diff options
author | Joe Thornber <thornber@redhat.com> | 2011-10-31 20:21:18 +0000 |
---|---|---|
committer | Alasdair G Kergon <agk@redhat.com> | 2011-10-31 20:21:18 +0000 |
commit | 991d9fa02da0dd1f843dc011376965e0c8c6c9b5 (patch) | |
tree | a64c94710246b77bb74cd77634581cea3d32cfe1 | |
parent | 3241b1d3e0aaafbfcd320f4d71ade629728cc4f4 (diff) |
dm: add thin provisioning target
Initial EXPERIMENTAL implementation of device-mapper thin provisioning
with snapshot support. The 'thin' target is used to create instances of
the virtual devices that are hosted in the 'thin-pool' target. The
thin-pool target provides data sharing among devices. This sharing is
made possible using the persistent-data library in the previous patch.
The main highlight of this implementation, compared to the previous
implementation of snapshots, is that it allows many virtual devices to
be stored on the same data volume, simplifying administration and
allowing sharing of data between volumes (thus reducing disk usage).
Another big feature is support for arbitrary depth of recursive
snapshots (snapshots of snapshots of snapshots ...). The previous
implementation of snapshots did this by chaining together lookup tables,
and so performance was O(depth). This new implementation uses a single
data structure so we don't get this degradation with depth.
For further information and examples of how to use this, please read
Documentation/device-mapper/thin-provisioning.txt
Signed-off-by: Joe Thornber <thornber@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
-rw-r--r-- | Documentation/device-mapper/thin-provisioning.txt | 285 | ||||
-rw-r--r-- | drivers/md/Kconfig | 28 | ||||
-rw-r--r-- | drivers/md/Makefile | 3 | ||||
-rw-r--r-- | drivers/md/dm-thin-metadata.c | 1391 | ||||
-rw-r--r-- | drivers/md/dm-thin-metadata.h | 156 | ||||
-rw-r--r-- | drivers/md/dm-thin.c | 2428 |
6 files changed, 4291 insertions, 0 deletions
diff --git a/Documentation/device-mapper/thin-provisioning.txt b/Documentation/device-mapper/thin-provisioning.txt new file mode 100644 index 000000000000..801d9d1cf82b --- /dev/null +++ b/Documentation/device-mapper/thin-provisioning.txt @@ -0,0 +1,285 @@ +Introduction +============ + +This document descibes a collection of device-mapper targets that +between them implement thin-provisioning and snapshots. + +The main highlight of this implementation, compared to the previous +implementation of snapshots, is that it allows many virtual devices to +be stored on the same data volume. This simplifies administration and +allows the sharing of data between volumes, thus reducing disk usage. + +Another significant feature is support for an arbitrary depth of +recursive snapshots (snapshots of snapshots of snapshots ...). The +previous implementation of snapshots did this by chaining together +lookup tables, and so performance was O(depth). This new +implementation uses a single data structure to avoid this degradation +with depth. Fragmentation may still be an issue, however, in some +scenarios. + +Metadata is stored on a separate device from data, giving the +administrator some freedom, for example to: + +- Improve metadata resilience by storing metadata on a mirrored volume + but data on a non-mirrored one. + +- Improve performance by storing the metadata on SSD. + +Status +====== + +These targets are very much still in the EXPERIMENTAL state. Please +do not yet rely on them in production. But do experiment and offer us +feedback. Different use cases will have different performance +characteristics, for example due to fragmentation of the data volume. + +If you find this software is not performing as expected please mail +dm-devel@redhat.com with details and we'll try our best to improve +things for you. + +Userspace tools for checking and repairing the metadata are under +development. + +Cookbook +======== + +This section describes some quick recipes for using thin provisioning. +They use the dmsetup program to control the device-mapper driver +directly. End users will be advised to use a higher-level volume +manager such as LVM2 once support has