forked from neilbrown/mdadm
-
Notifications
You must be signed in to change notification settings - Fork 0
/
mdadm.h
1569 lines (1426 loc) · 52.8 KB
/
mdadm.h
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
/*
* mdadm - manage Linux "md" devices aka RAID arrays.
*
* Copyright (C) 2001-2009 Neil Brown <neilb@suse.de>
*
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
* Author: Neil Brown
* Email: <neilb@suse.de>
*/
#define _GNU_SOURCE
#define _FILE_OFFSET_BITS 64
#include <unistd.h>
#ifdef __GLIBC__
extern __off64_t lseek64 __P ((int __fd, __off64_t __offset, int __whence));
#elif !defined(lseek64)
# if defined(__NO_STAT64) || __WORDSIZE != 32
# define lseek64 lseek
# endif
#endif
#include <sys/types.h>
#include <sys/stat.h>
#include <stdlib.h>
#include <time.h>
#include <sys/time.h>
#include <getopt.h>
#include <fcntl.h>
#include <stdio.h>
#include <errno.h>
#include <string.h>
#include <syslog.h>
#ifdef __dietlibc__
#include <strings.h>
/* dietlibc has deprecated random and srandom!! */
#define random rand
#define srandom srand
#endif
#include <linux/kdev_t.h>
/*#include <linux/fs.h> */
#include <sys/mount.h>
#include <asm/types.h>
#include <sys/ioctl.h>
#define MD_MAJOR 9
#define MdpMinorShift 6
#ifndef BLKGETSIZE64
#define BLKGETSIZE64 _IOR(0x12,114,size_t) /* return device size in bytes (u64 *arg) */
#endif
#define DEFAULT_CHUNK 512
#define DEFAULT_BITMAP_CHUNK 4096
#define DEFAULT_BITMAP_DELAY 5
#define DEFAULT_MAX_WRITE_BEHIND 256
/* MAP_DIR should be somewhere that persists across the pivotroot
* from early boot to late boot.
* /run seems to have emerged as the best standard.
*/
#ifndef MAP_DIR
#define MAP_DIR "/run/mdadm"
#endif /* MAP_DIR */
/* MAP_FILE is what we name the map file we put in MAP_DIR, in case you
* want something other than the default of "map"
*/
#ifndef MAP_FILE
#define MAP_FILE "map"
#endif /* MAP_FILE */
/* MDMON_DIR is where pid and socket files used for communicating
* with mdmon normally live. Best is /var/run/mdadm as
* mdmon is needed at early boot then it needs to write there prior
* to /var/run being mounted read/write, and it also then needs to
* persist beyond when /var/run is mounter read-only. So, to be
* safe, the default is somewhere that is read/write early in the
* boot process and stays up as long as possible during shutdown.
*/
#ifndef MDMON_DIR
#define MDMON_DIR "/run/mdadm"
#endif /* MDMON_DIR */
/* FAILED_SLOTS is where to save files storing recent removal of array
* member in order to allow future reuse of disk inserted in the same
* slot for array recovery
*/
#ifndef FAILED_SLOTS_DIR
#define FAILED_SLOTS_DIR "/run/mdadm/failed-slots"
#endif /* FAILED_SLOTS */
#include "md_u.h"
#include "md_p.h"
#include "bitmap.h"
#include "msg.h"
#include <endian.h>
/* Redhat don't like to #include <asm/byteorder.h>, and
* some time include <linux/byteorder/xxx_endian.h> isn't enough,
* and there is no standard conversion function so... */
/* And dietlibc doesn't think byteswap is ok, so.. */
/* #include <byteswap.h> */
#define bswap_16(x) (((x) & 0x00ffU) << 8 | \
((x) & 0xff00U) >> 8)
#define bswap_32(x) (((x) & 0x000000ffU) << 24 | \
