diff options
Diffstat (limited to 'include/linux/ceph')
-rw-r--r-- | include/linux/ceph/auth.h | 190 | ||||
-rw-r--r-- | include/linux/ceph/buffer.h | 39 | ||||
-rw-r--r-- | include/linux/ceph/ceph_debug.h | 39 | ||||
-rw-r--r-- | include/linux/ceph/ceph_features.h | 224 | ||||
-rw-r--r-- | include/linux/ceph/ceph_frag.h | 75 | ||||
-rw-r--r-- | include/linux/ceph/ceph_fs.h | 895 | ||||
-rw-r--r-- | include/linux/ceph/ceph_hash.h | 14 | ||||
-rw-r--r-- | include/linux/ceph/cls_lock_client.h | 58 | ||||
-rw-r--r-- | include/linux/ceph/debugfs.h | 14 | ||||
-rw-r--r-- | include/linux/ceph/decode.h | 398 | ||||
-rw-r--r-- | include/linux/ceph/libceph.h | 332 | ||||
-rw-r--r-- | include/linux/ceph/mdsmap.h | 72 | ||||
-rw-r--r-- | include/linux/ceph/messenger.h | 587 | ||||
-rw-r--r-- | include/linux/ceph/mon_client.h | 152 | ||||
-rw-r--r-- | include/linux/ceph/msgpool.h | 27 | ||||
-rw-r--r-- | include/linux/ceph/msgr.h | 234 | ||||
-rw-r--r-- | include/linux/ceph/osd_client.h | 562 | ||||
-rw-r--r-- | include/linux/ceph/osdmap.h | 339 | ||||
-rw-r--r-- | include/linux/ceph/pagelist.h | 72 | ||||
-rw-r--r-- | include/linux/ceph/rados.h | 551 | ||||
-rw-r--r-- | include/linux/ceph/string_table.h | 63 | ||||
-rw-r--r-- | include/linux/ceph/striper.h | 71 | ||||
-rw-r--r-- | include/linux/ceph/types.h | 31 |
23 files changed, 5039 insertions, 0 deletions
diff --git a/include/linux/ceph/auth.h b/include/linux/ceph/auth.h new file mode 100644 index 000000000..6b138fa97 --- /dev/null +++ b/include/linux/ceph/auth.h @@ -0,0 +1,190 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_AUTH_H +#define _FS_CEPH_AUTH_H + +#include <linux/ceph/types.h> +#include <linux/ceph/buffer.h> + +/* + * Abstract interface for communicating with the authenticate module. + * There is some handshake that takes place between us and the monitor + * to acquire the necessary keys. These are used to generate an + * 'authorizer' that we use when connecting to a service (mds, osd). + */ + +struct ceph_auth_client; +struct ceph_msg; + +struct ceph_authorizer { + void (*destroy)(struct ceph_authorizer *); +}; + +struct ceph_auth_handshake { + struct ceph_authorizer *authorizer; + void *authorizer_buf; + size_t authorizer_buf_len; + void *authorizer_reply_buf; + size_t authorizer_reply_buf_len; + int (*sign_message)(struct ceph_auth_handshake *auth, + struct ceph_msg *msg); + int (*check_message_signature)(struct ceph_auth_handshake *auth, + struct ceph_msg *msg); +}; + +struct ceph_auth_client_ops { + /* + * true if we are authenticated and can connect to + * services. + */ + int (*is_authenticated)(struct ceph_auth_client *ac); + + /* + * true if we should (re)authenticate, e.g., when our tickets + * are getting old and crusty. + */ + int (*should_authenticate)(struct ceph_auth_client *ac); + + /* + * build requests and process replies during monitor + * handshake. if handle_reply returns -EAGAIN, we build + * another request. + */ + int (*build_request)(struct ceph_auth_client *ac, void *buf, void *end); + int (*handle_reply)(struct ceph_auth_client *ac, u64 global_id, + void *buf, void *end, u8 *session_key, + int *session_key_len, u8 *con_secret, + int *con_secret_len); + + /* + * Create authorizer for connecting to a service, and verify + * the response to authenticate the service. + */ + int (*create_authorizer)(struct ceph_auth_client *ac, int peer_type, + struct ceph_auth_handshake *auth); + /* ensure that an existing authorizer is up to date */ + int (*update_authorizer)(struct ceph_auth_client *ac, int peer_type, + struct ceph_auth_handshake *auth); + int (*add_authorizer_challenge)(struct ceph_auth_client *ac, + struct ceph_authorizer *a, + void *challenge_buf, + int challenge_buf_len); + int (*verify_authorizer_reply)(struct ceph_auth_client *ac, + struct ceph_authorizer *a, + void *reply, int reply_len, + u8 *session_key, int *session_key_len, + u8 *con_secret, int *con_secret_len); + void (*invalidate_authorizer)(struct ceph_auth_client *ac, + int peer_type); + + /* reset when we (re)connect to a monitor */ + void (*reset)(struct ceph_auth_client *ac); + + void (*destroy)(struct ceph_auth_client *ac); + + int (*sign_message)(struct ceph_auth_handshake *auth, + struct ceph_msg *msg); + int (*check_message_signature)(struct ceph_auth_handshake *auth, + struct ceph_msg *msg); +}; + +struct ceph_auth_client { + u32 protocol; /* CEPH_AUTH_* */ + void *private; /* for use by protocol implementation */ + const struct ceph_auth_client_ops *ops; /* null iff protocol==0 */ + + bool negotiating; /* true if negotiating protocol */ + const char *name; /* entity name */ + u64 global_id; /* our unique id in system */ + const struct ceph_crypto_key *key; /* our secret key */ + unsigned want_keys; /* which services we want */ + + int preferred_mode; /* CEPH_CON_MODE_* */ + int fallback_mode; /* ditto */ + + struct mutex mutex; +}; + +void ceph_auth_set_global_id(struct ceph_auth_client *ac, u64 global_id); + +struct ceph_auth_client *ceph_auth_init(const char *name, + const struct ceph_crypto_key *key, + const int *con_modes); +extern void ceph_auth_destroy(struct ceph_auth_client *ac); + +extern void ceph_auth_reset(struct ceph_auth_client *ac); + +extern int ceph_auth_build_hello(struct ceph_auth_client *ac, + void *buf, size_t len); +extern int ceph_handle_auth_reply(struct ceph_auth_client *ac, + void *buf, size_t len, + void *reply_buf, size_t reply_len); +int ceph_auth_entity_name_encode(const char *name, void **p, void *end); + +extern int ceph_build_auth(struct ceph_auth_client *ac, + void *msg_buf, size_t msg_len); +extern int ceph_auth_is_authenticated(struct ceph_auth_client *ac); + +int __ceph_auth_get_authorizer(struct ceph_auth_client *ac, + struct ceph_auth_handshake *auth, + int peer_type, bool force_new, + int *proto, int *pref_mode, int *fallb_mode); +void ceph_auth_destroy_authorizer(struct ceph_authorizer *a); +int ceph_auth_add_authorizer_challenge(struct ceph_auth_client *ac, + struct ceph_authorizer *a, + void *challenge_buf, + int challenge_buf_len); +int ceph_auth_verify_authorizer_reply(struct ceph_auth_client *ac, + struct ceph_authorizer *a, + void *reply, int reply_len, + u8 *session_key, int *session_key_len, + u8 *con_secret, int *con_secret_len); +extern void ceph_auth_invalidate_authorizer(struct ceph_auth_client *ac, + int peer_type); + +static inline int ceph_auth_sign_message(struct ceph_auth_handshake *auth, + struct ceph_msg *msg) +{ + if (auth->sign_message) + return auth->sign_message(auth, msg); + return 0; +} + +static inline +int ceph_auth_check_message_signature(struct ceph_auth_handshake *auth, + struct ceph_msg *msg) +{ + if (auth->check_message_signature) + return auth->check_message_signature(auth, msg); + return 0; +} + +int ceph_auth_get_request(struct ceph_auth_client *ac, void *buf, int buf_len); +int ceph_auth_handle_reply_more(struct ceph_auth_client *ac, void *reply, + int reply_len, void *buf, int buf_len); +int ceph_auth_handle_reply_done(struct ceph_auth_client *ac, + u64 global_id, void *reply, int reply_len, + u8 *session_key, int *session_key_len, + u8 *con_secret, int *con_secret_len); +bool ceph_auth_handle_bad_method(struct ceph_auth_client *ac, + int used_proto, int result, + const int *allowed_protos, int proto_cnt, + const int *allowed_modes, int mode_cnt); + +int ceph_auth_get_authorizer(struct ceph_auth_client *ac, + struct ceph_auth_handshake *auth, + int peer_type, void *buf, int *buf_len); +int ceph_auth_handle_svc_reply_more(struct ceph_auth_client *ac, + struct ceph_auth_handshake *auth, + void *reply, int reply_len, + void *buf, int *buf_len); +int ceph_auth_handle_svc_reply_done(struct ceph_auth_client *ac, + struct ceph_auth_handshake *auth, + void *reply, int reply_len, + u8 *session_key, int *session_key_len, + u8 *con_secret, int *con_secret_len); +bool ceph_auth_handle_bad_authorizer(struct ceph_auth_client *ac, + int peer_type, int used_proto, int result, + const int *allowed_protos, int proto_cnt, + const int *allowed_modes, int mode_cnt); + +#endif diff --git a/include/linux/ceph/buffer.h b/include/linux/ceph/buffer.h new file mode 100644 index 000000000..11cdc7c60 --- /dev/null +++ b/include/linux/ceph/buffer.h @@ -0,0 +1,39 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __FS_CEPH_BUFFER_H +#define __FS_CEPH_BUFFER_H + +#include <linux/kref.h> +#include <linux/mm.h> +#include <linux/vmalloc.h> +#include <linux/types.h> +#include <linux/uio.h> + +/* + * a simple reference counted buffer. + * + * use kmalloc for smaller sizes, vmalloc for larger sizes. + */ +struct ceph_buffer { + struct kref kref; + struct kvec vec; + size_t alloc_len; +}; + +extern struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp); +extern void ceph_buffer_release(struct kref *kref); + +static inline struct ceph_buffer *ceph_buffer_get(struct ceph_buffer *b) +{ + kref_get(&b->kref); + return b; +} + +static inline void ceph_buffer_put(struct ceph_buffer *b) +{ + if (b) + kref_put(&b->kref, ceph_buffer_release); +} + +extern int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end); + +#endif diff --git a/include/linux/ceph/ceph_debug.h b/include/linux/ceph/ceph_debug.h new file mode 100644 index 000000000..d5a5da838 --- /dev/null +++ b/include/linux/ceph/ceph_debug.h @@ -0,0 +1,39 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_DEBUG_H +#define _FS_CEPH_DEBUG_H + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/string.h> + +#ifdef CONFIG_CEPH_LIB_PRETTYDEBUG + +/* + * wrap pr_debug to include a filename:lineno prefix on each line. + * this incurs some overhead (kernel size and execution time) due to + * the extra function call at each call site. + */ + +# if defined(DEBUG) || defined(CONFIG_DYNAMIC_DEBUG) +# define dout(fmt, ...) \ + pr_debug("%.