been added. + +Pool device +----------- + +The pool device ties together the metadata volume and the data volume. +It maps I/O linearly to the data volume and updates the metadata via +two mechanisms: + +- Function calls from the thin targets + +- Device-mapper 'messages' from userspace which control the creation of new + virtual devices amongst other things. + +Setting up a fresh pool device +------------------------------ + +Setting up a pool device requires a valid metadata device, and a +data device. If you do not have an existing metadata device you can +make one by zeroing the first 4k to indicate empty metadata. + + dd if=/dev/zero of=$metadata_dev bs=4096 count=1 + +The amount of metadata you need will vary according to how many blocks +are shared between thin devices (i.e. through snapshots). If you have +less sharing than average you'll need a larger-than-average metadata device. + +As a guide, we suggest you calculate the number of bytes to use in the +metadata device as 48 * $data_dev_size / $data_block_size but round it up +to 2MB if the answer is smaller. The largest size supported is 16GB. + +If you're creating large numbers of snapshots which are recording large +amounts of change, you may need find you need to increase this. + +Reloading a pool table +---------------------- + +You may reload a pool's table, indeed this is how the pool is resized +if it runs out of space. (N.B. While specifying a different metadata +device when reloading is not forbidden at the moment, things will go +wrong if it does not route I/O to exactly the same on-disk location as +previously.) + +Using an existing pool device +----------------------------- + + dmsetup create pool \ + --table "0 20971520 thin-pool $metadata_dev $data_dev \ + $data_block_size $low_water_mark" + +$data_block_size gives the smallest unit of disk space that can be +allocated at a time expressed in units of 512-byte sectors. People +primarily interested in thin provisioning may want to use a value such +as 1024 (512KB). People doing lots of snapshotting may want a smaller value +such as 128 (64KB). If you are not zeroing newly-allocated data, +a larger $data_block_size in the region of 256000 (128MB) is suggested. +$data_block_size must be the same for the lifetime of the +metadata device. + +$low_water_mark is expressed in blocks of size $data_block_size. If +free space on the data device drops below this level then a dm event +will be triggered which a userspace daemon should catch allowing it to +extend the pool device. Only one such event will be sent. +Resuming a device with a new table itself triggers an event so the +userspace daemon can use this to detect a situation where a new table +already exceeds the threshold. + +Thin provisioning +----------------- + +i) Creating a new thinly-provisioned volume. + + To create a new thinly- provisioned volume you must send a message to an + active pool device, /dev/mapper/pool in this example. + + dmsetup message /dev/mapper/pool 0 "create_thin 0" + + Here '0' is an identifier for the volume, a 24-bit number. It's up + to the caller to allocate and manage these identifiers. If the + identifier is already in use, the message will fail with -EEXIST. + +ii) Using a thinly-provisioned volume. + + Thinly-provisioned volumes are activated using the 'thin' target: + + dmsetup create thin --table "0 2097152 thin /dev/mapper/pool 0" + + The last parameter is the identifier for the thinp device. + +Internal snapshots +------------------ + +i) Creating an internal snapshot. + + Snapshots are created with another message to the pool. + + N.B. If the origin device that you wish to snapshot is active, you + must suspend it before creating the snapshot to avoid corruption. + This is NOT enforced at the moment, so please be careful! + + dmsetup suspend /dev/mapper/thin + dmsetup message /dev/mapper/pool 0 "create_snap 1 0" + dmsetup resume /dev/mapper/thin + + Here '1' is the identifier for the volume, a 24-bit number. '0' is the + identifier for the origin device. + +ii) Using an internal snapshot. + + Once created, the user