X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libbcachefs%2Fbcachefs_format.h;h=20e96daf9ca1960c66a2364638c115f73ce02677;hb=505c326cbe091b105dbdfbe66dae4ad6ee424906;hp=0a78d0f1d0c331354f562758a22271f5f6ade9e0;hpb=f9f57789de567726f7cfa46bd13df4b0815d137a;p=bcachefs-tools-debian

diff --git a/libbcachefs/bcachefs_format.h b/libbcachefs/bcachefs_format.h
index 0a78d0f..20e96da 100644
--- a/libbcachefs/bcachefs_format.h
+++ b/libbcachefs/bcachefs_format.h
@@ -76,6 +76,26 @@
 #include <asm/byteorder.h>
 #include <linux/kernel.h>
 #include <linux/uuid.h>
+#include "vstructs.h"
+
+#ifdef __KERNEL__
+typedef uuid_t __uuid_t;
+#endif
+
+#define BITMASK(name, type, field, offset, end)				\
+static const unsigned	name##_OFFSET = offset;				\
+static const 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;				\
@@ -131,7 +151,7 @@ struct bpos {
 #else
 #error edit for your odd byteorder.
 #endif
-} __attribute__((packed, aligned(4)));
+} __packed __aligned(4);
 
 #define KEY_INODE_MAX			((__u64)~0ULL)
 #define KEY_OFFSET_MAX			((__u64)~0ULL)
@@ -165,7 +185,7 @@ 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 +218,7 @@ struct bkey {
 
 	__u8		pad[1];
 #endif
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 struct bkey_packed {
 	__u64		_data[0];
@@ -232,7 +252,12 @@ 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
@@ -274,16 +299,8 @@ 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 KEY(_inode, _offset, _size)					\
@@ -302,7 +319,7 @@ 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
@@ -320,7 +337,7 @@ static inline void bkey_init(struct bkey *k)
  *   number.
  *
  * - WHITEOUT: for hash table btrees
-*/
+ */
 #define BCH_BKEY_TYPES()				\
 	x(deleted,		0)			\
 	x(whiteout,		1)			\
@@ -344,7 +361,16 @@ static inline void bkey_init(struct bkey *k)
 	x(indirect_inline_data,	19)			\
 	x(alloc_v2,		20)			\
 	x(subvolume,		21)			\
-	x(snapshot,		22)
+	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)
 
 enum bch_bkey_type {
 #define x(name, nr) KEY_TYPE_##name	= nr,
@@ -374,6 +400,10 @@ struct bch_hash_whiteout {
 	struct bch_val		v;
 };
 
+struct bch_set {
+	struct bch_val		v;
+};
+
 /* Extents */
 
 /*
@@ -451,15 +481,16 @@ struct bch_hash_whiteout {
 struct bch_csum {
 	__le64			lo;
 	__le64			hi;
-} __attribute__((packed, aligned(8)));
+} __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
+	x(stripe_ptr,		4)		\
+	x(rebalance,		5)
+#define BCH_EXTENT_ENTRY_MAX	6
 
 enum bch_extent_entry_type {
 #define x(f, n) BCH_EXTENT_ENTRY_##f = n,
@@ -488,7 +519,7 @@ struct bch_extent_crc32 {
 				_compressed_size:7,
 				type:2;
 #endif
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define CRC32_SIZE_MAX		(1U << 7)
 #define CRC32_NONCE_MAX		0
@@ -514,7 +545,7 @@ struct bch_extent_crc64 {
 				type:3;
 #endif
 	__u64			csum_lo;
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define CRC64_SIZE_MAX		(1U << 9)
 #define CRC64_NONCE_MAX		((1U << 10) - 1)
@@ -538,7 +569,7 @@ struct bch_extent_crc128 {
 				type:4;
 #endif
 	struct bch_csum		csum;
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define CRC128_SIZE_MAX		(1U << 13)
 #define CRC128_NONCE_MAX	((1U << 13) - 1)
@@ -551,7 +582,7 @@ struct bch_extent_ptr {
 	__u64			type:1,
 				cached:1,
 				unused:1,
-				reservation:1,
+				unwritten:1,
 				offset:44, /* 8 petabytes */
 				dev:8,
 				gen:8;
@@ -559,12 +590,12 @@ struct bch_extent_ptr {
 	__u64			gen:8,
 				dev:8,
 				offset:44,
-				reservation:1,
+				unwritten:1,
 				unused:1,
 				cached:1,
 				type:1;
 #endif
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 struct bch_extent_stripe_ptr {
 #if defined(__LITTLE_ENDIAN_BITFIELD)
@@ -594,6 +625,20 @@ struct bch_extent_reservation {
 #endif
 };
 
