#define MPB_SECTOR_CNT 2210
#define IMSM_RESERVED_SECTORS 4096
+#define NUM_BLOCKS_DIRTY_STRIPE_REGION 2056
#define SECT_PER_MB_SHIFT 11
/* Disk configuration info. */
* 0 - we return the first map
* 1 - we return the second map if it exists, else NULL
* -1 - we return the second map if it exists, else the first
+ * -2 - we return longer map /excluding uninitialized state/
*/
struct imsm_map *map = &dev->vol.map[0];
+ struct imsm_map *map2 = NULL;
- if (second_map == 1 && !dev->vol.migr_state)
- return NULL;
- else if (second_map == 1 ||
- (second_map < 0 && dev->vol.migr_state)) {
- void *ptr = map;
+ if (dev->vol.migr_state)
+ map2 = (void *)map + sizeof_imsm_map(map);
- return ptr + sizeof_imsm_map(map);
- } else
- return map;
+ switch (second_map) {
+ case 0:
+ break;
+ case 1:
+ map = map2;
+ break;
+ case -1:
+ if (map2)
+ map = map2;
+ break;
+ case -2:
+ if (map2
+ && map2->map_state != IMSM_T_STATE_UNINITIALIZED
+ && map2->num_members > map->num_members)
+ map = map2;
+ break;
+ default:
+ map = NULL;
+ }
+ return map;
}
return memberships;
}
+static __u32 imsm_min_reserved_sectors(struct intel_super *super);
+
static struct extent *get_extents(struct intel_super *super, struct dl *dl)
{
/* find a list of used extents on the given physical device */
struct extent *rv, *e;
int i;
int memberships = count_memberships(dl, super);
- __u32 reservation = MPB_SECTOR_CNT + IMSM_RESERVED_SECTORS;
+ __u32 reservation;
+
+ /* trim the reserved area for spares, so they can join any array
+ * regardless of whether the OROM has assigned sectors from the
+ * IMSM_RESERVED_SECTORS region
+ */
+ if (dl->index == -1)
+ reservation = imsm_min_reserved_sectors(super);
+ else
+ reservation = MPB_SECTOR_CNT + IMSM_RESERVED_SECTORS;
rv = malloc(sizeof(struct extent) * (memberships + 1));
if (!rv)
return (disk->status & FAILED_DISK) == FAILED_DISK;
}
+/* try to determine how much space is reserved for metadata from
+ * the last get_extents() entry on the smallest active disk,
+ * otherwise fallback to the default
+ */
+static __u32 imsm_min_reserved_sectors(struct intel_super *super)
+{
+ struct extent *e;
+ int i;
+ __u32 min_active, remainder;
+ __u32 rv = MPB_SECTOR_CNT + IMSM_RESERVED_SECTORS;
+ struct dl *dl, *dl_min = NULL;
+
+ if (!super)
+ return rv;
+
+ min_active = 0;
+ for (dl = super->disks; dl; dl = dl->next) {
+ if (dl->index < 0)
+ continue;
+ if (dl->disk.total_blocks < min_active || min_active == 0) {
+ dl_min = dl;
+ min_active = dl->disk.total_blocks;
+ }
+ }
+ if (!dl_min)
+ return rv;
+
+ /* find last lba used by subarrays on the smallest active disk */
+ e = get_extents(super, dl_min);
+ if (!e)
+ return rv;
+ for (i = 0; e[i].size; i++)
+ continue;
+
+ remainder = min_active - e[i].start;
+ free(e);
+
+ /* to give priority to recovery we should not require full
+ IMSM_RESERVED_SECTORS from the spare */
+ rv = MPB_SECTOR_CNT + NUM_BLOCKS_DIRTY_STRIPE_REGION;
+
+ /* if real reservation is smaller use that value */
+ return (remainder < rv) ? remainder : rv;
+}
+
/* Return minimum size of a spare that can be used in this array*/
static unsigned long long min_acceptable_spare_size_imsm(struct supertype *st)
{
if (i > 0)
rv = e[i-1].start + e[i-1].size;
free(e);
+
/* add the amount of space needed for metadata */
- rv = rv + MPB_SECTOR_CNT + IMSM_RESERVED_SECTORS;
+ rv = rv + imsm_min_reserved_sectors(super);
+
return rv * 512;
}
static __u64 blocks_per_migr_unit(struct intel_super *super,
struct imsm_dev *dev);
+static int is_gen_migration(struct imsm_dev *dev);
+
static void print_imsm_dev(struct intel_super *super,
struct imsm_dev *dev,
char *uuid,
struct imsm_map *map = get_imsm_map(dev, 1);
printf(" <-- %s", map_state_str[map->map_state]);
- printf("\n Checkpoint : %u (%llu)",
- __le32_to_cpu(dev->vol.curr_migr_unit),
- (unsigned long long)blocks_per_migr_unit(super, dev));
+ printf("\n Checkpoint : %u ",
+ __le32_to_cpu(dev->vol.curr_migr_unit));
+ if ((is_gen_migration(dev)) && (super->disks->index > 1))
+ printf("(N/A)");
+ else
+ printf("(%llu)", (unsigned long long)
+ blocks_per_migr_unit(super, dev));
}
printf("\n");
printf(" Dirty State : %s\n", dev->vol.dirty ? "dirty" : "clean");
}
-static void print_imsm_disk(struct imsm_super *mpb, int index, __u32 reserved)
+static void print_imsm_disk(struct imsm_disk *disk, int index, __u32 reserved)
