diff options
Diffstat (limited to 'include/linux/ceph')
| -rw-r--r-- | include/linux/ceph/auth.h | 4 | ||||
| -rw-r--r-- | include/linux/ceph/ceph_fs.h | 35 | ||||
| -rw-r--r-- | include/linux/ceph/libceph.h | 10 | ||||
| -rw-r--r-- | include/linux/ceph/messenger.h | 8 | ||||
| -rw-r--r-- | include/linux/ceph/osd_client.h | 64 | ||||
| -rw-r--r-- | include/linux/ceph/rados.h | 39 |
6 files changed, 127 insertions, 33 deletions
diff --git a/include/linux/ceph/auth.h b/include/linux/ceph/auth.h index 7fff521d7eb5..aa13392a7efb 100644 --- a/include/linux/ceph/auth.h +++ b/include/linux/ceph/auth.h @@ -67,12 +67,12 @@ struct ceph_auth_client { bool negotiating; /* true if negotiating protocol */ const char *name; /* entity name */ u64 global_id; /* our unique id in system */ - const char *secret; /* our secret key */ + const struct ceph_crypto_key *key; /* our secret key */ unsigned want_keys; /* which services we want */ }; extern struct ceph_auth_client *ceph_auth_init(const char *name, - const char *secret); + const struct ceph_crypto_key *key); extern void ceph_auth_destroy(struct ceph_auth_client *ac); extern void ceph_auth_reset(struct ceph_auth_client *ac); diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index c3c74aef289d..b8e995fbd867 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -43,6 +43,10 @@ #define CEPH_FEATURE_NOSRCADDR (1<<1) #define CEPH_FEATURE_MONCLOCKCHECK (1<<2) #define CEPH_FEATURE_FLOCK (1<<3) +#define CEPH_FEATURE_SUBSCRIBE2 (1<<4) +#define CEPH_FEATURE_MONNAMES (1<<5) +#define CEPH_FEATURE_RECONNECT_SEQ (1<<6) +#define CEPH_FEATURE_DIRLAYOUTHASH (1<<7) /* @@ -55,10 +59,10 @@ struct ceph_file_layout { __le32 fl_stripe_count; /* over this many objects */ __le32 fl_object_size; /* until objects are this big, then move to new objects */ - __le32 fl_cas_hash; /* 0 = none; 1 = sha256 */ + __le32 fl_cas_hash; /* UNUSED. 0 = none; 1 = sha256 */ /* pg -> disk layout */ - __le32 fl_object_stripe_unit; /* for per-object parity, if any */ + __le32 fl_object_stripe_unit; /* UNUSED. for per-object parity, if any */ /* object -> pg layout */ __le32 fl_pg_preferred; /* preferred primary for pg (-1 for none) */ @@ -69,6 +73,12 @@ struct ceph_file_layout { int ceph_file_layout_is_valid(const struct ceph_file_layout *layout); +struct ceph_dir_layout { + __u8 dl_dir_hash; /* see ceph_hash.h for ids */ + __u8 dl_unused1; + __u16 dl_unused2; + __u32 dl_unused3; +} __attribute__ ((packed)); /* crypto algorithms */ #define CEPH_CRYPTO_NONE 0x0 @@ -126,9 +136,18 @@ int ceph_file_layout_is_valid(const struct ceph_file_layout *layout); /* osd */ -#define CEPH_MSG_OSD_MAP 41 -#define CEPH_MSG_OSD_OP 42 -#define CEPH_MSG_OSD_OPREPLY 43 +#define CEPH_MSG_OSD_MAP 41 +#define CEPH_MSG_OSD_OP 42 +#define CEPH_MSG_OSD_OPREPLY 43 +#define CEPH_MSG_WATCH_NOTIFY 44 + + +/* watch-notify operations */ +enum { + WATCH_NOTIFY = 1, /* notifying watcher */ + WATCH_NOTIFY_COMPLETE = 2, /* notifier notified when done */ +}; + /* pool operations */ enum { @@ -203,8 +222,10 @@ struct ceph_client_mount { struct ceph_mon_request_header monhdr; } __attribute__ ((packed)); +#define CEPH_SUBSCRIBE_ONETIME 1 /* i want only 1 update after have */ + struct ceph_mon_subscribe_item { - __le64 have_version; __le64 have; + __le64 have_version; __le64 have; __u8 onetime; } __attribute__ ((packed)); @@ -457,7 +478,7 @@ struct ceph_mds_reply_inode { struct ceph_timespec rctime; struct ceph_frag_tree_head fragtree; /* (must be at end of struct) */ } __attribute__ ((packed)); -/* followed by frag array, then symlink string, then xattr