+struct bch_extent_rebalance {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	__u64			type:7,
+				unused:33,
+				compression:8,
+				target:16;
+#elif defined (__BIG_ENDIAN_BITFIELD)
+	__u64			target:16,
+				compression:8,
+				unused:33,
+				type:7;
+#endif
+};
+
 union bch_extent_entry {
 #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ ||  __BITS_PER_LONG == 64
 	unsigned long			type;
@@ -614,9 +659,9 @@ union bch_extent_entry {
 struct bch_btree_ptr {
 	struct bch_val		v;
 
-	struct bch_extent_ptr	start[0];
 	__u64			_data[0];
-} __attribute__((packed, aligned(8)));
+	struct bch_extent_ptr	start[];
+} __packed __aligned(8);
 
 struct bch_btree_ptr_v2 {
 	struct bch_val		v;
@@ -626,18 +671,18 @@ struct bch_btree_ptr_v2 {
 	__le16			sectors_written;
 	__le16			flags;
 	struct bpos		min_key;
-	struct bch_extent_ptr	start[0];
 	__u64			_data[0];
-} __attribute__((packed, aligned(8)));
+	struct bch_extent_ptr	start[];
+} __packed __aligned(8);
 
 LE16_BITMASK(BTREE_PTR_RANGE_UPDATED,	struct bch_btree_ptr_v2, flags, 0, 1);
 
 struct bch_extent {
 	struct bch_val		v;
 
-	union bch_extent_entry	start[0];
 	__u64			_data[0];
-} __attribute__((packed, aligned(8)));
+	union bch_extent_entry	start[];
+} __packed __aligned(8);
 
 struct bch_reservation {
 	struct bch_val		v;
@@ -645,12 +690,12 @@ struct bch_reservation {
 	__le32			generation;
 	__u8			nr_replicas;
 	__u8			pad[3];
-} __attribute__((packed, aligned(8)));
+} __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))
+	  sizeof(struct bch_extent_ptr)) / sizeof(__u64))
 
 /* Maximum possible size of an entire extent value: */
 #define BKEY_EXTENT_VAL_U64s_MAX				\
@@ -662,7 +707,7 @@ struct bch_reservation {
 /* 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))
+	  sizeof(struct bch_extent_ptr) * BCH_REPLICAS_MAX) / sizeof(__u64))
 #define BKEY_BTREE_PTR_U64s_MAX					\
 	(BKEY_U64s + BKEY_BTREE_PTR_VAL_U64s_MAX)
 
@@ -679,20 +724,45 @@ struct bch_inode {
 	__le32			bi_flags;
 	__le16			bi_mode;
 	__u8			fields[0];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
+
+struct bch_inode_v2 {
+	struct bch_val		v;
+
+	__le64			bi_journal_seq;
+	__le64			bi_hash_seed;
+	__le64			bi_flags;
+	__le16			bi_mode;
+	__u8			fields[0];
+} __packed __aligned(8);
+
+struct bch_inode_v3 {
+	struct bch_val		v;
+
+	__le64			bi_journal_seq;
+	__le64			bi_hash_seed;
+	__le64			bi_flags;
+	__le64			bi_sectors;
+	__le64			bi_size;
+	__le64			bi_version;
+	__u8			fields[0];
+} __packed __aligned(8);
+
+#define INODEv3_FIELDS_START_INITIAL	6
+#define INODEv3_FIELDS_START_CUR	(offsetof(struct bch_inode_v3, fields) / sizeof(__u64))
 
 struct bch_inode_generation {
 	struct bch_val		v;
 
 	__le32			bi_generation;
 	__le32			pad;
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /*
  * bi_subvol and bi_parent_subvol are only set for subvolume roots:
  */
 
-#define BCH_INODE_FIELDS()			\
+#define BCH_INODE_FIELDS_v2()			\
 	x(bi_atime,			96)	\
 	x(bi_ctime,			96)	\
 	x(bi_mtime,			96)	\
@@ -719,6 +789,32 @@ struct bch_inode_generation {
 	x(bi_subvol,			32)	\
 	x(bi_parent_subvol,		32)
 
