X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libbcachefs%2Fbcachefs_format.h;h=1bb24aa7352800a9660c513c028a865055e61ab0;hb=7a716b76b5963dc2d158883f4497bab221932412;hp=94b5418587e364591b29643fea73b3eb53a5103a;hpb=d2a118d921dfdf43adfa37aed1d9df62925bda66;p=bcachefs-tools-debian diff --git a/libbcachefs/bcachefs_format.h b/libbcachefs/bcachefs_format.h index 94b5418..1bb24aa 100644 --- a/libbcachefs/bcachefs_format.h +++ b/libbcachefs/bcachefs_format.h @@ -76,11 +76,31 @@ #include #include #include +#include "vstructs.h" + +#ifdef __KERNEL__ +typedef uuid_t __uuid_t; +#endif + +#define BITMASK(name, type, field, offset, end) \ +static const __maybe_unused unsigned name##_OFFSET = offset; \ +static const __maybe_unused unsigned name##_BITS = (end - offset); \ + \ +static inline __u64 name(const type *k) \ +{ \ + return (k->field >> offset) & ~(~0ULL << (end - offset)); \ +} \ + \ +static inline void SET_##name(type *k, __u64 v) \ +{ \ + k->field &= ~(~(~0ULL << (end - offset)) << offset); \ + k->field |= (v & ~(~0ULL << (end - offset))) << offset; \ +} #define LE_BITMASK(_bits, name, type, field, offset, end) \ -static const unsigned name##_OFFSET = offset; \ -static const unsigned name##_BITS = (end - offset); \ -static const __u##_bits name##_MAX = (1ULL << (end - offset)) - 1; \ +static const __maybe_unused unsigned name##_OFFSET = offset; \ +static const __maybe_unused unsigned name##_BITS = (end - offset); \ +static const __maybe_unused __u##_bits name##_MAX = (1ULL << (end - offset)) - 1;\ \ static inline __u64 name(const type *k) \ { \ @@ -131,26 +151,30 @@ struct bpos { #else #error edit for your odd byteorder. #endif -} __attribute__((packed, aligned(4))); +} __packed +#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ +__aligned(4) +#endif +; #define KEY_INODE_MAX ((__u64)~0ULL) #define KEY_OFFSET_MAX ((__u64)~0ULL) #define KEY_SNAPSHOT_MAX ((__u32)~0U) #define KEY_SIZE_MAX ((__u32)~0U) -static inline struct bpos POS(__u64 inode, __u64 offset) +static inline struct bpos SPOS(__u64 inode, __u64 offset, __u32 snapshot) { - struct bpos ret; - - ret.inode = inode; - ret.offset = offset; - ret.snapshot = 0; - - return ret; + return (struct bpos) { + .inode = inode, + .offset = offset, + .snapshot = snapshot, + }; } -#define POS_MIN POS(0, 0) -#define POS_MAX POS(KEY_INODE_MAX, KEY_OFFSET_MAX) +#define POS_MIN SPOS(0, 0, 0) +#define POS_MAX SPOS(KEY_INODE_MAX, KEY_OFFSET_MAX, 0) +#define SPOS_MAX SPOS(KEY_INODE_MAX, KEY_OFFSET_MAX, KEY_SNAPSHOT_MAX) +#define POS(_inode, _offset) SPOS(_inode, _offset, 0) /* Empty placeholder struct, for container_of() */ struct bch_val { @@ -165,7 +189,7 @@ struct bversion { __u32 hi; __u64 lo; #endif -} __attribute__((packed, aligned(4))); +} __packed __aligned(4); struct bkey { /* Size of combined key and value, in u64s */ @@ -198,7 +222,7 @@ struct bkey { __u8 pad[1]; #endif -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); struct bkey_packed { __u64 _data[0]; @@ -232,7 +256,12 @@ struct bkey_packed { * to the same size as struct bkey should hopefully be safest. */ __u8 pad[sizeof(struct bkey) - 3]; -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); + +typedef struct { + __le64 lo; + __le64 hi; +} bch_le128; #define BKEY_U64s (sizeof(struct bkey) / sizeof(__u64)) #define BKEY_U64s_MAX U8_MAX @@ -274,18 +303,17 @@ enum bch_bkey_fields { struct bkey_i { __u64 _data[0]; - union { - struct { - /* Size of combined key and value, in u64s */ - __u8 u64s; - }; - struct { - struct bkey k; - struct bch_val v; - }; - }; + struct bkey k; + struct bch_val v; }; +#define POS_KEY(_pos) \ +((struct bkey) { \ + .u64s = BKEY_U64s, \ + .format = KEY_FORMAT_CURRENT, \ + .p = _pos, \ +}) + #define KEY(_inode, _offset, _size) \ ((struct bkey) { \ .u64s = BKEY_U64s, \ @@ -302,7 +330,7 @@ static inline void bkey_init(struct bkey *k) #define bkey_bytes(_k) ((_k)->u64s * sizeof(__u64)) #define __BKEY_PADDED(key, pad) \ - struct { struct bkey_i key; __u64 key ## _pad[pad]; } + struct bkey_i key; __u64 key ## _pad[pad] /* * - DELETED keys are used internally to mark keys that should be ignored but @@ -320,13 +348,13 @@ static inline void bkey_init(struct bkey *k) * number. * * - WHITEOUT: for hash table btrees -*/ + */ #define BCH_BKEY_TYPES() \ x(deleted, 0) \ - x(discard, 1) \ + x(whiteout, 1) \ x(error, 2) \ x(cookie, 3) \ - x(whiteout, 4) \ + x(hash_whiteout, 4) \ x(btree_ptr, 5) \ x(extent, 6) \ x(reservation, 7) \ @@ -341,7 +369,21 @@ static inline void bkey_init(struct bkey *k) x(reflink_v, 16) \ x(inline_data, 17) \ x(btree_ptr_v2, 18) \ - x(indirect_inline_data, 19) + x(indirect_inline_data, 19) \ + x(alloc_v2, 20) \ + x(subvolume, 21) \ + x(snapshot, 22) \ + x(inode_v2, 23) \ + x(alloc_v3, 24) \ + x(set, 25) \ + x(lru, 26) \ + x(alloc_v4, 27) \ + x(backpointer, 28) \ + x(inode_v3, 29) \ + x(bucket_gens, 30) \ + x(snapshot_tree, 31) \ + x(logged_op_truncate, 32) \ + x(logged_op_finsert, 33) enum bch_bkey_type { #define x(name, nr) KEY_TYPE_##name = nr, @@ -350,556 +392,55 @@ enum bch_bkey_type { KEY_TYPE_MAX, }; -struct bch_cookie { +struct bch_deleted { struct bch_val v; - __le64 cookie; -}; - -/* Extents */ - -/* - * In extent bkeys, the value is a list of pointers (bch_extent_ptr), optionally - * preceded by checksum/compression information (bch_extent_crc32 or - * bch_extent_crc64). - * - * One major determining factor in the format of extents is how we handle and - * represent extents that have been partially overwritten and thus trimmed: - * - * If an extent is not checksummed or compressed, when the extent is trimmed we - * don't have to remember the extent we originally allocated and wrote: we can - * merely adjust ptr->offset to point to the start of the data that is currently - * live. The size field in struct bkey records the current (live) size of the - * extent, and is also used to mean "size of region on disk that we point to" in - * this case. - * - * Thus an extent that is not checksummed or compressed will consist only of a - * list of bch_extent_ptrs, with none of the fields in - * bch_extent_crc32/bch_extent_crc64. - * - * When an extent is checksummed or compressed, it's not possible to read only - * the data that is currently live: we have to read the entire extent that was - * originally written, and then return only the part of the extent that is - * currently live. - * - * Thus, in addition to the current size of the extent in struct bkey, we need - * to store the size of the originally allocated space - this is the - * compressed_size and uncompressed_size fields in bch_extent_crc32/64. Also, - * when the extent is trimmed, instead of modifying the offset field of the - * pointer, we keep a second smaller offset field - "offset into the original - * extent of the currently live region". - * - * The other major determining factor is replication and data migration: - * - * Each pointer may have its own bch_extent_crc32/64. When doing a replicated - * write, we will initially write all the replicas in the same format, with the - * same checksum type and compression format - however, when copygc runs later (or - * tiering/cache promotion, anything that moves data), it is not in general - * going to rewrite all the pointers at once - one of the replicas may be in a - * bucket on one device that has very little fragmentation while another lives - * in a bucket that has become heavily fragmented, and thus is being rewritten - * sooner than the rest. - * - * Thus it will only move a subset of the pointers (or in the case of - * tiering/cache promotion perhaps add a single pointer without dropping any - * current pointers), and if the extent has been partially overwritten it must - * write only the currently live portion (or copygc would not be able to reduce - * fragmentation!) - which necessitates a different bch_extent_crc format for - * the new pointer. - * - * But in the interests of space efficiency, we don't want to store one - * bch_extent_crc for each pointer if we don't have to. - * - * Thus, a bch_extent consists of bch_extent_crc32s, bch_extent_crc64s, and - * bch_extent_ptrs appended arbitrarily one after the other. We determine the - * type of a given entry with a scheme similar to utf8 (except we're encoding a - * type, not a size), encoding the type in the position of the first set bit: - * - * bch_extent_crc32 - 0b1 - * bch_extent_ptr - 0b10 - * bch_extent_crc64 - 0b100 - * - * We do it this way because bch_extent_crc32 is _very_ constrained on bits (and - * bch_extent_crc64 is the least constrained). - * - * Then, each bch_extent_crc32/64 applies to the pointers that follow after it, - * until the next bch_extent_crc32/64. - * - * If there are no bch_extent_crcs preceding a bch_extent_ptr, then that pointer - * is neither checksummed nor compressed. - */ - -/* 128 bits, sufficient for cryptographic MACs: */ -struct bch_csum { - __le64 lo; - __le64 hi; -} __attribute__((packed, aligned(8))); - -#define BCH_EXTENT_ENTRY_TYPES() \ - x(ptr, 0) \ - x(crc32, 1) \ - x(crc64, 2) \ - x(crc128, 3) \ - x(stripe_ptr, 4) -#define BCH_EXTENT_ENTRY_MAX 5 - -enum bch_extent_entry_type { -#define x(f, n) BCH_EXTENT_ENTRY_##f = n, - BCH_EXTENT_ENTRY_TYPES() -#undef x -}; - -/* Compressed/uncompressed size are stored biased by 1: */ -struct bch_extent_crc32 { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u32 type:2, - _compressed_size:7, - _uncompressed_size:7, - offset:7, - _unused:1, - csum_type:4, - compression_type:4; - __u32 csum; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u32 csum; - __u32 compression_type:4, - csum_type:4, - _unused:1, - offset:7, - _uncompressed_size:7, - _compressed_size:7, - type:2; -#endif -} __attribute__((packed, aligned(8))); - -#define CRC32_SIZE_MAX (1U << 7) -#define CRC32_NONCE_MAX 0 - -struct bch_extent_crc64 { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:3, - _compressed_size:9, - _uncompressed_size:9, - offset:9, - nonce:10, - csum_type:4, - compression_type:4, - csum_hi:16; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 csum_hi:16, - compression_type:4, - csum_type:4, - nonce:10, - offset:9, - _uncompressed_size:9, - _compressed_size:9, - type:3; -#endif - __u64 csum_lo; -} __attribute__((packed, aligned(8))); - -#define CRC64_SIZE_MAX (1U << 9) -#define CRC64_NONCE_MAX ((1U << 10) - 1) - -struct bch_extent_crc128 { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:4, - _compressed_size:13, - _uncompressed_size:13, - offset:13, - nonce:13, - csum_type:4, - compression_type:4; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 compression_type:4, - csum_type:4, - nonce:13, - offset:13, - _uncompressed_size:13, - _compressed_size:13, - type:4; -#endif - struct bch_csum csum; -} __attribute__((packed, aligned(8))); - -#define CRC128_SIZE_MAX (1U << 13) -#define CRC128_NONCE_MAX ((1U << 13) - 1) - -/* - * @reservation - pointer hasn't been written to, just reserved - */ -struct bch_extent_ptr { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:1, - cached:1, - unused:1, - reservation:1, - offset:44, /* 8 petabytes */ - dev:8, - gen:8; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 gen:8, - dev:8, - offset:44, - reservation:1, - unused:1, - cached:1, - type:1; -#endif -} __attribute__((packed, aligned(8))); - -struct bch_extent_stripe_ptr { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:5, - block:8, - idx:51; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 idx:51, - block:8, - type:5; -#endif -}; - -struct bch_extent_reservation { -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u64 type:6, - unused:22, - replicas:4, - generation:32; -#elif defined (__BIG_ENDIAN_BITFIELD) - __u64 generation:32, - replicas:4, - unused:22, - type:6; -#endif -}; - -union bch_extent_entry { -#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ || __BITS_PER_LONG == 64 - unsigned long type; -#elif __BITS_PER_LONG == 32 - struct { - unsigned long pad; - unsigned long type; - }; -#else -#error edit for your odd byteorder. -#endif - -#define x(f, n) struct bch_extent_##f f; - BCH_EXTENT_ENTRY_TYPES() -#undef x }; -struct bch_btree_ptr { +struct bch_whiteout { struct bch_val v; - - struct bch_extent_ptr start[0]; - __u64 _data[0]; -} __attribute__((packed, aligned(8))); - -struct bch_btree_ptr_v2 { - struct bch_val v; - - __u64 mem_ptr; - __le64 seq; - __le16 sectors_written; - /* In case we