blob */ +/* followed by frag array, symlink string, dir layout, xattr blob */ /* reply_lease follows dname, and reply_inode */ struct ceph_mds_reply_lease { diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h index f22b2e941686..6365f041745b 100644 --- a/include/linux/ceph/libceph.h +++ b/include/linux/ceph/libceph.h @@ -61,7 +61,7 @@ struct ceph_options { pointer type of args */ int num_mon; char *name; - char *secret; + struct ceph_crypto_key *key; }; /* @@ -71,7 +71,6 @@ struct ceph_options { #define CEPH_OSD_TIMEOUT_DEFAULT 60 /* seconds */ #define CEPH_OSD_KEEPALIVE_DEFAULT 5 #define CEPH_OSD_IDLE_TTL_DEFAULT 60 -#define CEPH_MOUNT_RSIZE_DEFAULT (512*1024) /* readahead */ #define CEPH_MSG_MAX_FRONT_LEN (16*1024*1024) #define CEPH_MSG_MAX_DATA_LEN (16*1024*1024) @@ -227,9 +226,10 @@ extern int ceph_open_session(struct ceph_client *client); extern void ceph_release_page_vector(struct page **pages, int num_pages); extern struct page **ceph_get_direct_page_vector(const char __user *data, - int num_pages, - loff_t off, size_t len); -extern void ceph_put_page_vector(struct page **pages, int num_pages); + int num_pages, + bool write_page); +extern void ceph_put_page_vector(struct page **pages, int num_pages, + bool dirty); extern void ceph_release_page_vector(struct page **pages, int num_pages); extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags); extern int ceph_copy_user_to_page_vector(struct page **pages, diff --git a/include/linux/ceph/messenger.h b/include/linux/ceph/messenger.h index 5956d62c3057..31d91a64838b 100644 --- a/include/linux/ceph/messenger.h +++ b/include/linux/ceph/messenger.h @@ -82,6 +82,7 @@ struct ceph_msg { struct ceph_buffer *middle; struct page **pages; /* data payload. NOT OWNER. */ unsigned nr_pages; /* size of page array */ + unsigned page_alignment; /* io offset in first page */ struct ceph_pagelist *pagelist; /* instead of pages */ struct list_head list_head; struct kref kref; @@ -109,17 +110,12 @@ struct ceph_msg_pos { /* * ceph_connection state bit flags - * - * QUEUED and BUSY are used together to ensure that only a single - * thread is currently opening, reading or writing data to the socket. */ #define LOSSYTX 0 /* we can close channel or drop messages on errors */ #define CONNECTING 1 #define NEGOTIATING 2 #define KEEPALIVE_PENDING 3 #define WRITE_PENDING 4 /* we have data ready to send */ -#define QUEUED 5 /* there is work queued on this connection */ -#define BUSY 6 /* work is being done */ #define STANDBY 8 /* no outgoing messages, socket closed. we keep * the ceph_connection around to maintain shared * state with the peer. */ @@ -127,6 +123,7 @@ struct ceph_msg_pos { #define SOCK_CLOSED 11 /* socket state changed to closed */ #define OPENING 13 /* open connection w/ (possibly new) peer */ #define DEAD 14 /* dead, about to kfree */ +#define BACKOFF 15 /* * A single connection with another host. @@ -164,7 +161,6 @@ struct ceph_connection { struct list_head out_queue; struct list_head out_sent; /* sending or sent but unacked */ u64 out_seq; /* last message queued for send */ - bool out_keepalive_pending; u64 in_seq, in_seq_acked; /* last message received, acked */ diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index 6c91fb032c39..f88eacb111d4 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -32,6 +32,7 @@ struct ceph_osd { struct rb_node o_node; struct ceph_connection o_con; struct list_head o_requests; + struct list_head o_linger_requests; struct list_head o_osd_lru; struct ceph_authorizer *o_authorizer; void *o_authorizer_buf, *o_authorizer_reply_buf; @@ -47,6 +48,8 @@ struct ceph_osd_request { struct rb_node r_node; struct list_head r_req_lru_item; struct list_head r_osd_item; + struct list_head r_linger_item; + struct list_head r_linger_osd; struct ceph_osd *r_osd; struct ceph_pg r_pgid; int r_pg_osds[CEPH_PG_MAX_SIZE]; @@ -59,6 +62,7 @@ struct ceph_osd_request { int r_flags; /* any additional flags for the osd */ u32 r_sent; /* >0 if r_request is sending/sent */ int r_got_reply; + int r_linger; struct ceph_osd_client *r_osdc; struct kref r_kref; @@ -74,11 +78,11 @@ struct ceph_osd_request { char r_oid[40]; /* object name */ int r_oid_len; unsigned long r_stamp; /* send OR check time */ - bool r_resend; /* msg send failed, needs retry */ struct ceph_file_layout r_file_layout; struct ceph_snap_context *r_snapc; /* snap context for writes */ unsigned r_num_pages; /* size of page array (follows) */ + unsigned r_page_alignment; /* io offset in first page */ struct page **r_pages; /* pages for data payload */ int r_pages_from_pool; int r_own_pages; /* if true, i own page list */ @@ -89,6 +93,26 @@ struct ceph_osd_request { struct ceph_pagelist *r_trail; /* trailing part of the data */ }; +struct ceph_osd_event { + u64 cookie; + int one_shot; + struct ceph_osd_client *osdc; + void (*cb)(u64, u64, u8, void *); + void *data; + struct rb_node node; + struct list_head osd_node; + struct kref kref; + struct completion completion; +}; + +struct ceph_osd_event_work { + struct work_struct work; + struct ceph_osd_event *event; + u64 ver; + u64 notify_id; + u8 opcode; +}; + struct ceph_osd_client { struct ceph_client *client; @@ -103,7 +127,10 @@ struct ceph_osd_client { u64 timeout_tid; /* tid of timeout triggering rq */ u64 last_tid; /* tid of last request */ struct rb_root requests; /* pending requests */ - struct list_head req_lru; /* pending requests lru */ + struct list_head req_lru; /* in-flight lru */ + struct list_head req_unsent; /* unsent/need-resend queue */ + struct list_head req_notarget; /* map to no osd */ + struct list_head req_linger; /* lingering requests */ int num_requests; struct delayed_work timeout_work; struct delayed_work osds_timeout_work; @@ -115,6 +142,12 @@ struct ceph_osd_client { struct ceph_msgpool msgpool_op; struct ceph_msgpool msgpool_op_reply; + + spinlock_t event_lock; + struct rb_root event_tree; + u64 event_count; + + struct workqueue_struct *notify_wq; }; struct ceph_osd_req_op { @@ -149,6 +182,13 @@ struct ceph_osd_req_op { struct { u64 snapid; } snap; + struct { + u64 cookie; + u64 ver; + __u8 flag; + u32 prot_ver; + u32 timeout; + } watch; }; u32 payload_len; }; @@ -194,7 +234,13 @@ extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *, int do_sync, u32 truncate_seq, u64 truncate_size, struct timespec *mtime, - bool use_mempool, int num_reply); + bool use_mempool, int num_reply, + int page_align); + +extern void ceph_osdc_set_request_linger(struct ceph_osd_client *osdc, + struct ceph_osd_request *req); +extern void ceph_osdc_unregister_linger_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req); static inline void ceph_osdc_get_request(struct ceph_osd_request *req) { @@ -218,7 +264,8 @@ extern int ceph_osdc_readpages(struct ceph_osd_client *osdc, struct ceph_file_layout *layout, u64 off, u64 *plen, u32 truncate_seq, u64 truncate_size, - struct page **pages, int nr_pages); + struct page **pages, int nr_pages, + int page_align); extern int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino, @@ -230,5 +277,14 @@ extern int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct page **pages, int nr_pages, int