{
- struct imsm_disk *disk = __get_imsm_disk(mpb, index);
char str[MAX_RAID_SERIAL_LEN + 1];
__u64 sz;
- if (index < 0 || !disk)
+ if (index < -1 || !disk)
return;
printf("\n");
snprintf(str, MAX_RAID_SERIAL_LEN + 1, "%s", disk->serial);
- printf(" Disk%02d Serial : %s\n", index, str);
+ if (index >= 0)
+ printf(" Disk%02d Serial : %s\n", index, str);
+ else
+ printf(" Disk Serial : %s\n", str);
printf(" State :%s%s%s\n", is_spare(disk) ? " spare" : "",
is_configured(disk) ? " active" : "",
is_failed(disk) ? " failed" : "");
human_size(sz * 512));
}
-static int is_gen_migration(struct imsm_dev *dev);
-
void examine_migr_rec_imsm(struct intel_super *super)
{
struct migr_record *migr_rec = super->migr_rec;
printf(" MPB Sectors : %d\n", mpb_sectors(mpb));
printf(" Disks : %d\n", mpb->num_disks);
printf(" RAID Devices : %d\n", mpb->num_raid_devs);
- print_imsm_disk(mpb, super->disks->index, reserved);
+ print_imsm_disk(__get_imsm_disk(mpb, super->disks->index), super->disks->index, reserved);
if (super->bbm_log) {
struct bbm_log *log = super->bbm_log;
for (i = 0; i < mpb->num_disks; i++) {
if (i == super->disks->index)
continue;
- print_imsm_disk(mpb, i, reserved);
+ print_imsm_disk(__get_imsm_disk(mpb, i), i, reserved);
}
- for (dl = super->disks ; dl; dl = dl->next) {
- struct imsm_disk *disk;
- char str[MAX_RAID_SERIAL_LEN + 1];
- __u64 sz;
-
- if (dl->index >= 0)
- continue;
- disk = &dl->disk;
- printf("\n");
- snprintf(str, MAX_RAID_SERIAL_LEN + 1, "%s", disk->serial);
- printf(" Disk Serial : %s\n", str);
- printf(" State :%s%s%s\n", is_spare(disk) ? " spare" : "",
- is_configured(disk) ? " active" : "",
- is_failed(disk) ? " failed" : "");
- printf(" Id : %08x\n", __le32_to_cpu(disk->scsi_id));
- sz = __le32_to_cpu(disk->total_blocks) - reserved;
- printf(" Usable Size : %llu%s\n", (unsigned long long)sz,
- human_size(sz * 512));
- }
+ for (dl = super->disks; dl; dl = dl->next)
+ if (dl->index == -1)
+ print_imsm_disk(&dl->disk, -1, reserved);
examine_migr_rec_imsm(super);
}
printf(" (%.*s)\n", MAX_RAID_SERIAL_LEN, buf);
else
printf(" ()\n");
+ close(fd);
}
- close(fd);
free(path);
path = NULL;
}
}
#endif /* MDASSEMBLE */
+/* spare/missing disks activations are not allowe when
+ * array/container performs reshape operation, because
+ * all arrays in container works on the same disks set
+ */
+int imsm_reshape_blocks_arrays_changes(struct intel_super *super)
+{
+ int rv = 0;
+ struct intel_dev *i_dev;
+ struct imsm_dev *dev;
+
+ /* check whole container
+ */
+ for (i_dev = super->devlist; i_dev; i_dev = i_dev->next) {
+ dev = i_dev->dev;
+ if (is_gen_migration(dev)) {
+ /* No repair during any migration in container
+ */
+ rv = 1;
+ break;
+ }
+ }
+ return rv;
+}
+
static void getinfo_super_imsm_volume(struct supertype *st, struct mdinfo *info, char *dmap)
{
struct intel_super *super = st->sb;
info->custom_array_size = __le32_to_cpu(dev->size_high);
info->custom_array_size <<= 32;
info->custom_array_size |= __le32_to_cpu(dev->size_low);
- if (prev_map && map->map_state == prev_map->map_state) {
+ info->recovery_blocked = imsm_reshape_blocks_arrays_changes(st->sb);
+
+ if (is_gen_migration(dev)) {
info->reshape_active = 1;
info->new_level = get_imsm_raid_level(map);
info->new_layout = imsm_level_to_layout(info->new_level);
/* this needs to be applied to every array
* in the container.
*/
- info->reshape_active = 2;
+ info->reshape_active = CONTAINER_RESHAPE;
}
/* We shape information that we give to md might have to be
* modify to cope with md's requirement for reshaping arrays.
info->reshape_progress = 0;
info->resync_start = MaxSector;
- if (map_to_analyse->map_state == IMSM_T_STATE_UNINITIALIZED ||
- dev->vol.dirty) {
+ if ((map_to_analyse->map_state == IMSM_T_STATE_UNINITIALIZED ||
+ dev->vol.dirty) &&
+ imsm_reshape_blocks_arrays_changes(super) == 0) {
info->resync_start = 0;
}
if (dev->vol.migr_state) {
info->disk.state = 0;
info->name[0] = 0;
info->recovery_start = MaxSector;
+ info->recovery_blocked = imsm_reshape_blocks_arrays_changes(st->sb);
/* do we have the all the insync disks that we expect? */
mpb = super->anchor;
failed = imsm_count_failed(super, dev);
state = imsm_check_degraded(super, dev, failed);
- map = get_imsm_map(dev, dev->vol.migr_state);
+ map = get_imsm_map(dev, 0);
/* any newly missing disks?