doesn't have to worry about any connection + between the origin and the snapshot. Indeed the snapshot is no + different from any other thinly-provisioned device and can be + snapshotted itself via the same method. It's perfectly legal to + have only one of them active, and there's no ordering requirement on + activating or removing them both. (This differs from conventional + device-mapper snapshots.) + + Activate it exactly the same way as any other thinly-provisioned volume: + + dmsetup create snap --table "0 2097152 thin /dev/mapper/pool 1" + +Deactivation +------------ + +All devices using a pool must be deactivated before the pool itself +can be. + + dmsetup remove thin + dmsetup remove snap + dmsetup remove pool + +Reference +========= + +'thin-pool' target +------------------ + +i) Constructor + + thin-pool <metadata dev> <data dev> <data block size (sectors)> \ + <low water mark (blocks)> [<number of feature args> [<arg>]*] + + Optional feature arguments: + - 'skip_block_zeroing': skips the zeroing of newly-provisioned blocks. + + Data block size must be between 64KB (128 sectors) and 1GB + (2097152 sectors) inclusive. + + +ii) Status + + <transaction id> <used metadata blocks>/<total metadata blocks> + <used data blocks>/<total data blocks> <held metadata root> + + + transaction id: + A 64-bit number used by userspace to help synchronise with metadata + from volume managers. + + used data blocks / total data blocks + If the number of free blocks drops below the pool's low water mark a + dm event will be sent to userspace. This event is edge-triggered and + it will occur only once after each resume so volume manager writers + should register for the event and then check the target's status. + + held metadata root: + The location, in sectors, of the metadata root that has been + 'held' for userspace read access. '-' indicates there is no + held root. This feature is not yet implemented so '-' is + always returned. + +iii) Messages + + create_thin <dev id> + + Create a new thinly-provisioned device. + <dev id> is an arbitrary unique 24-bit identifier chosen by + the caller. + + create_snap <dev id> <origin id> + + Create a new snapshot of another thinly-provisioned device. + <dev id> is an arbitrary unique 24-bit identifier chosen by + the caller. + <origin id> is the identifier of the thinly-provisioned device + of which the new device will be a snapshot. + + delete <dev id> + + Deletes a thin device. Irreversible. + + trim <dev id> <new size in sectors> + + Delete mappings from the end of a thin device. Irreversible. + You might want to use this if you're reducing the size of + your thinly-provisioned device. In many cases, due to the + sharing of blocks between devices, it is not possible to + determine in advance how much space 'trim' will release. (In + future a userspace tool might be able to perform this + calculation.) + + set_transaction_id <current id> <new id> + + Userland volume managers, such as LVM, need a way to + synchronise their external metadata with the internal metadata of the + pool target. The thin-pool target offers to store an + arbitrary 64-bit transaction id and return it on the target's + status line. To avoid races you must provide what you think + the current transaction id is when you change it with this + compare-and-swap message. + +'thin' target +------------- + +i) Constructor + + thin <pool dev> <dev id> + + pool dev: + the thin-pool device, e.g. /dev/mapper/my_pool or 253:0 + + dev id: + the internal device identifier of the device to be + activated. + +The pool doesn't store any size against the thin devices. If you +load a thin target that is smaller than you've been using previously, +then you'll have no access to blocks mapped beyond the end. If you +load a target that is bigger than before, then extra blocks will be +provisioned as and when needed. + +If you wish to reduce the size of your thin device and potentially +regain some space then send the 'trim' message to the pool. + +ii) Status + + <nr mapped sectors> <highest mapped sector> diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index