* Email: <neilb@suse.de>
*/
+#include <stddef.h>
#include "mdadm.h"
/*
* The version-1 superblock :
__u64 data_offset; /* sector start of data, often 0 */
__u64 data_size; /* sectors in this device that can be used for data */
__u64 super_offset; /* sector start of this superblock */
- __u64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */
+ union {
+ __u64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */
+ __u64 journal_tail;/* journal tail of journal device (from data_offset) */
+ };
__u32 dev_number; /* permanent identifier of this device - not role in raid */
__u32 cnt_corrected_read; /* number of read errors that were corrected by re-writing */
__u8 device_uuid[16]; /* user-space setable, ignored by kernel */
* backwards anyway.
*/
#define MD_FEATURE_NEW_OFFSET 64 /* new_offset must be honoured */
+#define MD_FEATURE_BITMAP_VERSIONED 256 /* bitmap version number checked properly */
+#define MD_FEATURE_JOURNAL 512 /* support write journal */
#define MD_FEATURE_ALL (MD_FEATURE_BITMAP_OFFSET \
|MD_FEATURE_RECOVERY_OFFSET \
|MD_FEATURE_RESHAPE_ACTIVE \
|MD_FEATURE_REPLACEMENT \
|MD_FEATURE_RESHAPE_BACKWARDS \
|MD_FEATURE_NEW_OFFSET \
+ |MD_FEATURE_BITMAP_VERSIONED \
+ |MD_FEATURE_JOURNAL \
)
-#ifndef offsetof
-#define offsetof(t,f) ((size_t)&(((t*)0)->f))
-#endif
+/* return how many bytes are needed for bitmap, for cluster-md each node
+ * should have it's own bitmap */
+static unsigned int calc_bitmap_size(bitmap_super_t *bms, unsigned int boundary)
+{
+ unsigned long long bits, bytes;
+
+ bits = __le64_to_cpu(bms->sync_size) / (__le32_to_cpu(bms->chunksize)>>9);
+ bytes = (bits+7) >> 3;
+ bytes += sizeof(bitmap_super_t);
+ bytes = ROUND_UP(bytes, boundary);
+
+ return bytes;
+}
+
static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb)
{
unsigned int disk_csum, csum;
if (!bsize || bsize > 4096 || len > 4096) {
if (!bsize)
- fprintf(stderr, "WARNING - aread() called with "
- "invalid block size\n");
+ fprintf(stderr, "WARNING - aread() called with invalid block size\n");
return -1;
}
b = ROUND_UP_PTR((char *)abuf, 4096);
bsize = afd->blk_sz;
if (!bsize || bsize > 4096 || len > 4096) {
if (!bsize)
- fprintf(stderr, "WARNING - awrite() called with "
- "invalid block size\n");
+ fprintf(stderr, "WARNING - awrite() called with invalid block size\n");
return -1;
}
b = ROUND_UP_PTR((char *)abuf, 4096);
static void examine_super1(struct supertype *st, char *homehost)
{
struct mdp_superblock_1 *sb = st->sb;
+ bitmap_super_t *bms = (bitmap_super_t*)(((char*)sb)+MAX_SB_SIZE);
time_t atime;
unsigned int d;
int role;
strncmp(sb->set_name, homehost, l) == 0)
printf(" (local to host %s)", homehost);
printf("\n");
+ if (bms->nodes > 0)
+ printf(" Cluster Name : %-64s\n", bms->cluster_name);
atime = __le64_to_cpu(sb->ctime) & 0xFFFFFFFFFFULL;
printf(" Creation Time : %.24s\n", ctime(&atime));
c=map_num(pers, __le32_to_cpu(sb->level));
/* This turns out to just be confusing */
printf(" Array Slot : %d (", __le32_to_cpu(sb->dev_number));
for (i= __le32_to_cpu(sb->max_dev); i> 0 ; i--)
- if (__le16_to_cpu(sb->dev_roles[i-1]) != 0xffff)