* (catches single-degraded vs double-degraded)
*/
for (j = 0; j < map->num_members; j++) {
- __u32 ord = get_imsm_ord_tbl_ent(dev, i, -1);
+ __u32 ord = get_imsm_ord_tbl_ent(dev, j, 0);
__u32 idx = ord_to_idx(ord);
if (!(ord & IMSM_ORD_REBUILD) &&
sprintf(path, "/sys/dev/block/%d:%d",
major(st.st_rdev), minor(st.st_rdev));
- rv = readlink(path, dname, sizeof(dname));
+ rv = readlink(path, dname, sizeof(dname)-1);
if (rv <= 0)
return;
dname[rv] = '\0';
nm = strrchr(dname, '/');
- nm++;
- snprintf(name, MAX_RAID_SERIAL_LEN, "/dev/%s", nm);
+ if (nm) {
+ nm++;
+ snprintf(name, MAX_RAID_SERIAL_LEN, "/dev/%s", nm);
+ }
}
extern int scsi_get_serial(int fd, void *buf, size_t buf_len);
strncpy((char *) dest, (char *) src, MAX_RAID_SERIAL_LEN);
}
-#ifndef MDASSEMBLE
static struct dl *serial_to_dl(__u8 *serial, struct intel_super *super)
{
struct dl *dl;
return dl;
}
-#endif
static struct imsm_disk *
__serial_to_disk(__u8 *serial, struct imsm_super *mpb, int *idx)
}
dev->vol.migr_state = 0;
- dev->vol.migr_type = 0;
+ set_migr_type(dev, 0);
dev->vol.curr_migr_unit = 0;
map->map_state = map_state;
}
return 0;
}
-#ifndef MDASSEMBLE
/* find_missing - helper routine for load_super_imsm_all that identifies
* disks that have disappeared from the system. This routine relies on
* the mpb being uptodate, which it is at load time.
return 0;
}
+#ifndef MDASSEMBLE
static struct intel_disk *disk_list_get(__u8 *serial, struct intel_disk *disk_list)
{
struct intel_disk *idisk = disk_list;
__func__);
free(super->buf);
free(super);
+ free(mpb_new);
return 0;
}
memcpy(mpb_new, mpb, size_old);
memset(mpb_new + size_old, 0, size_round - size_old);
}
super->current_vol = idx;
- /* when creating the first raid device in this container set num_disks
- * to zero, i.e. delete this spare and add raid member devices in
- * add_to_super_imsm_volume()
+
+ /* handle 'failed_disks' by either:
+ * a) create dummy disk entries in the table if this the first
+ * volume in the array. We add them here as this is the only
+ * opportunity to add them. add_to_super_imsm_volume()
+ * handles the non-failed disks and continues incrementing
+ * mpb->num_disks.
+ * b) validate that 'failed_disks' matches the current number
+ * of missing disks if the container is populated
*/
- if (super->current_vol == 0)
+ if (super->current_vol == 0) {
mpb->num_disks = 0;
+ for (i = 0; i < info->failed_disks; i++) {
+ struct imsm_disk *disk;
+
+ mpb->num_disks++;
+ disk = __get_imsm_disk(mpb, i);
+ disk->status = CONFIGURED_DISK | FAILED_DISK;
+ disk->scsi_id = __cpu_to_le32(~(__u32)0);
+ snprintf((char *) disk->serial, MAX_RAID_SERIAL_LEN,
+ "missing:%d", i);
+ }
+ find_missing(super);
+ } else {
+ int missing = 0;
+ struct dl *d;
+
+ for (d = super->missing; d; d = d->next)
+ missing++;
+ if (info->failed_disks > missing) {
+ fprintf(stderr, Name": unable to add 'missing' disk to container\n");
+ return 0;
+ }
+ }
if (!check_name(super, name, 0))
return 0;
vol = &dev->vol;
vol->migr_state = 0;
set_migr_type(dev, MIGR_INIT);
- vol->dirty = 0;
+ vol->dirty = !info->state;
vol->curr_migr_unit = 0;
map = get_imsm_map(dev, 0);
map->pba_of_lba0 = __cpu_to_le32(super->create_offset);
map->blocks_per_member = __cpu_to_le32(info_to_blocks_per_member(info));
map->blocks_per_strip = __cpu_to_le16(info_to_blocks_per_strip(info));
map->failed_disk_num = ~0;
- map->map_state = info->level ? IMSM_T_STATE_UNINITIALIZED :
- IMSM_T_STATE_NORMAL;
+ map->map_state = info->failed_disks ? IMSM_T_STATE_DEGRADED : IMSM_T_STATE_NORMAL;
map->ddf = 1;
if (info->level == 1 && info->raid_disks > 2) {
{
struct intel_super *super = st->sb;
struct imsm_super *mpb = super->anchor;
- struct dl *dl;
+ struct imsm_disk *_disk;
struct imsm_dev *dev;
struct imsm_map *map;
+ struct dl *dl, *df;
int slot;
dev = get_imsm_dev(super, super->current_vol);
set_imsm_ord_tbl_ent(map, dk->raid_disk, dl->index);
dl->disk.status = CONFIGURED_DISK;
+ /* update size of 'missing' disks to be at least as large as the
+ * largest acitve member (we only have dummy missing disks when
+ * creating the first volume)
+ */
+ if (super->current_vol == 0) {
+ for (df = super->missing; df; df = df->next) {
+ if (dl->disk.total_blocks > df->disk.total_blocks)
+ df->disk.total_blocks = dl->disk.total_blocks;
+ _disk = __get_imsm_disk(mpb, df->index);
+ *_disk = df->disk;
+ }
+ }
+
+ /* refresh unset/failed slots to point to valid 'missing' entries */
+ for (df = super->missing; df; df = df->next)
+ for (slot = 0; slot < mpb->num_disks; slot++) {
+ __u32 ord = get_imsm_ord_tbl_ent(dev, slot, -1);
+
+ if ((ord & IMSM_ORD_REBUILD) == 0)
+ continue;
+ set_imsm_ord_tbl_ent(map, slot, df->index | IMSM_ORD_REBUILD);
+ if (is_gen_migration(dev)) {
+ struct imsm_map *map2 = get_imsm_map(dev, 1);
+ if (slot < map2->num_members) {
+ __u32 ord2 = get_imsm_ord_tbl_ent(dev,
+ slot,
