Something went wrong. Try again.
Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Something went wrong. Try again.
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644/* SPDX-License-Identifier: GPL-2.0 */#ifndef _FS_CEPH_OSD_CLIENT_H#define _FS_CEPH_OSD_CLIENT_H
#include <linux/bitrev.h>#include <linux/completion.h>#include <linux/kref.h>#include <linux/mempool.h>#include <linux/rbtree.h>#include <linux/refcount.h>#include <linux/ktime.h>
#include <linux/ceph/types.h>#include <linux/ceph/osdmap.h>#include <linux/ceph/messenger.h>#include <linux/ceph/msgpool.h>#include <linux/ceph/auth.h>#include <linux/ceph/pagelist.h>
struct ceph_msg;struct ceph_snap_context;struct ceph_osd_request;struct ceph_osd_client;
/* * completion callback for async writepages */typedef void (*ceph_osdc_callback_t)(struct ceph_osd_request *);
#define CEPH_HOMELESS_OSD -1
/* * A single extent in a SPARSE_READ reply. * * Note that these come from the OSD as little-endian values. On BE arches, * we convert them in-place after receipt. */struct ceph_sparse_extent { u64 off; u64 len;} __packed;
/* Sparse read state machine state values */enum ceph_sparse_read_state { CEPH_SPARSE_READ_HDR = 0, CEPH_SPARSE_READ_EXTENTS, CEPH_SPARSE_READ_DATA_LEN, CEPH_SPARSE_READ_DATA_PRE, CEPH_SPARSE_READ_DATA,};
/* * A SPARSE_READ reply is a 32-bit count of extents, followed by an array of * 64-bit offset/length pairs, and then all of the actual file data * concatenated after it (sans holes). * * Unfortunately, we don't know how long the extent array is until we've * started reading the data section of the reply. The caller should send down * a destination buffer for the array, but we'll alloc one if it's too small * or if the caller doesn't. */struct ceph_sparse_read { enum ceph_sparse_read_state sr_state; /* state machine state */ u64 sr_req_off; /* orig request offset */ u64 sr_req_len; /* orig request length */ u64 sr_pos; /* current pos in buffer */ int sr_index; /* current extent index */ u32 sr_datalen; /* length of actual data */ u32 sr_count; /* extent count in reply */ int sr_ext_len; /* length of extent array */ struct ceph_sparse_extent *sr_extent; /* extent array */};
/* * A given osd we're communicating with. * * Note that the o_requests tree can be searched while holding the "lock" mutex * or the "o_requests_lock" spinlock. Insertion or removal requires both! */struct ceph_osd { refcount_t o_ref; int o_sparse_op_idx; struct ceph_osd_client *o_osdc; int o_osd; int o_incarnation; struct rb_node o_node; struct ceph_connection o_con; spinlock_t o_requests_lock; struct rb_root o_requests; struct rb_root o_linger_requests; struct rb_root o_backoff_mappings; struct rb_root o_backoffs_by_id; struct list_head o_osd_lru; struct ceph_auth_handshake o_auth; unsigned long lru_ttl; struct list_head o_keepalive_item; struct mutex lock; struct ceph_sparse_read o_sparse_read;};
#define CEPH_OSD_SLAB_OPS 2#define CEPH_OSD_MAX_OPS 16
enum ceph_osd_data_type { CEPH_OSD_DATA_TYPE_NONE = 0, CEPH_OSD_DATA_TYPE_PAGES, CEPH_OSD_DATA_TYPE_PAGELIST,#ifdef CONFIG_BLOCK CEPH_OSD_DATA_TYPE_BIO,#endif /* CONFIG_BLOCK */ CEPH_OSD_DATA_TYPE_BVECS, CEPH_OSD_DATA_TYPE_ITER,};
struct ceph_osd_data { enum ceph_osd_data_type type; union { struct { struct page **pages; u64 length; u32 alignment; bool pages_from_pool; bool own_pages; }; struct ceph_pagelist *pagelist;#ifdef CONFIG_BLOCK struct { struct ceph_bio_iter bio_pos; u32 bio_length; };#endif /* CONFIG_BLOCK */ struct { struct ceph_bvec_iter bvec_pos; u32 num_bvecs; }; struct iov_iter iter; };};
struct ceph_osd_req_op { u16 op; /* CEPH_OSD_OP_* */ u32 flags; /* CEPH_OSD_OP_FLAG_* */ u32 indata_len; /* request */ u32 outdata_len; /* reply */ s32 rval;
