[linux-2.6-block.git] / drivers / nvdimm / btt.h

/* SPDX-License-Identifier: GPL-2.0-only */
/*
 * Block Translation Table library
 * Copyright (c) 2014-2015, Intel Corporation.
 */

#ifndef _LINUX_BTT_H
#define _LINUX_BTT_H

#include <linux/badblocks.h>
#include <linux/types.h>

#define BTT_SIG_LEN 16
#define BTT_SIG "BTT_ARENA_INFO\0"
#define MAP_ENT_SIZE 4
#define MAP_TRIM_SHIFT 31
#define MAP_TRIM_MASK (1 << MAP_TRIM_SHIFT)
#define MAP_ERR_SHIFT 30
#define MAP_ERR_MASK (1 << MAP_ERR_SHIFT)
#define MAP_LBA_MASK (~((1 << MAP_TRIM_SHIFT) | (1 << MAP_ERR_SHIFT)))
#define MAP_ENT_NORMAL 0xC0000000
#define LOG_GRP_SIZE sizeof(struct log_group)
#define LOG_ENT_SIZE sizeof(struct log_entry)
#define ARENA_MIN_SIZE (1UL << 24)	/* 16 MB */
#define ARENA_MAX_SIZE (1ULL << 39)	/* 512 GB */
#define RTT_VALID (1UL << 31)
#define RTT_INVALID 0
#define BTT_PG_SIZE 4096
#define BTT_DEFAULT_NFREE ND_MAX_LANES
#define LOG_SEQ_INIT 1

#define IB_FLAG_ERROR 0x00000001
#define IB_FLAG_ERROR_MASK 0x00000001

#define ent_lba(ent) (ent & MAP_LBA_MASK)
#define ent_e_flag(ent) (!!(ent & MAP_ERR_MASK))
#define ent_z_flag(ent) (!!(ent & MAP_TRIM_MASK))
#define set_e_flag(ent) (ent |= MAP_ERR_MASK)
/* 'normal' is both e and z flags set */
#define ent_normal(ent) (ent_e_flag(ent) && ent_z_flag(ent))

enum btt_init_state {
	INIT_UNCHECKED = 0,
	INIT_NOTFOUND,
	INIT_READY
};

/*
 * A log group represents one log 'lane', and consists of four log entries.
 * Two of the four entries are valid entries, and the remaining two are
 * padding. Due to an old bug in the padding location, we need to perform a
 * test to determine the padding scheme being used, and use that scheme
 * thereafter.
 *
 * In kernels prior to 4.15, 'log group' would have actual log entries at
 * indices (0, 2) and padding at indices (1, 3), where as the correct/updated
 * format has log entries at indices (0, 1) and padding at indices (2, 3).
 *
 * Old (pre 4.15) format:
 * +-----------------+-----------------+
 * |      ent[0]     |      ent[1]     |
 * |       16B       |       16B       |
 * | lba/old/new/seq |       pad       |
 * +-----------------------------------+
 * |      ent[2]     |      ent[3]     |
 * |       16B       |       16B       |
 * | lba/old/new/seq |       pad       |
 * +-----------------+-----------------+
 *
 * New format:
 * +-----------------+-----------------+
 * |      ent[0]     |      ent[1]     |
 * |       16B       |       16B       |
 * | lba/old/new/seq | lba/old/new/seq |
 * +-----------------------------------+
 * |      ent[2]     |      ent[3]     |
 * |       16B       |       16B       |
 * |       pad       |       pad       |
 * +-----------------+-----------------+
 *
 * We detect during start-up which format is in use, and set
 * arena->log_index[(0, 1)] with the detected format.
 */

struct log_entry {
	__le32 lba;
	__le32 old_map;
	__le32 new_map;
	__le32 seq;
};

struct log_group {
	struct log_entry ent[4];
};

struct btt_sb {
	u8 signature[BTT_SIG_LEN];
	u8 uuid[16];
	u8 parent_uuid[16];
	__le32 flags;
	__le16 version_major;
	__le16 version_minor;
	__le32 external_lbasize;
	__le32 external_nlba;
	__le32 internal_lbasize;
	__le32 internal_nlba;
	__le32 nfree;
	__le32 infosize;
	__le64 nextoff;
	__le64 dataoff;
	__le64 mapoff;
	__le64 logoff;
	__le64 info2off;
	u8 padding[3968];
	__le64 checksum;
};

struct free_entry {
	u32 block;
	u8 sub;
	u8 seq;
	u8 has_err;
};

struct aligned_lock {
	union {
		spinlock_t lock;
		u8 cacheline_padding[L1_CACHE_BYTES];
	};
};