ever decide to do variable size btree nodes: */ - __le16 sectors; - struct bpos min_key; - struct bch_extent_ptr start[0]; - __u64 _data[0]; -} __attribute__((packed, aligned(8))); - -struct bch_extent { - struct bch_val v; - - union bch_extent_entry start[0]; - __u64 _data[0]; -} __attribute__((packed, aligned(8))); - -struct bch_reservation { - struct bch_val v; - - __le32 generation; - __u8 nr_replicas; - __u8 pad[3]; -} __attribute__((packed, aligned(8))); - -/* Maximum size (in u64s) a single pointer could be: */ -#define BKEY_EXTENT_PTR_U64s_MAX\ - ((sizeof(struct bch_extent_crc128) + \ - sizeof(struct bch_extent_ptr)) / sizeof(u64)) - -/* Maximum possible size of an entire extent value: */ -#define BKEY_EXTENT_VAL_U64s_MAX \ - (1 + BKEY_EXTENT_PTR_U64s_MAX * (BCH_REPLICAS_MAX + 1)) - -#define BKEY_PADDED(key) __BKEY_PADDED(key, BKEY_EXTENT_VAL_U64s_MAX) - -/* * Maximum possible size of an entire extent, key + value: */ -#define BKEY_EXTENT_U64s_MAX (BKEY_U64s + BKEY_EXTENT_VAL_U64s_MAX) - -/* Btree pointers don't carry around checksums: */ -#define BKEY_BTREE_PTR_VAL_U64s_MAX \ - ((sizeof(struct bch_btree_ptr_v2) + \ - sizeof(struct bch_extent_ptr) * BCH_REPLICAS_MAX) / sizeof(u64)) -#define BKEY_BTREE_PTR_U64s_MAX \ - (BKEY_U64s + BKEY_BTREE_PTR_VAL_U64s_MAX) - -/* Inodes */ - -#define BLOCKDEV_INODE_MAX 4096 - -#define BCACHEFS_ROOT_INO 4096 - -struct bch_inode { - struct bch_val v; - - __le64 bi_hash_seed; - __le32 bi_flags; - __le16 bi_mode; - __u8 fields[0]; -} __attribute__((packed, aligned(8))); - -struct bch_inode_generation { - struct bch_val v; - - __le32 bi_generation; - __le32 pad; -} __attribute__((packed, aligned(8))); - -#define BCH_INODE_FIELDS() \ - x(bi_atime, 96) \ - x(bi_ctime, 96) \ - x(bi_mtime, 96) \ - x(bi_otime, 96) \ - x(bi_size, 64) \ - x(bi_sectors, 64) \ - x(bi_uid, 32) \ - x(bi_gid, 32) \ - x(bi_nlink, 32) \ - x(bi_generation, 32) \ - x(bi_dev, 32) \ - x(bi_data_checksum, 8) \ - x(bi_compression, 8) \ - x(bi_project, 32) \ - x(bi_background_compression, 8) \ - x(bi_data_replicas, 8) \ - x(bi_promote_target, 16) \ - x(bi_foreground_target, 16) \ - x(bi_background_target, 16) \ - x(bi_erasure_code, 16) \ - x(bi_fields_set, 16) - -/* subset of BCH_INODE_FIELDS */ -#define BCH_INODE_OPTS() \ - x(data_checksum, 8) \ - x(compression, 8) \ - x(project, 32) \ - x(background_compression, 8) \ - x(data_replicas, 8) \ - x(promote_target, 16) \ - x(foreground_target, 16) \ - x(background_target, 16) \ - x(erasure_code, 16) - -enum inode_opt_id { -#define x(name, ...) \ - Inode_opt_##name, - BCH_INODE_OPTS() -#undef x - Inode_opt_nr, }; -enum { - /* - * User flags (get/settable with FS_IOC_*FLAGS, correspond to FS_*_FL - * flags) - */ - __BCH_INODE_SYNC = 0, - __BCH_INODE_IMMUTABLE = 1, - __BCH_INODE_APPEND = 2, - __BCH_INODE_NODUMP = 3, - __BCH_INODE_NOATIME = 4, - - __BCH_INODE_I_SIZE_DIRTY= 5, - __BCH_INODE_I_SECTORS_DIRTY= 6, - __BCH_INODE_UNLINKED = 7, - - /* bits 20+ reserved for packed fields below: */ -}; - -#define BCH_INODE_SYNC (1 << __BCH_INODE_SYNC) -#define BCH_INODE_IMMUTABLE (1 << __BCH_INODE_IMMUTABLE) -#define BCH_INODE_APPEND (1 << __BCH_INODE_APPEND) -#define BCH_INODE_NODUMP (1 << __BCH_INODE_NODUMP) -#define BCH_INODE_NOATIME (1 << __BCH_INODE_NOATIME) -#define BCH_INODE_I_SIZE_DIRTY (1 << __BCH_INODE_I_SIZE_DIRTY) -#define BCH_INODE_I_SECTORS_DIRTY (1 << __BCH_INODE_I_SECTORS_DIRTY) -#define BCH_INODE_UNLINKED (1 << __BCH_INODE_UNLINKED) - -LE32_BITMASK(INODE_STR_HASH, struct bch_inode, bi_flags, 20, 24); -LE32_BITMASK(INODE_NR_FIELDS, struct bch_inode, bi_flags, 24, 31); -LE32_BITMASK(INODE_NEW_VARINT, struct bch_inode, bi_flags, 31, 32); - -/* Dirents */ - -/* - * Dirents (and xattrs) have to implement string lookups; since our b-tree - * doesn't support arbitrary length strings for the key, we instead index by a - * 64 bit hash (currently truncated sha1) of the string, stored in the offset - * field of the key - using linear probing to resolve hash collisions. This also - * provides us with the readdir cookie posix requires. - * - * Linear probing requires us to use whiteouts for deletions, in the event of a - * collision: - */ - -struct bch_dirent { - struct bch_val v; - - /* Target inode number: */ - __le64 d_inum; - - /* - * Copy of mode bits 12-15 from the target inode - so userspace can get - * the filetype without having to do a stat() - */ - __u8 d_type; - - __u8 d_name[]; -} __attribute__((packed, aligned(8))); - -#define BCH_NAME_MAX (U8_MAX * sizeof(u64) - \ - sizeof(struct bkey) - \ - offsetof(struct bch_dirent, d_name)) - - -/* Xattrs */ - -#define KEY_TYPE_XATTR_INDEX_USER 0 -#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_ACCESS 1 -#define KEY_TYPE_XATTR_INDEX_POSIX_ACL_DEFAULT 2 -#define KEY_TYPE_XATTR_INDEX_TRUSTED 3 -#define KEY_TYPE_XATTR_INDEX_SECURITY 4 - -struct bch_xattr { +struct bch_error { struct bch_val v; - __u8 x_type; - __u8 x_name_len; - __le16 x_val_len; - __u8 x_name[]; -} __attribute__((packed, aligned(8))); - -/* Bucket/allocation information: */ - -struct bch_alloc { - struct bch_val v; - __u8 fields; - __u8 gen; - __u8 data[]; -} __attribute__((packed, aligned(8))); - -#define BCH_ALLOC_FIELDS() \ - x(read_time, 16) \ - x(write_time, 16) \ - x(data_type, 8) \ - x(dirty_sectors, 16) \ - x(cached_sectors, 16) \ - x(oldest_gen, 8) - -enum { -#define x(name, bytes) BCH_ALLOC_FIELD_##name, - BCH_ALLOC_FIELDS() -#undef x - BCH_ALLOC_FIELD_NR -}; - -static const unsigned BCH_ALLOC_FIELD_BYTES[] = { -#define x(name, bits) [BCH_ALLOC_FIELD_##name] = bits / 8, - BCH_ALLOC_FIELDS() -#undef x -}; - -#define x(name, bits) + (bits / 8) -static const unsigned BKEY_ALLOC_VAL_U64s_MAX = - DIV_ROUND_UP(offsetof(struct bch_alloc, data) - BCH_ALLOC_FIELDS(), sizeof(u64)); -#undef x - -#define BKEY_ALLOC_U64s_MAX (BKEY_U64s + BKEY_ALLOC_VAL_U64s_MAX) - -/* Quotas: */ - -enum quota_types { - QTYP_USR = 0, - QTYP_GRP = 1, - QTYP_PRJ = 2, - QTYP_NR = 3, }; -enum quota_counters { - Q_SPC = 0, - Q_INO = 1, - Q_COUNTERS = 2, +struct bch_cookie { + struct bch_val v; + __le64 cookie; }; -struct bch_quota_counter { - __le64 hardlimit; - __le64 softlimit; +struct bch_hash_whiteout { + struct bch_val