((x) & 0xff000000U) >> 24 | \
((x) & 0x0000ff00U) << 8 | \
((x) & 0x00ff0000U) >> 8)
#define bswap_64(x) (((x) & 0x00000000000000ffULL) << 56 | \
((x) & 0xff00000000000000ULL) >> 56 | \
((x) & 0x000000000000ff00ULL) << 40 | \
((x) & 0x00ff000000000000ULL) >> 40 | \
((x) & 0x0000000000ff0000ULL) << 24 | \
((x) & 0x0000ff0000000000ULL) >> 24 | \
((x) & 0x00000000ff000000ULL) << 8 | \
((x) & 0x000000ff00000000ULL) >> 8)
#if !defined(__KLIBC__)
#if BYTE_ORDER == LITTLE_ENDIAN
#define __cpu_to_le16(_x) (unsigned int)(_x)
#define __cpu_to_le32(_x) (unsigned int)(_x)
#define __cpu_to_le64(_x) (unsigned long long)(_x)
#define __le16_to_cpu(_x) (unsigned int)(_x)
#define __le32_to_cpu(_x) (unsigned int)(_x)
#define __le64_to_cpu(_x) (unsigned long long)(_x)
#define __cpu_to_be16(_x) bswap_16(_x)
#define __cpu_to_be32(_x) bswap_32(_x)
#define __cpu_to_be64(_x) bswap_64(_x)
#define __be16_to_cpu(_x) bswap_16(_x)
#define __be32_to_cpu(_x) bswap_32(_x)
#define __be64_to_cpu(_x) bswap_64(_x)
#elif BYTE_ORDER == BIG_ENDIAN
#define __cpu_to_le16(_x) bswap_16(_x)
#define __cpu_to_le32(_x) bswap_32(_x)
#define __cpu_to_le64(_x) bswap_64(_x)
#define __le16_to_cpu(_x) bswap_16(_x)
#define __le32_to_cpu(_x) bswap_32(_x)
#define __le64_to_cpu(_x) bswap_64(_x)
#define __cpu_to_be16(_x) (unsigned int)(_x)
#define __cpu_to_be32(_x) (unsigned int)(_x)
#define __cpu_to_be64(_x) (unsigned long long)(_x)
#define __be16_to_cpu(_x) (unsigned int)(_x)
#define __be32_to_cpu(_x) (unsigned int)(_x)
#define __be64_to_cpu(_x) (unsigned long long)(_x)
#else
# error "unknown endianness."
#endif
#endif /* __KLIBC__ */
/*
* min()/max()/clamp() macros that also do
* strict type-checking.. See the
* "unnecessary" pointer comparison.
*/
#define min(x, y) ({ \
typeof(x) _min1 = (x); \
typeof(y) _min2 = (y); \
(void) (&_min1 == &_min2); \
_min1 < _min2 ? _min1 : _min2; })
#define max(x, y) ({ \
typeof(x) _max1 = (x); \
typeof(y) _max2 = (y); \
(void) (&_max1 == &_max2); \
_max1 > _max2 ? _max1 : _max2; })
#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
/* general information that might be extracted from a superblock */
struct mdinfo {
mdu_array_info_t array;
mdu_disk_info_t disk;
__u64 events;
int uuid[4];
char name[33];
unsigned long long data_offset;
unsigned long long new_data_offset;
unsigned long long component_size; /* same as array.size, except in
* sectors and up to 64bits.
*/
unsigned long long custom_array_size; /* size for non-default sized
* arrays (in sectors)
*/
#define NO_RESHAPE 0
#define VOLUME_RESHAPE 1
#define CONTAINER_RESHAPE 2
#define RESHAPE_NO_BACKUP 16 /* Mask 'or'ed in */
int reshape_active;
unsigned long long reshape_progress;
int recovery_blocked; /* for external metadata it
* indicates that there is
* reshape in progress in
* container,
* for native metadata it is
* reshape_active field mirror
*/
/* During reshape we can sometimes change the data_offset to avoid
* over-writing still-valid data. We need to know if there is space.
* So getinfo_super will fill in space_before and space_after in sectors.
* data_offset can be increased or decreased by this amount.