b1a921497043..faa4741df6d3 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -216,6 +216,8 @@ config DM_BUFIO as a cache, holding recently-read blocks in memory and performing delayed writes. +source "drivers/md/persistent-data/Kconfig" + config DM_CRYPT tristate "Crypt target support" depends on BLK_DEV_DM @@ -241,6 +243,32 @@ config DM_SNAPSHOT ---help--- Allow volume managers to take writable snapshots of a device. +config DM_THIN_PROVISIONING + tristate "Thin provisioning target (EXPERIMENTAL)" + depends on BLK_DEV_DM && EXPERIMENTAL + select DM_PERSISTENT_DATA + ---help--- + Provides thin provisioning and snapshots that share a data store. + +config DM_DEBUG_BLOCK_STACK_TRACING + boolean "Keep stack trace of thin provisioning block lock holders" + depends on STACKTRACE_SUPPORT && DM_THIN_PROVISIONING + select STACKTRACE + ---help--- + Enable this for messages that may help debug problems with the + block manager locking used by thin provisioning. + + If unsure, say N. + +config DM_DEBUG_SPACE_MAPS + boolean "Extra validation for thin provisioning space maps" + depends on DM_THIN_PROVISIONING + ---help--- + Enable this for messages that may help debug problems with the + space maps used by thin provisioning. + + If unsure, say N. + config DM_MIRROR tristate "Mirror target" depends on BLK_DEV_DM diff --git a/drivers/md/Makefile b/drivers/md/Makefile index 56661c4272f2..046860c7a166 100644 --- a/drivers/md/Makefile +++ b/drivers/md/Makefile @@ -10,6 +10,7 @@ dm-snapshot-y += dm-snap.o dm-exception-store.o dm-snap-transient.o \ dm-mirror-y += dm-raid1.o dm-log-userspace-y \ += dm-log-userspace-base.o dm-log-userspace-transfer.o +dm-thin-pool-y += dm-thin.o dm-thin-metadata.o md-mod-y += md.o bitmap.o raid456-y += raid5.o @@ -35,10 +36,12 @@ obj-$(CONFIG_DM_MULTIPATH) += dm-multipath.o dm-round-robin.o obj-$(CONFIG_DM_MULTIPATH_QL) += dm-queue-length.o obj-$(CONFIG_DM_MULTIPATH_ST) += dm-service-time.o obj-$(CONFIG_DM_SNAPSHOT) += dm-snapshot.o +obj-$(CONFIG_DM_PERSISTENT_DATA) += persistent-data/ obj-$(CONFIG_DM_MIRROR) += dm-mirror.o dm-log.o dm-region-hash.o obj-$(CONFIG_DM_LOG_USERSPACE) += dm-log-userspace.o obj-$(CONFIG_DM_ZERO) += dm-zero.o obj-$(CONFIG_DM_RAID) += dm-raid.o +obj-$(CONFIG_DM_THIN_PROVISIONING) += dm-thin-pool.o ifeq ($(CONFIG_DM_UEVENT),y) dm-mod-objs += dm-uevent.o diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c new file mode 100644 index 000000000000..59c4f0446ffa --- /dev/null +++ b/drivers/md/dm-thin-metadata.c @@ -0,0 +1,1391 @@ +/* + * Copyright (C) 2011 Red Hat, Inc. + * + * This file is released under the GPL. + */ + +#include "dm-thin-metadata.h" +#include "persistent-data/dm-btree.h" +#include "persistent-data/dm-space-map.h" +#include "persistent-data/dm-space-map-disk.h" +#include "persistent-data/dm-transaction-manager.h" + +#include <linux/list.h> +#include <linux/device-mapper.h> +#include <linux/workqueue.h> + +/*-------------------------------------------------------------------------- + * As far as the metadata goes, there is: + * + * - A superblock in block zero, taking up fewer than 512 bytes for + * atomic writes. + * + * - A space map managing the metadata blocks. + * + * - A space map managing the data blocks. + * + * - A btree mapping our internal thin dev ids onto struct disk_device_details. + * + * - A hierarchical btree, with 2 levels which effectively maps (thin + * dev id, virtual block) -> block_time. Block time is a 64-bit + * field holding the time in the low 24 bits, and block in the top 48 + * bits. + * + * BTrees consist solely of btree_nodes, that fill a block. Some are + * internal nodes, as such their values are a __le64 pointing to other + * nodes. Leaf nodes can store data of any reasonable size (ie. much + * smaller than the block size). The nodes consist of the header, + * followed by an array of keys, followed by an array of values. We have + * to binary search on the keys so they're all held together to help the + * cpu cache. + * + * Space maps have 2 btrees: + * + * - One maps a uint64_t onto a struct index_entry. Which