v; }; -struct bch_quota { +struct bch_set { struct bch_val v; - struct bch_quota_counter c[Q_COUNTERS]; -} __attribute__((packed, aligned(8))); +}; -/* Erasure coding */ +/* 128 bits, sufficient for cryptographic MACs: */ +struct bch_csum { + __le64 lo; + __le64 hi; +} __packed __aligned(8); -struct bch_stripe { +struct bch_backpointer { struct bch_val v; - __le16 sectors; - __u8 algorithm; - __u8 nr_blocks; - __u8 nr_redundant; - - __u8 csum_granularity_bits; - __u8 csum_type; - __u8 pad; - - struct bch_extent_ptr ptrs[0]; -} __attribute__((packed, aligned(8))); + __u8 btree_id; + __u8 level; + __u8 data_type; + __u64 bucket_offset:40; + __u32 bucket_len; + struct bpos pos; +} __packed __aligned(8); -/* Reflink: */ +/* LRU btree: */ -struct bch_reflink_p { +struct bch_lru { struct bch_val v; __le64 idx; +} __packed __aligned(8); - __le32 reservation_generation; - __u8 nr_replicas; - __u8 pad[3]; -}; - -struct bch_reflink_v { - struct bch_val v; - __le64 refcount; - union bch_extent_entry start[0]; - __u64 _data[0]; -}; - -struct bch_indirect_inline_data { - struct bch_val v; - __le64 refcount; - u8 data[0]; -}; - -/* Inline data */ - -struct bch_inline_data { - struct bch_val v; - u8 data[0]; -}; +#define LRU_ID_STRIPES (1U << 16) /* Optional/variable size superblock sections: */ @@ -909,16 +450,35 @@ struct bch_sb_field { __le32 type; }; -#define BCH_SB_FIELDS() \ - x(journal, 0) \ - x(members, 1) \ - x(crypt, 2) \ - x(replicas_v0, 3) \ - x(quota, 4) \ - x(disk_groups, 5) \ - x(clean, 6) \ - x(replicas, 7) \ - x(journal_seq_blacklist, 8) +#define BCH_SB_FIELDS() \ + x(journal, 0) \ + x(members_v1, 1) \ + x(crypt, 2) \ + x(replicas_v0, 3) \ + x(quota, 4) \ + x(disk_groups, 5) \ + x(clean, 6) \ + x(replicas, 7) \ + x(journal_seq_blacklist, 8) \ + x(journal_v2, 9) \ + x(counters, 10) \ + x(members_v2, 11) \ + x(errors, 12) \ + x(ext, 13) \ + x(downgrade, 14) + +#include "alloc_background_format.h" +#include "extents_format.h" +#include "reflink_format.h" +#include "ec_format.h" +#include "inode_format.h" +#include "dirent_format.h" +#include "xattr_format.h" +#include "quota_format.h" +#include "logged_ops_format.h" +#include "snapshot_format.h" +#include "subvolume_format.h" +#include "sb-counters_format.h" enum bch_sb_field_type { #define x(f, nr) BCH_SB_FIELD_##f = nr, @@ -927,61 +487,114 @@ enum bch_sb_field_type { BCH_SB_FIELD_NR }; +/* + * Most superblock fields are replicated in all device's superblocks - a few are + * not: + */ +#define BCH_SINGLE_DEVICE_SB_FIELDS \ + ((1U << BCH_SB_FIELD_journal)| \ + (1U << BCH_SB_FIELD_journal_v2)) + /* BCH_SB_FIELD_journal: */ struct bch_sb_field_journal { struct bch_sb_field field; - __le64 buckets[0]; + __le64 buckets[]; }; -/* BCH_SB_FIELD_members: */ +struct bch_sb_field_journal_v2 { + struct bch_sb_field field; + + struct bch_sb_field_journal_v2_entry { + __le64 start; + __le64 nr; + } d[]; +}; + +/* BCH_SB_FIELD_members_v1: */ #define BCH_MIN_NR_NBUCKETS (1 << 6) +#define BCH_IOPS_MEASUREMENTS() \ + x(seqread, 0) \ + x(seqwrite, 1) \ + x(randread, 2) \ + x(randwrite, 3) + +enum bch_iops_measurement { +#define x(t, n) BCH_IOPS_##t = n, + BCH_IOPS_MEASUREMENTS() +#undef x + BCH_IOPS_NR +}; + +#define BCH_MEMBER_ERROR_TYPES() \ + x(read, 0) \ + x(write, 1) \ + x(checksum, 2) + +enum bch_member_error_type { +#define x(t, n) BCH_MEMBER_ERROR_##t = n, + BCH_MEMBER_ERROR_TYPES() +#undef x + BCH_MEMBER_ERROR_NR +}; + struct bch_member { - uuid_le uuid; + __uuid_t uuid; __le64 nbuckets; /* device size */ __le16 first_bucket; /* index of first bucket used */ __le16 bucket_size; /* sectors */ __le32 pad; __le64 last_mount; /* time_t */ - __le64 flags[2]; + __le64 flags; + __le32 iops[4]; + __le64 errors[BCH_MEMBER_ERROR_NR]; + __le64 errors_at_reset[BCH_MEMBER_ERROR_NR]; + __le64 errors_reset_time; + __le64 seq; }; -LE64_BITMASK(BCH_MEMBER_STATE, struct bch_member, flags[0], 0, 4) -/* 4-10 unused, was TIER, HAS_(META)DATA */ -LE64_BITMASK(BCH_MEMBER_REPLACEMENT, struct bch_member, flags[0], 10, 14) -LE64_BITMASK(BCH_MEMBER_DISCARD, struct bch_member, flags[0], 14, 15) -LE64_BITMASK(BCH_MEMBER_DATA_ALLOWED, struct bch_member, flags[0], 15, 20) -LE64_BITMASK(BCH_MEMBER_GROUP, struct bch_member, flags[0], 20, 28) -LE64_BITMASK(BCH_MEMBER_DURABILITY, struct bch_member, flags[0], 28, 30) +#define BCH_MEMBER_V1_BYTES 56 -#define BCH_TIER_MAX 4U +LE64_BITMASK(BCH_MEMBER_STATE, struct bch_member, flags, 0, 4) +/* 4-14 unused, was TIER, HAS_(META)DATA, REPLACEMENT */ +LE64_BITMASK(BCH_MEMBER_DISCARD, struct bch_member, flags, 14, 15) +LE64_BITMASK(BCH_MEMBER_DATA_ALLOWED, struct bch_member, flags, 15, 20) +LE64_BITMASK(BCH_MEMBER_GROUP, struct bch_member, flags, 20, 28) +LE64_BITMASK(BCH_MEMBER_DURABILITY, struct bch_member, flags, 28, 30) +LE64_BITMASK(BCH_MEMBER_FREESPACE_INITIALIZED, + struct bch_member, flags, 30, 31) #if 0 LE64_BITMASK(BCH_MEMBER_NR_READ_ERRORS, struct bch_member, flags[1], 0, 20); LE64_BITMASK(BCH_MEMBER_NR_WRITE_ERRORS,struct bch_member, flags[1], 20, 40); #endif +#define BCH_MEMBER_STATES() \ + x(rw, 0) \ + x(ro, 1) \ + x(failed, 2) \ + x(spare, 3) + enum bch_member_state { - BCH_MEMBER_STATE_RW = 0, - BCH_MEMBER_STATE_RO = 1, - BCH_MEMBER_STATE_FAILED = 2, - BCH_MEMBER_STATE_SPARE = 3, - BCH_MEMBER_STATE_NR = 4, +#define x(t, n) BCH_MEMBER_STATE_##t = n, + BCH_MEMBER_STATES() +#undef x + BCH_MEMBER_STATE_NR }; -enum cache_replacement { - CACHE_REPLACEMENT_LRU = 0, - CACHE_REPLACEMENT_FIFO = 1, - CACHE_REPLACEMENT_RANDOM = 2, - CACHE_REPLACEMENT_NR = 3, +struct bch_sb_field_members_v1 { + struct bch_sb_field field; + struct bch_member _members[]; //Members are now variable size }; -struct bch_sb_field_members { +struct bch_sb_field_members_v2 { struct bch_sb_field field; - struct bch_member members[0]; + __le16 member_bytes; //size of single member entry + u8 pad[6]; + struct bch_member _members[]; }; /* BCH_SB_FIELD_crypt: */ @@ -995,10 +608,10 @@ struct bch_key { }; #define BCH_KEY_MAGIC \ - (((u64) 'b' << 0)|((u64) 'c' << 8)| \ - ((u64) 'h' << 16)|((u64) '*' << 24)| \ - ((u64) '*' << 32)|((u64) 'k' << 40)| \ - ((u64) 'e' << 48)|((u64) 'y' << 56)) + (((__u64) 'b' << 0)|((__u64) 'c' << 8)| \ + ((__u64) 'h' << 