/**
 * struct arena_info - handle for an arena
 * @size:		Size in bytes this arena occupies on the raw device.
 *			This includes arena metadata.
 * @external_lba_start:	The first external LBA in this arena.
 * @internal_nlba:	Number of internal blocks available in the arena
 *			including nfree reserved blocks
 * @internal_lbasize:	Internal and external lba sizes may be different as
 *			we can round up 'odd' external lbasizes such as 520B
 *			to be aligned.
 * @external_nlba:	Number of blocks contributed by the arena to the number
 *			reported to upper layers. (internal_nlba - nfree)
 * @external_lbasize:	LBA size as exposed to upper layers.
 * @nfree:		A reserve number of 'free' blocks that is used to
 *			handle incoming writes.
 * @version_major:	Metadata layout version major.
 * @version_minor:	Metadata layout version minor.
 * @sector_size:	The Linux sector size - 512 or 4096
 * @nextoff:		Offset in bytes to the start of the next arena.
 * @infooff:		Offset in bytes to the info block of this arena.
 * @dataoff:		Offset in bytes to the data area of this arena.
 * @mapoff:		Offset in bytes to the map area of this arena.
 * @logoff:		Offset in bytes to the log area of this arena.
 * @info2off:		Offset in bytes to the backup info block of this arena.
 * @freelist:		Pointer to in-memory list of free blocks
 * @rtt:		Pointer to in-memory "Read Tracking Table"
 * @map_locks:		Spinlocks protecting concurrent map writes
 * @nd_btt:		Pointer to parent nd_btt structure.
 * @list:		List head for list of arenas
 * @debugfs_dir:	Debugfs dentry
 * @flags:		Arena flags - may signify error states.
 * @err_lock:		Mutex for synchronizing error clearing.
 * @log_index:		Indices of the valid log entries in a log_group
 *
 * arena_info is a per-arena handle. Once an arena is narrowed down for an
 * IO, this struct is passed around for the duration of the IO.
 */
struct arena_info {
	u64 size;			/* Total bytes for this arena */
	u64 external_lba_start;
	u32 internal_nlba;
	u32 internal_lbasize;
	u32 external_nlba;
	u32 external_lbasize;
	u32 nfree;
	u16 version_major;
	u16 version_minor;
	u32 sector_size;
	/* Byte offsets to the different on-media structures */
	u64 nextoff;
	u64 infooff;
	u64 dataoff;
	u64 mapoff;
	u64 logoff;
	u64 info2off;
	/* Pointers to other in-memory structures for this arena */
	struct free_entry *freelist;
	u32 *rtt;
	struct aligned_lock *map_locks;
	struct nd_btt *nd_btt;
	struct list_head list;
	struct dentry *debugfs_dir;
	/* Arena flags */
	u32 flags;
	struct mutex err_lock;
	int log_index[2];
};