+ 1);
+ if ((unsigned)df->index ==
+ ord_to_idx(ord2))
+ set_imsm_ord_tbl_ent(map2,
+ slot,
+ df->index |
+ IMSM_ORD_REBUILD);
+ }
+ }
+ dprintf("set slot:%d to missing disk:%d\n", slot, df->index);
+ break;
+ }
+
/* if we are creating the first raid device update the family number */
if (super->current_vol == 0) {
__u32 sum;
struct imsm_dev *_dev = __get_imsm_dev(mpb, 0);
- struct imsm_disk *_disk = __get_imsm_disk(mpb, dl->index);
+ _disk = __get_imsm_disk(mpb, dl->index);
if (!_dev || !_disk) {
fprintf(stderr, Name ": BUG mpb setup error\n");
return 1;
return 0;
}
+/* mark_spare()
+ * Function marks disk as spare and restores disk serial
+ * in case it was previously marked as failed by takeover operation
+ * reruns:
+ * -1 : critical error
+ * 0 : disk is marked as spare but serial is not set
+ * 1 : success
+ */
+int mark_spare(struct dl *disk)
+{
+ __u8 serial[MAX_RAID_SERIAL_LEN];
+ int ret_val = -1;
+
+ if (!disk)
+ return ret_val;
+
+ ret_val = 0;
+ if (!imsm_read_serial(disk->fd, NULL, serial)) {
+ /* Restore disk serial number, because takeover marks disk
+ * as failed and adds to serial ':0' before it becomes
+ * a spare disk.
+ */
+ serialcpy(disk->serial, serial);
+ serialcpy(disk->disk.serial, serial);
+ ret_val = 1;
+ }
+ disk->disk.status = SPARE_DISK;
+ disk->index = -1;
+
+ return ret_val;
+}
static int add_to_super_imsm(struct supertype *st, mdu_disk_info_t *dk,
int fd, char *devname)
memset(dd, 0, sizeof(*dd));
dd->major = major(stb.st_rdev);
dd->minor = minor(stb.st_rdev);
- dd->index = -1;
dd->devname = devname ? strdup(devname) : NULL;
dd->fd = fd;
dd->e = NULL;
size /= 512;
serialcpy(dd->disk.serial, dd->serial);
dd->disk.total_blocks = __cpu_to_le32(size);
- dd->disk.status = SPARE_DISK;
+ mark_spare(dd);
if (sysfs_disk_to_scsi_id(fd, &id) == 0)
dd->disk.scsi_id = __cpu_to_le32(id);
else
memset(dd, 0, sizeof(*dd));
dd->major = dk->major;
dd->minor = dk->minor;
- dd->index = -1;
dd->fd = -1;
- dd->disk.status = SPARE_DISK;
+ mark_spare(dd);
dd->action = DISK_REMOVE;
dd->next = super->disk_mgmt_list;
/* write the mpb for disks that compose raid devices */
for (d = super->disks; d ; d = d->next) {
- if (d->index < 0)
+ if (d->index < 0 || is_failed(&d->disk))
continue;
if (store_imsm_mpb(d->fd, mpb))
fprintf(stderr, "%s: failed for device %d:%d %s\n",
}
#define pr_vrb(fmt, arg...) (void) (verbose && fprintf(stderr, Name fmt, ##arg))
-/*
- * validate volume parameters with OROM/EFI capabilities
- */
static int
validate_geometry_imsm_orom(struct intel_super *super, int level, int layout,
int raiddisks, int *chunk, int verbose)
{
-#if DEBUG
- verbose = 1;
-#endif
- /* validate container capabilities */
- if (super->orom && raiddisks > super->orom->tds) {
- if (verbose)
- fprintf(stderr, Name ": %d exceeds maximum number of"
- " platform supported disks: %d\n",
- raiddisks, super->orom->tds);
+ /* check/set platform and metadata limits/defaults */
+ if (super->orom && raiddisks > super->orom->dpa) {
+ pr_vrb(": platform supports a maximum of %d disks per array\n",
+ super->orom->dpa);
return 0;
}
/* capabilities of OROM tested - copied from validate_geometry_imsm_volume */
- if (super->orom && (!is_raid_level_supported(super->orom, level,
- raiddisks))) {
+ if (!is_raid_level_supported(super->orom, level, raiddisks)) {
pr_vrb(": platform does not support raid%d with %d disk%s\n",
level, raiddisks, raiddisks > 1 ? "s" : "");
return 0;
{
struct stat stb;
struct intel_super *super = st->sb;
- struct imsm_super *mpb = super->anchor;
+ struct imsm_super *mpb;
struct dl *dl;
unsigned long long pos = 0;
unsigned long long maxsize;
if (!super)
return 0;
+ mpb = super->anchor;
+
+ if (mpb->num_raid_devs > 0 && mpb->num_disks != raiddisks) {
+ fprintf(stderr, Name ": the option-rom requires all "
+ "member disks to be a member of all volumes.\n");
+ return 0;
+ }
+
if (!validate_geometry_imsm_orom(super, level, layout, raiddisks, chunk, verbose)) {
fprintf(stderr, Name ": RAID gemetry validation failed. "
"Cannot proceed with the action(s).\n");
i += dl->extent_cnt;
maxsize = merge_extents(super, i);
+
+ if (!check_env("IMSM_NO_PLATFORM") &&
+ mpb->num_raid_devs > 0 && size && size != maxsize) {
+ fprintf(stderr, Name ": attempting to create a second "
+ "volume with size less then remaining space. "
+ "Aborting...\n");
+ return 0;
+ }
+
if (maxsize < size || maxsize == 0) {
if (verbose)
fprintf(stderr, Name ": not enough space after merge (%llu < %llu)\n",
}
if (!dev) {
- if (st->sb && freesize) {
+ if (st->sb) {
+ if (!validate_geometry_imsm_orom(st->sb, level, layout,
+ raiddisks, chunk,
+ verbose))
+ return 0;
/* we are being asked to automatically layout a
* new volume based on the current contents of
* the container. If the the parameters can be
* created. add_to_super and getinfo_super
* detect when autolayout is in progress.