*s %12.12s:%-4d : " fmt, \ + 8 - (int)sizeof(KBUILD_MODNAME), " ", \ + kbasename(__FILE__), __LINE__, ##__VA_ARGS__) +# else +/* faux printk call just to see any compiler warnings. */ +# define dout(fmt, ...) do { \ + if (0) \ + printk(KERN_DEBUG fmt, ##__VA_ARGS__); \ + } while (0) +# endif + +#else + +/* + * or, just wrap pr_debug + */ +# define dout(fmt, ...) pr_debug(" " fmt, ##__VA_ARGS__) + +#endif + +#endif diff --git a/include/linux/ceph/ceph_features.h b/include/linux/ceph/ceph_features.h new file mode 100644 index 000000000..3a47acd9c --- /dev/null +++ b/include/linux/ceph/ceph_features.h @@ -0,0 +1,224 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __CEPH_FEATURES +#define __CEPH_FEATURES + +/* + * Each time we reclaim bits for reuse we need to specify another bit + * that, if present, indicates we have the new incarnation of that + * feature. Base case is 1 (first use). + */ +#define CEPH_FEATURE_INCARNATION_1 (0ull) +#define CEPH_FEATURE_INCARNATION_2 (1ull<<57) // SERVER_JEWEL +#define CEPH_FEATURE_INCARNATION_3 ((1ull<<57)|(1ull<<28)) // SERVER_MIMIC + +#define DEFINE_CEPH_FEATURE(bit, incarnation, name) \ + static const uint64_t __maybe_unused CEPH_FEATURE_##name = (1ULL<<bit); \ + static const uint64_t __maybe_unused CEPH_FEATUREMASK_##name = \ + (1ULL<<bit | CEPH_FEATURE_INCARNATION_##incarnation); + +/* this bit is ignored but still advertised by release *when* */ +#define DEFINE_CEPH_FEATURE_DEPRECATED(bit, incarnation, name, when) \ + static const uint64_t __maybe_unused DEPRECATED_CEPH_FEATURE_##name = (1ULL<<bit); \ + static const uint64_t __maybe_unused DEPRECATED_CEPH_FEATUREMASK_##name = \ + (1ULL<<bit | CEPH_FEATURE_INCARNATION_##incarnation); + +/* + * this bit is ignored by release *unused* and not advertised by + * release *unadvertised* + */ +#define DEFINE_CEPH_FEATURE_RETIRED(bit, inc, name, unused, unadvertised) + + +/* + * test for a feature. this test is safer than a typical mask against + * the bit because it ensures that we have the bit AND the marker for the + * bit's incarnation. this must be used in any case where the features + * bits may include an old meaning of the bit. + */ +#define CEPH_HAVE_FEATURE(x, name) \ + (((x) & (CEPH_FEATUREMASK_##name)) == (CEPH_FEATUREMASK_##name)) + + +/* + * Notes on deprecation: + * + * A *major* release is a release through which all upgrades must pass + * (e.g., jewel). For example, no pre-jewel server will ever talk to + * a post-jewel server (mon, osd, etc). + * + * For feature bits used *only* on the server-side: + * + * - In the first phase we indicate that a feature is DEPRECATED as of + * a particular release. This is the first major release X (say, + * jewel) that does not depend on its peers advertising the feature. + * That is, it safely assumes its peers all have the feature. We + * indicate this with the DEPRECATED macro. For example, + * + * DEFINE_CEPH_FEATURE_DEPRECATED( 2, 1, MONCLOCKCHECK, JEWEL) + * + * because 10.2.z (jewel) did not care if its peers advertised this + * feature bit. + * + * - In the second phase we stop advertising the bit and call it + * RETIRED. This can normally be done in the *next* major release + * following the one in which we marked the feature DEPRECATED. In + * the above example, for 12.0.z (luminous) we can say: + * + * DEFINE_CEPH_FEATURE_RETIRED( 2, 1, MONCLOCKCHECK, JEWEL, LUMINOUS) + * + * - The bit can be reused in the first post-luminous release, 13.0.z + * (m). + * + * This ensures that no two versions who have different meanings for + * the bit ever speak to each other. + */ + +DEFINE_CEPH_FEATURE( 0, 1, UID) +DEFINE_CEPH_FEATURE( 1, 1, NOSRCADDR) +DEFINE_CEPH_FEATURE_RETIRED( 2, 1, MONCLOCKCHECK, JEWEL, LUMINOUS) +DEFINE_CEPH_FEATURE( 2, 3, SERVER_NAUTILUS) +DEFINE_CEPH_FEATURE( 3, 1, FLOCK) +DEFINE_CEPH_FEATURE( 4, 1, SUBSCRIBE2) +DEFINE_CEPH_FEATURE( 5, 1, MONNAMES) +DEFINE_CEPH_FEATURE( 6, 1, RECONNECT_SEQ) +DEFINE_CEPH_FEATURE( 7, 1, DIRLAYOUTHASH) +DEFINE_CEPH_FEATURE( 8, 1, OBJECTLOCATOR) +DEFINE_CEPH_FEATURE( 9, 1, PGID64) +DEFINE_CEPH_FEATURE(10, 1, INCSUBOSDMAP) +DEFINE_CEPH_FEATURE(11, 1, PGPOOL3) +DEFINE_CEPH_FEATURE(12, 1, OSDREPLYMUX) +DEFINE_CEPH_FEATURE(13, 1, OSDENC) +DEFINE_CEPH_FEATURE_RETIRED(14, 1, OMAP, HAMMER, JEWEL) +DEFINE_CEPH_FEATURE(14, 2, SERVER_KRAKEN) +DEFINE_CEPH_FEATURE(15, 1, MONENC) +DEFINE_CEPH_FEATURE_RETIRED(16, 1, QUERY_T, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE_RETIRED(17, 1, INDEP_PG_MAP, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE(18, 1, CRUSH_TUNABLES) +DEFINE_CEPH_FEATURE_RETIRED(19, 1, CHUNKY_SCRUB, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE_RETIRED(20, 1, MON_NULLROUTE, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE_RETIRED(21, 1, MON_GV, HAMMER, JEWEL) +DEFINE_CEPH_FEATURE(21, 2, SERVER_LUMINOUS) +DEFINE_CEPH_FEATURE(21, 2, RESEND_ON_SPLIT) // overlap +DEFINE_CEPH_FEATURE(21, 2, RADOS_BACKOFF) // overlap +DEFINE_CEPH_FEATURE(21, 2, OSDMAP_PG_UPMAP) // overlap +DEFINE_CEPH_FEATURE(21, 2, CRUSH_CHOOSE_ARGS) // overlap +DEFINE_CEPH_FEATURE_RETIRED(22, 1, BACKFILL_RESERVATION, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE(23, 1, MSG_AUTH) +DEFINE_CEPH_FEATURE_RETIRED(24, 1, RECOVERY_RESERVATION, JEWEL, LUNINOUS) + +DEFINE_CEPH_FEATURE(25, 1, CRUSH_TUNABLES2) +DEFINE_CEPH_FEATURE(26, 1, CREATEPOOLID) +DEFINE_CEPH_FEATURE(27, 1, REPLY_CREATE_INODE) +DEFINE_CEPH_FEATURE_RETIRED(28, 1, OSD_HBMSGS, HAMMER, JEWEL) +DEFINE_CEPH_FEATURE(28, 2, SERVER_MIMIC) +DEFINE_CEPH_FEATURE(29, 1, MDSENC) +DEFINE_CEPH_FEATURE(30, 1, OSDHASHPSPOOL) +DEFINE_CEPH_FEATURE(31, 1, MON_SINGLE_PAXOS) // deprecate me +DEFINE_CEPH_FEATURE_RETIRED(32, 1, OSD_SNAPMAPPER, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE_RETIRED(33, 1, MON_SCRUB, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE_RETIRED(34, 1, OSD_PACKED_RECOVERY, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE(35, 1, OSD_CACHEPOOL) +DEFINE_CEPH_FEATURE(36, 1, CRUSH_V2) +DEFINE_CEPH_FEATURE(37, 1, EXPORT_PEER) +DEFINE_CEPH_FEATURE(38, 1, OSD_ERASURE_CODES) +DEFINE_CEPH_FEATURE(38, 1, OSD_OSD_TMAP2OMAP) // overlap +DEFINE_CEPH_FEATURE(39, 1, OSDMAP_ENC) +DEFINE_CEPH_FEATURE(40, 1, MDS_INLINE_DATA) +DEFINE_CEPH_FEATURE(41, 1, CRUSH_TUNABLES3) +DEFINE_CEPH_FEATURE(41, 1, OSD_PRIMARY_AFFINITY) // overlap +DEFINE_CEPH_FEATURE(42, 1, MSGR_KEEPALIVE2) +DEFINE_CEPH_FEATURE(43, 1, OSD_POOLRESEND) +DEFINE_CEPH_FEATURE(44, 1, ERASURE_CODE_PLUGINS_V2) +DEFINE_CEPH_FEATURE_RETIRED(45, 1, OSD_SET_ALLOC_HINT, JEWEL, LUMINOUS) + +DEFINE_CEPH_FEATURE(46, 1, OSD_FADVISE_FLAGS) +DEFINE_CEPH_FEATURE_RETIRED(46, 1, OSD_REPOP, JEWEL, LUMINOUS) // overlap +DEFINE_CEPH_FEATURE_RETIRED(46, 1, OSD_OBJECT_DIGEST, JEWEL, LUMINOUS) // overlap +DEFINE_CEPH_FEATURE_RETIRED(46, 1, OSD_TRANSACTION_MAY_LAYOUT, JEWEL, LUMINOUS) // overlap + +DEFINE_CEPH_FEATURE(47, 1, MDS_QUOTA) +DEFINE_CEPH_FEATURE(48, 1, CRUSH_V4) +DEFINE_CEPH_FEATURE_RETIRED(49, 1, OSD_MIN_SIZE_RECOVERY, JEWEL, LUMINOUS) +DEFINE_CEPH_FEATURE_RETIRED(49, 1, OSD_PROXY_FEATURES, JEWEL, LUMINOUS) // overlap + +DEFINE_CEPH_FEATURE(50, 1, MON_METADATA) +DEFINE_CEPH_FEATURE(51, 1, OSD_BITWISE_HOBJ_SORT) +DEFINE_CEPH_FEATURE(52, 1, OSD_PROXY_WRITE_FEATURES) +DEFINE_CEPH_FEATURE(53, 1, ERASURE_CODE_PLUGINS_V3) +DEFINE_CEPH_FEATURE(54, 1, OSD_HITSET_GMT) +DEFINE_CEPH_FEATURE(55, 1, HAMMER_0_94_4) +DEFINE_CEPH_FEATURE(56, 1, NEW_OSDOP_ENCODING) +DEFINE_CEPH_FEATURE(57, 1, MON_STATEFUL_SUB) +DEFINE_CEPH_FEATURE(57, 1, MON_ROUTE_OSDMAP) // overlap +DEFINE_CEPH_FEATURE(57, 1, OSDSUBOP_NO_SNAPCONTEXT) // overlap +DEFINE_CEPH_FEATURE(57, 1, SERVER_JEWEL) // overlap +DEFINE_CEPH_FEATURE(58, 1, CRUSH_TUNABLES5) +DEFINE_CEPH_FEATURE(58, 1, NEW_OSDOPREPLY_ENCODING) // overlap +DEFINE_CEPH_FEATURE(58, 1, FS_FILE_LAYOUT_V2) // overlap +DEFINE_CEPH_FEATURE(59, 1, FS_BTIME) +DEFINE_CEPH_FEATURE(59, 1, FS_CHANGE_ATTR) // overlap +DEFINE_CEPH_FEATURE(59, 1, MSG_ADDR2) // overlap +DEFINE_CEPH_FEATURE(60, 1, OSD_RECOVERY_DELETES) // *do not share this bit* +DEFINE_CEPH_FEATURE(61, 1, CEPHX_V2) // *do