+#define BCH_INODE_FIELDS_v3()			\
+	x(bi_atime,			96)	\
+	x(bi_ctime,			96)	\
+	x(bi_mtime,			96)	\
+	x(bi_otime,			96)	\
+	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)	\
+	x(bi_dir,			64)	\
+	x(bi_dir_offset,		64)	\
+	x(bi_subvol,			32)	\
+	x(bi_parent_subvol,		32)	\
+	x(bi_nocow,			8)
+
 /* subset of BCH_INODE_FIELDS */
 #define BCH_INODE_OPTS()			\
 	x(data_checksum,		8)	\
@@ -729,7 +825,8 @@ struct bch_inode_generation {
 	x(promote_target,		16)	\
 	x(foreground_target,		16)	\
 	x(background_target,		16)	\
-	x(erasure_code,			16)
+	x(erasure_code,			16)	\
+	x(nocow,			8)
 
 enum inode_opt_id {
 #define x(name, ...)				\
@@ -744,16 +841,16 @@ 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_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,
-	__BCH_INODE_BACKPTR_UNTRUSTED = 8,
+	__BCH_INODE_I_SIZE_DIRTY	= 5,
+	__BCH_INODE_I_SECTORS_DIRTY	= 6,
+	__BCH_INODE_UNLINKED		= 7,
+	__BCH_INODE_BACKPTR_UNTRUSTED	= 8,
 
 	/* bits 20+ reserved for packed fields below: */
 };
@@ -772,6 +869,16 @@ 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);
 
+LE64_BITMASK(INODEv2_STR_HASH,	struct bch_inode_v2, bi_flags, 20, 24);
+LE64_BITMASK(INODEv2_NR_FIELDS,	struct bch_inode_v2, bi_flags, 24, 31);
+
+LE64_BITMASK(INODEv3_STR_HASH,	struct bch_inode_v3, bi_flags, 20, 24);
+LE64_BITMASK(INODEv3_NR_FIELDS,	struct bch_inode_v3, bi_flags, 24, 31);
+
+LE64_BITMASK(INODEv3_FIELDS_START,
+				struct bch_inode_v3, bi_flags, 31, 36);
+LE64_BITMASK(INODEv3_MODE,	struct bch_inode_v3, bi_flags, 36, 52);
+
 /* Dirents */
 
 /*
@@ -804,15 +911,12 @@ struct bch_dirent {
 	__u8			d_type;
 
 	__u8			d_name[];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define DT_SUBVOL	16
 #define BCH_DT_MAX	17
 
-#define BCH_NAME_MAX	(U8_MAX * sizeof(u64) -				\
-			 sizeof(struct bkey) -				\
-			 offsetof(struct bch_dirent, d_name))
-
+#define BCH_NAME_MAX	512
 
 /* Xattrs */
 
@@ -828,7 +932,7 @@ struct bch_xattr {
 	__u8			x_name_len;
 	__le16			x_val_len;
 	__u8			x_name[];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /* Bucket/allocation information: */
 
@@ -837,7 +941,7 @@ struct bch_alloc {
 	__u8			fields;
 	__u8			gen;
 	__u8			data[];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define BCH_ALLOC_FIELDS_V1()			\
 	x(read_time,		16)		\
@@ -849,6 +953,12 @@ struct bch_alloc {
 	x(stripe,		32)		\
 	x(stripe_redundancy,	8)
 
+enum {
+#define x(name, _bits) BCH_ALLOC_FIELD_V1_##name,
+	BCH_ALLOC_FIELDS_V1()
+#undef x
+};
+
 struct bch_alloc_v2 {
 	struct bch_val		v;
 	__u8			nr_fields;
@@ -856,22 +966,74 @@ struct bch_alloc_v2 {
 	__u8			oldest_gen;
 	__u8			data_type;
 	__u8			data[];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #define BCH_ALLOC_FIELDS_V2()			\
 	x(read_time,		64)		\
 	x(write_time,		64)		\
-	x(dirty_sectors,	16)		\
-	x(cached_sectors,	16)		\
+	x(dirty_sectors,	32)		\
+	x(cached_sectors,	32)		\
 	x(stripe,		32)		\
 	x(stripe_redundancy,	8)
 