+ if (__le16_to_cpu(sb->dev_roles[i-1]) != MD_DISK_ROLE_SPARE)
break;
for (d=0; d < i; d++) {
int role = __le16_to_cpu(sb->dev_roles[d]);
if (d) printf(", ");
- if (role == 0xffff) printf("empty");
- else if(role == 0xfffe) printf("failed");
+ if (role == MD_DISK_ROLE_SPARE) printf("empty");
+ else if(role == MD_DISK_ROLE_FAULTY) printf("failed");
else printf("%d", role);
}
printf(")\n");
if (d < __le32_to_cpu(sb->max_dev))
role = __le16_to_cpu(sb->dev_roles[d]);
else
- role = 0xFFFF;
- if (role >= 0xFFFE)
+ role = MD_DISK_ROLE_SPARE;
+ if (role >= MD_DISK_ROLE_FAULTY)
printf("spare\n");
+ else if (role == MD_DISK_ROLE_JOURNAL)
+ printf("Journal\n");
else if (sb->feature_map & __cpu_to_le32(MD_FEATURE_REPLACEMENT))
printf("Replacement device %d\n", role);
else
faulty = 0;
for (i=0; i< __le32_to_cpu(sb->max_dev); i++) {
int role = __le16_to_cpu(sb->dev_roles[i]);
- if (role == 0xFFFE)
+ if (role == MD_DISK_ROLE_FAULTY)
faulty++;
}
if (faulty) printf(" %d failed", faulty);
/* have the header, can calculate
* correct bitmap bytes */
bitmap_super_t *bms;
- int bits;
bms = (void*)buf;
- bits = __le64_to_cpu(bms->sync_size) / (__le32_to_cpu(bms->chunksize)>>9);
- bytes = (bits+7) >> 3;
- bytes += sizeof(bitmap_super_t);
- bytes = ROUND_UP(bytes, 512);
+ bytes = calc_bitmap_size(bms, 512);
if (n > bytes)
n = bytes;
}
static void detail_super1(struct supertype *st, char *homehost)
{
struct mdp_superblock_1 *sb = st->sb;
+ bitmap_super_t *bms = (bitmap_super_t*)(((char*)sb) + MAX_SB_SIZE);
int i;
int l = homehost ? strlen(homehost) : 0;
sb->set_name[l] == ':' &&
strncmp(sb->set_name, homehost, l) == 0)
printf(" (local to host %s)", homehost);
+ if (bms->nodes > 0)
+ printf("\n Cluster Name : %-64s", bms->cluster_name);
printf("\n UUID : ");
for (i=0; i<16; i++) {
if ((i&3)==0 && i != 0) printf(":");
size = __le32_to_cpu(sb->bblog_size)* 512;
if (posix_memalign((void**)&bbl, 4096, size) != 0) {
- pr_err("%s could not allocate badblocks list\n", __func__);
+ pr_err("could not allocate badblocks list\n");
return 0;
}
offset = __le64_to_cpu(sb->super_offset) +
info->array.state =
(__le64_to_cpu(sb->resync_offset) == MaxSector)
? 1 : 0;
+ if (__le32_to_cpu(bsb->nodes) > 1)
+ info->array.state |= (1 << MD_SB_CLUSTERED);
info->data_offset = __le64_to_cpu(sb->data_offset);
info->component_size = __le64_to_cpu(sb->size);
info->disk.number = __le32_to_cpu(sb->dev_number);
if (__le32_to_cpu(sb->dev_number) >= __le32_to_cpu(sb->max_dev) ||
__le32_to_cpu(sb->dev_number) >= MAX_DEVS)
- role = 0xfffe;
+ role = MD_DISK_ROLE_FAULTY;
else
role = __le16_to_cpu(sb->dev_roles[__le32_to_cpu(sb->dev_number)]);
info->disk.raid_disk = -1;
switch(role) {
- case 0xFFFF:
+ case MD_DISK_ROLE_SPARE:
info->disk.state = 0; /* spare: not active, not sync, not faulty */
break;
- case 0xFFFE:
+ case MD_DISK_ROLE_FAULTY:
info->disk.state = 1; /* faulty */
break;
+ case MD_DISK_ROLE_JOURNAL:
+ info->disk.state = (1 << MD_DISK_JOURNAL);
+ info->disk.raid_disk = role;
+ info->space_after = (misc->device_size - info->data_offset) % 8; /* journal uses all 4kB blocks*/
+ break;
default:
