
		if (!strncmp(name, nam2blk[i].name, strlen(nam2blk[i].name)))
			return (nam2blk[i].maj);
	return (-1);
}

char *
findblkname(int maj)
{
	int i;

	for (i = 0; nam2blk[i].name; i++)
		if (nam2blk[i].maj == maj)
			return (nam2blk[i].name);
	return (NULL);
}