points to a + * bitmap block, and has some details about how many free entries there + * are etc. + * + * - The bitmap blocks have a header (for the checksum). Then the rest + * of the block is pairs of bits. With the meaning being: + * + * 0 - ref count is 0 + * 1 - ref count is 1 + * 2 - ref count is 2 + * 3 - ref count is higher than 2 + * + * - If the count is higher than 2 then the ref count is entered in a + * second btree that directly maps the block_address to a uint32_t ref + * count. + * + * The space map metadata variant doesn't have a bitmaps btree. Instead + * it has one single blocks worth of index_entries. This avoids + * recursive issues with the bitmap btree needing to allocate space in + * order to insert. With a small data block size such as 64k the + * metadata support data devices that are hundreds of terrabytes. + * + * The space maps allocate space linearly from front to back. Space that + * is freed in a transaction is never recycled within that transaction. + * To try and avoid fragmenting _free_ space the allocator always goes + * back and fills in gaps. + * + * All metadata io is in THIN_METADATA_BLOCK_SIZE sized/aligned chunks + * from the block manager. + *--------------------------------------------------------------------------*/ + +#define DM_MSG_PREFIX "thin metadata" + +#define THIN_SUPERBLOCK_MAGIC 27022010 +#define THIN_SUPERBLOCK_LOCATION 0 +#define THIN_VERSION 1 +#define THIN_METADATA_CACHE_SIZE 64 +#define SECTOR_TO_BLOCK_SHIFT 3 + +/* This should be plenty */ +#define SPACE_MAP_ROOT_SIZE 128 + +/* + * Little endian on-disk superblock and device details. + */ +struct thin_disk_superblock { + __le32 csum; /* Checksum of superblock except for this field. */ + __le32 flags; + __le64 blocknr; /* This block number, dm_block_t. */ + + __u8 uuid[16]; + __le64 magic; + __le32 version; + __le32 time; + + __le64 trans_id; + + /* + * Root held by userspace transactions. + */ + __le64 held_root; + + __u8 data_space_map_root[SPACE_MAP_ROOT_SIZE]; + __u8 metadata_space_map_root[SPACE_MAP_ROOT_SIZE]; + + /* + * 2-level btree mapping (dev_id, (dev block, time)) -> data block + */ + __le64 data_mapping_root; + + /* + * Device detail root mapping dev_id -> device_details + */ + __le64 device_details_root; + + __le32 data_block_size; /* In 512-byte sectors. */ + + __le32 metadata_block_size; /* In 512-byte sectors. */ + __le64 metadata_nr_blocks; + + __le32 compat_flags; + __le32 compat_ro_flags; + __le32 incompat_flags; +} __packed; + +struct disk_device_details { + __le64 mapped_blocks; + __le64 transaction_id; /* When created. */ + __le32 creation_time; + __le32 snapshotted_time; +} __packed; + +struct dm_pool_metadata { + struct hlist_node hash; + + struct block_device *bdev; + struct dm_block_manager *bm; + struct dm_space_map *metadata_sm; + struct dm_space_map *data_sm; + struct dm_transaction_manager *tm; + struct dm_transaction_manager *nb_tm; + + /* + * Two-level btree. + * First level holds thin_dev_t. + * Second level holds mappings. + */ + struct dm_btree_info info; + + /* + * Non-blocking version of the above. + */ + struct dm_btree_info nb_info; + + /* + * Just the top level for deleting whole devices. + */ + struct dm_btree_info tl_info; + + /* + * Just the bottom level for creating new devices. + */ + struct dm_btree_info bl_info; + + /* + * Describes the device details btree. + */ + struct dm_btree_info details_info; + + struct rw_semaphore root_lock; + uint32_t time; + int need_commit; + dm_block_t root; + dm_block_t details_root; + struct list_head thin_devices; + uint64_t trans_id; + unsigned long flags; + sector_t data_block_size; +}; + +struct dm_thin_device { + struct list_head list; + struct dm_pool_metadata *pmd; + dm_thin_id id; + + int open_count; + int changed; + uint64_t mapped_blocks; + uint64_t transaction_id; + uint32_t creation_time; + uint32_t snapshotted_time; +}; + +/*---------------------------------------------------------------- + * superblock validator + *--------------------------------------------------------------*/ + +#define SUPERBLOCK_CSUM_XOR 160774 + +static