info->disk.state = 6; /* active and in sync */
info->disk.raid_disk = role;
map[i] = 0;
for (i = 0; i < __le32_to_cpu(sb->max_dev); i++) {
role = __le16_to_cpu(sb->dev_roles[i]);
- if (/*role == 0xFFFF || */role < (unsigned) info->array.raid_disks) {
+ if (/*role == MD_DISK_ROLE_SPARE || */role < (unsigned) info->array.raid_disks) {
working++;
if (map && role < map_disks)
map[role] = 1;
if (info->disk.state & (1<<MD_DISK_ACTIVE))
want = info->disk.raid_disk;
else
- want = 0xFFFF;
+ want = MD_DISK_ROLE_SPARE;
if (sb->dev_roles[d] != __cpu_to_le16(want)) {
sb->dev_roles[d] = __cpu_to_le16(want);
rv = 1;
unsigned int max = __le32_to_cpu(sb->max_dev);
for (i=0 ; i < max ; i++)
- if (__le16_to_cpu(sb->dev_roles[i]) >= 0xfffe)
+ if (__le16_to_cpu(sb->dev_roles[i]) >= MD_DISK_ROLE_FAULTY)
break;
sb->dev_number = __cpu_to_le32(i);
info->disk.number = i;
/* set data_size to device size less data_offset */
struct misc_dev_info *misc = (struct misc_dev_info*)
(st->sb + MAX_SB_SIZE + BM_SUPER_SIZE);
- printf("Size was %llu\n", (unsigned long long)
- __le64_to_cpu(sb->data_size));
sb->data_size = __cpu_to_le64(
misc->device_size - __le64_to_cpu(sb->data_offset));
- printf("Size is %llu\n", (unsigned long long)
- __le64_to_cpu(sb->data_size));
} else if (strcmp(update, "revert-reshape") == 0) {
rv = -2;
if (!(sb->feature_map & __cpu_to_le32(MD_FEATURE_RESHAPE_ACTIVE)))
* So we reject a revert-reshape unless the
* alignment is good.
*/
- reshape_sectors = __le64_to_cpu(sb->reshape_position);
- reshape_chunk = __le32_to_cpu(sb->new_chunk);
- reshape_chunk *= __le32_to_cpu(sb->raid_disks) - __le32_to_cpu(sb->delta_disks) -
- (__le32_to_cpu(sb->level)==6 ? 2 : 1);
- if (reshape_sectors % reshape_chunk) {
- pr_err("Reshape position is not suitably aligned.\n");
- pr_err("Try normal assembly as stop again\n");
- return -2;
+ if (__le32_to_cpu(sb->level) >= 4 &&
+ __le32_to_cpu(sb->level) <= 6) {
+ reshape_sectors = __le64_to_cpu(sb->reshape_position);
+ reshape_chunk = __le32_to_cpu(sb->new_chunk);
+ reshape_chunk *= __le32_to_cpu(sb->raid_disks) - __le32_to_cpu(sb->delta_disks) -
+ (__le32_to_cpu(sb->level)==6 ? 2 : 1);
+ if (reshape_sectors % reshape_chunk) {
+ pr_err("Reshape position is not suitably aligned.\n");
+ pr_err("Try normal assembly and stop again\n");
+ return -2;
+ }
}
sb->raid_disks = __cpu_to_le32(__le32_to_cpu(sb->raid_disks) -
__le32_to_cpu(sb->delta_disks));
sb->chunksize = temp;
if (sb->feature_map & __cpu_to_le32(MD_FEATURE_NEW_OFFSET)) {
- sb->data_offset = __cpu_to_le64(__le64_to_cpu(sb->data_offset) +
- (long)(int32_t)__le32_to_cpu(sb->new_offset));
- sb->new_offset = __cpu_to_le32(-(int32_t)__le32_to_cpu(sb->new_offset));
+ long offset_delta = (int32_t)__le32_to_cpu(sb->new_offset);
+ sb->data_offset = __cpu_to_le64(__le64_to_cpu(sb->data_offset) + offset_delta);
+ sb->new_offset = __cpu_to_le32(-offset_delta);
+ sb->data_size = __cpu_to_le64(__le64_to_cpu(sb->data_size) - offset_delta);
}
}
} else if (strcmp(update, "_reshape_progress")==0)
int sbsize;
if (posix_memalign((void**)&sb, 4096, SUPER1_SIZE) != 0) {