/**
 * struct btt - handle for a BTT instance
 * @btt_disk:		Pointer to the gendisk for BTT device
 * @btt_queue:		Pointer to the request queue for the BTT device
 * @arena_list:		Head of the list of arenas
 * @debugfs_dir:	Debugfs dentry
 * @nd_btt:		Parent nd_btt struct
 * @nlba:		Number of logical blocks exposed to the	upper layers
 *			after removing the amount of space needed by metadata
 * @rawsize:		Total size in bytes of the available backing device
 * @lbasize:		LBA size as requested and presented to upper layers.
 *			This is sector_size + size of any metadata.
 * @sector_size:	The Linux sector size - 512 or 4096
 * @lanes:		Per-lane spinlocks
 * @init_lock:		Mutex used for the BTT initialization
 * @init_state:		Flag describing the initialization state for the BTT
 * @num_arenas:		Number of arenas in the BTT instance
 * @phys_bb:		Pointer to the namespace's badblocks structure
 */
struct btt {
	struct gendisk *btt_disk;
	struct request_queue *btt_queue;
	struct list_head arena_list;
	struct dentry *debugfs_dir;
	struct nd_btt *nd_btt;
	u64 nlba;
	unsigned long long rawsize;
	u32 lbasize;
	u32 sector_size;
	struct nd_region *nd_region;
	struct mutex init_lock;
	int init_state;
	int num_arenas;
	struct badblocks *phys_bb;
};

bool nd_btt_arena_is_valid(struct nd_btt *nd_btt, struct btt_sb *super);
int nd_btt_version(struct nd_btt *nd_btt, struct nd_namespace_common *ndns,
		struct btt_sb *btt_sb);