*/
- if (!validate_geometry_imsm_orom(st->sb, level, layout,
- raiddisks, chunk,
- verbose))
- return 0;
- return reserve_space(st, raiddisks, size,
- chunk?*chunk:0, freesize);
+ if (freesize)
+ return reserve_space(st, raiddisks, size,
+ chunk?*chunk:0, freesize);
}
return 1;
}
return validate_geometry_imsm_volume(st, level, layout,
raiddisks, chunk,
size, dev,
- freesize, verbose);
+ freesize, 1)
+ ? 1 : -1;
}
}
struct dl *d;
for (d = super->disks; d; d = d->next)
- if (d->index > -2) {
- d->index = -1;
- d->disk.status = SPARE_DISK;
- }
+ if (d->index > -2)
+ mark_spare(d);
}
super->updates_pending++;
struct imsm_super *mpb = super->anchor;
struct mdinfo *rest = NULL;
unsigned int i;
- int bbm_errors = 0;
+ int sb_errors = 0;
struct dl *d;
int spare_disks = 0;
/* do not assemble arrays when not all attributes are supported */
if (imsm_check_attributes(mpb->attributes) == 0) {
- fprintf(stderr, Name ": IMSM metadata loading not allowed "
- "due to attributes incompatibility.\n");
- return NULL;
+ sb_errors = 1;
+ fprintf(stderr, Name ": Unsupported attributes in IMSM metadata."
+ "Arrays activation is blocked.\n");
}
/* check for bad blocks */
- if (imsm_bbm_log_size(super->anchor))
- bbm_errors = 1;
+ if (imsm_bbm_log_size(super->anchor)) {
+ fprintf(stderr, Name ": BBM log found in IMSM metadata."
+ "Arrays activation is blocked.\n");
+ sb_errors = 1;
+ }
+
/* count spare devices, not used in maps
*/
*/
chunk = __le16_to_cpu(map->blocks_per_strip) >> 1;
-#ifndef MDASSEMBLE
- if (!validate_geometry_imsm_orom(super,
- get_imsm_raid_level(map), /* RAID level */
- imsm_level_to_layout(get_imsm_raid_level(map)),
- map->num_members, /* raid disks */
- &chunk,
- 1 /* verbose */)) {
- fprintf(stderr, Name ": RAID gemetry validation failed. "
- "Cannot proceed with the action(s).\n");
- continue;
- }
-#endif /* MDASSEMBLE */
this = malloc(sizeof(*this));
if (!this) {
fprintf(stderr, Name ": failed to allocate %zu bytes\n",
super->current_vol = i;
getinfo_super_imsm_volume(st, this, NULL);
this->next = rest;
+#ifndef MDASSEMBLE
+ /* mdadm does not support all metadata features- set the bit in all arrays state */
+ if (!validate_geometry_imsm_orom(super,
+ get_imsm_raid_level(map), /* RAID level */
+ imsm_level_to_layout(get_imsm_raid_level(map)),
+ map->num_members, /* raid disks */
+ &chunk,
+ 1 /* verbose */)) {
+ fprintf(stderr, Name ": IMSM RAID geometry validation"
+ " failed. Array %s activation is blocked.\n",
+ dev->volume);
+ this->array.state |=
+ (1<<MD_SB_BLOCK_CONTAINER_RESHAPE) |
+ (1<<MD_SB_BLOCK_VOLUME);
+ }
+#endif
+
+ /* if array has bad blocks, set suitable bit in all arrays state */
+ if (sb_errors)
+ this->array.state |=
+ (1<<MD_SB_BLOCK_CONTAINER_RESHAPE) |
+ (1<<MD_SB_BLOCK_VOLUME);
+
for (slot = 0 ; slot < map->num_members; slot++) {
unsigned long long recovery_start;
struct mdinfo *info_d;
rest = this;
}
- /* if array has bad blocks, set suitable bit in array status */
- if (bbm_errors)
- rest->array.state |= (1<<MD_SB_BBM_ERRORS);
-
return rest;
}
struct imsm_disk *disk;
struct imsm_map *map = get_imsm_map(dev, 0);
struct imsm_map *prev = get_imsm_map(dev, dev->vol.migr_state);
+ struct imsm_map *map_for_loop;
__u32 ord;
int idx;
* map[0]. So we look through all the disks we started with and
* see if any failures are still present, or if any new ones
* have arrived
- *
- * FIXME add support for online capacity expansion and
- * raid-level-migration
*/
- for (i = 0; i < prev->num_members; i++) {
- ord = __le32_to_cpu(prev->disk_ord_tbl[i]);
- ord |= __le32_to_cpu(map->disk_ord_tbl[i]);
+ map_for_loop = prev;
+ if (is_gen_migration(dev))
+ if (prev && (map->num_members > prev->num_members))
+ map_for_loop = map;
+
+ for (i = 0; i < map_for_loop->num_members; i++) {