16)|((__u64) '*' << 24)| \ + ((__u64) '*' << 32)|((__u64) 'k' << 40)| \ + ((__u64) 'e' << 48)|((__u64) 'y' << 56)) struct bch_encrypted_key { __le64 magic; @@ -1035,13 +648,16 @@ LE64_BITMASK(BCH_KDF_SCRYPT_P, struct bch_sb_field_crypt, kdf_flags, 32, 48); /* BCH_SB_FIELD_replicas: */ #define BCH_DATA_TYPES() \ - x(none, 0) \ + x(free, 0) \ x(sb, 1) \ x(journal, 2) \ x(btree, 3) \ x(user, 4) \ x(cached, 5) \ - x(parity, 6) + x(parity, 6) \ + x(stripe, 7) \ + x(need_gc_gens, 8) \ + x(need_discard, 9) enum bch_data_type { #define x(t, n) BCH_DATA_##t, @@ -1050,48 +666,54 @@ enum bch_data_type { BCH_DATA_NR }; +static inline bool data_type_is_empty(enum bch_data_type type) +{ + switch (type) { + case BCH_DATA_free: + case BCH_DATA_need_gc_gens: + case BCH_DATA_need_discard: + return true; + default: + return false; + } +} + +static inline bool data_type_is_hidden(enum bch_data_type type) +{ + switch (type) { + case BCH_DATA_sb: + case BCH_DATA_journal: + return true; + default: + return false; + } +} + struct bch_replicas_entry_v0 { __u8 data_type; __u8 nr_devs; - __u8 devs[0]; -} __attribute__((packed)); + __u8 devs[]; +} __packed; struct bch_sb_field_replicas_v0 { struct bch_sb_field field; - struct bch_replicas_entry_v0 entries[0]; -} __attribute__((packed, aligned(8))); + struct bch_replicas_entry_v0 entries[]; +} __packed __aligned(8); -struct bch_replicas_entry { +struct bch_replicas_entry_v1 { __u8 data_type; __u8 nr_devs; __u8 nr_required; - __u8 devs[0]; -} __attribute__((packed)); + __u8 devs[]; +} __packed; #define replicas_entry_bytes(_i) \ (offsetof(typeof(*(_i)), devs) + (_i)->nr_devs) struct bch_sb_field_replicas { struct bch_sb_field field; - struct bch_replicas_entry entries[0]; -} __attribute__((packed, aligned(8))); - -/* BCH_SB_FIELD_quota: */ - -struct bch_sb_quota_counter { - __le32 timelimit; - __le32 warnlimit; -}; - -struct bch_sb_quota_type { - __le64 flags; - struct bch_sb_quota_counter c[Q_COUNTERS]; -}; - -struct bch_sb_field_quota { - struct bch_sb_field field; - struct bch_sb_quota_type q[QTYP_NR]; -} __attribute__((packed, aligned(8))); + struct bch_replicas_entry_v1 entries[]; +} __packed __aligned(8); /* BCH_SB_FIELD_disk_groups: */ @@ -1100,7 +722,7 @@ struct bch_sb_field_quota { struct bch_disk_group { __u8 label[BCH_SB_LABEL_SIZE]; __le64 flags[2]; -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); LE64_BITMASK(BCH_GROUP_DELETED, struct bch_disk_group, flags[0], 0, 1) LE64_BITMASK(BCH_GROUP_DATA_ALLOWED, struct bch_disk_group, flags[0], 1, 6) @@ -1108,8 +730,8 @@ LE64_BITMASK(BCH_GROUP_PARENT, struct bch_disk_group, flags[0], 6, 24) struct bch_sb_field_disk_groups { struct bch_sb_field field; - struct bch_disk_group entries[0]; -} __attribute__((packed, aligned(8))); + struct bch_disk_group entries[]; +} __packed __aligned(8); /* * On clean shutdown, store btree roots and current journal sequence number in @@ -1122,24 +744,20 @@ struct jset_entry { __u8 type; /* designates what this jset holds */ __u8 pad[3]; - union { - struct bkey_i start[0]; - __u64 _data[0]; - }; + struct bkey_i start[0]; + __u64 _data[]; }; struct bch_sb_field_clean { struct bch_sb_field field; __le32 flags; - __le16 read_clock; - __le16 write_clock; + __le16 _read_clock; /* no longer used */ + __le16 _write_clock; __le64 journal_seq; - union { - struct jset_entry start[0]; - __u64 _data[0]; - }; + struct jset_entry start[0]; + __u64 _data[]; }; struct journal_seq_blacklist_entry { @@ -1149,11 +767,36 @@ struct journal_seq_blacklist_entry { struct bch_sb_field_journal_seq_blacklist { struct bch_sb_field field; + struct journal_seq_blacklist_entry start[]; +}; - union { - struct journal_seq_blacklist_entry start[0]; - __u64 _data[0]; - }; +struct bch_sb_field_errors { + struct bch_sb_field field; + struct bch_sb_field_error_entry { + __le64 v; + __le64 last_error_time; + } entries[]; +}; + +LE64_BITMASK(BCH_SB_ERROR_ENTRY_ID, struct bch_sb_field_error_entry, v, 0, 16); +LE64_BITMASK(BCH_SB_ERROR_ENTRY_NR, struct bch_sb_field_error_entry, v, 16, 64); + +struct bch_sb_field_ext { + struct bch_sb_field field; + __le64 recovery_passes_required[2]; + __le64 errors_silent[8]; +}; + +struct bch_sb_field_downgrade_entry { + __le16 version; + __le64 recovery_passes[2]; + __le16 nr_errors; + __le16 errors[] __counted_by(nr_errors); +} __packed __aligned(2); + +struct bch_sb_field_downgrade { + struct bch_sb_field field; + struct bch_sb_field_downgrade_entry entries[]; }; /* Superblock: */ @@ -1163,30 +806,68 @@ struct bch_sb_field_journal_seq_blacklist { * One common version number for all on disk data structures - superblock, btree * nodes, journal entries */ -#define BCH_JSET_VERSION_OLD 2 -#define BCH_BSET_VERSION_OLD 3 +#define BCH_VERSION_MAJOR(_v) ((__u16) ((_v) >> 10)) +#define BCH_VERSION_MINOR(_v) ((__u16) ((_v) & ~(~0U << 10))) +#define BCH_VERSION(_major, _minor) (((_major) << 10)|(_minor) << 0) + +/* + * field 1: version name + * field 2: BCH_VERSION(major, minor) + * field 3: recovery passess required on upgrade + */ +#define BCH_METADATA_VERSIONS() \ + x(bkey_renumber, BCH_VERSION(0, 10)) \ + x(inode_btree_change, BCH_VERSION(0, 11)) \ + x(snapshot, BCH_VERSION(0, 12)) \ + x(inode_backpointers, BCH_VERSION(0, 13)) \ + x(btree_ptr_sectors_written, BCH_VERSION(0, 14)) \ + x(snapshot_2, BCH_VERSION(0, 15)) \ + x(reflink_p_fix, BCH_VERSION(0, 16)) \ + x(subvol_dirent, BCH_VERSION(0, 17)) \ + x(inode_v2, BCH_VERSION(0, 18)) \ + x(freespace, BCH_VERSION(0, 19)) \ + x(alloc_v4, BCH_VERSION(0, 20)) \ + x(new_data_types, BCH_VERSION(0, 21)) \ + x(backpointers, BCH_VERSION(0, 22)) \ + x(inode_v3, BCH_VERSION(0, 23)) \ + x(unwritten_extents, BCH_VERSION(0, 24)) \ + x(bucket_gens, BCH_VERSION(0, 25)) \ + x(lru_v2, BCH_VERSION(0, 26)) \ + x(fragmentation_lru, BCH_VERSION(0, 27)) \ + x(no_bps_in_alloc_keys, BCH_VERSION(0, 28)) \ + x(snapshot_trees, BCH_VERSION(0, 29)) \ + x(major_minor, BCH_VERSION(1, 0)) \ + x(snapshot_skiplists, BCH_VERSION(1, 1)) \ + x(deleted_inodes, BCH_VERSION(1, 2)) \ + x(rebalance_work, BCH_VERSION(1, 3)) \ + x(member_seq, BCH_VERSION(1, 4)) \ + x(subvolume_fs_parent, BCH_VERSION(1, 5)) \ + x(btree_subvolume_children, BCH_VERSION(1, 