#endif
Commit	Line	Data
2025cf9e	1	/* SPDX-License-Identifier: GPL-2.0-only */
8c2f7e86 DW	2	/*
	3	* Block Translation Table library
	4	* Copyright (c) 2014-2015, Intel Corporation.
8c2f7e86 DW	5	*/
	6
	7	#ifndef _LINUX_BTT_H
	8	#define _LINUX_BTT_H
	9
d9b83c75	10	#include <linux/badblocks.h>
8c2f7e86 DW	11	#include <linux/types.h>
	12
	13	#define BTT_SIG_LEN 16
	14	#define BTT_SIG "BTT_ARENA_INFO\0"
5212e11f VV	15	#define MAP_ENT_SIZE 4
	16	#define MAP_TRIM_SHIFT 31
	17	#define MAP_TRIM_MASK (1 << MAP_TRIM_SHIFT)
	18	#define MAP_ERR_SHIFT 30
	19	#define MAP_ERR_MASK (1 << MAP_ERR_SHIFT)
	20	#define MAP_LBA_MASK (~((1 << MAP_TRIM_SHIFT) \| (1 << MAP_ERR_SHIFT)))
	21	#define MAP_ENT_NORMAL 0xC0000000
24e3a7fb	22	#define LOG_GRP_SIZE sizeof(struct log_group)
5212e11f VV	23	#define LOG_ENT_SIZE sizeof(struct log_entry)
	24	#define ARENA_MIN_SIZE (1UL << 24) /* 16 MB */
	25	#define ARENA_MAX_SIZE (1ULL << 39) /* 512 GB */
	26	#define RTT_VALID (1UL << 31)
	27	#define RTT_INVALID 0
5212e11f VV	28	#define BTT_PG_SIZE 4096
	29	#define BTT_DEFAULT_NFREE ND_MAX_LANES
	30	#define LOG_SEQ_INIT 1
	31
	32	#define IB_FLAG_ERROR 0x00000001
	33	#define IB_FLAG_ERROR_MASK 0x00000001
	34
0595d539 VV	35	#define ent_lba(ent) (ent & MAP_LBA_MASK)
	36	#define ent_e_flag(ent) (!!(ent & MAP_ERR_MASK))
	37	#define ent_z_flag(ent) (!!(ent & MAP_TRIM_MASK))
d9b83c75	38	#define set_e_flag(ent) (ent \|= MAP_ERR_MASK)
9dedc73a VV	39	/* 'normal' is both e and z flags set */
9dedc73a VV	40	#define ent_normal(ent) (ent_e_flag(ent) && ent_z_flag(ent))
0595d539	41
5212e11f VV	42	enum btt_init_state {
	43	INIT_UNCHECKED = 0,
	44	INIT_NOTFOUND,
	45	INIT_READY
	46	};
	47
24e3a7fb VV	48	/*
	49	* A log group represents one log 'lane', and consists of four log entries.
	50	* Two of the four entries are valid entries, and the remaining two are
	51	* padding. Due to an old bug in the padding location, we need to perform a
	52	* test to determine the padding scheme being used, and use that scheme
	53	* thereafter.
	54	*
	55	* In kernels prior to 4.15, 'log group' would have actual log entries at
	56	* indices (0, 2) and padding at indices (1, 3), where as the correct/updated
	57	* format has log entries at indices (0, 1) and padding at indices (2, 3).
	58	*
	59	* Old (pre 4.15) format:
	60	* +-----------------+-----------------+
	61	* \| ent[0] \| ent[1] \|
	62	* \| 16B \| 16B \|
	63	* \| lba/old/new/seq \| pad \|
	64	* +-----------------------------------+
	65	* \| ent[2] \| ent[3] \|
	66	* \| 16B \| 16B \|
	67	* \| lba/old/new/seq \| pad \|
	68	* +-----------------+-----------------+
	69	*
	70	* New format:
	71	* +-----------------+-----------------+
	72	* \| ent[0] \| ent[1] \|
	73	* \| 16B \| 16B \|
	74	* \| lba/old/new/seq \| lba/old/new/seq \|
	75	* +-----------------------------------+
	76	* \| ent[2] \| ent[3] \|
	77	* \| 16B \| 16B \|
	78	* \| pad \| pad \|
	79	* +-----------------+-----------------+
	80	*
	81	* We detect during start-up which format is in use, and set
	82	* arena->log_index[(0, 1)] with the detected format.
	83	*/
	84
5212e11f VV	85	struct log_entry {
	86	__le32 lba;
	87	__le32 old_map;
	88	__le32 new_map;
	89	__le32 seq;
24e3a7fb VV	90	};
	91
	92	struct log_group {
	93	struct log_entry ent[4];
5212e11f	94	};
8c2f7e86 DW	95
	96	struct btt_sb {
	97	u8 signature[BTT_SIG_LEN];
	98	u8 uuid[16];
	99	u8 parent_uuid[16];
	100	__le32 flags;
	101	__le16 version_major;
	102	__le16 version_minor;
	103	__le32 external_lbasize;
	104	__le32 external_nlba;
	105	__le32 internal_lbasize;
	106	__le32 internal_nlba;
	107	__le32 nfree;
	108	__le32 infosize;
	109	__le64 nextoff;
	110	__le64 dataoff;
	111	__le64 mapoff;
	112	__le64 logoff;
	113	__le64 info2off;
	114	u8 padding[3968];
	115	__le64 checksum;
	116	};
	117
5212e11f VV	118	struct free_entry {
	119	u32 block;
	120	u8 sub;
	121	u8 seq;
d9b83c75	122	u8 has_err;
5212e11f VV	123	};
	124
	125	struct aligned_lock {
	126	union {
	127	spinlock_t lock;
	128	u8 cacheline_padding[L1_CACHE_BYTES];
	129	};
	130	};
	131
	132	/**
	133	* struct arena_info - handle for an arena
	134	* @size: Size in bytes this arena occupies on the raw device.
	135	* This includes arena metadata.
	136	* @external_lba_start: The first external LBA in this arena.
	137	* @internal_nlba: Number of internal blocks available in the arena