+ ord = 0;
+ if (i < prev->num_members)
+ ord |= __le32_to_cpu(prev->disk_ord_tbl[i]);
+ if (i < map->num_members)
+ ord |= __le32_to_cpu(map->disk_ord_tbl[i]);
idx = ord_to_idx(ord);
disk = get_imsm_disk(super, idx);
__u32 ord;
int slot;
struct imsm_map *map;
+ char buf[MAX_RAID_SERIAL_LEN+3];
+ unsigned int len, shift = 0;
/* new failures are always set in map[0] */
map = get_imsm_map(dev, 0);
if (is_failed(disk) && (ord & IMSM_ORD_REBUILD))
return 0;
+ memcpy(buf, disk->serial, MAX_RAID_SERIAL_LEN);
+ buf[MAX_RAID_SERIAL_LEN] = '\000';
+ strcat(buf, ":0");
+ if ((len = strlen(buf)) >= MAX_RAID_SERIAL_LEN)
+ shift = len - MAX_RAID_SERIAL_LEN + 1;
+ strncpy((char *)disk->serial, &buf[shift], MAX_RAID_SERIAL_LEN);
+
disk->status |= FAILED_DISK;
set_imsm_ord_tbl_ent(map, slot, idx | IMSM_ORD_REBUILD);
+ if (is_gen_migration(dev)) {
+ struct imsm_map *map2 = get_imsm_map(dev, 1);
+ if (slot < map2->num_members)
+ set_imsm_ord_tbl_ent(map2, slot,
+ idx | IMSM_ORD_REBUILD);
+ }
if (map->failed_disk_num == 0xff)
map->failed_disk_num = slot;
return 1;
static void handle_missing(struct intel_super *super, struct imsm_dev *dev)
{
- __u8 map_state;
struct dl *dl;
- int failed;
if (!super->missing)
return;
- failed = imsm_count_failed(super, dev);
- map_state = imsm_check_degraded(super, dev, failed);
dprintf("imsm: mark missing\n");
- end_migration(dev, map_state);
for (dl = super->missing; dl; dl = dl->next)
mark_missing(dev, &dl->disk, dl->index);
super->updates_pending++;
map->num_members = prev_disks;
dev->vol.migr_state = 1;
dev->vol.curr_migr_unit = 0;
- dev->vol.migr_type = MIGR_GEN_MIGR;
+ set_migr_type(dev, MIGR_GEN_MIGR);
for (i = prev_num_members;
i < map->num_members; i++)
set_imsm_ord_tbl_ent(map, i, i);
if (0) {
struct imsm_map *map2 = get_imsm_map(dev, 1);
dev->vol.migr_state = 0;
- dev->vol.migr_type = 0;
+ set_migr_type(dev, 0);
dev->vol.curr_migr_unit = 0;
memcpy(map, map2, sizeof_imsm_map(map2));
super->updates_pending++;
super->updates_pending++;
a->last_checkpoint = 0;
}
- } else if (!is_resyncing(dev) && !failed) {
+ } else if ((!is_resyncing(dev) && !failed) &&
+ (imsm_reshape_blocks_arrays_changes(super) == 0)) {
/* mark the start of the init process if nothing is failed */
dprintf("imsm: mark resync start\n");
if (map->map_state == IMSM_T_STATE_UNINITIALIZED)
dprintf("imsm: set_disk %d:%x\n", n, state);
- ord = get_imsm_ord_tbl_ent(dev, n, -1);
+ ord = get_imsm_ord_tbl_ent(dev, n, -2);
disk = get_imsm_disk(super, ord_to_idx(ord));
/* check for new failures */
a->last_checkpoint = 0;
} else if (is_gen_migration(dev)) {
dprintf("imsm: Detected General Migration in state: ");
- if (map_state == IMSM_T_STATE_NORMAL) {
- end_migration(dev, map_state);
+
+ switch (map_state) {
+ case IMSM_T_STATE_NORMAL:
+ dprintf("normal\n");
+ if (a->last_checkpoint >= a->info.component_size)
+ end_migration(dev, map_state);
map = get_imsm_map(dev, 0);
map->failed_disk_num = ~0;
- dprintf("normal\n");
- } else {
- if (map_state == IMSM_T_STATE_DEGRADED) {
- printf("degraded\n");
+ break;
+ case IMSM_T_STATE_DEGRADED:
+ dprintf("degraded\n");
+ if (a->last_checkpoint >= a->info.component_size)
end_migration(dev, map_state);
- } else {
- dprintf("failed\n");
- }
- map->map_state = map_state;
+ break;
+ default:
+ dprintf("failed\n");
}
+ map->map_state = map_state;
super->updates_pending++;
}
}
dprintf("imsm: activate spare: inst=%d failed=%d (%d) level=%d\n",
inst, failed, a->info.array.raid_disks, a->info.array.level);
- if (dev->vol.migr_state &&
- dev->vol.migr_type == MIGR_GEN_MIGR)
- /* No repair during migration */
- return NULL;
+ if (imsm_reshape_blocks_arrays_changes(super))
+ return NULL;
if (a->info.array.level == 4)
/* No repair for takeovered array
* are removed from container.