void sb_prepare_for_write(struct dm_block_validator *v, + struct dm_block *b, + size_t block_size) +{ + struct thin_disk_superblock *disk_super = dm_block_data(b); + + disk_super->blocknr = cpu_to_le64(dm_block_location(b)); + disk_super->csum = cpu_to_le32(dm_bm_checksum(&disk_super->flags, + block_size - sizeof(__le32), + SUPERBLOCK_CSUM_XOR)); +} + +static int sb_check(struct dm_block_validator *v, + struct dm_block *b, + size_t block_size) +{ + struct thin_disk_superblock *disk_super = dm_block_data(b); + __le32 csum_le; + + if (dm_block_location(b) != le64_to_cpu(disk_super->blocknr)) { + DMERR("sb_check failed: blocknr %llu: " + "wanted %llu", le64_to_cpu(disk_super->blocknr), + (unsigned long long)dm_block_location(b)); + return -ENOTBLK; + } + + if (le64_to_cpu(disk_super->magic) != THIN_SUPERBLOCK_MAGIC) { + DMERR("sb_check failed: magic %llu: " + "wanted %llu", le64_to_cpu(disk_super->magic), + (unsigned long long)THIN_SUPERBLOCK_MAGIC); + return -EILSEQ; + } + + csum_le = cpu_to_le32(dm_bm_checksum(&disk_super->flags, + block_size - sizeof(__le32), + SUPERBLOCK_CSUM_XOR)); + if (csum_le != disk_super->csum) { + DMERR("sb_check failed: csum %u: wanted %u", + le32_to_cpu(csum_le), le32_to_cpu(disk_super->csum)); + return -EILSEQ; + } + + return 0; +} + +static struct dm_block_validator sb_validator = { + .name = "superblock", + .prepare_for_write = sb_prepare_for_write, + .check = sb_check +}; + +/*---------------------------------------------------------------- + * Methods for the btree value types + *--------------------------------------------------------------*/ + +static uint64_t pack_block_time(dm_block_t b, uint32_t t) +{ + return (b << 24) | t; +} + +static void unpack_block_time(uint64_t v, dm_block_t *b, uint32_t *t) +{ + *b = v >> 24; + *t = v & ((1 << 24) - 1); +} + +static void data_block_inc(void *context, void *value_le) +{ + struct dm_space_map *sm = context; + __le64 v_le; + uint64_t b; + uint32_t t; + + memcpy(&v_le, value_le, sizeof(v_le)); + unpack_block_time(le64_to_cpu(v_le), &b, &t); + dm_sm_inc_block(sm, b); +} + +static void data_block_dec(void *context, void *value_le) +{ + struct dm_space_map *sm = context; + __le64 v_le; + uint64_t b; + uint32_t t; + + memcpy(&v_le, value_le, sizeof(v_le)); + unpack_block_time(le64_to_cpu(v_le), &b, &t); + dm_sm_dec_block(sm, b); +} + +static int data_block_equal(void *context, void *value1_le, void *value2_le) +{ + __le64 v1_le, v2_le; + uint64_t b1, b2; + uint32_t t; + + memcpy(&v1_le, value1_le, sizeof(v1_le)); + memcpy(&v2_le, value2_le, sizeof(v2_le)); + unpack_block_time(le64_to_cpu(v1_le), &b1, &t); + unpack_block_time(le64_to_cpu(v2_le), &b2, &t); + + return b1 == b2; +} + +static void subtree_inc(void *context, void *value) +{ + struct dm_btree_info *info = context; + __le64 root_le; + uint64_t root; + + memcpy(&root_le, value, sizeof(root_le)); + root = le64_to_cpu(root_le); + dm_tm_inc(info->tm, root); +} + +static void subtree_dec(void *context, void *value) +{ + struct dm_btree_info *info = context; + __le64 root_le; + uint64_t root; + + memcpy(&root_le, value, sizeof(root_le)); + root = le64_to_cpu(root_le); + if (dm_btree_del(info, root)) + DMERR("btree delete failed\n"); +} + +static int subtree_equal(void *context, void *value1_le, void *value2_le) +{ + __le64 v1_le, v2_le; + memcpy(&v1_le, value1_le, sizeof(v1_le)); + memcpy(&v2_le, value2_le, sizeof(v2_le)); + + return v1_le == v2_le; +} + +/*----------------------------------------------------------------*/ + +static int superblock_all_zeroes(struct dm_block_manager *bm, int *result) +{ + int r; + unsigned i; + struct dm_block *b; + __le64 *data_le, zero = cpu_to_le64(0); + unsigned block_size = dm_bm_block_size(bm) / sizeof(__le64); + + /* + * We can't use a validator here - it may be all zeroes. + */ + r = dm_bm_read_lock(bm, THIN_SUPERBLOCK_LOCATION, NULL, &b); + if (r) + return r; + + data_le = dm_block_data(b); + *result = 1; + for (i = 0; i < block_size; i++) { + if (data_le[i] != zero) { + *result = 0; + break; + } + } + + return dm_bm_unlock(b); +} + +static int init_pmd(struct