- pr_err("%s could not allocate superblock\n", __func__);
+ pr_err("could not allocate superblock\n");
return 0;
}
memset(sb, 0, SUPER1_SIZE);
if ((dk->state & 6) == 6) /* active, sync */
*rp = __cpu_to_le16(dk->raid_disk);
+ else if (dk->state & (1<<MD_DISK_JOURNAL))
+ *rp = MD_DISK_ROLE_JOURNAL;
else if ((dk->state & ~2) == 0) /* active or idle -> spare */
- *rp = 0xffff;
+ *rp = MD_DISK_ROLE_SPARE;
else
- *rp = 0xfffe;
+ *rp = MD_DISK_ROLE_FAULTY;
if (dk->number >= (int)__le32_to_cpu(sb->max_dev) &&
__le32_to_cpu(sb->max_dev) < MAX_DEVS)
static void free_super1(struct supertype *st);
+#define META_BLOCK_SIZE 4096
+unsigned long crc32(
+ unsigned long crc,
+ const unsigned char *buf,
+ unsigned len);
+
+static int write_empty_r5l_meta_block(struct supertype *st, int fd)
+{
+ struct r5l_meta_block *mb;
+ struct mdp_superblock_1 *sb = st->sb;
+ struct align_fd afd;
+ __u32 crc;
+
+ init_afd(&afd, fd);
+
+ if (posix_memalign((void**)&mb, 4096, META_BLOCK_SIZE) != 0) {
+ pr_err("Could not allocate memory for the meta block.\n");
+ return 1;
+ }
+
+ memset(mb, 0, META_BLOCK_SIZE);
+
+ mb->magic = __cpu_to_le32(R5LOG_MAGIC);
+ mb->version = R5LOG_VERSION;
+ mb->meta_size = __cpu_to_le32(sizeof(struct r5l_meta_block));
+ mb->seq = __cpu_to_le64(random32());
+ mb->position = __cpu_to_le64(0);
+
+ crc = crc32(0xffffffff, sb->set_uuid, sizeof(sb->set_uuid));
+ crc = crc32(crc, (void *)mb, META_BLOCK_SIZE);
+ mb->checksum = __cpu_to_le32(crc);
+
+ if (lseek64(fd, (sb->data_offset) * 512, 0) < 0LL) {
+ pr_err("cannot seek to offset of the meta block\n");
+ goto fail_to_write;
+ }
+
+ if (awrite(&afd, mb, META_BLOCK_SIZE) != META_BLOCK_SIZE) {
+ pr_err("failed to store write the meta block \n");
+ goto fail_to_write;
+ }
+ fsync(fd);
+
+ free(mb);
+ return 0;
+
+fail_to_write:
+ free(mb);
+ return 1;
+}
+
#ifndef MDASSEMBLE
static int write_init_super1(struct supertype *st)
{
int rfd;
int rv = 0;
unsigned long long bm_space;
- unsigned long long reserved;
struct devinfo *di;
unsigned long long dsize, array_size;
- unsigned long long sb_offset, headroom;
+ unsigned long long sb_offset;
unsigned long long data_offset;
for (di = st->info; di; di = di->next) {
- if (di->disk.state == 1)
+ if (di->disk.state & (1 << MD_DISK_JOURNAL))
+ sb->feature_map |= MD_FEATURE_JOURNAL;
+ }
+
+ for (di = st->info; di; di = di->next) {
+ if (di->disk.state & (1 << MD_DISK_FAULTY))
continue;
if (di->fd < 0)
continue;
* 0: At least 8K, but less than 12K, from end of device
* 1: At start of device
* 2: 4K from start of device.
- * Depending on the array size, we might leave extra space
- * for a bitmap.
- * Also leave 4K for bad-block log.
+ * data_offset has already been set.
*/
array_size = __le64_to_cpu(sb->size);
/* work out how much space we left for a bitmap,
* Add 8 sectors for bad block log */
bm_space = choose_bm_space(array_size) + 8;
- /* We try to leave 0.1% at the start for reshape
- * operations, but limit this to 128Meg (0.1% of 10Gig)
- * which is plenty for efficient reshapes
- * However we make it at least 2 chunks as one chunk
- * is minimum needed for reshape.