*/
unsigned long long space_before, space_after;
union {
unsigned long long resync_start; /* per-array resync position */
unsigned long long recovery_start; /* per-device rebuild position */
#define MaxSector (~0ULL) /* resync/recovery complete position */
};
long bitmap_offset; /* 0 == none, 1 == a file */
unsigned long safe_mode_delay; /* ms delay to mark clean */
int new_level, delta_disks, new_layout, new_chunk;
int errors;
unsigned long cache_size; /* size of raid456 stripe cache*/
int mismatch_cnt;
char text_version[50];
int container_member; /* for assembling external-metatdata arrays
* This is to be used internally by metadata
* handler only */
int container_enough; /* flag external handlers can set to
* indicate that subarrays have not enough (-1),
* enough to start (0), or all expected disks (1) */
char sys_name[20];
struct mdinfo *devs;
struct mdinfo *next;
/* Device info for mdmon: */
int recovery_fd;
int state_fd;
#define DS_FAULTY 1
#define DS_INSYNC 2
#define DS_WRITE_MOSTLY 4
#define DS_SPARE 8
#define DS_BLOCKED 16
#define DS_REMOVE 1024
#define DS_UNBLOCK 2048
int prev_state, curr_state, next_state;
};
struct createinfo {
int uid;
int gid;
int autof;
int mode;
int symlinks;
int names;
struct supertype *supertype;
};
#define Name "mdadm"
enum mode {
ASSEMBLE=1,
BUILD,
CREATE,
MANAGE,
MISC,
MONITOR,
GROW,
INCREMENTAL,
AUTODETECT,
mode_count
};
extern char short_options[];
extern char short_bitmap_options[];
extern char short_bitmap_auto_options[];
extern struct option long_options[];
extern char Version[], Usage[], Help[], OptionHelp[],
*mode_help[],
Help_create[], Help_build[], Help_assemble[], Help_grow[],
Help_incr[],
Help_manage[], Help_misc[], Help_monitor[], Help_config[];
/* for option that don't have short equivilents, we assign arbitrary
* numbers later than any 'short' character option.
*/
enum special_options {
AssumeClean = 300,
BitmapChunk,
WriteBehind,
ReAdd,
NoDegraded,
Sparc22,
BackupFile,
HomeHost,
AutoHomeHost,
Symlinks,
AutoDetect,
Waitclean,
DetailPlatform,
KillSubarray,
UpdateSubarray,
IncrementalPath,
NoSharing,
HelpOptions,
Brief,
ManageOpt,
Add,
AddSpare,
Remove,
Fail,
Replace,
With,
MiscOpt,
WaitOpt,
ConfigFile,
ChunkSize,
WriteMostly,
Layout,
Auto,
Force,
SuperMinor,
EMail,
ProgramOpt,
Increment,
Fork,
Bitmap,
RebuildMapOpt,
InvalidBackup,
UdevRules,
FreezeReshape,
Continue,
OffRootOpt,
Prefer,
KillOpt,
DataOffset,
ExamineBB,
Dump,
Restore,
Action,
};
enum prefix_standard {
JEDEC,
IEC
};
/* structures read from config file */
/* List of mddevice names and identifiers
* Identifiers can be:
* uuid=128-hex-uuid
* super-minor=decimal-minor-number-from-superblock
* devices=comma,separated,list,of,device,names,with,wildcards
*
* If multiple fields are present, the intersection of all matching
* devices is considered
*/
#define UnSet (0xfffe)
struct mddev_ident {
char *devname;
int uuid_set;
int uuid[4];
char name[33];
int super_minor;
char *devices; /* comma separated list of device
* names with wild cards
*/
int level;
int raid_disks;
int spare_disks;
struct supertype *st;
int autof; /* 1 for normal, 2 for partitioned */
char *spare_group;
char *bitmap_file;
int bitmap_fd;
char *container; /* /dev/whatever name of container, or
* uuid of container. You would expect
* this to be the 'devname' or UUID
* of some other entry.
*/
char *member; /* subarray within a container */
struct mddev_ident *next;
union {
/* fields needed by different users of this structure */
int assembled; /* set when assembly succeeds */
};
};
struct context {
int readonly;
int runstop;
int verbose;
int brief;
int force;
char *homehost;
int require_homehost;
char *prefer;
int export;
int test;
char *subarray;
char *update;
int scan;
int SparcAdjust;
int autof;
int delay;
int freeze_reshape;
char *backup_file;
int invalid_backup;
char *action;
};
struct shape {
int raiddisks;
int sparedisks;
int level;
int layout;
char *layout_str;
int chunk;
int bitmap_chunk;
char *bitmap_file;
int assume_clean;
int write_behind;
unsigned long long size;
};
/* List of device names - wildcards expanded */
struct mddev_dev {
char *devname;
int disposition; /* 'a' for add, 'r' for remove, 'f' for fail,
* 'A' for re_add.