union { struct ceph_osd_data raw_data_in; struct { u64 offset, length; u64 truncate_size; u32 truncate_seq; int sparse_ext_cnt; struct ceph_sparse_extent *sparse_ext; struct ceph_osd_data osd_data; } extent; struct { u32 name_len; u32 value_len; __u8 cmp_op; /* CEPH_OSD_CMPXATTR_OP_* */ __u8 cmp_mode; /* CEPH_OSD_CMPXATTR_MODE_* */ struct ceph_osd_data osd_data; } xattr; struct { const char *class_name; const char *method_name; struct ceph_osd_data request_info; struct ceph_osd_data request_data; struct ceph_osd_data response_data; __u8 class_len; __u8 method_len; u32 indata_len; } cls; struct { u64 cookie; __u8 op; /* CEPH_OSD_WATCH_OP_ */ u32 gen; } watch; struct { struct ceph_osd_data request_data; } notify_ack; struct { u64 cookie; struct ceph_osd_data request_data; struct ceph_osd_data response_data; } notify; struct { struct ceph_osd_data response_data; } list_watchers; struct { u64 expected_object_size; u64 expected_write_size; u32 flags; /* CEPH_OSD_OP_ALLOC_HINT_FLAG_* */ } alloc_hint; struct { u64 snapid; u64 src_version; u8 flags; u32 src_fadvise_flags; struct ceph_osd_data osd_data; } copy_from; struct { u64 ver; } assert_ver; };};
struct ceph_osd_request_target { struct ceph_object_id base_oid; struct ceph_object_locator base_oloc; struct ceph_object_id target_oid; struct ceph_object_locator target_oloc;
struct ceph_pg pgid; /* last raw pg we mapped to */ struct ceph_spg spgid; /* last actual spg we mapped to */ u32 pg_num; u32 pg_num_mask; struct ceph_osds acting; struct ceph_osds up; int size; int min_size; bool sort_bitwise; bool recovery_deletes;
unsigned int flags; /* CEPH_OSD_FLAG_* */ bool used_replica; bool paused;
u32 epoch; u32 last_force_resend;
int osd;};
/* an in-flight request */struct ceph_osd_request { u64 r_tid; /* unique for this client */ struct rb_node r_node; struct rb_node r_mc_node; /* map check */ struct work_struct r_complete_work; struct ceph_osd *r_osd;
struct ceph_osd_request_target r_t;#define r_base_oid r_t.base_oid#define r_base_oloc r_t.base_oloc#define r_flags r_t.flags
struct ceph_msg *r_request, *r_reply; u32 r_sent; /* >0 if r_request is sending/sent */
/* request osd ops array */ unsigned int r_num_ops;
int r_result;
struct ceph_osd_client *r_osdc; struct kref r_kref; bool r_mempool; bool r_linger; /* don't resend on failure */ struct completion r_completion; /* private to osd_client.c */ ceph_osdc_callback_t r_callback;
struct inode *r_inode; /* for use by callbacks */ struct list_head r_private_item; /* ditto */ void *r_priv; /* ditto */
/* set by submitter */ u64 r_snapid; /* for reads, CEPH_NOSNAP o/w */ struct ceph_snap_context *r_snapc; /* for writes */ struct timespec64 r_mtime; /* ditto */ u64 r_data_offset; /* ditto */
/* internal */ u64 r_version; /* data version sent in reply */ unsigned long r_stamp; /* jiffies, send or check time */ unsigned long r_start_stamp; /* jiffies */ ktime_t r_start_latency; /* ktime_t */ ktime_t r_end_latency; /* ktime_t */ int r_attempts; u32 r_map_dne_bound;
struct ceph_osd_req_op r_ops[] __counted_by(r_num_ops);};
struct ceph_request_redirect { struct ceph_object_locator oloc;};
/* * osd request identifier * * caller name + incarnation# + tid to unique identify this request */struct ceph_osd_reqid { struct ceph_entity_name name; __le64 tid; __le32 inc;} __packed;
struct ceph_blkin_trace_info { __le64 trace_id; __le64 span_id; __le64 parent_span_id;} __packed;
typedef void (*rados_watchcb2_t)(void *arg, u64 notify_id, u64 cookie, u64 notifier_id, void *data, size_t data_len);typedef void (*rados_watcherrcb_t)(void *arg, u64 cookie, int err);
struct ceph_osd_linger_request { struct ceph_osd_client *osdc; u64 linger_id; bool committed; bool is_watch; /* watch or notify */
struct ceph_osd *osd; struct ceph_osd_request *reg_req; struct ceph_osd_request *ping_req; unsigned long ping_sent; unsigned long watch_valid_thru; struct list_head pending_lworks;
struct ceph_osd_request_target t; u32 map_dne_bound;
struct timespec64 mtime;
struct kref kref; struct mutex lock; struct rb_node node; /* osd */ struct rb_node osdc_node; /* osdc */ struct rb_node mc_node; /* map check */ struct list_head scan_item;
struct completion reg_commit_wait; struct completion notify_finish_wait; int reg_commit_error; int notify_finish_error; int last_error;