-enum {
-#define x(name, _bits) BCH_ALLOC_FIELD_V1_##name,
-	BCH_ALLOC_FIELDS_V1()
-#undef x
-	BCH_ALLOC_FIELD_NR
-};
+struct bch_alloc_v3 {
+	struct bch_val		v;
+	__le64			journal_seq;
+	__le32			flags;
+	__u8			nr_fields;
+	__u8			gen;
+	__u8			oldest_gen;
+	__u8			data_type;
+	__u8			data[];
+} __packed __aligned(8);
+
+LE32_BITMASK(BCH_ALLOC_V3_NEED_DISCARD,struct bch_alloc_v3, flags,  0,  1)
+LE32_BITMASK(BCH_ALLOC_V3_NEED_INC_GEN,struct bch_alloc_v3, flags,  1,  2)
+
+struct bch_alloc_v4 {
+	struct bch_val		v;
+	__u64			journal_seq;
+	__u32			flags;
+	__u8			gen;
+	__u8			oldest_gen;
+	__u8			data_type;
+	__u8			stripe_redundancy;
+	__u32			dirty_sectors;
+	__u32			cached_sectors;
+	__u64			io_time[2];
+	__u32			stripe;
+	__u32			nr_external_backpointers;
+	__u64			fragmentation_lru;
+} __packed __aligned(8);
+
+#define BCH_ALLOC_V4_U64s_V0	6
+#define BCH_ALLOC_V4_U64s	(sizeof(struct bch_alloc_v4) / sizeof(__u64))
+
+BITMASK(BCH_ALLOC_V4_NEED_DISCARD,	struct bch_alloc_v4, flags,  0,  1)
+BITMASK(BCH_ALLOC_V4_NEED_INC_GEN,	struct bch_alloc_v4, flags,  1,  2)
+BITMASK(BCH_ALLOC_V4_BACKPOINTERS_START,struct bch_alloc_v4, flags,  2,  8)
+BITMASK(BCH_ALLOC_V4_NR_BACKPOINTERS,	struct bch_alloc_v4, flags,  8,  14)
+
+#define BCH_ALLOC_V4_NR_BACKPOINTERS_MAX	40
+
+struct bch_backpointer {
+	struct bch_val		v;
+	__u8			btree_id;
+	__u8			level;
+	__u8			data_type;
+	__u64			bucket_offset:40;
+	__u32			bucket_len;
+	struct bpos		pos;
+} __packed __aligned(8);
+
+#define KEY_TYPE_BUCKET_GENS_BITS	8
+#define KEY_TYPE_BUCKET_GENS_NR		(1U << KEY_TYPE_BUCKET_GENS_BITS)
+#define KEY_TYPE_BUCKET_GENS_MASK	(KEY_TYPE_BUCKET_GENS_NR - 1)
+
+struct bch_bucket_gens {
+	struct bch_val		v;
+	u8			gens[KEY_TYPE_BUCKET_GENS_NR];
+} __packed __aligned(8);
 
 /* Quotas: */
 
@@ -896,7 +1058,7 @@ struct bch_quota_counter {
 struct bch_quota {
 	struct bch_val		v;
 	struct bch_quota_counter c[Q_COUNTERS];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /* Erasure coding */
 
@@ -911,8 +1073,8 @@ struct bch_stripe {
 	__u8			csum_type;
 	__u8			pad;
 
-	struct bch_extent_ptr	ptrs[0];
-} __attribute__((packed, aligned(8)));
+	struct bch_extent_ptr	ptrs[];
+} __packed __aligned(8);
 
 /* Reflink: */
 
@@ -929,14 +1091,14 @@ struct bch_reflink_p {
 	 */
 	__le32			front_pad;
 	__le32			back_pad;
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 struct bch_reflink_v {
 	struct bch_val		v;
 	__le64			refcount;
 	union bch_extent_entry	start[0];
 	__u64			_data[0];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 struct bch_indirect_inline_data {
 	struct bch_val		v;
@@ -962,6 +1124,14 @@ struct bch_subvolume {
 	__le32			flags;
 	__le32			snapshot;
 	__le64			inode;
+	/*
+	 * Snapshot subvolumes form a tree, separate from the snapshot nodes
+	 * tree - if this subvolume is a snapshot, this is the ID of the
+	 * subvolume it was created from:
+	 */
+	__le32			parent;
+	__le32			pad;
+	bch_le128		otime;
 };
 
 LE32_BITMASK(BCH_SUBVOLUME_RO,		struct bch_subvolume, flags,  0,  1)
@@ -970,6 +1140,7 @@ LE32_BITMASK(BCH_SUBVOLUME_RO,		struct bch_subvolume, flags,  0,  1)
  * can delete it (or whether it should just be rm -rf'd)
  */
 LE32_BITMASK(BCH_SUBVOLUME_SNAP,	struct bch_subvolume, flags,  1,  2)
+LE32_BITMASK(BCH_SUBVOLUME_UNLINKED,	struct bch_subvolume, flags,  2,  3)
 