* Not set for names read from .config
*/
char writemostly; /* 1 for 'set writemostly', 2 for 'clear writemostly' */
int used; /* set when used */
long long data_offset;
struct mddev_dev *next;
};
typedef struct mapping {
char *name;
int num;
} mapping_t;
struct mdstat_ent {
char *dev;
char devnm[32];
int active;
char *level;
char *pattern; /* U or up, _ for down */
int percent; /* -1 if no resync */
int resync; /* 3 if check, 2 if reshape, 1 if resync, 0 if recovery */
int devcnt;
int raid_disks;
char * metadata_version;
struct dev_member {
char *name;
struct dev_member *next;
} *members;
struct mdstat_ent *next;
};
extern struct mdstat_ent *mdstat_read(int hold, int start);
extern void mdstat_close(void);
extern void free_mdstat(struct mdstat_ent *ms);
extern void mdstat_wait(int seconds);
extern void mdstat_wait_fd(int fd, const sigset_t *sigmask);
extern int mddev_busy(char *devnm);
extern struct mdstat_ent *mdstat_by_component(char *name);
extern struct mdstat_ent *mdstat_by_subdev(char *subdev, char *container);
struct map_ent {
struct map_ent *next;
char devnm[32];
char metadata[20];
int uuid[4];
int bad;
char *path;
};
extern int map_update(struct map_ent **mpp, char *devnm, char *metadata,
int uuid[4], char *path);
extern void map_remove(struct map_ent **map, char *devnm);
extern struct map_ent *map_by_uuid(struct map_ent **map, int uuid[4]);
#ifdef MDASSEMBLE
static inline struct map_ent *map_by_devnm(struct map_ent **map, char *name)
{
return NULL;
}
static inline void map_free(struct map_ent *map)
{
}
#else
extern struct map_ent *map_by_devnm(struct map_ent **map, char *devnm);
extern void map_free(struct map_ent *map);
#endif
extern struct map_ent *map_by_name(struct map_ent **map, char *name);
extern void map_read(struct map_ent **melp);
extern int map_write(struct map_ent *mel);
extern void map_delete(struct map_ent **mapp, char *devnm);
extern void map_add(struct map_ent **melp,
char *devnm, char *metadata, int uuid[4], char *path);
extern int map_lock(struct map_ent **melp);
extern void map_unlock(struct map_ent **melp);
extern void map_fork(void);
/* various details can be requested */
enum sysfs_read_flags {
GET_LEVEL = (1 << 0),
GET_LAYOUT = (1 << 1),
GET_COMPONENT = (1 << 2),
GET_CHUNK = (1 << 3),
GET_CACHE = (1 << 4),
GET_MISMATCH = (1 << 5),
GET_VERSION = (1 << 6),
GET_DISKS = (1 << 7),
GET_DEGRADED = (1 << 8),
GET_SAFEMODE = (1 << 9),
GET_BITMAP_LOCATION = (1 << 10),
GET_DEVS = (1 << 20), /* gets role, major, minor */
GET_OFFSET = (1 << 21),
GET_SIZE = (1 << 22),
GET_STATE = (1 << 23),
GET_ERROR = (1 << 24),
};
/* If fd >= 0, get the array it is open on,
* else use devnm.