not share this bit* + +DEFINE_CEPH_FEATURE(62, 1, RESERVED) // do not use; used as a sentinal +DEFINE_CEPH_FEATURE_DEPRECATED(63, 1, RESERVED_BROKEN, LUMINOUS) // client-facing + + +/* + * Features supported. + */ +#define CEPH_FEATURES_SUPPORTED_DEFAULT \ + (CEPH_FEATURE_NOSRCADDR | \ + CEPH_FEATURE_SERVER_NAUTILUS | \ + CEPH_FEATURE_FLOCK | \ + CEPH_FEATURE_SUBSCRIBE2 | \ + CEPH_FEATURE_MONNAMES | \ + CEPH_FEATURE_RECONNECT_SEQ | \ + CEPH_FEATURE_DIRLAYOUTHASH | \ + CEPH_FEATURE_PGID64 | \ + CEPH_FEATURE_PGPOOL3 | \ + CEPH_FEATURE_OSDENC | \ + CEPH_FEATURE_MONENC | \ + CEPH_FEATURE_CRUSH_TUNABLES | \ + CEPH_FEATURE_SERVER_LUMINOUS | \ + CEPH_FEATURE_RESEND_ON_SPLIT | \ + CEPH_FEATURE_RADOS_BACKOFF | \ + CEPH_FEATURE_OSDMAP_PG_UPMAP | \ + CEPH_FEATURE_CRUSH_CHOOSE_ARGS | \ + CEPH_FEATURE_MSG_AUTH | \ + CEPH_FEATURE_CRUSH_TUNABLES2 | \ + CEPH_FEATURE_REPLY_CREATE_INODE | \ + CEPH_FEATURE_SERVER_MIMIC | \ + CEPH_FEATURE_MDSENC | \ + CEPH_FEATURE_OSDHASHPSPOOL | \ + CEPH_FEATURE_OSD_CACHEPOOL | \ + CEPH_FEATURE_CRUSH_V2 | \ + CEPH_FEATURE_EXPORT_PEER | \ + CEPH_FEATURE_OSDMAP_ENC | \ + CEPH_FEATURE_MDS_INLINE_DATA | \ + CEPH_FEATURE_CRUSH_TUNABLES3 | \ + CEPH_FEATURE_OSD_PRIMARY_AFFINITY | \ + CEPH_FEATURE_MSGR_KEEPALIVE2 | \ + CEPH_FEATURE_OSD_POOLRESEND | \ + CEPH_FEATURE_MDS_QUOTA | \ + CEPH_FEATURE_CRUSH_V4 | \ + CEPH_FEATURE_NEW_OSDOP_ENCODING | \ + CEPH_FEATURE_SERVER_JEWEL | \ + CEPH_FEATURE_MON_STATEFUL_SUB | \ + CEPH_FEATURE_CRUSH_TUNABLES5 | \ + CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING | \ + CEPH_FEATURE_MSG_ADDR2 | \ + CEPH_FEATURE_CEPHX_V2) + +#define CEPH_FEATURES_REQUIRED_DEFAULT 0 + +#endif diff --git a/include/linux/ceph/ceph_frag.h b/include/linux/ceph/ceph_frag.h new file mode 100644 index 000000000..97bab0adc --- /dev/null +++ b/include/linux/ceph/ceph_frag.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef FS_CEPH_FRAG_H +#define FS_CEPH_FRAG_H + +/* + * "Frags" are a way to describe a subset of a 32-bit number space, + * using a mask and a value to match against that mask. Any given frag + * (subset of the number space) can be partitioned into 2^n sub-frags. + * + * Frags are encoded into a 32-bit word: + * 8 upper bits = "bits" + * 24 lower bits = "value" + * (We could go to 5+27 bits, but who cares.) + * + * We use the _most_ significant bits of the 24 bit value. This makes + * values logically sort. + * + * Unfortunately, because the "bits" field is still in the high bits, we + * can't sort encoded frags numerically. However, it does allow you + * to feed encoded frags as values into frag_contains_value. + */ +static inline __u32 ceph_frag_make(__u32 b, __u32 v) +{ + return (b << 24) | + (v & (0xffffffu << (24-b)) & 0xffffffu); +} +static inline __u32 ceph_frag_bits(__u32 f) +{ + return f >> 24; +} +static inline __u32 ceph_frag_value(__u32 f) +{ + return f & 0xffffffu; +} +static inline __u32 ceph_frag_mask(__u32 f) +{ + return (0xffffffu << (24-ceph_frag_bits(f))) & 0xffffffu; +} +static inline __u32 ceph_frag_mask_shift(__u32 f) +{ + return 24 - ceph_frag_bits(f); +} + +static inline bool ceph_frag_contains_value(__u32 f, __u32 v) +{ + return (v & ceph_frag_mask(f)) == ceph_frag_value(f); +} + +static inline __u32 ceph_frag_make_child(__u32 f, int by, int i) +{ + int newbits = ceph_frag_bits(f) + by; + return ceph_frag_make(newbits, + ceph_frag_value(f) | (i << (24 - newbits))); +} +static inline bool ceph_frag_is_leftmost(__u32 f) +{ + return ceph_frag_value(f) == 0; +} +static inline bool ceph_frag_is_rightmost(__u32 f) +{ + return ceph_frag_value(f) == ceph_frag_mask(f); +} +static inline __u32 ceph_frag_next(__u32 f) +{ + return ceph_frag_make(ceph_frag_bits(f), + ceph_frag_value(f) + (0x1000000 >> ceph_frag_bits(f))); +} + +/* + * comparator to sort frags logically, as when traversing the + * number space in ascending order... + */ +int ceph_frag_compare(__u32 a, __u32 b); + +#endif diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h new file mode 100644 index 000000000..49586ff26 --- /dev/null +++ b/include/linux/ceph/ceph_fs.h @@ -0,0 +1,895 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * ceph_fs.h - Ceph constants and data types to share between kernel and + * user space. + * + * Most types in this file are defined as little-endian, and are + * primarily intended to describe data structures that pass over the + * wire or that are stored on disk. + * + * LGPL2 + */ + +#ifndef CEPH_FS_H +#define CEPH_FS_H + +#include <linux/ceph/msgr.h> +#include <linux/ceph/rados.h> + +/* + * subprotocol versions. when specific messages types or high-level + * protocols change, bump the affected components. we keep rev + * internal cluster protocols separately from the public, + * client-facing protocol. + */ +#define CEPH_OSDC_PROTOCOL 24 /* server/client */ +#define CEPH_MDSC_PROTOCOL 32 /* server/client */ +#define CEPH_MONC_PROTOCOL 15 /* server/client */ + + +#define CEPH_INO_ROOT 1 +#define CEPH_INO_CEPH 2 /* hidden .ceph dir */ +#define CEPH_INO_GLOBAL_SNAPREALM 3 /* global dummy snaprealm */ + +/* arbitrary limit on max # of monitors (cluster of 3 is typical) */ +#define CEPH_MAX_MON 31 + +/* + * legacy ceph_file_layoute + */ +struct ceph_file_layout_legacy { + /* file -> object mapping */ + __le32 fl_stripe_unit; /* stripe unit, in bytes. must be multiple + of page size. */ + __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; /* UNUSED. 0 = none; 1 = sha256 */ + + /* pg -> disk layout */ + __le32 fl_object_stripe_unit; /* UNUSED. for per-object parity, if any */ + + /* object -> pg layout */ + __le32 fl_unused; /* unused; used to be preferred primary for pg (-1 for none) */ + __le32 fl_pg_pool; /* namespace, crush ruleset, rep level */ +} __attribute__ ((packed)); + +struct ceph_string; +/* + * ceph_file_layout - describe data layout for a file/inode + */ +struct ceph_file_layout { + /* file -> object mapping */ + u32 stripe_unit; /* stripe unit, in bytes */ + u32 stripe_count; /* over this many objects */ + u32 object_size; /* until objects are this big */ + s64 pool_id; /* rados pool id */ + struct ceph_string __rcu *pool_ns; /* rados pool namespace */ +}; + +extern int ceph_file_layout_is_valid(const struct ceph_file_layout *layout); +extern void ceph_file_layout_from_legacy(struct ceph_file_layout *fl, + struct ceph_file_layout_legacy *legacy); +extern void ceph_file_layout_to_legacy(struct ceph_file_layout *fl, + struct ceph_file_layout_legacy *legacy); + +#define CEPH_MIN_STRIPE_UNIT 65536 + +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 +#define CEPH_CRYPTO_AES 0x1 + +#define CEPH_AES_IV "cephsageyudagreg" + +/* security/authentication protocols */ +#define CEPH_AUTH_UNKNOWN 0x0 +#define CEPH_AUTH_NONE 0x1 +#define CEPH_AUTH_CEPHX 0x2 + +#define CEPH_AUTH_MODE_NONE 0 +#define CEPH_AUTH_MODE_AUTHORIZER 1 +#define CEPH_AUTH_MODE_MON 10 + +/* msgr2 protocol modes */ +#define CEPH_CON_MODE_UNKNOWN 0x0 +#define CEPH_CON_MODE_CRC 0x1 +#define CEPH_CON_MODE_SECURE 0x2 + +#define CEPH_AUTH_UID_DEFAULT ((__u64) -1) + +const char *ceph_auth_proto_name(int proto); +const char *ceph_con_mode_name(int mode); + +/********************************************* + * message layer + */ + +/* + * message types + */ + +/* misc */ +#define CEPH_MSG_SHUTDOWN 1 +#define CEPH_MSG_PING 2 + +/* client <-> monitor */ +#define CEPH_MSG_MON_MAP 4 +#define CEPH_MSG_MON_GET_MAP 5 +#define CEPH_MSG_STATFS 13 +#define CEPH_MSG_STATFS_REPLY 14 +#define CEPH_MSG_MON_SUBSCRIBE 15 +#define CEPH_MSG_MON_SUBSCRIBE_ACK 16 +#define CEPH_MSG_AUTH 17 +#define CEPH_MSG_AUTH_REPLY 18 +#define CEPH_MSG_MON_GET_VERSION 19 +#define CEPH_MSG_MON_GET_VERSION_REPLY 20 + +/* client <-> mds */ +#define CEPH_MSG_MDS_MAP 21 +#define CEPH_MSG_FS_MAP_USER 103 + +#define CEPH_MSG_CLIENT_SESSION 22 +#define CEPH_MSG_CLIENT_RECONNECT 23 + +#define CEPH_MSG_CLIENT_REQUEST 24 +#define CEPH_MSG_CLIENT_REQUEST_FORWARD 25 +#define CEPH_MSG_CLIENT_REPLY 26 +#define CEPH_MSG_CLIENT_METRICS 29 +#define CEPH_MSG_CLIENT_CAPS 0x310 +#define CEPH_MSG_CLIENT_LEASE 0x311 +#define CEPH_MSG_CLIENT_SNAP 0x312 +#define CEPH_MSG_CLIENT_CAPRELEASE 0x313 +#define CEPH_MSG_CLIENT_QUOTA 0x314 + +/* pool ops */ +#define CEPH_MSG_POOLOP_REPLY 48 +#define CEPH_MSG_POOLOP 49 + +/* mon commands */ +#define CEPH_MSG_MON_COMMAND 50 +#define CEPH_MSG_MON_COMMAND_ACK 51 + +/* osd */ +#define CEPH_MSG_OSD_MAP 41 +#define CEPH_MSG_OSD_OP 42 +#define CEPH_MSG_OSD_OPREPLY 43 +#define CEPH_MSG_WATCH_NOTIFY 44 +#define CEPH_MSG_OSD_BACKOFF 61 + + +/* watch-notify operations */ +enum { + CEPH_WATCH_EVENT_NOTIFY = 1, /* notifying watcher */ + CEPH_WATCH_EVENT_NOTIFY_COMPLETE = 2, /* notifier notified when done */ + CEPH_WATCH_EVENT_DISCONNECT = 3, /* we were disconnected */ +}; + + +struct ceph_mon_request_header { + __le64 have_version; + __le16 session_mon; + __le64 session_mon_tid; +} __attribute__ ((packed)); + +struct ceph_mon_statfs { + struct ceph_mon_request_header monhdr; + struct ceph_fsid fsid; + __u8 contains_data_pool; + __le64 data_pool; +} __attribute__ ((packed)); + +struct ceph_statfs { + __le64 kb, kb_used, kb_avail; + __le64 num_objects; +} __attribute__ ((packed)); + +struct ceph_mon_statfs_reply { + struct ceph_fsid fsid; + __le64 version; + struct ceph_statfs st; +} __attribute__ ((packed)); + +struct ceph_mon_command { + struct ceph_mon_request_header monhdr; + struct ceph_fsid fsid; + __le32 num_strs; /* always 1 */ + __le32 str_len; + char str[]; +} __attribute__ ((packed)); + +struct ceph_osd_getmap { + struct ceph_mon_request_header monhdr; + struct ceph_fsid fsid; + __le32 start; +} __attribute__ ((packed)); + +struct ceph_mds_getmap { + struct ceph_mon_request_header monhdr; + struct ceph_fsid fsid; +} __attribute__ ((packed)); + +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 start; + __u8 flags; +} __attribute__ ((packed)); + +struct ceph_mon_subscribe_ack { + __le32 duration; /* seconds */ + struct ceph_fsid fsid; +} __attribute__ ((packed)); + +#define CEPH_FS_CLUSTER_ID_NONE -1 + +/* + * mdsmap flags + */ +#define CEPH_MDSMAP_DOWN (1<<0) /* cluster deliberately down */ + +/* + * mds states + * > 0 -> in + * <= 0 -> out + */ +#define CEPH_MDS_STATE_DNE 0 /* down, does not exist. */ +#define CEPH_MDS_STATE_STOPPED -1 /* down, once existed, but no subtrees. + empty log. */ +#define CEPH_MDS_STATE_BOOT -4 /* up, boot announcement. */ +#define CEPH_MDS_STATE_STANDBY -5 /* up, idle. waiting for assignment. */ +#define CEPH_MDS_STATE_CREATING -6 /* up, creating MDS instance. */ +#define CEPH_MDS_STATE_STARTING -7 /* up, starting previously stopped mds */ +#define CEPH_MDS_STATE_STANDBY_REPLAY -8 /* up, tailing active node's journal */ +#define CEPH_MDS_STATE_REPLAYONCE -9 /* up, replaying an active node's journal */ + +#define CEPH_MDS_STATE_REPLAY 8 /* up, replaying journal. */ +#define CEPH_MDS_STATE_RESOLVE 9 /* up, disambiguating distributed + operations (import, rename, etc.) */ +#define CEPH_MDS_STATE_RECONNECT 10 /* up, reconnect to clients */ +#define CEPH_MDS_STATE_REJOIN 11 /* up, rejoining distributed cache */ +#define CEPH_MDS_STATE_CLIENTREPLAY 12 /* up, replaying client operations */ +#define CEPH_MDS_STATE_ACTIVE 13 /* up, active */ +#define CEPH_MDS_STATE_STOPPING 14 /* up, but exporting metadata */ + +extern const char *ceph_mds_state_name(int s); + + +/* + * metadata lock types. + * - these are bitmasks.. we can compose them + * - they also define the lock ordering by the MDS + * - a few of these are internal to the mds + */ +#define CEPH_LOCK_DVERSION 1 +#define CEPH_LOCK_DN 2 +#define CEPH_LOCK_ISNAP 16 +#define CEPH_LOCK_IVERSION 32 /* mds internal */ +#define CEPH_LOCK_IFILE 64 +#define CEPH_LOCK_IAUTH 128 +#define CEPH_LOCK_ILINK 256 +#define CEPH_LOCK_IDFT 512 /* dir frag tree */ +#define CEPH_LOCK_INEST 1024 /* mds internal */ +#define CEPH_LOCK_IXATTR 2048 +#define CEPH_LOCK_IFLOCK 4096 /* advisory file locks */ +#define CEPH_LOCK_INO 8192 /* immutable inode bits; not a lock */ +#define CEPH_LOCK_IPOLICY 16384 /* policy lock on dirs. MDS internal */ + +/* client_session ops */ +enum { + CEPH_SESSION_REQUEST_OPEN, + CEPH_SESSION_OPEN, + CEPH_SESSION_REQUEST_CLOSE, + CEPH_SESSION_CLOSE, + CEPH_SESSION_REQUEST_RENEWCAPS, + CEPH_SESSION_RENEWCAPS, + CEPH_SESSION_STALE, + CEPH_SESSION_RECALL_STATE, + CEPH_SESSION_FLUSHMSG, + CEPH_SESSION_FLUSHMSG_ACK, + CEPH_SESSION_FORCE_RO, + CEPH_SESSION_REJECT, + CEPH_SESSION_REQUEST_FLUSH_MDLOG, +}; + +#define CEPH_SESSION_BLOCKLISTED (1 << 0) /* session blocklisted */ + +extern const char *ceph_session_op_name(int op); + +struct ceph_mds_session_head { + __le32 op; + __le64 seq; + struct ceph_timespec stamp; + __le32 max_caps, max_leases; +} __attribute__ ((packed)); + +/* client_request */ +/* + * metadata ops. + * & 0x001000 -> write op + * & 0x010000 -> follow symlink (e.g. stat(), not lstat()). + & & 0x100000 -> use weird ino/path trace + */ +#define CEPH_MDS_OP_WRITE 0x001000 +enum { + CEPH_MDS_OP_LOOKUP = 0x00100, + CEPH_MDS_OP_GETATTR = 0x00101, + CEPH_MDS_OP_LOOKUPHASH = 0x00102, + CEPH_MDS_OP_LOOKUPPARENT = 0x00103, + CEPH_MDS_OP_LOOKUPINO = 0x00104, + CEPH_MDS_OP_LOOKUPNAME = 0x00105, + CEPH_MDS_OP_GETVXATTR = 0x00106, + + CEPH_MDS_OP_SETXATTR = 0x01105, + CEPH_MDS_OP_RMXATTR = 0x01106, + CEPH_MDS_OP_SETLAYOUT = 0x01107, + CEPH_MDS_OP_SETATTR = 0x01108, + CEPH_MDS_OP_SETFILELOCK= 0x01109, + CEPH_MDS_OP_GETFILELOCK= 0x00110, + CEPH_MDS_OP_SETDIRLAYOUT=0x0110a, + + CEPH_MDS_OP_MKNOD = 0x01201, + CEPH_MDS_OP_LINK = 0x01202, + CEPH_MDS_OP_UNLINK = 0x01203, + CEPH_MDS_OP_RENAME = 0x01204, + CEPH_MDS_OP_MKDIR = 0x01220, + CEPH_MDS_OP_RMDIR = 0x01221, + CEPH_MDS_OP_SYMLINK = 0x01222, + + CEPH_MDS_OP_CREATE = 0x01301, + CEPH_MDS_OP_OPEN = 0x00302, + CEPH_MDS_OP_READDIR = 0x00305, + + CEPH_MDS_OP_LOOKUPSNAP = 0x00400, + CEPH_MDS_OP_MKSNAP = 0x01400, + CEPH_MDS_OP_RMSNAP = 0x01401, + CEPH_MDS_OP_LSSNAP = 0x00402, + CEPH_MDS_OP_RENAMESNAP = 0x01403, +}; + +extern const char *ceph_mds_op_name(int op); + + +#define CEPH_SETATTR_MODE 1 +#define CEPH_SETATTR_UID 2 +#define CEPH_SETATTR_GID 4 +#define CEPH_SETATTR_MTIME 8 +#define CEPH_SETATTR_ATIME 16 +#define CEPH_SETATTR_SIZE 32 +#define CEPH_SETATTR_CTIME 64 + +/* + * Ceph setxattr request flags. + */ +#define CEPH_XATTR_CREATE (1 << 0) +#define CEPH_XATTR_REPLACE (1 << 1) +#define CEPH_XATTR_REMOVE (1 << 31) + +/* + * readdir request flags; + */ +#define CEPH_READDIR_REPLY_BITFLAGS (1<<0) + +/* + * readdir reply flags. + */ +#define CEPH_READDIR_FRAG_END (1<<0) +#define CEPH_READDIR_FRAG_COMPLETE (1<<8) +#define CEPH_READDIR_HASH_ORDER (1<<9) +#define CEPH_READDIR_OFFSET_HASH (1<<10) + +/* + * open request flags + */ +#define CEPH_O_RDONLY 00000000 +#define CEPH_O_WRONLY 00000001 +#define CEPH_O_RDWR 00000002 +#define CEPH_O_CREAT 00000100 +#define CEPH_O_EXCL 00000200 +#define CEPH_O_TRUNC 00001000 +#define CEPH_O_DIRECTORY 00200000 +#define CEPH_O_NOFOLLOW 00400000 + +union ceph_mds_request_args { + struct { + __le32 mask; /* CEPH_CAP_* */ + } __attribute__ ((packed)) getattr; + struct { + __le32 mode; + __le32 uid; + __le32 gid; + struct ceph_timespec mtime; + struct ceph_timespec atime; + __le64 size, old_size; /* old_size needed by truncate */ + __le32 mask; /* CEPH_SETATTR_* */ + } __attribute__ ((packed)) setattr; + struct { + __le32 frag; /* which dir fragment */ + __le32 max_entries; /* how many dentries to grab */ + __le32 max_bytes; + __le16 flags; + __le32 offset_hash; + } __attribute__ ((packed)) readdir; + struct { + __le32 mode; + __le32 rdev; + } __attribute__ ((packed)) mknod; + struct { + __le32 mode; + } __attribute__ ((packed)) mkdir; + struct { + __le32 flags; + __le32 mode; + __le32 stripe_unit; /* layout for newly created file */ + __le32 stripe_count; /* ... */ + __le32 object_size; + __le32 pool; + __le32 mask; /* CEPH_CAP_* */ + __le64 old_size; + } __attribute__ ((packed)) open; + struct { + __le32 flags; + __le32 osdmap_epoch; /* used for setting file/dir layouts */ + } __attribute__ ((packed)) setxattr; + struct { + struct ceph_file_layout_legacy layout; + } __attribute__ ((packed)) setlayout; + struct { + __u8 rule; /* currently fcntl or flock */ + __u8 type; /* shared, exclusive, remove*/ + __le64 owner; /* owner of the lock */ + __le64 pid; /* process id requesting the lock */ + __le64 start; /* initial location to lock */ + __le64 length; /* num bytes to lock from start */ + __u8 wait; /* will caller wait for lock to become available? */ + } __attribute__ ((packed)) filelock_change; + struct { + __le32 mask; /* CEPH_CAP_* */ + __le64 snapid; + __le64 parent; + __le32 hash; + } __attribute__ ((packed)) lookupino; +} __attribute__ ((packed)); + +union ceph_mds_request_args_ext { + union ceph_mds_request_args old; + struct { + __le32 mode; + __le32 uid; + __le32 gid; + struct ceph_timespec mtime; + struct ceph_timespec atime; + __le64 size, old_size; /* old_size needed by truncate */ + __le32 mask; /* CEPH_SETATTR_* */ + struct ceph_timespec btime; + } __attribute__ ((packed)) setattr_ext; +}; + +#define CEPH_MDS_FLAG_REPLAY 1 /* this is a replayed op */ +#define CEPH_MDS_FLAG_WANT_DENTRY 2 /* want dentry in reply */ +#define CEPH_MDS_FLAG_ASYNC 4 /* request is asynchronous */ + +struct ceph_mds_request_head_old { + __le64 oldest_client_tid; + __le32 mdsmap_epoch; /* on client */ + __le32 flags; /* CEPH_MDS_FLAG_* */ + __u8 num_retry, num_fwd; /* count retry, fwd attempts */ + __le16 num_releases; /* # include cap/lease release records */ + __le32 op; /* mds op code */ + __le32 caller_uid, caller_gid; + __le64 ino; /* use this ino for openc, mkdir, mknod, + etc. (if replaying) */ + union ceph_mds_request_args args; +} __attribute__ ((packed)); + +#define CEPH_MDS_REQUEST_HEAD_VERSION 1 + +struct ceph_mds_request_head { + __le16 version; /* struct version */ + __le64 oldest_client_tid; + __le32 