 /* Snapshots */
 
@@ -979,7 +1150,9 @@ struct bch_snapshot {
 	__le32			parent;
 	__le32			children[2];
 	__le32			subvol;
-	__le32			pad;
+	__le32			tree;
+	__le32			depth;
+	__le32			skip[3];
 };
 
 LE32_BITMASK(BCH_SNAPSHOT_DELETED,	struct bch_snapshot, flags,  0,  1)
@@ -987,6 +1160,28 @@ LE32_BITMASK(BCH_SNAPSHOT_DELETED,	struct bch_snapshot, flags,  0,  1)
 /* True if a subvolume points to this snapshot node: */
 LE32_BITMASK(BCH_SNAPSHOT_SUBVOL,	struct bch_snapshot, flags,  1,  2)
 
+/*
+ * Snapshot trees:
+ *
+ * The snapshot_trees btree gives us persistent indentifier for each tree of
+ * bch_snapshot nodes, and allow us to record and easily find the root/master
+ * subvolume that other snapshots were created from:
+ */
+struct bch_snapshot_tree {
+	struct bch_val		v;
+	__le32			master_subvol;
+	__le32			root_snapshot;
+};
+
+/* LRU btree: */
+
+struct bch_lru {
+	struct bch_val		v;
+	__le64			idx;
+} __packed __aligned(8);
+
+#define LRU_ID_STRIPES		(1U << 16)
+
 /* Optional/variable size superblock sections: */
 
 struct bch_sb_field {
@@ -995,16 +1190,18 @@ 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,	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)
 
 enum bch_sb_field_type {
 #define x(f, nr)	BCH_SB_FIELD_##f = nr,
@@ -1013,6 +1210,14 @@ 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 {
@@ -1020,12 +1225,21 @@ struct bch_sb_field_journal {
 	__le64			buckets[0];
 };
 
+struct bch_sb_field_journal_v2 {
+	struct bch_sb_field	field;
+
+	struct bch_sb_field_journal_v2_entry {
+		__le64		start;
+		__le64		nr;
+	}			d[0];
+};
+
 /* BCH_SB_FIELD_members: */
 
 #define BCH_MIN_NR_NBUCKETS	(1 << 6)
 
 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 */
@@ -1036,12 +1250,13 @@ struct bch_member {
 };
 
 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)
+/* 4-14 unused, was TIER, HAS_(META)DATA, REPLACEMENT */
 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)
+LE64_BITMASK(BCH_MEMBER_FREESPACE_INITIALIZED,
+					struct bch_member, flags[0], 30, 31)
 
 #if 0
 LE64_BITMASK(BCH_MEMBER_NR_READ_ERRORS,	struct bch_member, flags[1], 0,  20);
@@ -1061,18 +1276,6 @@ enum bch_member_state {
 	BCH_MEMBER_STATE_NR
 };
 
-#define BCH_CACHE_REPLACEMENT_POLICIES()	\
-	x(lru,		0)			\
-	x(fifo,		1)			\
-	x(random,	2)
-
-enum bch_cache_replacement_policies {
-#define x(t, n) BCH_CACHE_REPLACEMENT_##t = n,
-	BCH_CACHE_REPLACEMENT_POLICIES()
-#undef x
-	BCH_CACHE_REPLACEMENT_NR
-};
-
 struct bch_sb_field_members {
 	struct bch_sb_field	field;
 	struct bch_member	members[0];
@@ -1089,10 +1292,10 @@ 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;
@@ -1129,13 +1332,16 @@ 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,
@@ -1144,23 +1350,46 @@ 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));
+} __packed;
 
 struct bch_sb_field_replicas_v0 {
 	struct bch_sb_field	field;
 	struct bch_replicas_entry_v0 entries[0];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 struct bch_replicas_entry {
 	__u8			data_type;
 	__u8			nr_devs;
 	__u8			nr_required;
 	__u8			devs[0];
-} __attribute__((packed));
+} __packed;
 
 #define replicas_entry_bytes(_i)					\
 	(offsetof(typeof(*(_i)), devs) + (_i)->nr_devs)
@@ -1168,7 +1397,7 @@ struct bch_replicas_entry {
 struct bch_sb_field_replicas {
 	struct bch_sb_field	field;
 	struct bch_replicas_entry entries[0];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /* BCH_SB_FIELD_quota: */
 