*/
extern int sysfs_open(char *devnm, char *devname, char *attr);
extern void sysfs_init(struct mdinfo *mdi, int fd, char *devnm);
extern void sysfs_free(struct mdinfo *sra);
extern struct mdinfo *sysfs_read(int fd, char *devnm, unsigned long options);
extern int sysfs_attr_match(const char *attr, const char *str);
extern int sysfs_match_word(const char *word, char **list);
extern int sysfs_set_str(struct mdinfo *sra, struct mdinfo *dev,
char *name, char *val);
extern int sysfs_set_num(struct mdinfo *sra, struct mdinfo *dev,
char *name, unsigned long long val);
extern int sysfs_set_num_signed(struct mdinfo *sra, struct mdinfo *dev,
char *name, long long val);
extern int sysfs_uevent(struct mdinfo *sra, char *event);
extern int sysfs_get_fd(struct mdinfo *sra, struct mdinfo *dev,
char *name);
extern int sysfs_fd_get_ll(int fd, unsigned long long *val);
extern int sysfs_get_ll(struct mdinfo *sra, struct mdinfo *dev,
char *name, unsigned long long *val);
extern int sysfs_fd_get_two(int fd, unsigned long long *v1, unsigned long long *v2);
extern int sysfs_get_two(struct mdinfo *sra, struct mdinfo *dev,
char *name, unsigned long long *v1, unsigned long long *v2);
extern int sysfs_fd_get_str(int fd, char *val, int size);
extern int sysfs_attribute_available(struct mdinfo *sra, struct mdinfo *dev,
char *name);
extern int sysfs_get_str(struct mdinfo *sra, struct mdinfo *dev,
char *name, char *val, int size);
extern int sysfs_set_safemode(struct mdinfo *sra, unsigned long ms);
extern int sysfs_set_array(struct mdinfo *info, int vers);
extern int sysfs_add_disk(struct mdinfo *sra, struct mdinfo *sd, int resume);
extern int sysfs_disk_to_scsi_id(int fd, __u32 *id);
extern int sysfs_unique_holder(char *devnm, long rdev);
extern int sysfs_freeze_array(struct mdinfo *sra);
extern int sysfs_wait(int fd, int *msec);
extern int load_sys(char *path, char *buf);
extern int reshape_prepare_fdlist(char *devname,
struct mdinfo *sra,
int raid_disks,
int nrdisks,
unsigned long blocks,
char *backup_file,
int *fdlist,
unsigned long long *offsets);
extern void reshape_free_fdlist(int *fdlist,
unsigned long long *offsets,
int size);
extern int reshape_open_backup_file(char *backup,
int fd,
char *devname,
long blocks,
int *fdlist,
unsigned long long *offsets,
char *sysfs_name,
int restart);
extern unsigned long compute_backup_blocks(int nchunk, int ochunk,
unsigned int ndata, unsigned int odata);
extern char *locate_backup(char *name);
extern char *make_backup(char *name);
extern int save_stripes(int *source, unsigned long long *offsets,
int raid_disks, int chunk_size, int level, int layout,
int nwrites, int *dest,
unsigned long long start, unsigned long long length,
char *buf);
extern int restore_stripes(int *dest, unsigned long long *offsets,
int raid_disks, int chunk_size, int level, int layout,
int source, unsigned long long read_offset,
unsigned long long start, unsigned long long length,
char *src_buf);
#ifndef Sendmail
#define Sendmail "/usr/lib/sendmail -t"
#endif
#define SYSLOG_FACILITY LOG_DAEMON
extern char *map_num(mapping_t *map, int num);
extern int map_name(mapping_t *map, char *name);
extern mapping_t r5layout[], r6layout[], pers[], modes[], faultylayout[];
extern char *map_dev_preferred(int major, int minor, int create,
char *prefer);
static inline char *map_dev(int major, int minor, int create)
{
return map_dev_preferred(major, minor, create, NULL);
}
struct active_array;
struct metadata_update;
/* 'struct reshape' records the intermediate states of
* a general reshape.
* The starting geometry is converted to the 'before' geometry
* by at most an atomic level change. They could be the same.
* Similarly the 'after' geometry is converted to the final
* geometry by at most a level change.
* Note that 'before' and 'after' must have the same level.
* 'blocks' is the minimum number of sectors for a reshape unit.
* This will be a multiple of the stripe size in each of the
* 'before' and 'after' geometries.
* If 'blocks' is 0, no restriping is necessary.
* 'min_offset_change' is the minimum change to data_offset to
* allow the reshape to happen. It is at least the larger of
* the old and new chunk sizes, and typically the same as 'blocks'
* divided by number of data disks.
*/
struct reshape {
int level;
int parity; /* number of parity blocks/devices */
struct {
int layout;
int data_disks;
} before, after;
unsigned long long backup_blocks;
unsigned long long min_offset_change;
unsigned long long stripes; /* number of old stripes that comprise 'blocks'*/
unsigned long long new_size; /* New size of array in sectors */
};
/* A superswitch provides entry point the a metadata handler.
*
* The superswitch primarily operates on some "metadata" that
* is accessed via the 'supertype'.
* This metadata has one of three possible sources.
* 1/ It is read from a single device. In this case it may not completely
* describe the array or arrays as some information might be on other
* devices.
* 2/ It is read from all devices in a container. In this case all
* information is present.