*/
if (failed) {
- dprintf("found failed disks in %s, check if there another"
+ dprintf("found failed disks in %.*s, check if there another"
"failed sub-array.\n",
- dev->volume);
+ MAX_RAID_SERIAL_LEN, dev->volume);
/* check if states of the other volumes allow for rebuild */
for (i = 0; i < super->anchor->num_raid_devs; i++) {
if (i != inst) {
*/
dl = imsm_readd(super, i, a);
if (!dl)
- dl = imsm_add_spare(super, i, a, 0, NULL);
+ dl = imsm_add_spare(super, i, a, 0, rv);
if (!dl)
- dl = imsm_add_spare(super, i, a, 1, NULL);
+ dl = imsm_add_spare(super, i, a, 1, rv);
if (!dl)
continue;
num_spares++;
dprintf("%x:%x to be %d at %llu\n", dl->major, dl->minor,
i, di->data_offset);
-
- break;
}
if (!rv)
return ret_val;
}
+static int apply_update_activate_spare(struct imsm_update_activate_spare *u,
+ struct intel_super *super,
+ struct active_array *active_array)
+{
+ struct imsm_super *mpb = super->anchor;
+ struct imsm_dev *dev = get_imsm_dev(super, u->array);
+ struct imsm_map *map = get_imsm_map(dev, 0);
+ struct imsm_map *migr_map;
+ struct active_array *a;
+ struct imsm_disk *disk;
+ __u8 to_state;
+ struct dl *dl;
+ unsigned int found;
+ int failed;
+ int victim;
+ int i;
+ int second_map_created = 0;
+
+ for (; u; u = u->next) {
+ victim = get_imsm_disk_idx(dev, u->slot, -1);
+
+ if (victim < 0)
+ return 0;
+
+ for (dl = super->disks; dl; dl = dl->next)
+ if (dl == u->dl)
+ break;
+
+ if (!dl) {
+ fprintf(stderr, "error: imsm_activate_spare passed "
+ "an unknown disk (index: %d)\n",
+ u->dl->index);
+ return 0;
+ }
+
+ /* count failures (excluding rebuilds and the victim)
+ * to determine map[0] state
+ */
+ failed = 0;
+ for (i = 0; i < map->num_members; i++) {
+ if (i == u->slot)
+ continue;
+ disk = get_imsm_disk(super,
+ get_imsm_disk_idx(dev, i, -1));
+ if (!disk || is_failed(disk))
+ failed++;
+ }
+
+ /* adding a pristine spare, assign a new index */
+ if (dl->index < 0) {
+ dl->index = super->anchor->num_disks;
+ super->anchor->num_disks++;
+ }
+ disk = &dl->disk;
+ disk->status |= CONFIGURED_DISK;
+ disk->status &= ~SPARE_DISK;
+
+ /* mark rebuild */
+ to_state = imsm_check_degraded(super, dev, failed);
+ if (!second_map_created) {
+ second_map_created = 1;
+ map->map_state = IMSM_T_STATE_DEGRADED;
+ migrate(dev, super, to_state, MIGR_REBUILD);
+ } else
+ map->map_state = to_state;
+ migr_map = get_imsm_map(dev, 1);
+ set_imsm_ord_tbl_ent(map, u->slot, dl->index);
+ set_imsm_ord_tbl_ent(migr_map, u->slot,
+ dl->index | IMSM_ORD_REBUILD);
+
+ /* update the family_num to mark a new container
+ * generation, being careful to record the existing
+ * family_num in orig_family_num to clean up after
+ * earlier mdadm versions that neglected to set it.