flags, int do_sync, bool nofail); +/* watch/notify events */ +extern int ceph_osdc_create_event(struct ceph_osd_client *osdc, + void (*event_cb)(u64, u64, u8, void *), + int one_shot, void *data, + struct ceph_osd_event **pevent); +extern void ceph_osdc_cancel_event(struct ceph_osd_event *event); +extern int ceph_osdc_wait_event(struct ceph_osd_event *event, + unsigned long timeout); +extern void ceph_osdc_put_event(struct ceph_osd_event *event); #endif diff --git a/include/linux/ceph/rados.h b/include/linux/ceph/rados.h index 6d5247f2e81b..0a99099801a4 100644 --- a/include/linux/ceph/rados.h +++ b/include/linux/ceph/rados.h @@ -12,9 +12,9 @@ * osdmap encoding versions */ #define CEPH_OSDMAP_INC_VERSION 5 -#define CEPH_OSDMAP_INC_VERSION_EXT 5 +#define CEPH_OSDMAP_INC_VERSION_EXT 6 #define CEPH_OSDMAP_VERSION 5 -#define CEPH_OSDMAP_VERSION_EXT 5 +#define CEPH_OSDMAP_VERSION_EXT 6 /* * fs id @@ -181,9 +181,17 @@ enum { /* read */ CEPH_OSD_OP_READ = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 1, CEPH_OSD_OP_STAT = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 2, + CEPH_OSD_OP_MAPEXT = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 3, /* fancy read */ - CEPH_OSD_OP_MASKTRUNC = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 4, + CEPH_OSD_OP_MASKTRUNC = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 4, + CEPH_OSD_OP_SPARSE_READ = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 5, + + CEPH_OSD_OP_NOTIFY = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 6, + CEPH_OSD_OP_NOTIFY_ACK = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 7, + + /* versioning */ + CEPH_OSD_OP_ASSERT_VER = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 8, /* write */ CEPH_OSD_OP_WRITE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 1, @@ -205,6 +213,8 @@ enum { CEPH_OSD_OP_CREATE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 13, CEPH_OSD_OP_ROLLBACK= CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 14, + CEPH_OSD_OP_WATCH = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 15, + /** attrs **/ /* read */ CEPH_OSD_OP_GETXATTR = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 1, @@ -218,11 +228,14 @@ enum { CEPH_OSD_OP_RMXATTR = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 4, /** subop **/ - CEPH_OSD_OP_PULL = CEPH_OSD_OP_MODE_SUB | 1, - CEPH_OSD_OP_PUSH = CEPH_OSD_OP_MODE_SUB | 2, - CEPH_OSD_OP_BALANCEREADS = CEPH_OSD_OP_MODE_SUB | 3, - CEPH_OSD_OP_UNBALANCEREADS = CEPH_OSD_OP_MODE_SUB | 4, - CEPH_OSD_OP_SCRUB = CEPH_OSD_OP_MODE_SUB | 5, + CEPH_OSD_OP_PULL = CEPH_OSD_OP_MODE_SUB | 1, + CEPH_OSD_OP_PUSH = CEPH_OSD_OP_MODE_SUB | 2, + CEPH_OSD_OP_BALANCEREADS = CEPH_OSD_OP_MODE_SUB | 3, + CEPH_OSD_OP_UNBALANCEREADS = CEPH_OSD_OP_MODE_SUB | 4, + CEPH_OSD_OP_SCRUB = CEPH_OSD_OP_MODE_SUB | 5, + CEPH_OSD_OP_SCRUB_RESERVE = CEPH_OSD_OP_MODE_SUB | 6, + CEPH_OSD_OP_SCRUB_UNRESERVE = CEPH_OSD_OP_MODE_SUB | 7, + CEPH_OSD_OP_SCRUB_STOP = CEPH_OSD_OP_MODE_SUB | 8, /** lock **/ CEPH_OSD_OP_WRLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 1, @@ -328,6 +341,8 @@ enum { CEPH_OSD_CMPXATTR_MODE_U64 = 2 }; +#define RADOS_NOTIFY_VER 1 + /* * an individual object operation. each may be accompanied by some data * payload @@ -359,7 +374,12 @@ struct ceph_osd_op { struct { __le64 snapid; } __attribute__ ((packed)) snap; - }; + struct { + __le64 cookie; + __le64 ver; + __u8 flag; /* 0 = unwatch, 1 = watch */ + } __attribute__ ((packed)) watch; +}; __le32 payload_len; } __attribute__ ((packed)); @@ -402,4 +422,5 @@ struct ceph_osd_reply_head { } __attribute__ ((packed)); + #endif |