u32 register_gen; u64 notify_id;
rados_watchcb2_t wcb; rados_watcherrcb_t errcb; void *data;
struct ceph_pagelist *request_pl; struct page **notify_id_pages;
struct page ***preply_pages; size_t *preply_len;};
struct ceph_watch_item { struct ceph_entity_name name; u64 cookie; struct ceph_entity_addr addr;};
struct ceph_spg_mapping { struct rb_node node; struct ceph_spg spgid;
struct rb_root backoffs;};
struct ceph_hobject_id { void *key; size_t key_len; void *oid; size_t oid_len; u64 snapid; u32 hash; u8 is_max; void *nspace; size_t nspace_len; s64 pool;
/* cache */ u32 hash_reverse_bits;};
static inline void ceph_hoid_build_hash_cache(struct ceph_hobject_id *hoid){ hoid->hash_reverse_bits = bitrev32(hoid->hash);}
/* * PG-wide backoff: [begin, end) * per-object backoff: begin == end */struct ceph_osd_backoff { struct rb_node spg_node; struct rb_node id_node;
struct ceph_spg spgid; u64 id; struct ceph_hobject_id *begin; struct ceph_hobject_id *end;};
#define CEPH_LINGER_ID_START 0xffff000000000000ULL
struct ceph_osd_client { struct ceph_client *client;
struct ceph_osdmap *osdmap; /* current map */ struct rw_semaphore lock;
struct rb_root osds; /* osds */ struct list_head osd_lru; /* idle osds */ spinlock_t osd_lru_lock; u32 epoch_barrier; struct ceph_osd homeless_osd; atomic64_t last_tid; /* tid of last request */ u64 last_linger_id; struct rb_root linger_requests; /* lingering requests */ struct rb_root map_checks; struct rb_root linger_map_checks; atomic_t num_requests; atomic_t num_homeless; int abort_err; struct delayed_work timeout_work; struct delayed_work osds_timeout_work;#ifdef CONFIG_DEBUG_FS struct dentry *debugfs_file;#endif
mempool_t *req_mempool;
struct ceph_msgpool msgpool_op; struct ceph_msgpool msgpool_op_reply;
struct workqueue_struct *notify_wq; struct workqueue_struct *completion_wq;};
static inline bool ceph_osdmap_flag(struct ceph_osd_client *osdc, int flag){ return osdc->osdmap->flags & flag;}
extern int ceph_osdc_setup(void);extern void ceph_osdc_cleanup(void);
extern int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client);extern void ceph_osdc_stop(struct ceph_osd_client *osdc);extern void ceph_osdc_reopen_osds(struct ceph_osd_client *osdc);
extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg);void ceph_osdc_update_epoch_barrier(struct ceph_osd_client *osdc, u32 eb);void ceph_osdc_abort_requests(struct ceph_osd_client *osdc, int err);void ceph_osdc_clear_abort_err(struct ceph_osd_client *osdc);
#define osd_req_op_data(oreq, whch, typ, fld) \({ \ struct ceph_osd_request *__oreq = (oreq); \ unsigned int __whch = (whch); \ BUG_ON(__whch >= __oreq->r_num_ops); \ &__oreq->r_ops[__whch].typ.fld; \})
struct ceph_osd_req_op *osd_req_op_init(struct ceph_osd_request *osd_req, unsigned int which, u16 opcode, u32 flags);
extern void osd_req_op_raw_data_in_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages);
extern void osd_req_op_extent_init(struct ceph_osd_request *osd_req, unsigned int which, u16 opcode, u64 offset, u64 length, u64 truncate_size, u32 truncate_seq);extern void osd_req_op_extent_update(struct ceph_osd_request *osd_req, unsigned int which, u64 length);extern void osd_req_op_extent_dup_last(struct ceph_osd_request *osd_req, unsigned int which, u64 offset_inc);
extern struct ceph_osd_data *osd_req_op_extent_osd_data( struct ceph_osd_request *osd_req, unsigned int which);
extern void osd_req_op_extent_osd_data_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages);#ifdef CONFIG_BLOCKvoid osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req, unsigned int which, struct ceph_bio_iter *bio_pos, u32 bio_length);#endif /* CONFIG_BLOCK */void osd_req_op_extent_osd_data_bvecs(struct ceph_osd_request *osd_req, unsigned int which, struct bio_vec *bvecs, u32 num_bvecs, u32 bytes);void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req, unsigned int which, struct ceph_bvec_iter *bvec_pos);void osd_req_op_extent_osd_iter(struct ceph_osd_request *osd_req, unsigned int which, struct iov_iter *iter);