+ */
+ if (mpb->orig_family_num == 0)
+ mpb->orig_family_num = mpb->family_num;
+ mpb->family_num += super->random;
+
+ /* count arrays using the victim in the metadata */
+ found = 0;
+ for (a = active_array; a ; a = a->next) {
+ dev = get_imsm_dev(super, a->info.container_member);
+ map = get_imsm_map(dev, 0);
+
+ if (get_imsm_disk_slot(map, victim) >= 0)
+ found++;
+ }
+
+ /* delete the victim if it is no longer being
+ * utilized anywhere
+ */
+ if (!found) {
+ struct dl **dlp;
+
+ /* We know that 'manager' isn't touching anything,
+ * so it is safe to delete
+ */
+ for (dlp = &super->disks; *dlp; dlp = &(*dlp)->next)
+ if ((*dlp)->index == victim)
+ break;
+
+ /* victim may be on the missing list */
+ if (!*dlp)
+ for (dlp = &super->missing; *dlp;
+ dlp = &(*dlp)->next)
+ if ((*dlp)->index == victim)
+ break;
+ imsm_delete(super, dlp, victim);
+ }
+ }
+
+ return 1;
+}
static int apply_reshape_container_disks_update(struct imsm_update_reshape *u,
struct intel_super *super,
devices_to_reshape--;
newdev->vol.migr_state = 1;
newdev->vol.curr_migr_unit = 0;
- newdev->vol.migr_type = MIGR_GEN_MIGR;
+ set_migr_type(newdev, MIGR_GEN_MIGR);
newmap->num_members = u->new_raid_disks;
for (i = 0; i < delta_disks; i++) {
set_imsm_ord_tbl_ent(newmap,
if (du->index > idx)
du->index--;
/* mark as spare disk */
- dm->disk.status = SPARE_DISK;
- dm->index = -1;
+ mark_spare(dm);
}
}
/* update map */
}
case update_activate_spare: {
struct imsm_update_activate_spare *u = (void *) update->buf;
- struct imsm_dev *dev = get_imsm_dev(super, u->array);
- struct imsm_map *map = get_imsm_map(dev, 0);
- struct imsm_map *migr_map;
- struct active_array *a;
- struct imsm_disk *disk;
- __u8 to_state;
- struct dl *dl;
- unsigned int found;
- int failed;
- int victim = get_imsm_disk_idx(dev, u->slot, -1);
- int i;
-
- for (dl = super->disks; dl; dl = dl->next)
- if (dl == u->dl)
- break;
-
- if (!dl) {
- fprintf(stderr, "error: imsm_activate_spare passed "
- "an unknown disk (index: %d)\n",
- u->dl->index);
- return;
- }
-
- super->updates_pending++;
- /* count failures (excluding rebuilds and the victim)
- * to determine map[0] state
- */
- failed = 0;
- for (i = 0; i < map->num_members; i++) {
- if (i == u->slot)
- continue;
- disk = get_imsm_disk(super,
- get_imsm_disk_idx(dev, i, -1));
- if (!disk || is_failed(disk))
- failed++;
- }
-
- /* adding a pristine spare, assign a new index */
- if (dl->index < 0) {
- dl->index = super->anchor->num_disks;
- super->anchor->num_disks++;
- }
- disk = &dl->disk;
- disk->status |= CONFIGURED_DISK;
- disk->status &= ~SPARE_DISK;
-
- /* mark rebuild */
- to_state = imsm_check_degraded(super, dev, failed);
- map->map_state = IMSM_T_STATE_DEGRADED;
- migrate(dev, super, to_state, MIGR_REBUILD);
- migr_map = get_imsm_map(dev, 1);
- set_imsm_ord_tbl_ent(map, u->slot, dl->index);
- set_imsm_ord_tbl_ent(migr_map, u->slot, dl->index | IMSM_ORD_REBUILD);
-
- /* update the family_num to mark a new container
- * generation, being careful to record the existing
- * family_num in orig_family_num to clean up after
- * earlier mdadm versions that neglected to set it.
- */
- if (mpb->orig_family_num == 0)
- mpb->orig_family_num = mpb->family_num;
- mpb->family_num += super->random;
-
- /* count arrays using the victim in the metadata */
- found = 0;
- for (a = st->arrays; a ; a = a->next) {
- dev = get_imsm_dev(super, a->info.container_member);
- map = get_imsm_map(dev, 0);
-
- if (get_imsm_disk_slot(map, victim) >= 0)
- found++;
- }
-
- /* delete the victim if it is no longer being
- * utilized anywhere
- */
- if (!found) {
- struct dl **dlp;
-
- /* We know that 'manager' isn't touching anything,
- * so it is safe to delete
- */
- for (dlp = &super->disks; *dlp; dlp = &(*dlp)->next)
- if ((*dlp)->index == victim)
- break;
-
- /* victim may be on the missing list */
- if (!*dlp)
- for (dlp = &super->missing; *dlp; dlp = &(*dlp)->next)
- if ((*dlp)->index == victim)
- break;
- imsm_delete(super, dlp, victim);
- }
+ if (apply_update_activate_spare(u, super, st->arrays))
+ super->updates_pending++;
break;
}
case update_create_array: {
int open_backup_targets(struct mdinfo *info, int raid_disks, int *raid_fds)
{
struct mdinfo *sd;
+ int i;
+
+ for (i = 0; i < raid_disks; i++)
+ raid_fds[i] = -1;
for (sd = info->devs ; sd ; sd = sd->next) {
char *dn;
if (!targets)
goto abort;
- open_backup_targets(info, new_disks, targets);
+ if (open_backup_targets(info, new_disks, targets)) {
+ fprintf(stderr,
+ Name ": Cannot open some devices belonging to array.\n");
+ goto abort;
+ }
for (i = 0; i < new_disks; i++) {
if (targets[i] < 0) {
int change = -1;
int check_devs = 0;
int chunk;
+ int devNumChange=0;
+ int layout = -1;
getinfo_super_imsm_volume(st, &info, NULL);
if ((geo->level != info.array.level) &&
change = -1;
goto analyse_change_exit;
}
+ layout = geo->layout;
check_devs = 1;
- }
- if (geo->level == 10) {
+ devNumChange = 1; /* parity disk added */
+ } else if (geo->level == 10) {
change = CH_TAKEOVER;
check_devs = 1;
+ devNumChange = 2; /* two mirrors added */
+ layout = 0x102; /* imsm supported layout */
}
break;
case 1:
- if (geo->level == 0) {
- change = CH_TAKEOVER;
- check_devs = 1;
- }
- break;
case 10:
if (geo->level == 0) {
change = CH_TAKEOVER;
check_devs = 1;
+ devNumChange = -(geo->raid_disks/2);
+ layout = 0; /* imsm raid0 layout */
}
break;
}
chunk = geo->chunksize / 1024;
if (!validate_geometry_imsm(st,
geo->level,
- geo->layout,
- geo->raid_disks,
+ layout,
+ geo->raid_disks + devNumChange,
&chunk,
geo->size,
0, 0, 1))
projects / thirdparty / mdadm.git / blobdiff