- */
- headroom = 128 * 1024 * 2;
- while (headroom << 10 > array_size &&
- headroom/2 >= __le32_to_cpu(sb->chunksize) * 2)
- headroom >>= 1;
-
data_offset = di->data_offset;
+ if (data_offset == INVALID_SECTORS)
+ data_offset = st->data_offset;
switch(st->minor_version) {
case 0:
+ if (data_offset == INVALID_SECTORS)
+ data_offset = 0;
sb_offset = dsize;
sb_offset -= 8*2;
sb_offset &= ~(4*2-1);
+ sb->data_offset = __cpu_to_le64(data_offset);
sb->super_offset = __cpu_to_le64(sb_offset);
- if (data_offset == INVALID_SECTORS)
- sb->data_offset = 0;
if (sb_offset < array_size + bm_space)
bm_space = sb_offset - array_size;
sb->data_size = __cpu_to_le64(sb_offset - bm_space);
break;
case 1:
sb->super_offset = __cpu_to_le64(0);
- if (data_offset == INVALID_SECTORS) {
- reserved = bm_space + 4*2;
- if (reserved < headroom)
- reserved = headroom;
- if (reserved + array_size > dsize)
- reserved = dsize - array_size;
- /* Try for multiple of 1Meg so it is nicely aligned */
- #define ONE_MEG (2*1024)
- if (reserved > ONE_MEG)
- reserved = (reserved/ONE_MEG) * ONE_MEG;
-
- /* force 4K alignment */
- reserved &= ~7ULL;
-
- } else
- reserved = data_offset;
-
- sb->data_offset = __cpu_to_le64(reserved);
- sb->data_size = __cpu_to_le64(dsize - reserved);
- if (reserved >= 8 + 32*2 + 8) {
+ if (data_offset == INVALID_SECTORS)
+ data_offset = 16;
+
+ sb->data_offset = __cpu_to_le64(data_offset);
+ sb->data_size = __cpu_to_le64(dsize - data_offset);
+ if (data_offset >= 8 + 32*2 + 8) {
sb->bblog_size = __cpu_to_le16(8);
sb->bblog_offset = __cpu_to_le32(8 + 32*2);
- } else if (reserved >= 16) {
+ } else if (data_offset >= 16) {
sb->bblog_size = __cpu_to_le16(8);
- sb->bblog_offset = __cpu_to_le32(reserved-8);
+ sb->bblog_offset = __cpu_to_le32(data_offset-8);
}
break;
case 2:
sb_offset = 4*2;
- sb->super_offset = __cpu_to_le64(4*2);
- if (data_offset == INVALID_SECTORS) {
- if (4*2 + 4*2 + bm_space + array_size
- > dsize)
- bm_space = dsize - array_size
- - 4*2 - 4*2;
-
- reserved = bm_space + 4*2 + 4*2;
- if (reserved < headroom)
- reserved = headroom;
- if (reserved + array_size > dsize)
- reserved = dsize - array_size;
- /* Try for multiple of 1Meg so it is nicely aligned */
- #define ONE_MEG (2*1024)
- if (reserved > ONE_MEG)
- reserved = (reserved/ONE_MEG) * ONE_MEG;
-
- /* force 4K alignment */
- reserved &= ~7ULL;
-
- } else
- reserved = data_offset;
-
- sb->data_offset = __cpu_to_le64(reserved);
- sb->data_size = __cpu_to_le64(dsize - reserved);
- if (reserved >= 16 + 32*2 + 8) {
+ sb->super_offset = __cpu_to_le64(sb_offset);
+ if (data_offset == INVALID_SECTORS)
+ data_offset = 24;
+
+ sb->data_offset = __cpu_to_le64(data_offset);
+ sb->data_size = __cpu_to_le64(dsize - data_offset);
+ if (data_offset >= 16 + 32*2 + 8) {
sb->bblog_size = __cpu_to_le16(8);
sb->bblog_offset = __cpu_to_le32(8 + 32*2);
- } else if (reserved >= 16+16) {
+ } else if (data_offset >= 16+16) {
sb->bblog_size = __cpu_to_le16(8);
/* '8' sectors for the bblog, and another '8'
* because we want offset from superblock, not
* start of device.