* 3/ It is created by ->init_super / ->add_to_super. In this case it will
* be complete once enough ->add_to_super calls have completed.
*
* When creating an array inside a container, the metadata will be
* formed by a combination of 2 and 3. The metadata or the array is read,
* then new information is added.
*
* The metadata must sometimes have a concept of a 'current' array
* and a 'current' device.
* The 'current' array is set by init_super to be the newly created array,
* or is set by super_by_fd when it finds it is looking at an array inside
* a container.
*
* The 'current' device is either the device that the metadata was read from
* in case 1, or the last device added by add_to_super in case 3.
* Case 2 does not identify a 'current' device.
*/
extern struct superswitch {
/* Used to report details of metadata read from a component
* device. ->load_super has been called.
*/
void (*examine_super)(struct supertype *st, char *homehost);
void (*brief_examine_super)(struct supertype *st, int verbose);
void (*brief_examine_subarrays)(struct supertype *st, int verbose);
void (*export_examine_super)(struct supertype *st);
int (*examine_badblocks)(struct supertype *st, int fd, char *devname);
int (*copy_metadata)(struct supertype *st, int from, int to);
/* Used to report details of an active array.
* ->load_super was possibly given a 'component' string.
*/
void (*detail_super)(struct supertype *st, char *homehost);
void (*brief_detail_super)(struct supertype *st);
void (*export_detail_super)(struct supertype *st);
/* Optional: platform hardware / firmware details */
int (*detail_platform)(int verbose, int enumerate_only, char *controller_path);
int (*export_detail_platform)(int verbose, char *controller_path);
/* Used:
* to get uuid to storing in bitmap metadata
* and 'reshape' backup-data metadata
* To see if a device is being re-added to an array it was part of.
*/
void (*uuid_from_super)(struct supertype *st, int uuid[4]);
/* Extract generic details from metadata. This could be details about
* the container, or about an individual array within the container.
* The determination is made either by:
* load_super being given a 'component' string.
* validate_geometry determining what to create.
* The info includes both array information and device information.
* The particular device should be:
* The last device added by add_to_super
* The device the metadata was loaded from by load_super
* If 'map' is present, then it is an array raid_disks long
* (raid_disk must already be set and correct) and it is filled
* with 1 for slots that are thought to be active and 0 for slots which
* appear to be failed/missing.
* *info is zeroed out before data is added.
*/
void (*getinfo_super)(struct supertype *st, struct mdinfo *info, char *map);
struct mdinfo *(*getinfo_super_disks)(struct supertype *st);
/* Check if the given metadata is flagged as belonging to "this"
* host. 0 for 'no', 1 for 'yes', -1 for "Don't record homehost"
*/
int (*match_home)(struct supertype *st, char *homehost);
/* Make one of several generic modifications to metadata
* prior to assembly (or other times).
* sparc2.2 - first bug in early 0.90 metadata
* super-minor - change name of 0.90 metadata
* summaries - 'correct' any redundant data
* resync - mark array as dirty to trigger a resync.
* uuid - set new uuid - only 0.90 or 1.x
* name - change the name of the array (where supported)
* homehost - change which host this array is tied to.
* devicesize - If metadata is at start of device, change recorded
* device size to match actual device size
* byteorder - swap bytes for 0.90 metadata
*
* force-one - mark that device as uptodate, not old or failed.
* force-array - mark array as clean if it would not otherwise
* assemble
* assemble - not sure how this is different from force-one...
* linear-grow-new - add a new device to a linear array, but don't
* change the size: so superblock still matches
* linear-grow-update - now change the size of the array.
* writemostly - set the WriteMostly1 bit in the superblock devflags
* readwrite - clear the WriteMostly1 bit in the superblock devflags
* no-bitmap - clear any record that a bitmap is present.
* bbl - add a bad-block-log if possible
* no-bbl - remove and bad-block-log is it is empty.
* revert-reshape - If a reshape is in progress, modify metadata so
* it will resume going in the opposite direction.
*/
int (*update_super)(struct supertype *st, struct mdinfo *info,
char *update,
char *devname, int verbose,
int uuid_set, char *homehost);
/* Create new metadata for new array as described. This could
* be a new container, or an array in a pre-existing container.
* Also used to zero metadata prior to writing it to invalidate old
* metadata.