6)) enum bcachefs_metadata_version { - bcachefs_metadata_version_min = 9, - bcachefs_metadata_version_new_versioning = 10, - bcachefs_metadata_version_bkey_renumber = 10, - bcachefs_metadata_version_inode_btree_change = 11, - bcachefs_metadata_version_max = 12, + bcachefs_metadata_version_min = 9, +#define x(t, n) bcachefs_metadata_version_##t = n, + BCH_METADATA_VERSIONS() +#undef x + bcachefs_metadata_version_max }; +static const __maybe_unused +unsigned bcachefs_metadata_required_upgrade_below = bcachefs_metadata_version_rebalance_work; + #define bcachefs_metadata_version_current (bcachefs_metadata_version_max - 1) #define BCH_SB_SECTOR 8 #define BCH_SB_MEMBERS_MAX 64 /* XXX kill */ struct bch_sb_layout { - uuid_le magic; /* bcachefs superblock UUID */ + __uuid_t magic; /* bcachefs superblock UUID */ __u8 layout_type; __u8 sb_max_size_bits; /* base 2 of 512 byte sectors */ __u8 nr_superblocks; __u8 pad[5]; __le64 sb_offset[61]; -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); #define BCH_SB_LAYOUT_SECTOR 7 @@ -1196,7 +877,7 @@ struct bch_sb_layout { * @version_min - Oldest metadata version this filesystem contains; so we can * safely drop compatibility code and refuse to mount filesystems * we'd need it for - * @magic - identifies as a bcachefs superblock (BCACHE_MAGIC) + * @magic - identifies as a bcachefs superblock (BCHFS_MAGIC) * @seq - incremented each time superblock is written * @uuid - used for generating various magic numbers and identifying * member devices, never changes @@ -1211,9 +892,9 @@ struct bch_sb { __le16 version; __le16 version_min; __le16 pad[2]; - uuid_le magic; - uuid_le uuid; - uuid_le user_uuid; + __uuid_t magic; + __uuid_t uuid; + __uuid_t user_uuid; __u8 label[BCH_SB_LABEL_SIZE]; __le64 offset; __le64 seq; @@ -1227,17 +908,16 @@ struct bch_sb { __le32 time_base_hi; __le32 time_precision; - __le64 flags[8]; + __le64 flags[7]; + __le64 write_time; __le64 features[2]; __le64 compat[2]; struct bch_sb_layout layout; - union { - struct bch_sb_field start[0]; - __le64 _data[0]; - }; -} __attribute__((packed, aligned(8))); + struct bch_sb_field start[0]; + __le64 _data[]; +} __packed __aligned(8); /* * Flags: @@ -1275,13 +955,12 @@ LE64_BITMASK(BCH_SB_GRPQUOTA, struct bch_sb, flags[0], 58, 59); LE64_BITMASK(BCH_SB_PRJQUOTA, struct bch_sb, flags[0], 59, 60); LE64_BITMASK(BCH_SB_HAS_ERRORS, struct bch_sb, flags[0], 60, 61); +LE64_BITMASK(BCH_SB_HAS_TOPOLOGY_ERRORS,struct bch_sb, flags[0], 61, 62); -LE64_BITMASK(BCH_SB_REFLINK, struct bch_sb, flags[0], 61, 62); - -/* 61-64 unused */ +LE64_BITMASK(BCH_SB_BIG_ENDIAN, struct bch_sb, flags[0], 62, 63); LE64_BITMASK(BCH_SB_STR_HASH_TYPE, struct bch_sb, flags[1], 0, 4); -LE64_BITMASK(BCH_SB_COMPRESSION_TYPE, struct bch_sb, flags[1], 4, 8); +LE64_BITMASK(BCH_SB_COMPRESSION_TYPE_LO,struct bch_sb, flags[1], 4, 8); LE64_BITMASK(BCH_SB_INODE_32BIT, struct bch_sb, flags[1], 8, 9); LE64_BITMASK(BCH_SB_128_BIT_MACS, struct bch_sb, flags[1], 9, 10); @@ -1301,11 +980,51 @@ LE64_BITMASK(BCH_SB_PROMOTE_TARGET, struct bch_sb, flags[1], 28, 40); LE64_BITMASK(BCH_SB_FOREGROUND_TARGET, struct bch_sb, flags[1], 40, 52); LE64_BITMASK(BCH_SB_BACKGROUND_TARGET, struct bch_sb, flags[1], 52, 64); -LE64_BITMASK(BCH_SB_BACKGROUND_COMPRESSION_TYPE, +LE64_BITMASK(BCH_SB_BACKGROUND_COMPRESSION_TYPE_LO, struct bch_sb, flags[2], 0, 4); LE64_BITMASK(BCH_SB_GC_RESERVE_BYTES, struct bch_sb, flags[2], 4, 64); LE64_BITMASK(BCH_SB_ERASURE_CODE, struct bch_sb, flags[3], 0, 16); +LE64_BITMASK(BCH_SB_METADATA_TARGET, struct bch_sb, flags[3], 16, 28); +LE64_BITMASK(BCH_SB_SHARD_INUMS, struct bch_sb, flags[3], 28, 29); +LE64_BITMASK(BCH_SB_INODES_USE_KEY_CACHE,struct bch_sb, flags[3], 29, 30); +LE64_BITMASK(BCH_SB_JOURNAL_FLUSH_DELAY,struct bch_sb, flags[3], 30, 62); +LE64_BITMASK(BCH_SB_JOURNAL_FLUSH_DISABLED,struct bch_sb, flags[3], 62, 63); +LE64_BITMASK(BCH_SB_JOURNAL_RECLAIM_DELAY,struct bch_sb, flags[4], 0, 32); +LE64_BITMASK(BCH_SB_JOURNAL_TRANSACTION_NAMES,struct bch_sb, flags[4], 32, 33); +LE64_BITMASK(BCH_SB_NOCOW, struct bch_sb, flags[4], 33, 34); +LE64_BITMASK(BCH_SB_WRITE_BUFFER_SIZE, struct bch_sb, flags[4], 34, 54); +LE64_BITMASK(BCH_SB_VERSION_UPGRADE, struct bch_sb, flags[4], 54, 56); + +LE64_BITMASK(BCH_SB_COMPRESSION_TYPE_HI,struct bch_sb, flags[4], 56, 60); +LE64_BITMASK(BCH_SB_BACKGROUND_COMPRESSION_TYPE_HI, + struct bch_sb, flags[4], 60, 64); + +LE64_BITMASK(BCH_SB_VERSION_UPGRADE_COMPLETE, + struct bch_sb, flags[5], 0, 16); + +static inline __u64 BCH_SB_COMPRESSION_TYPE(const struct bch_sb *sb) +{ + return BCH_SB_COMPRESSION_TYPE_LO(sb) | (BCH_SB_COMPRESSION_TYPE_HI(sb) << 4); +} + +static inline void SET_BCH_SB_COMPRESSION_TYPE(struct bch_sb *sb, __u64 v) +{ + SET_BCH_SB_COMPRESSION_TYPE_LO(sb, v); + SET_BCH_SB_COMPRESSION_TYPE_HI(sb, v >> 4); +} + +static inline __u64 BCH_SB_BACKGROUND_COMPRESSION_TYPE(const struct bch_sb *sb) +{ + return BCH_SB_BACKGROUND_COMPRESSION_TYPE_LO(sb) | + (BCH_SB_BACKGROUND_COMPRESSION_TYPE_HI(sb) << 4); +} + +static inline void SET_BCH_SB_BACKGROUND_COMPRESSION_TYPE(struct bch_sb *sb, __u64 v) +{ + SET_BCH_SB_BACKGROUND_COMPRESSION_TYPE_LO(sb, v); + SET_BCH_SB_BACKGROUND_COMPRESSION_TYPE_HI(sb, v >> 4); +} /* * Features: @@ -1313,7 +1032,7 @@ LE64_BITMASK(BCH_SB_ERASURE_CODE, struct bch_sb, flags[3], 0, 16); * journal_seq_blacklist_v3: gates BCH_SB_FIELD_journal_seq_blacklist * reflink: gates KEY_TYPE_reflink * inline_data: gates KEY_TYPE_inline_data - * new_siphash: gates BCH_STR_HASH_SIPHASH + * new_siphash: gates BCH_STR_HASH_siphash * new_extent_overwrite: gates BTREE_NODE_NEW_EXTENT_OVERWRITE */ #define BCH_SB_FEATURES() \ @@ -1332,14 +1051,24 @@ LE64_BITMASK(BCH_SB_ERASURE_CODE, struct bch_sb, flags[3], 0, 16); x(extents_above_btree_updates, 12) \ x(btree_updates_journalled, 13) \ x(reflink_inline_data, 14) \ - x(new_varint, 15) + x(new_varint, 15) \ + x(journal_no_flush, 16) \ + x(alloc_v2, 17) \ + x(extents_across_btree_nodes, 18) + +#define BCH_SB_FEATURES_ALWAYS \ + ((1ULL << BCH_FEATURE_new_extent_overwrite)| \ + (1ULL << BCH_FEATURE_extents_above_btree_updates)|\ + (1ULL << BCH_FEATURE_btree_updates_journalled)|\ + (1ULL << BCH_FEATURE_alloc_v2)|\ + (1ULL << BCH_FEATURE_extents_across_btree_nodes)) #define BCH_SB_FEATURES_ALL \ - ((1ULL << BCH_FEATURE_new_siphash)| \ - (1ULL << BCH_FEATURE_new_extent_overwrite)| \ + (BCH_SB_FEATURES_ALWAYS| \ + (1ULL << BCH_FEATURE_new_siphash)| \ (1ULL << BCH_FEATURE_btree_ptr_v2)| \ - (1ULL << BCH_FEATURE_extents_above_btree_updates)|\ - (1ULL << BCH_FEATURE_new_varint))\ + (1ULL << BCH_FEATURE_new_varint)| \ + (1ULL << BCH_FEATURE_journal_no_flush)) enum bch_sb_feature { #define x(f, n) BCH_FEATURE_##f, @@ -1348,74 +1077,123 @@ enum bch_sb_feature { BCH_FEATURE_NR, }; +#define BCH_SB_COMPAT() \ + x(alloc_info, 0) \ + x(alloc_metadata, 1) \ + x(extents_above_btree_updates_done, 2) \ + x(bformat_overflow_done, 3) + enum bch_sb_compat { - BCH_COMPAT_FEAT_ALLOC_INFO = 0, - BCH_COMPAT_FEAT_ALLOC_METADATA = 1, +#define x(f, n) BCH_COMPAT_##f, + BCH_SB_COMPAT() +#undef x + BCH_COMPAT_NR, }; /* options: */ +#define BCH_VERSION_UPGRADE_OPTS() \ + x(compatible, 0) \ + x(incompatible, 1) \ + x(none, 2) + +enum bch_version_upgrade_opts { +#define x(t, n) BCH_VERSION_UPGRADE_##t = n, + BCH_VERSION_UPGRADE_OPTS() +#undef x +}; + #define BCH_REPLICAS_MAX 4U +#define BCH_BKEY_PTRS_MAX 16U + +#define BCH_ERROR_ACTIONS() \ + x(continue, 0) \ + x(ro, 1) \ + x(panic, 2) + enum bch_error_actions { - BCH_ON_ERROR_CONTINUE = 0, - BCH_ON_ERROR_RO = 1, - BCH_ON_ERROR_PANIC = 2, - BCH_NR_ERROR_ACTIONS = 3, +#define x(t, n) BCH_ON_ERROR_##t = n, + BCH_ERROR_ACTIONS() +#undef x + BCH_ON_ERROR_NR }; +#define BCH_STR_HASH_TYPES() \ + x(crc32c, 0) \ + x(crc64, 1) \ + x(siphash_old, 2) \ + x(siphash, 3) + enum bch_str_hash_type { - BCH_STR_HASH_CRC32C = 0, - BCH_STR_HASH_CRC64 = 1, - BCH_STR_HASH_SIPHASH_OLD = 2, - BCH_STR_HASH_SIPHASH = 3, - BCH_STR_HASH_NR = 4, +#define x(t, n) BCH_STR_HASH_##t = n, + BCH_STR_HASH_TYPES() +#undef x + BCH_STR_HASH_NR }; +#define BCH_STR_HASH_OPTS() \ + x(crc32c, 0) \ + x(crc64, 1) \ + x(siphash, 2) + enum bch_str_hash_opts { - BCH_STR_HASH_OPT_CRC32C = 0, - BCH_STR_HASH_OPT_CRC64 = 1, - BCH_STR_HASH_OPT_SIPHASH = 2, - BCH_STR_HASH_OPT_NR = 3, +#define x(t, n) BCH_STR_HASH_OPT_##t = n, + BCH_STR_HASH_OPTS() +#undef x + BCH_STR_HASH_OPT_NR }; +#define BCH_CSUM_TYPES() \ + x(none, 0) \ + x(crc32c_nonzero, 1) \ + x(crc64_nonzero, 2) \ + x(chacha20_poly1305_80, 3) \ + x(chacha20_poly1305_128, 4) \ + x(crc32c, 5) \ + x(crc64, 6) \ + x(xxhash, 7) + enum bch_csum_type { - BCH_CSUM_NONE = 0, - BCH_CSUM_CRC32C_NONZERO = 1, - BCH_CSUM_CRC64_NONZERO = 2, - BCH_CSUM_CHACHA20_POLY1305_80 = 3, - BCH_CSUM_CHACHA20_POLY1305_128 = 4, - BCH_CSUM_CRC32C = 5, - BCH_CSUM_CRC64 = 6, - BCH_CSUM_NR = 7, +#define x(t, n) BCH_CSUM_##t = n, + BCH_CSUM_TYPES() +#undef x + BCH_CSUM_NR }; -static const unsigned bch_crc_bytes[] = { - [BCH_CSUM_NONE] = 0, - [BCH_CSUM_CRC32C_NONZERO] = 4, - [BCH_CSUM_CRC32C] = 4, - [BCH_CSUM_CRC64_NONZERO] = 8, - [BCH_CSUM_CRC64] = 8, - [BCH_CSUM_CHACHA20_POLY1305_80] = 10, - [BCH_CSUM_CHACHA20_POLY1305_128] = 16, +static const __maybe_unused unsigned bch_crc_bytes[] = { + [BCH_CSUM_none] = 0, + [BCH_CSUM_crc32c_nonzero] = 4, + [BCH_CSUM_crc32c] = 4, + [BCH_CSUM_crc64_nonzero] = 8, + [BCH_CSUM_crc64] = 8, + [BCH_CSUM_xxhash] = 8, + [BCH_CSUM_chacha20_poly1305_80] = 10, + [BCH_CSUM_chacha20_poly1305_128] = 16, }; static inline _Bool bch2_csum_type_is_encryption(enum bch_csum_type type) { switch (type) { - case BCH_CSUM_CHACHA20_POLY1305_80: - case BCH_CSUM_CHACHA20_POLY1305_128: + case BCH_CSUM_chacha20_poly1305_80: + case BCH_CSUM_chacha20_poly1305_128: return true; default: return false; } } +#define BCH_CSUM_OPTS() \ + x(none, 0) \ + x(crc32c, 1) \ + x(crc64, 2) \ + x(xxhash, 3) + enum bch_csum_opts { - BCH_CSUM_OPT_NONE = 0, - BCH_CSUM_OPT_CRC32C = 1, - BCH_CSUM_OPT_CRC64 = 2, - BCH_CSUM_OPT_NR = 3, +#define x(t, n) BCH_CSUM_OPT_##t = n, + BCH_CSUM_OPTS() +#undef x + BCH_CSUM_OPT_NR }; #define BCH_COMPRESSION_TYPES() \ @@ -1427,7 +1205,7 @@ enum bch_csum_opts { x(incompressible, 5) enum bch_compression_type { -#define x(t, n) BCH_COMPRESSION_TYPE_##t, +#define x(t, n) BCH_COMPRESSION_TYPE_##t = n, BCH_COMPRESSION_TYPES() #undef x BCH_COMPRESSION_TYPE_NR @@ -1440,7 +1218,7 @@ enum bch_compression_type { x(zstd, 3) enum bch_compression_opts { -#define x(t, n) BCH_COMPRESSION_OPT_##t, +#define x(t, n) BCH_COMPRESSION_OPT_##t = n, BCH_COMPRESSION_OPTS() #undef x BCH_COMPRESSION_OPT_NR @@ -1454,8 +1232,11 @@ enum bch_compression_opts { */ #define BCACHE_MAGIC \ - UUID_LE(0xf67385c6, 0x1a4e, 0xca45, \ - 0x82, 0x65, 0xf5, 0x7f, 0x48, 0xba, 0x6d, 0x81) + UUID_INIT(0xc68573f6, 0x4e1a, 0x45ca, \ + 0x82, 0x65, 0xf5, 0x7f, 0x48, 0xba, 0x6d, 0x81) +#define BCHFS_MAGIC \ + UUID_INIT(0xc68573f6, 0x66ce, 0x90a9, \ + 0xd9, 0x6a, 0x60, 0xcf, 0x80, 0x3d, 0xf7, 0xef) #define BCACHEFS_STATFS_MAGIC 0xca451a4e @@ -1465,6 +1246,7 @@ enum bch_compression_opts { static inline __le64 __bch2_sb_magic(struct bch_sb *sb) { __le64 ret; + memcpy(&ret, &sb->uuid, sizeof(ret)); return ret; } @@ -1490,7 +1272,13 @@ static inline __u64 __bset_magic(struct bch_sb *sb) x(blacklist, 3) \ x(blacklist_v2, 4) \ x(usage, 5) \ - x(data_usage, 6) + x(data_usage, 6) \ + x(clock, 7) \ + x(dev_usage, 8) \ + x(log, 9) \ + x(overwrite, 10) \ + x(write_buffer_keys, 11) \ + x(datetime, 12) enum { #define x(f, nr) BCH_JSET_ENTRY_##f = nr, @@ -1499,6 +1287,19 @@ enum { BCH_JSET_ENTRY_NR }; +static inline bool jset_entry_is_key(struct jset_entry *e) +{ + switch (e->type) { + case BCH_JSET_ENTRY_btree_keys: + case BCH_JSET_ENTRY_btree_root: + case BCH_JSET_ENTRY_overwrite: + case BCH_JSET_ENTRY_write_buffer_keys: + return true; + } + + return false; +} + /* * Journal sequence numbers can be blacklisted: bsets record the max sequence * number of all the journal entries they contain updates for, so that on @@ -1520,23 +1321,68 @@ struct jset_entry_blacklist_v2 { __le64 end; }; +#define BCH_FS_USAGE_TYPES() \ + x(reserved, 0) \ + x(inodes, 1) \ + x(key_version, 2) + enum { - FS_USAGE_RESERVED = 0, - FS_USAGE_INODES = 1, - FS_USAGE_KEY_VERSION = 2, - FS_USAGE_NR = 3 +#define x(f, nr) BCH_FS_USAGE_##f = nr, + BCH_FS_USAGE_TYPES() +#undef x + BCH_FS_USAGE_NR }; struct jset_entry_usage { struct jset_entry entry; __le64 v; -} __attribute__((packed)); +} __packed; struct