mdsmap_epoch; /* on client */ + __le32 flags; /* CEPH_MDS_FLAG_* */ + __u8 num_retry, num_fwd; /* count retry, fwd attempts */ + __le16 num_releases; /* # include cap/lease release records */ + __le32 op; /* mds op code */ + __le32 caller_uid, caller_gid; + __le64 ino; /* use this ino for openc, mkdir, mknod, + etc. (if replaying) */ + union ceph_mds_request_args_ext args; +} __attribute__ ((packed)); + +/* cap/lease release record */ +struct ceph_mds_request_release { + __le64 ino, cap_id; /* ino and unique cap id */ + __le32 caps, wanted; /* new issued, wanted */ + __le32 seq, issue_seq, mseq; + __le32 dname_seq; /* if releasing a dentry lease, a */ + __le32 dname_len; /* string follows. */ +} __attribute__ ((packed)); + +/* client reply */ +struct ceph_mds_reply_head { + __le32 op; + __le32 result; + __le32 mdsmap_epoch; + __u8 safe; /* true if committed to disk */ + __u8 is_dentry, is_target; /* true if dentry, target inode records + are included with reply */ +} __attribute__ ((packed)); + +/* one for each node split */ +struct ceph_frag_tree_split { + __le32 frag; /* this frag splits... */ + __le32 by; /* ...by this many bits */ +} __attribute__ ((packed)); + +struct ceph_frag_tree_head { + __le32 nsplits; /* num ceph_frag_tree_split records */ + struct ceph_frag_tree_split splits[]; +} __attribute__ ((packed)); + +/* capability issue, for bundling with mds reply */ +struct ceph_mds_reply_cap { + __le32 caps, wanted; /* caps issued, wanted */ + __le64 cap_id; + __le32 seq, mseq; + __le64 realm; /* snap realm */ + __u8 flags; /* CEPH_CAP_FLAG_* */ +} __attribute__ ((packed)); + +#define CEPH_CAP_FLAG_AUTH (1 << 0) /* cap is issued by auth mds */ +#define CEPH_CAP_FLAG_RELEASE (1 << 1) /* release the cap */ + +/* inode record, for bundling with mds reply */ +struct ceph_mds_reply_inode { + __le64 ino; + __le64 snapid; + __le32 rdev; + __le64 version; /* inode version */ + __le64 xattr_version; /* version for xattr blob */ + struct ceph_mds_reply_cap cap; /* caps issued for this inode */ + struct ceph_file_layout_legacy layout; + struct ceph_timespec ctime, mtime, atime; + __le32 time_warp_seq; + __le64 size, max_size, truncate_size; + __le32 truncate_seq; + __le32 mode, uid, gid; + __le32 nlink; + __le64 files, subdirs, rbytes, rfiles, rsubdirs; /* dir stats */ + struct ceph_timespec rctime; + struct ceph_frag_tree_head fragtree; /* (must be at end of struct) */ +} __attribute__ ((packed)); +/* followed by frag array, symlink string, dir layout, xattr blob */ + +/* reply_lease follows dname, and reply_inode */ +struct ceph_mds_reply_lease { + __le16 mask; /* lease type(s) */ + __le32 duration_ms; /* lease duration */ + __le32 seq; +} __attribute__ ((packed)); + +#define CEPH_LEASE_VALID (1 | 2) /* old and new bit values */ +#define CEPH_LEASE_PRIMARY_LINK 4 /* primary linkage */ + +struct ceph_mds_reply_dirfrag { + __le32 frag; /* fragment */ + __le32 auth; /* auth mds, if this is a delegation point */ + __le32 ndist; /* number of mds' this is replicated on */ + __le32 dist[]; +} __attribute__ ((packed)); + +#define CEPH_LOCK_FCNTL 1 +#define CEPH_LOCK_FLOCK 2 +#define CEPH_LOCK_FCNTL_INTR 3 +#define CEPH_LOCK_FLOCK_INTR 4 + + +#define CEPH_LOCK_SHARED 1 +#define CEPH_LOCK_EXCL 2 +#define CEPH_LOCK_UNLOCK 4 + +struct ceph_filelock { + __le64 start;/* file offset to start lock at */ + __le64 length; /* num bytes to lock; 0 for all following start */ + __le64 client; /* which client holds the lock */ + __le64 owner; /* owner the lock */ + __le64 pid; /* process id holding the lock on the client */ + __u8 type; /* shared lock, exclusive lock, or unlock */ +} __attribute__ ((packed)); + + +/* file access modes */ +#define CEPH_FILE_MODE_PIN 0 +#define CEPH_FILE_MODE_RD 1 +#define CEPH_FILE_MODE_WR 2 +#define CEPH_FILE_MODE_RDWR 3 /* RD | WR */ +#define CEPH_FILE_MODE_LAZY 4 /* lazy io */ +#define CEPH_FILE_MODE_BITS 4 +#define CEPH_FILE_MODE_MASK ((1 << CEPH_FILE_MODE_BITS) - 1) + +int ceph_flags_to_mode(int flags); + +#define CEPH_INLINE_NONE ((__u64)-1) + +/* capability bits */ +#define CEPH_CAP_PIN 1 /* no specific capabilities beyond the pin */ + +/* generic cap bits */ +#define CEPH_CAP_GSHARED 1 /* client can reads */ +#define CEPH_CAP_GEXCL 2 /* client can read and update */ +#define CEPH_CAP_GCACHE 4 /* (file) client can cache reads */ +#define CEPH_CAP_GRD 8 /* (file) client can read */ +#define CEPH_CAP_GWR 16 /* (file) client can write */ +#define CEPH_CAP_GBUFFER 32 /* (file) client can buffer writes */ +#define CEPH_CAP_GWREXTEND 64 /* (file) client can extend EOF */ +#define CEPH_CAP_GLAZYIO 128 /* (file) client can perform lazy io */ + +#define CEPH_CAP_SIMPLE_BITS 2 +#define CEPH_CAP_FILE_BITS 8 + +/* per-lock shift */ +#define CEPH_CAP_SAUTH 2 +#define CEPH_CAP_SLINK 4 +#define CEPH_CAP_SXATTR 6 +#define CEPH_CAP_SFILE 8 +#define CEPH_CAP_SFLOCK 20 + +#define CEPH_CAP_BITS 22 + +/* composed values */ +#define CEPH_CAP_AUTH_SHARED (CEPH_CAP_GSHARED << CEPH_CAP_SAUTH) +#define CEPH_CAP_AUTH_EXCL (CEPH_CAP_GEXCL << CEPH_CAP_SAUTH) +#define CEPH_CAP_LINK_SHARED (CEPH_CAP_GSHARED << CEPH_CAP_SLINK) +#define CEPH_CAP_LINK_EXCL (CEPH_CAP_GEXCL << CEPH_CAP_SLINK) +#define CEPH_CAP_XATTR_SHARED (CEPH_CAP_GSHARED << CEPH_CAP_SXATTR) +#define CEPH_CAP_XATTR_EXCL (CEPH_CAP_GEXCL << CEPH_CAP_SXATTR) +#define CEPH_CAP_FILE(x) (x << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_SHARED (CEPH_CAP_GSHARED << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_EXCL (CEPH_CAP_GEXCL << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_CACHE (CEPH_CAP_GCACHE << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_RD (CEPH_CAP_GRD << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_WR (CEPH_CAP_GWR << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_BUFFER (CEPH_CAP_GBUFFER << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_WREXTEND (CEPH_CAP_GWREXTEND << CEPH_CAP_SFILE) +#define CEPH_CAP_FILE_LAZYIO (CEPH_CAP_GLAZYIO << CEPH_CAP_SFILE) +#define CEPH_CAP_FLOCK_SHARED (CEPH_CAP_GSHARED << CEPH_CAP_SFLOCK) +#define CEPH_CAP_FLOCK_EXCL (CEPH_CAP_GEXCL << CEPH_CAP_SFLOCK) + + +/* cap masks (for getattr) */ +#define CEPH_STAT_CAP_INODE CEPH_CAP_PIN +#define CEPH_STAT_CAP_TYPE CEPH_CAP_PIN /* mode >> 12 */ +#define CEPH_STAT_CAP_SYMLINK CEPH_CAP_PIN +#define CEPH_STAT_CAP_UID CEPH_CAP_AUTH_SHARED +#define CEPH_STAT_CAP_GID CEPH_CAP_AUTH_SHARED +#define CEPH_STAT_CAP_MODE CEPH_CAP_AUTH_SHARED +#define CEPH_STAT_CAP_NLINK CEPH_CAP_LINK_SHARED +#define CEPH_STAT_CAP_LAYOUT CEPH_CAP_FILE_SHARED +#define CEPH_STAT_CAP_MTIME CEPH_CAP_FILE_SHARED +#define CEPH_STAT_CAP_SIZE CEPH_CAP_FILE_SHARED +#define CEPH_STAT_CAP_ATIME CEPH_CAP_FILE_SHARED /* fixme */ +#define CEPH_STAT_CAP_XATTR CEPH_CAP_XATTR_SHARED +#define CEPH_STAT_CAP_INODE_ALL (CEPH_CAP_PIN | \ + CEPH_CAP_AUTH_SHARED | \ + CEPH_CAP_LINK_SHARED | \ + CEPH_CAP_FILE_SHARED | \ + CEPH_CAP_XATTR_SHARED) +#define CEPH_STAT_CAP_INLINE_DATA (CEPH_CAP_FILE_SHARED | \ + CEPH_CAP_FILE_RD) +#define CEPH_STAT_RSTAT CEPH_CAP_FILE_WREXTEND + +#define CEPH_CAP_ANY_SHARED (CEPH_CAP_AUTH_SHARED | \ + CEPH_CAP_LINK_SHARED | \ + CEPH_CAP_XATTR_SHARED | \ + CEPH_CAP_FILE_SHARED) +#define CEPH_CAP_ANY_RD (CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_RD | \ + CEPH_CAP_FILE_CACHE) + +#define CEPH_CAP_ANY_EXCL (CEPH_CAP_AUTH_EXCL | \ + CEPH_CAP_LINK_EXCL | \ + CEPH_CAP_XATTR_EXCL | \ + CEPH_CAP_FILE_EXCL) +#define CEPH_CAP_ANY_FILE_RD (CEPH_CAP_FILE_RD | CEPH_CAP_FILE_CACHE | \ + CEPH_CAP_FILE_SHARED) +#define CEPH_CAP_ANY_FILE_WR (CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER | \ + CEPH_CAP_FILE_EXCL) +#define CEPH_CAP_ANY_WR (CEPH_CAP_ANY_EXCL | CEPH_CAP_ANY_FILE_WR) +#define CEPH_CAP_ANY (CEPH_CAP_ANY_RD | CEPH_CAP_ANY_EXCL | \ + CEPH_CAP_ANY_FILE_WR | CEPH_CAP_FILE_LAZYIO | \ + CEPH_CAP_PIN) +#define CEPH_CAP_ALL_FILE (CEPH_CAP_PIN | CEPH_CAP_ANY_SHARED | \ + CEPH_CAP_AUTH_EXCL | CEPH_CAP_XATTR_EXCL | \ + CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR) + +#define CEPH_CAP_LOCKS (CEPH_LOCK_IFILE | CEPH_LOCK_IAUTH | CEPH_LOCK_ILINK | \ + CEPH_LOCK_IXATTR) + +/* cap masks async dir operations */ +#define CEPH_CAP_DIR_CREATE CEPH_CAP_FILE_CACHE +#define CEPH_CAP_DIR_UNLINK CEPH_CAP_FILE_RD +#define CEPH_CAP_ANY_DIR_OPS (CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_RD | \ + CEPH_CAP_FILE_WREXTEND | CEPH_CAP_FILE_LAZYIO) + +int ceph_caps_for_mode(int mode); + +enum { + CEPH_CAP_OP_GRANT, /* mds->client grant */ + CEPH_CAP_OP_REVOKE, /* mds->client revoke */ + CEPH_CAP_OP_TRUNC, /* mds->client trunc notify */ + CEPH_CAP_OP_EXPORT, /* mds has exported the cap */ + CEPH_CAP_OP_IMPORT, /* mds has imported the cap */ + CEPH_CAP_OP_UPDATE, /* client->mds update */ + CEPH_CAP_OP_DROP, /* client->mds drop cap bits */ + CEPH_CAP_OP_FLUSH, /* client->mds cap writeback */ + CEPH_CAP_OP_FLUSH_ACK, /* mds->client flushed */ + CEPH_CAP_OP_FLUSHSNAP, /* client->mds flush