*/
int (*init_super)(struct supertype *st, mdu_array_info_t *info,
unsigned long long size, char *name,
char *homehost, int *uuid,
unsigned long long data_offset);
/* update the metadata to include new device, either at create or
* when hot-adding a spare.
*/
int (*add_to_super)(struct supertype *st, mdu_disk_info_t *dinfo,
int fd, char *devname,
unsigned long long data_offset);
/* update the metadata to delete a device,
* when hot-removing.
*/
int (*remove_from_super)(struct supertype *st, mdu_disk_info_t *dinfo);
/* Write metadata to one device when fixing problems or adding
* a new device.
*/
int (*store_super)(struct supertype *st, int fd);
/* Write all metadata for this array.
*/
int (*write_init_super)(struct supertype *st);
/* Check if metadata read from one device is compatible with an array,
* used when assembling an array, or pseudo-assembling was with
* "--examine --brief"
* If "st" has not yet been loaded the superblock from, "tst" is
* moved in, otherwise the superblock in 'st' is compared with
* 'tst'.
*/
int (*compare_super)(struct supertype *st, struct supertype *tst);
/* Load metadata from a single device. If 'devname' is not NULL
* print error messages as appropriate */
int (*load_super)(struct supertype *st, int fd, char *devname);
/* 'fd' is a 'container' md array - load array metadata from the
* whole container.
*/
int (*load_container)(struct supertype *st, int fd, char *devname);
/* If 'arg' is a valid name of this metadata type, allocate and
* return a 'supertype' for the particular minor version */
struct supertype * (*match_metadata_desc)(char *arg);
/* If a device has the given size, and the data_offset has been
* requested - work out how much space is available for data.
* This involves adjusting for reserved space (e.g. bitmaps)
* and for any rounding.
* 'mdadm' only calls this for existing arrays where a possible
* spare is being added. However some super-handlers call it
* internally from validate_geometry when creating an array.
*/
__u64 (*avail_size)(struct supertype *st, __u64 size,
unsigned long long data_offset);
/* This is similar to 'avail_size' in purpose, but is used for
* containers for which there is no 'component size' to compare.
* This reports that whole-device size which is a minimum
*/
unsigned long long (*min_acceptable_spare_size)(struct supertype *st);
/* Find somewhere to put a bitmap - possibly auto-size it - and
* update the metadata to record this. The array may be newly
* created, in which case data_size may be updated, or it might
* already exist. Metadata handler can know if init_super
* has been called, but not write_init_super.
*/
int (*add_internal_bitmap)(struct supertype *st, int *chunkp,
int delay, int write_behind,
unsigned long long size, int may_change, int major);
/* Seek 'fd' to start of write-intent-bitmap. Must be an
* md-native format bitmap
*/
void (*locate_bitmap)(struct supertype *st, int fd);
/* if add_internal_bitmap succeeded for existing array, this
* writes it out.
*/
int (*write_bitmap)(struct supertype *st, int fd);
/* Free the superblock and any other allocated data */
void (*free_super)(struct supertype *st);
/* validate_geometry is called with an st returned by
* match_metadata_desc.
* It should check that the geometry described is compatible with
* the metadata type. It will be called repeatedly as devices
* added to validate changing size and new devices. If there are
* inter-device dependencies, it should record sufficient details
* so these can be validated.
* Both 'size' and '*freesize' are in sectors. chunk is KiB.
* Return value is:
* 1: everything is OK
* 0: not OK for some reason - if 'verbose', then error was reported.
* -1: st->sb was NULL, 'subdev' is a member of a container of this
* type, but array is not acceptable for some reason
* message was reported even if verbose is 0.
*/
int (*validate_geometry)(struct supertype *st, int level, int layout,
int raiddisks,
int *chunk, unsigned long long size,
unsigned long long data_offset,
char *subdev, unsigned long long *freesize,
int verbose);
/* Return a linked list of 'mdinfo' structures for all arrays
* in the container. For non-containers, it is like
* getinfo_super with an allocated mdinfo.*/
struct mdinfo *(*container_content)(struct supertype *st, char *subarray);
/* query the supertype for default geometry */
void (*default_geometry)(struct supertype *st, int *level, int *layout, int *chunk); /* optional */
/* Permit subarray's to be deleted from inactive containers */
int (*kill_subarray)(struct supertype *st); /* optional */
/* Permit subarray's to be modified */
int (*update_subarray)(struct supertype *st, char *subarray,
char *update, struct mddev_ident *ident); /* optional */
/* Check if reshape is supported for this external format.