*/
- sb->bblog_offset = __cpu_to_le32(reserved-8-8);
+ sb->bblog_offset = __cpu_to_le32(data_offset-8-8);
}
break;
default:
- pr_err("Failed to write invalid "
- "metadata format 1.%i to %s\n",
+ pr_err("Failed to write invalid metadata format 1.%i to %s\n",
st->minor_version, di->devname);
rv = -EINVAL;
goto out;
}
+ /* Disable badblock log on clusters, or when explicitly requested */
+ if (st->nodes > 0 || conf_get_create_info()->bblist == 0) {
+ sb->bblog_size = 0;
+ sb->bblog_offset = 0;
+ }
sb->sb_csum = calc_sb_1_csum(sb);
rv = store_super1(st, di->fd);
+
+ if (rv == 0 && (di->disk.state & (1 << MD_DISK_JOURNAL))) {
+ rv = write_empty_r5l_meta_block(st, di->fd);
+ if (rv)
+ goto error_out;
+ }
+
if (rv == 0 && (__le32_to_cpu(sb->feature_map) & 1))
- rv = st->ss->write_bitmap(st, di->fd);
+ rv = st->ss->write_bitmap(st, di->fd, NoUpdate);
close(di->fd);
di->fd = -1;
if (rv)
if (!first) {
if (posix_memalign((void**)&first, 4096, SUPER1_SIZE) != 0) {
- pr_err("%s could not allocate superblock\n", __func__);
+ pr_err("could not allocate superblock\n");
return 1;
}
memcpy(first, second, SUPER1_SIZE);
}
if (posix_memalign((void**)&super, 4096, SUPER1_SIZE) != 0) {
- pr_err("%s could not allocate superblock\n",
- __func__);
+ pr_err("could not allocate superblock\n");
return 1;
}
misc = (struct misc_dev_info*) (((char*)super)+MAX_SB_SIZE+BM_SUPER_SIZE);
misc->device_size = dsize;
+ if (st->data_offset == INVALID_SECTORS)
+ st->data_offset = __le64_to_cpu(super->data_offset);
/* Now check on the bitmap superblock */
if ((__le32_to_cpu(super->feature_map)&MD_FEATURE_BITMAP_OFFSET) == 0)
st->ss = &super1;
st->max_devs = MAX_DEVS;
st->sb = NULL;
+ st->data_offset = INVALID_SECTORS;
/* leading zeros can be safely ignored. --detail generates them. */
while (*arg == '0')
arg++;
* superblock type st, and reserving 'reserve' sectors for
* a possible bitmap
*/
-static __u64 _avail_size1(struct supertype *st, __u64 devsize,
- unsigned long long data_offset, int chunksize)
+static __u64 avail_size1(struct supertype *st, __u64 devsize,
+ unsigned long long data_offset)
{
struct mdp_superblock_1 *super = st->sb;
int bmspace = 0;
+ int bbspace = 0;
if (devsize < 24)
return 0;
- if (super == NULL)
- /* creating: allow suitable space for bitmap */
- bmspace = choose_bm_space(devsize);
#ifndef MDASSEMBLE
- else if (__le32_to_cpu(super->feature_map)&MD_FEATURE_BITMAP_OFFSET) {
+ if (__le32_to_cpu(super->feature_map)&MD_FEATURE_BITMAP_OFFSET) {
/* hot-add. allow for actual size of bitmap */
struct bitmap_super_s *bsb;
bsb = (struct bitmap_super_s *)(((char*)super)+MAX_SB_SIZE);
}
#endif
/* Allow space for bad block log */
- if (super && super->bblog_size)
- devsize -= __le16_to_cpu(super->bblog_size);
- else
- devsize -= 8;
+ if (super->bblog_size)
+ bbspace = __le16_to_cpu(super->bblog_size);
if (st->minor_version < 0)
/* not specified, so time to set default */
st->minor_version = 2;
+ if (data_offset == INVALID_SECTORS)
+ data_offset = st->data_offset;
+
if (data_offset != INVALID_SECTORS)
switch(st->minor_version) {
case 0:
- return devsize - data_offset - 8*2;
+ return devsize - data_offset - 8*2 - bbspace;
case 1:
case 2:
return devsize - data_offset;
devsize -= bmspace;
- if (super == NULL && st->minor_version > 0) {
- /* haven't committed to a size yet, so allow some
- * slack for space for reshape.
- * Limit slack to 128M, but aim for about 0.1%
- */
- unsigned long long headroom = 128*1024*2;
- while ((headroom << 10) > devsize &&
- (chunksize == 0 ||
- headroom / 2 >= ((unsigned)chunksize*2)*2))
- headroom >>= 1;
- devsize -= headroom;
- }
switch(st->minor_version) {
case 0:
/* at end */
- return ((devsize - 8*2 ) & ~(4*2-1));
+ return ((devsize - 8*2 - bbspace ) & ~(4*2-1));
case 1:
/* at start, 4K for superblock and possible bitmap */
- return devsize - 4*2;
+ return devsize - 4*2 - bbspace;
case 2:
/* 4k from start, 4K for superblock and possible bitmap */
- return devsize - (4+4)*2;
+ return devsize - (4+4)*2 - bbspace;
}
return 0;
}
-static __u64 avail_size1(struct supertype *st, __u64 devsize,
- unsigned long long data_offset)
-{
- return _avail_size1(st, devsize, data_offset, 0);
-}
static int
add_internal_bitmap1(struct supertype *st,
* been left.