@@ -1185,7 +1414,7 @@ struct bch_sb_quota_type {
 struct bch_sb_field_quota {
 	struct bch_sb_field		field;
 	struct bch_sb_quota_type	q[QTYP_NR];
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /* BCH_SB_FIELD_disk_groups: */
 
@@ -1194,7 +1423,7 @@ 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)
@@ -1203,7 +1432,100 @@ 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)));
+} __packed __aligned(8);
+
+/* BCH_SB_FIELD_counters */
+
+#define BCH_PERSISTENT_COUNTERS()				\
+	x(io_read,					0)	\
+	x(io_write,					1)	\
+	x(io_move,					2)	\
+	x(bucket_invalidate,				3)	\
+	x(bucket_discard,				4)	\
+	x(bucket_alloc,					5)	\
+	x(bucket_alloc_fail,				6)	\
+	x(btree_cache_scan,				7)	\
+	x(btree_cache_reap,				8)	\
+	x(btree_cache_cannibalize,			9)	\
+	x(btree_cache_cannibalize_lock,			10)	\
+	x(btree_cache_cannibalize_lock_fail,		11)	\
+	x(btree_cache_cannibalize_unlock,		12)	\
+	x(btree_node_write,				13)	\
+	x(btree_node_read,				14)	\
+	x(btree_node_compact,				15)	\
+	x(btree_node_merge,				16)	\
+	x(btree_node_split,				17)	\
+	x(btree_node_rewrite,				18)	\
+	x(btree_node_alloc,				19)	\
+	x(btree_node_free,				20)	\
+	x(btree_node_set_root,				21)	\
+	x(btree_path_relock_fail,			22)	\
+	x(btree_path_upgrade_fail,			23)	\
+	x(btree_reserve_get_fail,			24)	\
+	x(journal_entry_full,				25)	\
+	x(journal_full,					26)	\
+	x(journal_reclaim_finish,			27)	\
+	x(journal_reclaim_start,			28)	\
+	x(journal_write,				29)	\
+	x(read_promote,					30)	\
+	x(read_bounce,					31)	\
+	x(read_split,					33)	\
+	x(read_retry,					32)	\
+	x(read_reuse_race,				34)	\
+	x(move_extent_read,				35)	\
+	x(move_extent_write,				36)	\
+	x(move_extent_finish,				37)	\
+	x(move_extent_fail,				38)	\
+	x(move_extent_alloc_mem_fail,			39)	\
+	x(copygc,					40)	\
+	x(copygc_wait,					41)	\
+	x(gc_gens_end,					42)	\
+	x(gc_gens_start,				43)	\
+	x(trans_blocked_journal_reclaim,		44)	\
+	x(trans_restart_btree_node_reused,		45)	\
+	x(trans_restart_btree_node_split,		46)	\
+	x(trans_restart_fault_inject,			47)	\
+	x(trans_restart_iter_upgrade,			48)	\
+	x(trans_restart_journal_preres_get,		49)	\
+	x(trans_restart_journal_reclaim,		50)	\
+	x(trans_restart_journal_res_get,		51)	\
+	x(trans_restart_key_cache_key_realloced,	52)	\
+	x(trans_restart_key_cache_raced,		53)	\
+	x(trans_restart_mark_replicas,			54)	\
+	x(trans_restart_mem_realloced,			55)	\
+	x(trans_restart_memory_allocation_failure,	56)	\
+	x(trans_restart_relock,				57)	\
+	x(trans_restart_relock_after_fill,		58)	\
+	x(trans_restart_relock_key_cache_fill,		59)	\
+	x(trans_restart_relock_next_node,		60)	\
+	x(trans_restart_relock_parent_for_fill,		61)	\
+	x(trans_restart_relock_path,			62)	\
+	x(trans_restart_relock_path_intent,		63)	\
+	x(trans_restart_too_many_iters,			64)	\
+	x(trans_restart_traverse,			65)	\
+	x(trans_restart_upgrade,			66)	\
+	x(trans_restart_would_deadlock,			67)	\
+	x(trans_restart_would_deadlock_write,		68)	\
+	x(trans_restart_injected,			69)	\
+	x(trans_restart_key_cache_upgrade,		70)	\
+	x(trans_traverse_all,				71)	\
+	x(transaction_commit,				72)	\
+	x(write_super,					73)	\
+	x(trans_restart_would_deadlock_recursion_limit,	74)	\
+	x(trans_restart_write_buffer_flush,		75)	\
+	x(trans_restart_split_race,			76)
+
+enum bch_persistent_counters {
+#define x(t, n, ...) BCH_COUNTER_##t,
+	BCH_PERSISTENT_COUNTERS()
+#undef x
+	BCH_COUNTER_NR
+};
+
+struct bch_sb_field_counters {
+	struct bch_sb_field	field;
+	__le64			d[0];
+};
 