* st is obtained from super_by_fd() where st->subarray[0] is
* initialized to indicate if reshape is being performed at the
* container or subarray level
*/
#define APPLY_METADATA_CHANGES 1
#define ROLLBACK_METADATA_CHANGES 0
int (*reshape_super)(struct supertype *st,
unsigned long long size, int level,
int layout, int chunksize, int raid_disks,
int delta_disks, char *backup, char *dev,
int direction,
int verbose); /* optional */
int (*manage_reshape)( /* optional */
int afd, struct mdinfo *sra, struct reshape *reshape,
struct supertype *st, unsigned long blocks,
int *fds, unsigned long long *offsets,
int dests, int *destfd, unsigned long long *destoffsets);
/* for mdmon */
int (*open_new)(struct supertype *c, struct active_array *a,
char *inst);
/* Tell the metadata handler the current state of the array.
* This covers whether it is known to be consistent (no pending writes)
* and how far along a resync is known to have progressed
* (in a->resync_start).
* resync status is really irrelevant if the array is not consistent,
* but some metadata (DDF!) have a place to record the distinction.
* If 'consistent' is '2', then the array can mark it dirty if a
* resync/recovery/whatever is required, or leave it clean if not.
* Return value is 0 dirty (not consistent) and 1 if clean.
* it is only really important if consistent is passed in as '2'.
*/
int (*set_array_state)(struct active_array *a, int consistent);
/* When the state of a device might have changed, we call set_disk to
* tell the metadata what the current state is.
* Typically this happens on spare->in_sync and (spare|in_sync)->faulty
* transitions.
* set_disk might be called when the state of the particular disk has
* not in fact changed.
*/
void (*set_disk)(struct active_array *a, int n, int state);
void (*sync_metadata)(struct supertype *st);
void (*process_update)(struct supertype *st,
struct metadata_update *update);
void (*prepare_update)(struct supertype *st,
struct metadata_update *update);
/* activate_spare will check if the array is degraded and, if it
* is, try to find some spare space in the container.
* On success, it add appropriate updates (For process_update) to
* to the 'updates' list and returns a list of 'mdinfo' identifying
* the device, or devices as there might be multiple missing
* devices and multiple spares available.
*/
struct mdinfo *(*activate_spare)(struct active_array *a,
struct metadata_update **updates);
/*
* Return statically allocated string that represents metadata specific
* controller domain of the disk. The domain is used in disk domain
* matching functions. Disks belong to the same domain if the they have
* the same domain from mdadm.conf and belong the same metadata domain.
* Returning NULL or not providing this handler means that metadata
* does not distinguish the differences between disks that belong to
* different controllers. They are in the domain specified by
* configuration file (mdadm.conf).
* In case when the metadata has the notion of domains based on disk
* it shall return NULL for disks that do not belong to the controller
* the supported domains. Such disks will form another domain and won't
* be mixed with supported ones.
*/
const char *(*get_disk_controller_domain)(const char *path);
/* for external backup area */
int (*recover_backup)(struct supertype *st, struct mdinfo *info);
int swapuuid; /* true if uuid is bigending rather than hostendian */
int external;
const char *name; /* canonical metadata name */
} *superlist[];
extern struct superswitch super0, super1;
extern struct superswitch super_imsm, super_ddf;
extern struct superswitch mbr, gpt;
struct metadata_update {
int len;
char *buf;
void *space; /* allocated space that monitor will use */
void **space_list; /* list of allocated spaces that monitor can
* use or that it returned.
*/
struct metadata_update *next;
};
/* A supertype holds a particular collection of metadata.
* It identifies the metadata type by the superswitch, and the particular
* sub-version of that metadata type.
* metadata read in or created is stored in 'sb' and 'info'.
* There are also fields used by mdmon to track containers.
*
* A supertype may refer to:
* Just an array, possibly in a container
* A container, not identifying any particular array
* Info read from just one device, not yet fully describing the array/container.
*
*
* A supertype is created by:
* super_by_fd
* guess_super
* dup_super
*/
struct supertype {
struct superswitch *ss;