*/
offset = 0;
- room = choose_bm_space(__le64_to_cpu(sb->size));
bbl_size = 8;
+ room = choose_bm_space(__le64_to_cpu(sb->size)) + bbl_size;
} else {
room = __le64_to_cpu(sb->super_offset)
- __le64_to_cpu(sb->data_offset)
bbl_size = -bbl_offset;
if (!may_change || (room < 3*2 &&
- __le32_to_cpu(sb->max_dev) <= 384)) {
+ __le32_to_cpu(sb->max_dev) <= 384)) {
room = 3*2;
offset = 1*2;
bbl_size = 0;
case 2: /* between superblock and data */
if (creating) {
offset = 4*2;
- room = choose_bm_space(__le64_to_cpu(sb->size));
bbl_size = 8;
+ room = choose_bm_space(__le64_to_cpu(sb->size)) + bbl_size;
} else {
room = __le64_to_cpu(sb->data_offset)
- __le64_to_cpu(sb->super_offset);
/* Limit to 128K of bitmap when chunk size not requested */
room = 128*2;
+ if (room <= 1)
+ /* No room for a bitmap */
+ return 0;
+
max_bits = (room * 512 - sizeof(bitmap_super_t)) * 8;
min_chunk = 4096; /* sub-page chunks don't work yet.. */
bms->daemon_sleep = __cpu_to_le32(delay);
bms->sync_size = __cpu_to_le64(size);
bms->write_behind = __cpu_to_le32(write_behind);
+ bms->nodes = __cpu_to_le32(st->nodes);
+ if (st->nodes)
+ sb->feature_map = __cpu_to_le32(__le32_to_cpu(sb->feature_map)
+ | MD_FEATURE_BITMAP_VERSIONED);
+ if (st->cluster_name)
+ strncpy((char *)bms->cluster_name,
+ st->cluster_name, strlen(st->cluster_name));
*chunkp = chunk;
return 1;
lseek64(fd, offset<<9, 0);
}
-static int write_bitmap1(struct supertype *st, int fd)
+static int write_bitmap1(struct supertype *st, int fd, enum bitmap_update update)
{
struct mdp_superblock_1 *sb = st->sb;
bitmap_super_t *bms = (bitmap_super_t*)(((char*)sb)+MAX_SB_SIZE);
void *buf;
int towrite, n;
struct align_fd afd;
+ unsigned int i = 0;
+ unsigned long long total_bm_space, bm_space_per_node;
+
+ switch (update) {
+ case NameUpdate:
+ /* update cluster name */
+ if (st->cluster_name) {
+ memset((char *)bms->cluster_name, 0, sizeof(bms->cluster_name));
+ strncpy((char *)bms->cluster_name, st->cluster_name, 64);
+ }
+ break;
+ case NodeNumUpdate:
+ /* cluster md only supports superblock 1.2 now */
+ if (st->minor_version != 2) {
+ pr_err("Warning: cluster md only works with superblock 1.2\n");
+ return -EINVAL;
+ }
+
+ /* Each node has an independent bitmap, it is necessary to calculate the
+ * space is enough or not, first get how many bytes for the total bitmap */
+ bm_space_per_node = calc_bitmap_size(bms, 4096);
+
+ total_bm_space = 512 * (__le64_to_cpu(sb->data_offset) - __le64_to_cpu(sb->super_offset));
+ total_bm_space = total_bm_space - 4096; /* leave another 4k for superblock */
+
+ if (bm_space_per_node * st->nodes > total_bm_space) {
+ pr_err("Warning: The max num of nodes can't exceed %llu\n",
+ total_bm_space / bm_space_per_node);
+ return -ENOMEM;
+ }
+
+ bms->nodes = __cpu_to_le32(st->nodes);
+ break;
+ case NoUpdate:
+ default:
+ break;
+ }
init_afd(&afd, fd);
if (posix_memalign(&buf, 4096, 4096))
return -ENOMEM;
- memset(buf, 0xff, 4096);
- memcpy(buf, (char *)bms, sizeof(bitmap_super_t));
-
- towrite = __le64_to_cpu(bms->sync_size) / (__le32_to_cpu(bms->chunksize)>>9);
- towrite = (towrite+7) >> 3; /* bits to bytes */