snapped metadata */ + CEPH_CAP_OP_FLUSHSNAP_ACK, /* mds->client flushed snapped metadata */ + CEPH_CAP_OP_RELEASE, /* client->mds release (clean) cap */ + CEPH_CAP_OP_RENEW, /* client->mds renewal request */ +}; + +extern const char *ceph_cap_op_name(int op); + +/* flags field in client cap messages (version >= 10) */ +#define CEPH_CLIENT_CAPS_SYNC (1<<0) +#define CEPH_CLIENT_CAPS_NO_CAPSNAP (1<<1) +#define CEPH_CLIENT_CAPS_PENDING_CAPSNAP (1<<2) + +/* + * caps message, used for capability callbacks, acks, requests, etc. + */ +struct ceph_mds_caps { + __le32 op; /* CEPH_CAP_OP_* */ + __le64 ino, realm; + __le64 cap_id; + __le32 seq, issue_seq; + __le32 caps, wanted, dirty; /* latest issued/wanted/dirty */ + __le32 migrate_seq; + __le64 snap_follows; + __le32 snap_trace_len; + + /* authlock */ + __le32 uid, gid, mode; + + /* linklock */ + __le32 nlink; + + /* xattrlock */ + __le32 xattr_len; + __le64 xattr_version; + + /* a union of non-export and export bodies. */ + __le64 size, max_size, truncate_size; + __le32 truncate_seq; + struct ceph_timespec mtime, atime, ctime; + struct ceph_file_layout_legacy layout; + __le32 time_warp_seq; +} __attribute__ ((packed)); + +struct ceph_mds_cap_peer { + __le64 cap_id; + __le32 seq; + __le32 mseq; + __le32 mds; + __u8 flags; +} __attribute__ ((packed)); + +/* cap release msg head */ +struct ceph_mds_cap_release { + __le32 num; /* number of cap_items that follow */ +} __attribute__ ((packed)); + +struct ceph_mds_cap_item { + __le64 ino; + __le64 cap_id; + __le32 migrate_seq, seq; +} __attribute__ ((packed)); + +#define CEPH_MDS_LEASE_REVOKE 1 /* mds -> client */ +#define CEPH_MDS_LEASE_RELEASE 2 /* client -> mds */ +#define CEPH_MDS_LEASE_RENEW 3 /* client <-> mds */ +#define CEPH_MDS_LEASE_REVOKE_ACK 4 /* client -> mds */ + +extern const char *ceph_lease_op_name(int o); + +/* lease msg header */ +struct ceph_mds_lease { + __u8 action; /* CEPH_MDS_LEASE_* */ + __le16 mask; /* which lease */ + __le64 ino; + __le64 first, last; /* snap range */ + __le32 seq; + __le32 duration_ms; /* duration of renewal */ +} __attribute__ ((packed)); +/* followed by a __le32+string for dname */ + +/* client reconnect */ +struct ceph_mds_cap_reconnect { + __le64 cap_id; + __le32 wanted; + __le32 issued; + __le64 snaprealm; + __le64 pathbase; /* base ino for our path to this ino */ + __le32 flock_len; /* size of flock state blob, if any */ +} __attribute__ ((packed)); +/* followed by flock blob */ + +struct ceph_mds_cap_reconnect_v1 { + __le64 cap_id; + __le32 wanted; + __le32 issued; + __le64 size; + struct ceph_timespec mtime, atime; + __le64 snaprealm; + __le64 pathbase; /* base ino for our path to this ino */ +} __attribute__ ((packed)); + +struct ceph_mds_snaprealm_reconnect { + __le64 ino; /* snap realm base */ + __le64 seq; /* snap seq for this snap realm */ + __le64 parent; /* parent realm */ +} __attribute__ ((packed)); + +/* + * snaps + */ +enum { + CEPH_SNAP_OP_UPDATE, /* CREATE or DESTROY */ + CEPH_SNAP_OP_CREATE, + CEPH_SNAP_OP_DESTROY, + CEPH_SNAP_OP_SPLIT, +}; + +extern const char *ceph_snap_op_name(int o); + +/* snap msg header */ +struct ceph_mds_snap_head { + __le32 op; /* CEPH_SNAP_OP_* */ + __le64 split; /* ino to split off, if any */ + __le32 num_split_inos; /* # inos belonging to new child realm */ + __le32 num_split_realms; /* # child realms udner new child realm */ + __le32 trace_len; /* size of snap trace blob */ +} __attribute__ ((packed)); +/* followed by split ino list, then split realms, then the trace blob */ + +/* + * encode info about a snaprealm, as viewed by a client + */ +struct ceph_mds_snap_realm { + __le64 ino; /* ino */ + __le64 created; /* snap: when created */ + __le64 parent; /* ino: parent realm */ + __le64 parent_since; /* snap: same parent since */ + __le64 seq; /* snap: version */ + __le32 num_snaps; + __le32 num_prior_parent_snaps; +} __attribute__ ((packed)); +/* followed by my snap list, then prior parent snap list */ + +/* + * quotas + */ +struct ceph_mds_quota { + __le64 ino; /* ino */ + struct ceph_timespec rctime; + __le64 rbytes; /* dir stats */ + __le64 rfiles; + __le64 rsubdirs; + __u8 struct_v; /* compat */ + __u8 struct_compat; + __le32 struct_len; + __le64 max_bytes; /* quota max. bytes */ + __le64 max_files; /* quota max. files */ +} __attribute__ ((packed)); + +#endif diff --git a/include/linux/ceph/ceph_hash.h b/include/linux/ceph/ceph_hash.h new file mode 100644 index 000000000..fda474c7a --- /dev/null +++ b/include/linux/ceph/ceph_hash.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef FS_CEPH_HASH_H +#define FS_CEPH_HASH_H + +#define CEPH_STR_HASH_LINUX 0x1 /* linux dcache hash */ +#define CEPH_STR_HASH_RJENKINS 0x2 /* robert jenkins' */ + +extern unsigned ceph_str_hash_linux(const char *s, unsigned len); +extern unsigned ceph_str_hash_rjenkins(const char *s, unsigned len); + +extern unsigned ceph_str_hash(int type, const char *s, unsigned len); +extern const char *ceph_str_hash_name(int type); + +#endif diff --git a/include/linux/ceph/cls_lock_client.h b/include/linux/ceph/cls_lock_client.h new file mode 100644 index 000000000..17bc7584d --- /dev/null +++ b/include/linux/ceph/cls_lock_client.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_CEPH_CLS_LOCK_CLIENT_H +#define _LINUX_CEPH_CLS_LOCK_CLIENT_H + +#include <linux/ceph/osd_client.h> + +enum ceph_cls_lock_type { + CEPH_CLS_LOCK_NONE = 0, + CEPH_CLS_LOCK_EXCLUSIVE = 1, + CEPH_CLS_LOCK_SHARED = 2, +}; + +struct ceph_locker_id { + struct ceph_entity_name name; /* locker's client name */ + char *cookie; /* locker's cookie */ +}; + +struct ceph_locker_info { + struct ceph_entity_addr addr; /* locker's address */ +}; + +struct ceph_locker { + struct ceph_locker_id id; + struct ceph_locker_info info; +}; + +int ceph_cls_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 type, char *cookie, + char *tag, char *desc, u8 flags); +int ceph_cls_unlock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie); +int ceph_cls_break_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie, + struct ceph_entity_name *locker); +int ceph_cls_set_cookie(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 type, char *old_cookie, + char *tag, char *new_cookie); + +void ceph_free_lockers(struct ceph_locker *lockers, u32 num_lockers); + +int ceph_cls_lock_info(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 *type, char **tag, + struct ceph_locker **lockers, u32 *num_lockers); + +int ceph_cls_assert_locked(struct ceph_osd_request *req, int which, + char *lock_name, u8 type, char *cookie, char *tag); + +#endif diff --git a/include/linux/ceph/debugfs.h b/include/linux/ceph/debugfs.h new file mode 100644 index 000000000..8b3a1a7a9 --- /dev/null +++ b/include/linux/ceph/debugfs.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_DEBUGFS_H +#define _FS_CEPH_DEBUGFS_H + +#include <linux/ceph/types.h> + +/* debugfs.c */ +extern void ceph_debugfs_init(void); +extern void ceph_debugfs_cleanup(void); +extern void ceph_debugfs_client_init(struct ceph_client *client); +extern void ceph_debugfs_client_cleanup(struct ceph_client *client); + +#endif + diff --git a/include/linux/ceph/decode.h b/include/linux/ceph/decode.h new file mode 100644 index 000000000..04f3ace57 --- /dev/null +++ b/include/linux/ceph/decode.h @@ -0,0 +1,398 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __CEPH_DECODE_H +#define __CEPH_DECODE_H + +#include <linux/err.h> +#include <linux/bug.h> +#include <linux/slab.h> +#include <linux/time.h> +#include <asm/unaligned.h> + +#include <linux/ceph/types.h> + +/* + * in all cases, + * void **p pointer to position pointer + * void *end pointer to end of buffer (last byte + 1) + */ + +static inline u64 ceph_decode_64(void **p) +{ + u64 v = get_unaligned_le64(*p); + *p += sizeof(u64); + return v; +} +static inline u32 ceph_decode_32(void **p) +{ + u32 v = get_unaligned_le32(*p); + *p += sizeof(u32); + return v; +} +static inline u16 ceph_decode_16(void **p) +{ + u16 v = get_unaligned_le16(*p); + *p += sizeof(u16); + return v; +} +static inline u8 ceph_decode_8(void **p) +{ + u8 v = *(u8 *)*p; + (*p)++; + return v; +} +static inline void ceph_decode_copy(void **p, void *pv, size_t n) +{ + memcpy(pv, *p, n); + *p += n; +} + +/* + * bounds check input. + */ +static inline bool ceph_has_room(void **p, void *end, size_t n) +{ + return end >= *p && n <= end - *p; +} + +#define ceph_decode_need(p, end, n, bad) \ + do { \ + if (!likely(ceph_has_room(p, end, n))) \ + goto bad; \ + } while (0) + +#define ceph_decode_64_safe(p, end, v, bad) \ + do { \ + ceph_decode_need(p, end, sizeof(u64), bad); \ + v = ceph_decode_64(p); \ + } while (0) +#define ceph_decode_32_safe(p, end, v, bad) \ + do { \ + ceph_decode_need(p, end, sizeof(u32), bad); \ + v = ceph_decode_32(p); \ + } while (0) +#define ceph_decode_16_safe(p, end, v, bad) \ + do { \ + ceph_decode_need(p, end, sizeof(u16), bad); \ + v = ceph_decode_16(p); \ + } while (0) +#define ceph_decode_8_safe(p, end, v, bad) \ + do { \ + ceph_decode_need(p, end, sizeof(u8), bad); \ + v = ceph_decode_8(p); \ + } while (0) + +#define ceph_decode_copy_safe(p, end, pv, n, bad) \ + do { \ + ceph_decode_need(p, end, n, bad); \ + ceph_decode_copy(p, pv, n); \ + } while (0) + +/* + * Allocate a buffer big enough to hold the wire-encoded string, and + * decode the string into it. The resulting string will always be + * terminated with '\0'. If successful, *p will be advanced + * past the decoded data. Also, if lenp is not a null pointer, the + * length (not including the terminating '\0') will be recorded in + * *lenp. Note that a zero-length string is a valid return value. + * + * Returns a pointer to the newly-allocated string buffer, or a + * pointer-coded errno if an error occurs. Neither *p nor *lenp + * will have been updated if an error is returned. + * + * There are two possible failures: + * - converting the string would require accessing memory at or + * beyond the "end" pointer provided (-ERANGE) + * - memory could not be allocated for the result (-ENOMEM) + */ +static inline char *ceph_extract_encoded_string(void **p, void *end, + size_t *lenp, gfp_t gfp) +{ + u32 len; + void *sp = *p; + char *buf; + + ceph_decode_32_safe(&sp, end, len, bad); + if (!ceph_has_room(&sp, end, len)) + goto bad; + + buf = kmalloc(len + 1, gfp); + if (!buf) + return ERR_PTR(-ENOMEM); + + if (len) + memcpy(buf, sp, len); + buf[len] = '\0'; + + *p = (char *) *p + sizeof (u32) + len; + if (lenp) + *lenp = (size_t) len; + + return buf; + +bad: + return ERR_PTR(-ERANGE); +} + +/* + * skip helpers + */ +#define ceph_decode_skip_n(p, end, n, bad) \ + do { \ + ceph_decode_need(p, end, n, bad); \ + *p += n; \ + } while (0) + +#define ceph_decode_skip_64(p, end, bad) \ +ceph_decode_skip_n(p, end, sizeof(u64), bad) + +#define ceph_decode_skip_32(p, end, bad) \ +ceph_decode_skip_n(p, end, sizeof(u32), bad) + +#define ceph_decode_skip_16(p, end, bad) \ +ceph_decode_skip_n(p, end, sizeof(u16), bad) + +#define ceph_decode_skip_8(p, end, bad) \ +ceph_decode_skip_n(p, end, sizeof(u8), bad) + +#define ceph_decode_skip_string(p, end, bad) \ + do { \ + u32 len; \ + \ + ceph_decode_32_safe(p, end, len, bad); \ + ceph_decode_skip_n(p, end, len, bad); \ + } while (0) + +#define ceph_decode_skip_set(p, end, type, bad) \ + do { \ + u32 len; \ + \ + ceph_decode_32_safe(p, end, len, bad); \ + while (len--) \ + ceph_decode_skip_##type(p, end, bad); \ + } while (0) + +#define ceph_decode_skip_map(p, end, ktype, vtype, bad) \ + do { \ + u32 len; \ + \ + ceph_decode_32_safe(p, end, len, bad); \ + while (len--) { \ + ceph_decode_skip_##ktype(p, end, bad); \ + ceph_decode_skip_##vtype(p, end, bad); \ + } \ + } while (0) + +#define ceph_decode_skip_map_of_map(p, end, ktype1, ktype2, vtype2, bad) \ + do { \ + u32 len; \ + \ + ceph_decode_32_safe(p, end, len, bad); \ + while (len--) { \ + ceph_decode_skip_##ktype1(p, end, bad); \ + ceph_decode_skip_map(p, end, ktype2, vtype2, bad); \ + } \ + } while (0) + +/* + * struct ceph_timespec <-> struct timespec64 + */ +static inline void ceph_decode_timespec64(struct timespec64 *ts, + const struct ceph_timespec *tv) +{ + /* + * This will still overflow in year 2106. We could extend + * the protocol to steal two more bits from tv_nsec to + * add three more 136 year epochs after that the way ext4 + * does if necessary. + */ + ts->tv_sec = (time64_t)le32_to_cpu(tv->tv_sec); + ts->tv_nsec = (long)le32_to_cpu(tv->tv_nsec); +} +static inline void ceph_encode_timespec64(struct ceph_timespec *tv, + const struct timespec64 *ts) +{ + tv->tv_sec = cpu_to_le32((u32)ts->tv_sec); + tv->tv_nsec = cpu_to_le32((u32)ts->tv_nsec); +} + +/* + * sockaddr_storage <-> ceph_sockaddr + */ +#define CEPH_ENTITY_ADDR_TYPE_NONE 0 +#define CEPH_ENTITY_ADDR_TYPE_LEGACY __cpu_to_le32(1) +#define CEPH_ENTITY_ADDR_TYPE_MSGR2 __cpu_to_le32(2) +#define CEPH_ENTITY_ADDR_TYPE_ANY __cpu_to_le32(3) + +static inline void ceph_encode_banner_addr(struct ceph_entity_addr *a) +{ + __be16 ss_family = htons(a->in_addr.ss_family); + a->in_addr.ss_family = *(__u16 *)&ss_family; + + /* Banner addresses require TYPE_NONE */ + a->type = CEPH_ENTITY_ADDR_TYPE_NONE; +} +static inline void ceph_decode_banner_addr(struct ceph_entity_addr *a) +{ + __be16 ss_family = *(__be16 *)&a->in_addr.ss_family; + a->in_addr.ss_family = ntohs(ss_family); + WARN_ON(a->in_addr.ss_family == 512); + a->type = CEPH_ENTITY_ADDR_TYPE_LEGACY; +} + +extern int ceph_decode_entity_addr(void **p, void *end, + struct ceph_entity_addr *addr); +int ceph_decode_entity_addrvec(void **p, void *end, bool msgr2, + struct ceph_entity_addr *addr); + +int ceph_entity_addr_encoding_len(const struct ceph_entity_addr *addr); +void ceph_encode_entity_addr(void **p, const struct ceph_entity_addr *addr); + +/* + * encoders + */ +static inline void ceph_encode_64(void **p, u64 v) +{ + put_unaligned_le64(v, (__le64 *)*p); + *p += sizeof(u64); +} +static inline void ceph_encode_32(void **p, u32 v) +{ + put_unaligned_le32(v, (__le32 *)*p); + *p += sizeof(u32); +} +static inline void ceph_encode_16(void **p, u16 v) +{ + put_unaligned_le16(v, (__le16 *)*p); + *p += sizeof(u16); +} +static inline void ceph_encode_8(void **p, u8 v) +{ + *(u8 *)*p = v; + (*p)++; +} +static inline void ceph_encode_copy(void **p, const void *s, int len) +{ + memcpy(*p, s, len); + *p += len; +} + +/* + * filepath, string encoders + */ +static inline void ceph_encode_filepath(void **p, void *end, + u64 ino, const char *path) +{ + u32 len = path ? strlen(path) : 0; + BUG_ON(*p + 1 + sizeof(ino) + sizeof(len) + len > end); + ceph_encode_8(p, 1); + ceph_encode_64(p, ino); + ceph_encode_32(p, len); + if (len) + memcpy(*p, path, len); + *p += len; +} + +static inline void ceph_encode_string(void **p, void *end, + const char *s, u32 len) +{ + BUG_ON(*p + sizeof(len) + len > end); + ceph_encode_32(p, len); + if (len) + memcpy(*p, s, len); + *p += len; +} + +/* + * version and length starting block encoders/decoders + */ + +/* current code version (u8) + compat code version (u8) + len of struct (u32) */ +#define CEPH_ENCODING_START_BLK_LEN 6 + +/** + * ceph_start_encoding - start encoding block + * @struct_v: current (code) version of the encoding + * @struct_compat: oldest code version that can decode it + * @struct_len: length of struct encoding + */ +static inline void ceph_start_encoding(void **p, u8 struct_v, u8 struct_compat, + u32 struct_len) +{ + ceph_encode_8(p, struct_v); + ceph_encode_8(p, struct_compat); + ceph_encode_32(p, struct_len); +} + +/** + * ceph_start_decoding - start decoding block + * @v: current version of the encoding that the code supports + * @name: name of the struct (free-form) + * @struct_v: out param for the encoding version + * @struct_len: out param for the length of struct encoding + * + * Validates the length of struct encoding, so unsafe ceph_decode_* + * variants can be used for decoding. + */ +static inline int ceph_start_decoding(void **p, void *end, u8 v, + const char *name, u8 *struct_v, + u32 *struct_len) +{ + u8 struct_compat; + + ceph_decode_need(p, end, CEPH_ENCODING_START_BLK_LEN, bad); + *struct_v = ceph_decode_8(p); + struct_compat = ceph_decode_8(p); + if (v < struct_compat) { + pr_warn("got struct_v %d struct_compat %d > %d of %s\n", + *struct_v, struct_compat, v, name); + return -EINVAL; + } + + *struct_len = ceph_decode_32(p); + ceph_decode_need(p, end, *struct_len, bad); + return 0; + +bad: + return -ERANGE; +} + +#define ceph_encode_need(p, end, n, bad) \ + do { \ + if (!likely(ceph_has_room(p, end, n))) \ + goto bad; \ + } while (0) + +#define ceph_encode_64_safe(p, end, v, bad) \ + do { \ + ceph_encode_need(p, end, sizeof(u64), bad); \ + ceph_encode_64(p, v); \ + } while (0) +#define ceph_encode_32_safe(p, end, v, bad) \ + do { \ + ceph_encode_need(p, end, sizeof(u32), bad); \ + ceph_encode_32(p, v); \ + } while (0) +#define ceph_encode_16_safe(p, end, v, bad) \ + do { \ + ceph_encode_need(p, end, sizeof(u16), bad); \ + ceph_encode_16(p, v); \ + } while (0) +#define ceph_encode_8_safe(p, end, v, bad) \ + do { \ + ceph_encode_need(p, end, sizeof(u8), bad); \ + ceph_encode_8(p, v); \ + } while (0) + +#define ceph_encode_copy_safe(p, end, pv, n, bad) \ + do { \ + ceph_encode_need(p, end, n, bad); \ + ceph_encode_copy(p, pv, n); \ + } while (0) +#define ceph_encode_string_safe(p, end, s, n, bad) \ + do { \ + ceph_encode_need(p, end, n, bad); \ + ceph_encode_string(p, end, s, n); \ + } while (0) + + +#endif diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h new file mode 100644 index 000000000..4497d0a67 --- /dev/null +++ b/include/linux/ceph/libceph.h @@ -0,0 +1,332 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_LIBCEPH_H +#define _FS_CEPH_LIBCEPH_H + +#include <linux/ceph/ceph_debug.h> + +#include <asm/unaligned.h> +#include <linux/backing-dev.h> +#include <linux/completion.h> +#include <linux/exportfs.h> +#include <linux/bug.h> +#include <linux/fs.h> +#include <linux/mempool.h> +#include <linux/pagemap.h> +#include <linux/wait.h> +#include <linux/writeback.h> +#include <linux/slab.h> +#include <linux/refcount.h> + +#include <linux/ceph/types.h> +#include <linux/ceph/messenger.h> +#include <linux/ceph/msgpool.h> +#include <linux/ceph/mon_client.h> +#include <linux/ceph/osd_client.h> +#include <linux/ceph/ceph_fs.h> +#include <linux/ceph/string_table.h> + +/* + * mount options + */ +#define CEPH_OPT_FSID (1<<0) +#define CEPH_OPT_NOSHARE (1<<1) /* don't share client with other sbs */ +#define CEPH_OPT_MYIP (1<<2) /* specified my ip */ +#define CEPH_OPT_NOCRC (1<<3) /* no data crc on writes (msgr1) */ +#define CEPH_OPT_TCP_NODELAY (1<<4) /* TCP_NODELAY on TCP sockets */ +#define CEPH_OPT_NOMSGSIGN (1<<5) /* don't sign msgs (msgr1) */ +#define CEPH_OPT_ABORT_ON_FULL (1<<6) /* abort w/ ENOSPC when full */ +#define CEPH_OPT_RXBOUNCE (1<<7) /* double-buffer read data */ + +#define CEPH_OPT_DEFAULT (CEPH_OPT_TCP_NODELAY) + +#define ceph_set_opt(client, opt) \ + (client)->options->flags |= CEPH_OPT_##opt; +#define ceph_test_opt(client, opt) \ + (!!