dm_pool_metadata *pmd, + struct dm_block_manager *bm, + dm_block_t nr_blocks, int create) +{ + int r; + struct dm_space_map *sm, *data_sm; + struct dm_transaction_manager *tm; + struct dm_block *sblock; + + if (create) { + r = dm_tm_create_with_sm(bm, THIN_SUPERBLOCK_LOCATION, + &sb_validator, &tm, &sm, &sblock); + if (r < 0) { + DMERR("tm_create_with_sm failed"); + return r; + } + + data_sm = dm_sm_disk_create(tm, nr_blocks); + if (IS_ERR(data_sm)) { + DMERR("sm_disk_create failed"); + r = PTR_ERR(data_sm); + goto bad; + } + } else { + struct thin_disk_superblock *disk_super = NULL; + size_t space_map_root_offset = + offsetof(struct thin_disk_superblock, metadata_space_map_root); + + r = dm_tm_open_with_sm(bm, THIN_SUPERBLOCK_LOCATION, + &sb_validator, space_map_root_offset, + SPACE_MAP_ROOT_SIZE, &tm, &sm, &sblock); + if (r < 0) { + DMERR("tm_open_with_sm failed"); + return r; + } + + disk_super = dm_block_data(sblock); + data_sm = dm_sm_disk_open(tm, disk_super->data_space_map_root, + sizeof(disk_super->data_space_map_root)); + if (IS_ERR(data_sm)) { + DMERR("sm_disk_open failed"); + r = PTR_ERR(data_sm); + goto bad; + } + } + + + r = dm_tm_unlock(tm, sblock); + if (r < 0) { + DMERR("couldn't unlock superblock"); + goto bad_data_sm; + } + + pmd->bm = bm; + pmd->metadata_sm = sm; + pmd->data_sm = data_sm; + pmd->tm = tm; + pmd->nb_tm = dm_tm_create_non_blocking_clone(tm); + if (!pmd->nb_tm) { + DMERR("could not create clone tm"); + r = -ENOMEM; + goto bad_data_sm; + } + + pmd->info.tm = tm; + pmd->info.levels = 2; + pmd->info.value_type.context = pmd->data_sm; + pmd->info.value_type.size = sizeof(__le64); + pmd->info.value_type.inc = data_block_inc; + pmd->info.value_type.dec = data_block_dec; + pmd->info.value_type.equal = data_block_equal; + + memcpy(&pmd->nb_info, &pmd->info, sizeof(pmd->nb_info)); + pmd->nb_info.tm = pmd->nb_tm; + + pmd->tl_info.tm = tm; + pmd->tl_info.levels = 1; + pmd->tl_info.value_type.context = &pmd->info; + pmd->tl_info.value_type.size = sizeof(__le64); + pmd->tl_info.value_type.inc = subtree_inc; + pmd->tl_info.value_type.dec = subtree_dec; + pmd->tl_info.value_type.equal = subtree_equal; + + pmd->bl_info.tm = tm; + pmd->bl_info.levels = 1; + pmd->bl_info.value_type.context = pmd->data_sm; + pmd->bl_info.value_type.size = sizeof(__le64); + pmd->bl_info.value_type.inc = data_block_inc; + pmd->bl_info.value_type.dec = data_block_dec; + pmd->bl_info.value_type.equal = data_block_equal; + + pmd->details_info.tm = tm; + pmd->details_info.levels = 1; + pmd->details_info.value_type.context = NULL; + pmd->details_info.value_type.size = sizeof(struct disk_device_details); + pmd->details_info.value_type.inc = NULL; + pmd->details_info.value_type.dec = NULL; + pmd->details_info.value_type.equal = NULL; + + pmd->root = 0; + + init_rwsem(&pmd->root_lock); + pmd->time = 0; + pmd->need_commit = 0; + pmd->details_root = 0; + pmd->trans_id = 0; + pmd->flags = 0; + INIT_LIST_HEAD(&pmd->thin_devices); + + return 0; + +bad_data_sm: + dm_sm_destroy(data_sm); +bad: + dm_tm_destroy(tm); + dm_sm_destroy(sm); + + return r; +} + +static int __begin_transaction(struct dm_pool_metadata *pmd) +{ + int r; + u32 features; + struct thin_disk_superblock *disk_super; + struct dm_block *sblock; + + /* + * __maybe_commit_transaction() resets these + */ + WARN_ON(pmd->need_commit); + + /* + * We re-read the superblock every time. Shouldn't need to do this + * really. + */ + r = dm_bm_read_lock(pmd->bm, THIN_SUPERBLOCK_LOCATION, + &sb_validator, &sblock); + if (r) + return r; + + disk_super = dm_block_data(sblock); + pmd->time = le32_to_cpu(disk_super->time); + pmd->root = le64_to_cpu(disk_super->data_mapping_root); + pmd->details_root = le64_to_cpu(disk_super->device_details_root); + pmd->trans_id = le64_to_cpu(disk_super->trans_id); + pmd->flags = le32_to_cpu(disk_super->flags); + pmd->data_block_size = le32_to_cpu(disk_super->data_block_size); + + features = le32_to_cpu(disk_super->incompat_flags) & ~THIN_FEATURE_INCOMPAT_SUPP; + if (features) { + DMERR("could not access metadata due to " + "unsupported optional features (%lx).", + (unsigned long)features); + r = -EINVAL; + goto out; + } + + /* + * Check for read-only metadata to skip the following RDWR checks. + */ + if (get_disk_ro(pmd->bdev->bd_disk)) + goto out; + + features = le32_to_cpu(disk_super->compat_ro_flags) & ~THIN_FEATURE_COMPAT_RO_SUPP; + if (features) { + DMERR("could not access metadata RDWR due to " + "unsupported optional features (%lx).", + (unsigned long)features); + r = -EINVAL; + } + +out: + dm_bm_unlock(sblock); + return r; +} + +static int __write_changed_details(struct dm_pool_metadata *pmd) +{ + int r; + struct dm_thin_device *td, *tmp; + struct disk_device_details details; + uint64_t key; + + list_for_each_entry_safe(td, tmp, &pmd->thin_devices, list) { + if (!td->changed) + continue; + + key = td->id; + + details.mapped_blocks = cpu_to_le64(td->mapped_blocks); + details.transaction_id = cpu_to_le64(td->transaction_id); + details.creation_time = cpu_to_le32(td->creation_time); + details.snapshotted_time = cpu_to_le32(td->snapshotted_time); + __dm_bless_for_disk(&details); + + r = dm_btree_insert(&pmd->details_info, pmd->details_root, + &key, &details, &pmd->details_root); + if (r) + return r; + + if (td->open_count) + td->changed = 0; + else { + list_del(&td->list); + kfree(td); + } + + pmd->need_commit = 1; + } + + return 0; +} + +static int __commit_transaction(struct dm_pool_metadata *pmd) +{ + /* + * FIXME: Associated pool should be made read-only on failure. + */ + int r; + size_t metadata_len, data_len; + struct thin_disk_superblock *disk_super; + struct dm_block *sblock; + + /* + * We need to know if the thin_disk_superblock exceeds a 512-byte sector. + */ + BUILD_BUG_ON(sizeof(struct thin_disk_superblock) > 512); + + r = __write_changed_details(pmd); + if (r < 0) + goto out; + + if (!pmd->need_commit) + goto out; + + r = dm_sm_commit(pmd->data_sm); + if (r < 0) + goto out; + + r = dm_tm_pre_commit(pmd->tm); + if (r < 0) + goto out; + + r = dm_sm_root_size(pmd->metadata_sm, &metadata_len); + if (r < 0) + goto out; + + r = dm_sm_root_size(pmd->metadata_sm, &data_len); + if (r < 0) + goto out; + + r = dm_bm_write_lock(pmd->bm, THIN_SUPERBLOCK_LOCATION, + &sb_validator, &sblock); + if (r) + goto out; + + disk_super = dm_block_data(sblock); + disk_super->time = cpu_to_le32(pmd->time); + disk_super->data_mapping_root = cpu_to_le64(pmd->root); + disk_super->device_details_root = cpu_to_le64(pmd->details_root); + disk_super->trans_id = cpu_to_le64(pmd->trans_id); + disk_super->flags = cpu_to_le32(pmd->flags); + + r = dm_sm_copy_root(pmd->metadata_sm, &disk_super->metadata_space_map_root, + metadata_len); + if (r < 0) + goto out_locked; + + r = dm_sm_copy_root(pmd->data_sm, &disk_super->data_space_map_root, + data_len); + if (r < 0) + goto out_locked; + + r = dm_tm_commit(pmd->tm, sblock); + if (!r) + pmd->need_commit = 0; + +out: + return r; + +out_locked: + dm_bm_unlock(sblock); + return r; +} + +struct dm_pool_metadata *dm_pool_metadata_open(struct block_device *bdev, + sector_t data_block_size) +{ + int r; + struct thin_disk_superblock *disk_super; + struct dm_pool_metadata *pmd; + sector_t bdev_size = i_size_read(bdev->bd_inode) >> SECTOR_SHIFT; + struct dm_block_manager *bm; + int create; + struct dm_block *sblock; + + pmd = kmalloc(sizeof(*pmd), GFP_KERNEL); + if (!pmd) { + DMERR("could not allocate metadata struct"); + return ERR_PTR(-ENOMEM); + } + + /* + * Max hex locks: + * 3 for btree insert + + * 2 for btree lookup used within space map + */ + bm = dm_block_manager_create(bdev, THIN_METADATA_BLOCK_SIZE, + THIN_METADATA_CACHE_SIZE, 5); + if (!bm) { + DMERR("could not create block manager"); + kfree(pmd); + return ERR_PTR(-ENOMEM); + } + + r = superblock_all_zeroes(bm, &create); + if (r) { + dm_block_manager_destroy(bm); + kfree(pmd); + return ERR_PTR(r); + } + + + r = init_pmd(pmd, bm, 0, create); + if (r) { + dm_block_manager_destroy(bm); + kfree(pmd); + return ERR_PTR(r); + } + pmd->bdev = bdev; + + if (!create) { + r = __begin_transaction(pmd); + if (r < 0) + goto bad; + return pmd; + } + + /* + * Create. + */ + r = dm_bm_write_lock(pmd->bm, THIN_SUPERBLOCK_ |