- towrite += sizeof(bitmap_super_t);
- towrite = ROUND_UP(towrite, 512);
- while (towrite > 0) {
- n = towrite;
- if (n > 4096)
- n = 4096;
- n = awrite(&afd, buf, n);
- if (n > 0)
- towrite -= n;
+ do {
+ /* Only the bitmap[0] should resync
+ * whole device on initial assembly
+ */
+ if (i)
+ memset(buf, 0x00, 4096);
else
+ memset(buf, 0xff, 4096);
+ memcpy(buf, (char *)bms, sizeof(bitmap_super_t));
+
+ towrite = calc_bitmap_size(bms, 4096);
+ while (towrite > 0) {
+ n = towrite;
+ if (n > 4096)
+ n = 4096;
+ n = awrite(&afd, buf, n);
+ if (n > 0)
+ towrite -= n;
+ else
+ break;
+ if (i)
+ memset(buf, 0x00, 4096);
+ else
+ memset(buf, 0xff, 4096);
+ }
+ fsync(fd);
+ if (towrite) {
+ rv = -2;
break;
- memset(buf, 0xff, 4096);
- }
- fsync(fd);
- if (towrite)
- rv = -2;
+ }
+ } while (++i < __le32_to_cpu(bms->nodes));
free(buf);
return rv;
char *subdev, unsigned long long *freesize,
int verbose)
{
- unsigned long long ldsize;
+ unsigned long long ldsize, devsize;
+ int bmspace;
+ unsigned long long headroom;
int fd;
if (level == LEVEL_CONTAINER) {
pr_err("1.x metadata does not support containers\n");
return 0;
}
- if (chunk && *chunk == UnSet)
+ if (*chunk == UnSet)
*chunk = DEFAULT_CHUNK;
if (!subdev)
return 1;
+ if (st->minor_version < 0)
+ /* not specified, so time to set default */
+ st->minor_version = 2;
+
fd = open(subdev, O_RDONLY|O_EXCL, 0);
if (fd < 0) {
if (verbose)
}
close(fd);
- *freesize = _avail_size1(st, ldsize >> 9, data_offset, *chunk);
+ devsize = ldsize >> 9;
+ if (devsize < 24) {
+ *freesize = 0;
+ return 0;
+ }
+
+ /* creating: allow suitable space for bitmap */
+ bmspace = choose_bm_space(devsize);
+
+ if (data_offset == INVALID_SECTORS)
+ data_offset = st->data_offset;
+ if (data_offset == INVALID_SECTORS)
+ switch (st->minor_version) {
+ case 0:
+ data_offset = 0;
+ break;
+ case 1:
+ case 2:
+ /* Choose data offset appropriate for this device
+ * and use as default for whole array.
+ * The data_offset must allow for bitmap space
+ * and base metadata, should allow for some headroom
+ * for reshape, and should be rounded to multiple
+ * of 1M.
+ * Headroom is limited to 128M, but aim for about 0.1%
+ */
+ headroom = 128*1024*2;
+ while ((headroom << 10) > devsize &&
+ (*chunk == 0 ||
+ headroom / 2 >= ((unsigned)(*chunk)*2)*2))
+ headroom >>= 1;
+ data_offset = 12*2 + bmspace + headroom;
+ #define ONE_MEG (2*1024)
+ if (data_offset > ONE_MEG)
+ data_offset = (data_offset / ONE_MEG) * ONE_MEG;
+ break;
+ }
+ if (st->data_offset == INVALID_SECTORS)
+ st->data_offset = data_offset;
+ switch(st->minor_version) {
+ case 0: /* metadata at end. Round down and subtract space to reserve */
+ devsize = (devsize & ~(4ULL*2-1));
+ /* space for metadata, bblog, bitmap */
+ devsize -= 8*2 + 8 + bmspace;
+ break;
+ case 1:
+ case 2:
+ devsize -= data_offset;
+ break;
+ }
+ *freesize = devsize;
return 1;
}
#endif /* MDASSEMBLE */
for (i = 0; i < MD_SB_DISKS; i++) {
int state = sb0->disks[i].state;
- sb->dev_roles[i] = 0xFFFF;
+ sb->dev_roles[i] = MD_DISK_ROLE_SPARE;
if ((state & (1<<MD_DISK_SYNC)) &&
!(state & (1<<MD_DISK_FAULTY)))
sb->dev_roles[i] = __cpu_to_le16(sb0->disks[i].raid_disk);