	138	* including nfree reserved blocks
	139	* @internal_lbasize: Internal and external lba sizes may be different as
	140	* we can round up 'odd' external lbasizes such as 520B
	141	* to be aligned.
	142	* @external_nlba: Number of blocks contributed by the arena to the number
	143	* reported to upper layers. (internal_nlba - nfree)
	144	* @external_lbasize: LBA size as exposed to upper layers.
	145	* @nfree: A reserve number of 'free' blocks that is used to
	146	* handle incoming writes.
	147	* @version_major: Metadata layout version major.
	148	* @version_minor: Metadata layout version minor.
75892004	149	* @sector_size: The Linux sector size - 512 or 4096
5212e11f VV	150	* @nextoff: Offset in bytes to the start of the next arena.
	151	* @infooff: Offset in bytes to the info block of this arena.
	152	* @dataoff: Offset in bytes to the data area of this arena.
	153	* @mapoff: Offset in bytes to the map area of this arena.
	154	* @logoff: Offset in bytes to the log area of this arena.
	155	* @info2off: Offset in bytes to the backup info block of this arena.
	156	* @freelist: Pointer to in-memory list of free blocks
	157	* @rtt: Pointer to in-memory "Read Tracking Table"
	158	* @map_locks: Spinlocks protecting concurrent map writes
	159	* @nd_btt: Pointer to parent nd_btt structure.
	160	* @list: List head for list of arenas
	161	* @debugfs_dir: Debugfs dentry
	162	* @flags: Arena flags - may signify error states.
13b7954c	163	* @err_lock: Mutex for synchronizing error clearing.
24e3a7fb	164	* @log_index: Indices of the valid log entries in a log_group
5212e11f VV	165	*
	166	* arena_info is a per-arena handle. Once an arena is narrowed down for an
	167	* IO, this struct is passed around for the duration of the IO.
	168	*/
	169	struct arena_info {
	170	u64 size; /* Total bytes for this arena */
	171	u64 external_lba_start;
	172	u32 internal_nlba;
	173	u32 internal_lbasize;
	174	u32 external_nlba;
	175	u32 external_lbasize;
	176	u32 nfree;
	177	u16 version_major;
	178	u16 version_minor;
75892004	179	u32 sector_size;
5212e11f VV	180	/* Byte offsets to the different on-media structures */
	181	u64 nextoff;
	182	u64 infooff;
	183	u64 dataoff;
	184	u64 mapoff;
	185	u64 logoff;
	186	u64 info2off;
	187	/* Pointers to other in-memory structures for this arena */
	188	struct free_entry *freelist;
	189	u32 *rtt;
	190	struct aligned_lock *map_locks;
	191	struct nd_btt *nd_btt;
	192	struct list_head list;
	193	struct dentry *debugfs_dir;
	194	/* Arena flags */
	195	u32 flags;
d9b83c75	196	struct mutex err_lock;
24e3a7fb	197	int log_index[2];
5212e11f VV	198	};
	199
	200	/**
	201	* struct btt - handle for a BTT instance
	202	* @btt_disk: Pointer to the gendisk for BTT device
	203	* @btt_queue: Pointer to the request queue for the BTT device
	204	* @arena_list: Head of the list of arenas
	205	* @debugfs_dir: Debugfs dentry
	206	* @nd_btt: Parent nd_btt struct
	207	* @nlba: Number of logical blocks exposed to the upper layers
	208	* after removing the amount of space needed by metadata
	209	* @rawsize: Total size in bytes of the available backing device
	210	* @lbasize: LBA size as requested and presented to upper layers.
	211	* This is sector_size + size of any metadata.
	212	* @sector_size: The Linux sector size - 512 or 4096
	213	* @lanes: Per-lane spinlocks
	214	* @init_lock: Mutex used for the BTT initialization
	215	* @init_state: Flag describing the initialization state for the BTT
	216	* @num_arenas: Number of arenas in the BTT instance
13b7954c	217	* @phys_bb: Pointer to the namespace's badblocks structure
5212e11f VV	218	*/
	219	struct btt {
	220	struct gendisk *btt_disk;
	221	struct request_queue *btt_queue;
	222	struct list_head arena_list;
	223	struct dentry *debugfs_dir;
	224	struct nd_btt *nd_btt;
	225	u64 nlba;
	226	unsigned long long rawsize;
	227	u32 lbasize;
	228	u32 sector_size;
	229	struct nd_region *nd_region;
	230	struct mutex init_lock;
	231	int init_state;
	232	int num_arenas;
d9b83c75	233	struct badblocks *phys_bb;
5212e11f	234	};
ab45e763 VV	235
ab45e763 VV	236	bool nd_btt_arena_is_valid(struct nd_btt nd_btt, struct btt_sb super);
14e49454 VV	237	int nd_btt_version(struct nd_btt nd_btt, struct nd_namespace_common ndns,
14e49454 VV	238	struct btt_sb *btt_sb);
ab45e763	239
8c2f7e86	240	#endif