 /*
  * On clean shutdown, store btree roots and current journal sequence number in
@@ -1257,36 +1579,86 @@ 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)
+
+#define RECOVERY_PASS_ALL_FSCK		(1ULL << 63)
+
+#define BCH_METADATA_VERSIONS()						\
+	x(bkey_renumber,		BCH_VERSION(0, 10),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(inode_btree_change,		BCH_VERSION(0, 11),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(snapshot,			BCH_VERSION(0, 12),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(inode_backpointers,		BCH_VERSION(0, 13),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(btree_ptr_sectors_written,	BCH_VERSION(0, 14),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(snapshot_2,			BCH_VERSION(0, 15),		\
+	  BIT_ULL(BCH_RECOVERY_PASS_fs_upgrade_for_subvolumes)|		\
+	  BIT_ULL(BCH_RECOVERY_PASS_initialize_subvolumes)|		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(reflink_p_fix,		BCH_VERSION(0, 16),		\
+	  BIT_ULL(BCH_RECOVERY_PASS_fix_reflink_p))			\
+	x(subvol_dirent,		BCH_VERSION(0, 17),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(inode_v2,			BCH_VERSION(0, 18),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(freespace,			BCH_VERSION(0, 19),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(alloc_v4,			BCH_VERSION(0, 20),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(new_data_types,		BCH_VERSION(0, 21),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(backpointers,			BCH_VERSION(0, 22),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(inode_v3,			BCH_VERSION(0, 23),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(unwritten_extents,		BCH_VERSION(0, 24),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(bucket_gens,			BCH_VERSION(0, 25),		\
+	  BIT_ULL(BCH_RECOVERY_PASS_bucket_gens_init)|			\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(lru_v2,			BCH_VERSION(0, 26),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(fragmentation_lru,		BCH_VERSION(0, 27),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(no_bps_in_alloc_keys,		BCH_VERSION(0, 28),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(snapshot_trees,		BCH_VERSION(0, 29),		\
+	  RECOVERY_PASS_ALL_FSCK)					\
+	x(major_minor,			BCH_VERSION(1,  0),		\
+	  0)								\
+	x(snapshot_skiplists,		BCH_VERSION(1,  1),		\
+	  BIT_ULL(BCH_RECOVERY_PASS_check_snapshots))			\
+	x(deleted_inodes,		BCH_VERSION(1,  2),		\
+	  BIT_ULL(BCH_RECOVERY_PASS_check_inodes))
 
 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_snapshot		= 12,
-	bcachefs_metadata_version_inode_backpointers	= 13,
-	bcachefs_metadata_version_btree_ptr_sectors_written = 14,
-	bcachefs_metadata_version_snapshot_2		= 15,
-	bcachefs_metadata_version_reflink_p_fix		= 16,
-	bcachefs_metadata_version_subvol_dirent		= 17,
-	bcachefs_metadata_version_max			= 18,
+	bcachefs_metadata_version_min = 9,
+#define x(t, n, upgrade_passes)	bcachefs_metadata_version_##t = n,
+	BCH_METADATA_VERSIONS()
+#undef x
+	bcachefs_metadata_version_max
 };
 
+static const unsigned bcachefs_metadata_required_upgrade_below = bcachefs_metadata_version_major_minor;
+
 #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
 
@@ -1296,7 +1668,7 @@ 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
@@ -1311,9 +1683,9 @@ 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;
@@ -1337,7 +1709,7 @@ struct bch_sb {
 		struct bch_sb_field start[0];
 		__le64		_data[0];
 	};
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 /*
  * Flags:
@@ -1380,7 +1752,7 @@ LE64_BITMASK(BCH_SB_HAS_TOPOLOGY_ERRORS,struct bch_sb, flags[0], 61, 62);
 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);
@@ -1400,7 +1772,7 @@ 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);
 
@@ -1408,6 +1780,43 @@ 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:
@@ -1415,7 +1824,7 @@ LE64_BITMASK(BCH_SB_INODES_USE_KEY_CACHE,struct bch_sb, flags[3], 29, 30);
  * 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()			\
@@ -1475,6 +1884,17 @@ enum bch_sb_compat {
 