jset_entry_data_usage { struct jset_entry entry; __le64 v; - struct bch_replicas_entry r; -} __attribute__((packed)); + struct bch_replicas_entry_v1 r; +} __packed; + +struct jset_entry_clock { + struct jset_entry entry; + __u8 rw; + __u8 pad[7]; + __le64 time; +} __packed; + +struct jset_entry_dev_usage_type { + __le64 buckets; + __le64 sectors; + __le64 fragmented; +} __packed; + +struct jset_entry_dev_usage { + struct jset_entry entry; + __le32 dev; + __u32 pad; + + __le64 _buckets_ec; /* No longer used */ + __le64 _buckets_unavailable; /* No longer used */ + + struct jset_entry_dev_usage_type d[]; +}; + +static inline unsigned jset_entry_dev_usage_nr_types(struct jset_entry_dev_usage *u) +{ + return (vstruct_bytes(&u->entry) - sizeof(struct jset_entry_dev_usage)) / + sizeof(struct jset_entry_dev_usage_type); +} + +struct jset_entry_log { + struct jset_entry entry; + u8 d[]; +} __packed __aligned(8); + +struct jset_entry_datetime { + struct jset_entry entry; + __le64 seconds; +} __packed __aligned(8); /* * On disk format for a journal entry: @@ -1560,38 +1406,96 @@ struct jset { __u8 encrypted_start[0]; - __le16 read_clock; - __le16 write_clock; + __le16 _read_clock; /* no longer used */ + __le16 _write_clock; /* Sequence number of oldest dirty journal entry */ __le64 last_seq; - union { - struct jset_entry start[0]; - __u64 _data[0]; - }; -} __attribute__((packed, aligned(8))); + struct jset_entry start[0]; + __u64 _data[]; +} __packed __aligned(8); LE32_BITMASK(JSET_CSUM_TYPE, struct jset, flags, 0, 4); LE32_BITMASK(JSET_BIG_ENDIAN, struct jset, flags, 4, 5); +LE32_BITMASK(JSET_NO_FLUSH, struct jset, flags, 5, 6); #define BCH_JOURNAL_BUCKETS_MIN 8 /* Btree: */ -#define BCH_BTREE_IDS() \ - x(EXTENTS, 0, "extents") \ - x(INODES, 1, "inodes") \ - x(DIRENTS, 2, "dirents") \ - x(XATTRS, 3, "xattrs") \ - x(ALLOC, 4, "alloc") \ - x(QUOTAS, 5, "quotas") \ - x(EC, 6, "stripes") \ - x(REFLINK, 7, "reflink") +enum btree_id_flags { + BTREE_ID_EXTENTS = BIT(0), + BTREE_ID_SNAPSHOTS = BIT(1), + BTREE_ID_SNAPSHOT_FIELD = BIT(2), + BTREE_ID_DATA = BIT(3), +}; + +#define BCH_BTREE_IDS() \ + x(extents, 0, BTREE_ID_EXTENTS|BTREE_ID_SNAPSHOTS|BTREE_ID_DATA,\ + BIT_ULL(KEY_TYPE_whiteout)| \ + BIT_ULL(KEY_TYPE_error)| \ + BIT_ULL(KEY_TYPE_cookie)| \ + BIT_ULL(KEY_TYPE_extent)| \ + BIT_ULL(KEY_TYPE_reservation)| \ + BIT_ULL(KEY_TYPE_reflink_p)| \ + BIT_ULL(KEY_TYPE_inline_data)) \ + x(inodes, 1, BTREE_ID_SNAPSHOTS, \ + BIT_ULL(KEY_TYPE_whiteout)| \ + BIT_ULL(KEY_TYPE_inode)| \ + BIT_ULL(KEY_TYPE_inode_v2)| \ + BIT_ULL(KEY_TYPE_inode_v3)| \ + BIT_ULL(KEY_TYPE_inode_generation)) \ + x(dirents, 2, BTREE_ID_SNAPSHOTS, \ + BIT_ULL(KEY_TYPE_whiteout)| \ + BIT_ULL(KEY_TYPE_hash_whiteout)| \ + BIT_ULL(KEY_TYPE_dirent)) \ + x(xattrs, 3, BTREE_ID_SNAPSHOTS, \ + BIT_ULL(KEY_TYPE_whiteout)| \ + BIT_ULL(KEY_TYPE_cookie)| \ + BIT_ULL(KEY_TYPE_hash_whiteout)| \ + BIT_ULL(KEY_TYPE_xattr)) \ + x(alloc, 4, 0, \ + BIT_ULL(KEY_TYPE_alloc)| \ + BIT_ULL(KEY_TYPE_alloc_v2)| \ + BIT_ULL(KEY_TYPE_alloc_v3)| \ + BIT_ULL(KEY_TYPE_alloc_v4)) \ + x(quotas, 5, 0, \ + BIT_ULL(KEY_TYPE_quota)) \ + x(stripes, 6, 0, \ + BIT_ULL(KEY_TYPE_stripe)) \ + x(reflink, 7, BTREE_ID_EXTENTS|BTREE_ID_DATA, \ + BIT_ULL(KEY_TYPE_reflink_v)| \ + BIT_ULL(KEY_TYPE_indirect_inline_data)) \ + x(subvolumes, 8, 0, \ + BIT_ULL(KEY_TYPE_subvolume)) \ + x(snapshots, 9, 0, \ + BIT_ULL(KEY_TYPE_snapshot)) \ + x(lru, 10, 0, \ + BIT_ULL(KEY_TYPE_set)) \ + x(freespace, 11, BTREE_ID_EXTENTS, \ + BIT_ULL(KEY_TYPE_set)) \ + x(need_discard, 12, 0, \ + BIT_ULL(KEY_TYPE_set)) \ + x(backpointers, 13, 0, \ + BIT_ULL(KEY_TYPE_backpointer)) \ + x(bucket_gens, 14, 0, \ + BIT_ULL(KEY_TYPE_bucket_gens)) \ + x(snapshot_trees, 15, 0, \ + BIT_ULL(KEY_TYPE_snapshot_tree)) \ + x(deleted_inodes, 16, BTREE_ID_SNAPSHOT_FIELD, \ + BIT_ULL(KEY_TYPE_set)) \ + x(logged_ops, 17, 0, \ + BIT_ULL(KEY_TYPE_logged_op_truncate)| \ + BIT_ULL(KEY_TYPE_logged_op_finsert)) \ + x(rebalance_work, 18, BTREE_ID_SNAPSHOT_FIELD, \ + BIT_ULL(KEY_TYPE_set)|BIT_ULL(KEY_TYPE_cookie)) \ + x(subvolume_children, 19, 0, \ + BIT_ULL(KEY_TYPE_set)) enum btree_id { -#define x(kwd, val, name) BTREE_ID_##kwd = val, +#define x(name, nr, ...) BTREE_ID_##name = nr, BCH_BTREE_IDS() #undef x BTREE_ID_NR @@ -1623,11 +1527,9 @@ struct bset { __le16 version; __le16 u64s; /* count of d[] in u64s */ - union { - struct bkey_packed start[0]; - __u64 _data[0]; - }; -} __attribute__((packed, aligned(8))); + struct bkey_packed start[0]; + __u64 _data[]; +} __packed __aligned(8); LE32_BITMASK(BSET_CSUM_TYPE, struct bset, flags, 0, 4); @@ -1635,6 +1537,9 @@ LE32_BITMASK(BSET_BIG_ENDIAN, struct bset, flags, 4, 5); LE32_BITMASK(BSET_SEPARATE_WHITEOUTS, struct bset, flags, 5, 6); +/* Sector offset within the btree node: */ +LE32_BITMASK(BSET_OFFSET, struct bset, flags, 16, 32); + struct btree_node { struct bch_csum csum; __le64 magic; @@ -1645,7 +1550,7 @@ struct btree_node { /* Closed interval: */ struct bpos min_key; struct bpos max_key; - struct bch_extent_ptr ptr; + struct bch_extent_ptr _ptr; /* not used anymore */ struct bkey_format format; union { @@ -1657,15 +1562,27 @@ struct btree_node { }; }; -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); -LE64_BITMASK(BTREE_NODE_ID, struct btree_node, flags, 0, 4); +LE64_BITMASK(BTREE_NODE_ID_LO, struct btree_node, flags, 0, 4); LE64_BITMASK(BTREE_NODE_LEVEL, struct btree_node, flags, 4, 8); LE64_BITMASK(BTREE_NODE_NEW_EXTENT_OVERWRITE, struct btree_node, flags, 8, 9); -/* 9-32 unused */ +LE64_BITMASK(BTREE_NODE_ID_HI, struct btree_node, flags, 9, 25); +/* 25-32 unused */ LE64_BITMASK(BTREE_NODE_SEQ, struct btree_node, flags, 32, 64); +static inline __u64 BTREE_NODE_ID(struct btree_node *n) +{ + return BTREE_NODE_ID_LO(n) | (BTREE_NODE_ID_HI(n) << 4); +} + +static inline void SET_BTREE_NODE_ID(struct btree_node *n, __u64 v) +{ + SET_BTREE_NODE_ID_LO(n, v); + SET_BTREE_NODE_ID_HI(n, v >> 4); +} + struct btree_node_entry { struct bch_csum csum; @@ -1675,9 +1592,8 @@ struct btree_node_entry { __u8 pad[22]; __le16 u64s; __u64 _data[0]; - }; }; -} __attribute__((packed, aligned(8))); +} __packed __aligned(8); #endif /* _BCACHEFS_FORMAT_H */