extern void osd_req_op_cls_request_data_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages);void osd_req_op_cls_request_data_bvecs(struct ceph_osd_request *osd_req, unsigned int which, struct bio_vec *bvecs, u32 num_bvecs, u32 bytes);extern void osd_req_op_cls_response_data_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages);int osd_req_op_cls_init(struct ceph_osd_request *osd_req, unsigned int which, const char *class, const char *method);extern int osd_req_op_xattr_init(struct ceph_osd_request *osd_req, unsigned int which, u16 opcode, const char *name, const void *value, size_t size, u8 cmp_op, u8 cmp_mode);extern void osd_req_op_alloc_hint_init(struct ceph_osd_request *osd_req, unsigned int which, u64 expected_object_size, u64 expected_write_size, u32 flags);extern int osd_req_op_copy_from_init(struct ceph_osd_request *req, u64 src_snapid, u64 src_version, struct ceph_object_id *src_oid, struct ceph_object_locator *src_oloc, u32 src_fadvise_flags, u32 dst_fadvise_flags, u32 truncate_seq, u64 truncate_size, u8 copy_from_flags);
extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc, struct ceph_snap_context *snapc, unsigned int num_ops, bool use_mempool, gfp_t gfp_flags);int ceph_osdc_alloc_messages(struct ceph_osd_request *req, gfp_t gfp);
extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *, struct ceph_file_layout *layout, struct ceph_vino vino, u64 offset, u64 *len, unsigned int which, int num_ops, int opcode, int flags, struct ceph_snap_context *snapc, u32 truncate_seq, u64 truncate_size, bool use_mempool);
int __ceph_alloc_sparse_ext_map(struct ceph_osd_req_op *op, int cnt);
/* * How big an extent array should we preallocate for a sparse read? This is * just a starting value. If we get more than this back from the OSD, the * receiver will reallocate. */#define CEPH_SPARSE_EXT_ARRAY_INITIAL 16
static inline int ceph_alloc_sparse_ext_map(struct ceph_osd_req_op *op, int cnt){ if (!cnt) cnt = CEPH_SPARSE_EXT_ARRAY_INITIAL;
return __ceph_alloc_sparse_ext_map(op, cnt);}
extern void ceph_osdc_get_request(struct ceph_osd_request *req);extern void ceph_osdc_put_request(struct ceph_osd_request *req);
void ceph_osdc_start_request(struct ceph_osd_client *osdc, struct ceph_osd_request *req);extern void ceph_osdc_cancel_request(struct ceph_osd_request *req);extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc, struct ceph_osd_request *req);extern void ceph_osdc_sync(struct ceph_osd_client *osdc);
extern void ceph_osdc_flush_notifies(struct ceph_osd_client *osdc);void ceph_osdc_maybe_request_map(struct ceph_osd_client *osdc);
int ceph_osdc_call(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, const char *class, const char *method, unsigned int flags, struct page *req_page, size_t req_len, struct page **resp_pages, size_t *resp_len);
/* watch/notify */struct ceph_osd_linger_request *ceph_osdc_watch(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, rados_watchcb2_t wcb, rados_watcherrcb_t errcb, void *data);int ceph_osdc_unwatch(struct ceph_osd_client *osdc, struct ceph_osd_linger_request *lreq);
int ceph_osdc_notify_ack(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, u64 notify_id, u64 cookie, void *payload, u32 payload_len);int ceph_osdc_notify(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, void *payload, u32 payload_len, u32 timeout, struct page ***preply_pages, size_t *preply_len);int ceph_osdc_list_watchers(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, struct ceph_watch_item **watchers, u32 *num_watchers);
/* Find offset into the buffer of the end of the extent map */static inline u64 ceph_sparse_ext_map_end(struct ceph_osd_req_op *op){ struct ceph_sparse_extent *ext;
/* No extents? No data */ if (op->extent.sparse_ext_cnt == 0) return 0;
ext = &op->extent.sparse_ext[op->extent.sparse_ext_cnt - 1];
return ext->off + ext->len - op->extent.offset;}
#endif