 /* 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
@@ -1491,12 +1911,17 @@ enum bch_error_actions {
 	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()		\
@@ -1511,34 +1936,39 @@ enum bch_str_hash_opts {
 	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_XXHASH			= 7,
-	BCH_CSUM_NR			= 8,
+#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_XXHASH]			= 8,
-	[BCH_CSUM_CHACHA20_POLY1305_80]		= 10,
-	[BCH_CSUM_CHACHA20_POLY1305_128]	= 16,
+	[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;
@@ -1594,8 +2024,11 @@ 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
 
@@ -1605,6 +2038,7 @@ enum bch_compression_opts {
 static inline __le64 __bch2_sb_magic(struct bch_sb *sb)
 {
 	__le64 ret;
+
 	memcpy(&ret, &sb->uuid, sizeof(ret));
 	return ret;
 }
@@ -1632,7 +2066,9 @@ static inline __u64 __bset_magic(struct bch_sb *sb)
 	x(usage,		5)		\
 	x(data_usage,		6)		\
 	x(clock,		7)		\
-	x(dev_usage,		8)
+	x(dev_usage,		8)		\
+	x(log,			9)		\
+	x(overwrite,		10)
 
 enum {
 #define x(f, nr)	BCH_JSET_ENTRY_##f	= nr,
@@ -1662,36 +2098,41 @@ 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));
+} __packed;
 
 struct jset_entry_clock {
 	struct jset_entry	entry;
 	__u8			rw;
 	__u8			pad[7];
 	__le64			time;
-} __attribute__((packed));
+} __packed;
 
 struct jset_entry_dev_usage_type {
 	__le64			buckets;
 	__le64			sectors;
 	__le64			fragmented;
-} __attribute__((packed));
+} __packed;
 
 struct jset_entry_dev_usage {
 	struct jset_entry	entry;
@@ -1699,10 +2140,21 @@ struct jset_entry_dev_usage {
 	__u32			pad;
 
 	__le64			buckets_ec;
-	__le64			buckets_unavailable;
+	__le64			_buckets_unavailable; /* No longer used */
 
 	struct jset_entry_dev_usage_type d[];
-} __attribute__((packed));
+};
+
+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;
 
 /*
  * On disk format for a journal entry:
@@ -1737,7 +2189,7 @@ struct jset {
 		struct jset_entry start[0];
 		__u64		_data[0];
 	};
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 LE32_BITMASK(JSET_CSUM_TYPE,	struct jset, flags, 0, 4);
 LE32_BITMASK(JSET_BIG_ENDIAN,	struct jset, flags, 4, 5);
@@ -1747,20 +2199,69 @@ LE32_BITMASK(JSET_NO_FLUSH,	struct jset, flags, 5, 6);
 
 /* Btree: */
 
-#define BCH_BTREE_IDS()				\
-	x(extents,	0)			\
-	x(inodes,	1)			\
-	x(dirents,	2)			\
-	x(xattrs,	3)			\
-	x(alloc,	4)			\
-	x(quotas,	5)			\
-	x(stripes,	6)			\
-	x(reflink,	7)			\
-	x(subvolumes,	8)			\
-	x(snapshots,	9)
+enum btree_id_flags {
+	BTREE_ID_EXTENTS	= BIT(0),
+	BTREE_ID_SNAPSHOTS	= BIT(1),
+	BTREE_ID_DATA		= BIT(2),
+};
+
+#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_SNAPSHOTS,			\
+	  BIT_ULL(KEY_TYPE_set))
 
 enum btree_id {
-#define x(kwd, val) BTREE_ID_##kwd = val,
+#define x(name, nr, ...) BTREE_ID_##name = nr,
 	BCH_BTREE_IDS()
 #undef x
 	BTREE_ID_NR
@@ -1796,7 +2297,7 @@ struct bset {
 		struct bkey_packed start[0];
 		__u64		_data[0];
 	};
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 LE32_BITMASK(BSET_CSUM_TYPE,	struct bset, flags, 0, 4);
 
@@ -1829,15 +2330,27 @@ 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;
 
@@ -1847,9 +2360,8 @@ struct btree_node_entry {
 		__u8		pad[22];
 		__le16		u64s;
 		__u64		_data[0];
-
 	};
 	};
-} __attribute__((packed, aligned(8)));
+} __packed __aligned(8);
 
 #endif /* _BCACHEFS_FORMAT_H */