((client)->options->flags & CEPH_OPT_##opt)) + +struct ceph_options { + int flags; + struct ceph_fsid fsid; + struct ceph_entity_addr my_addr; + unsigned long mount_timeout; /* jiffies */ + unsigned long osd_idle_ttl; /* jiffies */ + unsigned long osd_keepalive_timeout; /* jiffies */ + unsigned long osd_request_timeout; /* jiffies */ + u32 read_from_replica; /* CEPH_OSD_FLAG_BALANCE/LOCALIZE_READS */ + int con_modes[2]; /* CEPH_CON_MODE_* */ + + /* + * any type that can't be simply compared or doesn't need + * to be compared should go beyond this point, + * ceph_compare_options() should be updated accordingly + */ + + struct ceph_entity_addr *mon_addr; /* should be the first + pointer type of args */ + int num_mon; + char *name; + struct ceph_crypto_key *key; + struct rb_root crush_locs; +}; + +/* + * defaults + */ +#define CEPH_MOUNT_TIMEOUT_DEFAULT msecs_to_jiffies(60 * 1000) +#define CEPH_OSD_KEEPALIVE_DEFAULT msecs_to_jiffies(5 * 1000) +#define CEPH_OSD_IDLE_TTL_DEFAULT msecs_to_jiffies(60 * 1000) +#define CEPH_OSD_REQUEST_TIMEOUT_DEFAULT 0 /* no timeout */ +#define CEPH_READ_FROM_REPLICA_DEFAULT 0 /* read from primary */ + +#define CEPH_MONC_HUNT_INTERVAL msecs_to_jiffies(3 * 1000) +#define CEPH_MONC_PING_INTERVAL msecs_to_jiffies(10 * 1000) +#define CEPH_MONC_PING_TIMEOUT msecs_to_jiffies(30 * 1000) +#define CEPH_MONC_HUNT_BACKOFF 2 +#define CEPH_MONC_HUNT_MAX_MULT 10 + +#define CEPH_MSG_MAX_CONTROL_LEN (16*1024*1024) +#define CEPH_MSG_MAX_FRONT_LEN (16*1024*1024) +#define CEPH_MSG_MAX_MIDDLE_LEN (16*1024*1024) + +/* + * The largest possible rbd data object is 32M. + * The largest possible rbd object map object is 64M. + * + * There is no limit on the size of cephfs objects, but it has to obey + * rsize and wsize mount options anyway. + */ +#define CEPH_MSG_MAX_DATA_LEN (64*1024*1024) + +#define CEPH_AUTH_NAME_DEFAULT "guest" + +static inline unsigned long ceph_timeout_jiffies(unsigned long timeout) +{ + return timeout ?: MAX_SCHEDULE_TIMEOUT; +} + +struct ceph_mds_client; + +/* + * per client state + * + * possibly shared by multiple mount points, if they are + * mounting the same ceph filesystem/cluster. + */ +struct ceph_client { + struct ceph_fsid fsid; + bool have_fsid; + + void *private; + + struct ceph_options *options; + + struct mutex mount_mutex; /* serialize mount attempts */ + wait_queue_head_t auth_wq; + int auth_err; + + int (*extra_mon_dispatch)(struct ceph_client *, struct ceph_msg *); + + u64 supported_features; + u64 required_features; + + struct ceph_messenger msgr; /* messenger instance */ + struct ceph_mon_client monc; + struct ceph_osd_client osdc; + +#ifdef CONFIG_DEBUG_FS + struct dentry *debugfs_dir; + struct dentry *debugfs_monmap; + struct dentry *debugfs_osdmap; + struct dentry *debugfs_options; +#endif +}; + +#define from_msgr(ms) container_of(ms, struct ceph_client, msgr) + +static inline bool ceph_msgr2(struct ceph_client *client) +{ + return client->options->con_modes[0] != CEPH_CON_MODE_UNKNOWN; +} + +/* + * snapshots + */ + +/* + * A "snap context" is the set of existing snapshots when we + * write data. It is used by the OSD to guide its COW behavior. + * + * The ceph_snap_context is refcounted, and attached to each dirty + * page, indicating which context the dirty data belonged when it was + * dirtied. + */ +struct ceph_snap_context { + refcount_t nref; + u64 seq; + u32 num_snaps; + u64 snaps[]; +}; + +extern struct ceph_snap_context *ceph_create_snap_context(u32 snap_count, + gfp_t gfp_flags); +extern struct ceph_snap_context *ceph_get_snap_context( + struct ceph_snap_context *sc); +extern void ceph_put_snap_context(struct ceph_snap_context *sc); + +/* + * calculate the number of pages a given length and offset map onto, + * if we align the data. + */ +static inline int calc_pages_for(u64 off, u64 len) +{ + return ((off+len+PAGE_SIZE-1) >> PAGE_SHIFT) - + (off >> PAGE_SHIFT); +} + +#define RB_BYVAL(a) (a) +#define RB_BYPTR(a) (&(a)) +#define RB_CMP3WAY(a, b) ((a) < (b) ? -1 : (a) > (b)) + +#define DEFINE_RB_INSDEL_FUNCS2(name, type, keyfld, cmpexp, keyexp, nodefld) \ +static bool __insert_##name(struct rb_root *root, type *t) \ +{ \ + struct rb_node **n = &root->rb_node; \ + struct rb_node *parent = NULL; \ + \ + BUG_ON(!RB_EMPTY_NODE(&t->nodefld)); \ + \ + while (*n) { \ + type *cur = rb_entry(*n, type, nodefld); \ + int cmp; \ + \ + parent = *n; \ + cmp = cmpexp(keyexp(t->keyfld), keyexp(cur->keyfld)); \ + if (cmp < 0) \ + n = &(*n)->rb_left; \ + else if (cmp > 0) \ + n = &(*n)->rb_right; \ + else \ + return false; \ + } \ + \ + rb_link_node(&t->nodefld, parent, n); \ + rb_insert_color(&t->nodefld, root); \ + return true; \ +} \ +static void __maybe_unused insert_##name(struct rb_root *root, type *t) \ +{ \ + if (!__insert_##name(root, t)) \ + BUG(); \ +} \ +static void erase_##name(struct rb_root *root, type *t) \ +{ \ + BUG_ON(RB_EMPTY_NODE(&t->nodefld)); \ + rb_erase(&t->nodefld, root); \ + RB_CLEAR_NODE(&t->nodefld); \ +} + +/* + * @lookup_param_type is a parameter and not constructed from (@type, + * @keyfld) with typeof() because adding const is too unwieldy. + */ +#define DEFINE_RB_LOOKUP_FUNC2(name, type, keyfld, cmpexp, keyexp, \ + lookup_param_type, nodefld) \ +static type *lookup_##name(struct rb_root *root, lookup_param_type key) \ +{ \ + struct rb_node *n = root->rb_node; \ + \ + while (n) { \ + type *cur = rb_entry(n, type, nodefld); \ + int cmp; \ + \ + cmp = cmpexp(key, keyexp(cur->keyfld)); \ + if (cmp < 0) \ + n = n->rb_left; \ + else if (cmp > 0) \ + n = n->rb_right; \ + else \ + return cur; \ + } \ + \ + return NULL; \ +} + +#define DEFINE_RB_FUNCS2(name, type, keyfld, cmpexp, keyexp, \ + lookup_param_type, nodefld) \ +DEFINE_RB_INSDEL_FUNCS2(name, type, keyfld, cmpexp, keyexp, nodefld) \ +DEFINE_RB_LOOKUP_FUNC2(name, type, keyfld, cmpexp, keyexp, \ + lookup_param_type, nodefld) + +/* + * Shorthands for integer keys. + */ +#define DEFINE_RB_INSDEL_FUNCS(name, type, keyfld, nodefld) \ +DEFINE_RB_INSDEL_FUNCS2(name, type, keyfld, RB_CMP3WAY, RB_BYVAL, nodefld) + +#define DEFINE_RB_LOOKUP_FUNC(name, type, keyfld, nodefld) \ +extern type __lookup_##name##_key; \ +DEFINE_RB_LOOKUP_FUNC2(name, type, keyfld, RB_CMP3WAY, RB_BYVAL, \ + typeof(__lookup_##name##_key.keyfld), nodefld) + +#define DEFINE_RB_FUNCS(name, type, keyfld, nodefld) \ +DEFINE_RB_INSDEL_FUNCS(name, type, keyfld, nodefld) \ +DEFINE_RB_LOOKUP_FUNC(name, type, keyfld, nodefld) + +extern struct kmem_cache *ceph_inode_cachep; +extern struct kmem_cache *ceph_cap_cachep; +extern struct kmem_cache *ceph_cap_snap_cachep; +extern struct kmem_cache *ceph_cap_flush_cachep; +extern struct kmem_cache *ceph_dentry_cachep; +extern struct kmem_cache *ceph_file_cachep; +extern struct kmem_cache *ceph_dir_file_cachep; +extern struct kmem_cache *ceph_mds_request_cachep; +extern mempool_t *ceph_wb_pagevec_pool; + +/* ceph_common.c */ +extern bool libceph_compatible(void *data); + +extern const char *ceph_msg_type_name(int type); +extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid); +extern int ceph_parse_fsid(const char *str, struct ceph_fsid *fsid); + +struct fs_parameter; +struct fc_log; +struct ceph_options *ceph_alloc_options(void); +int ceph_parse_mon_ips(const char *buf, size_t len, struct ceph_options *opt, + struct fc_log *l, char delim); +int ceph_parse_param(struct fs_parameter *param, struct ceph_options *opt, + struct fc_log *l); +int ceph_print_client_options(struct seq_file *m, struct ceph_client *client, + bool show_all); +extern void ceph_destroy_options(struct ceph_options *opt); +extern int ceph_compare_options(struct ceph_options *new_opt, + struct ceph_client *client); +struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private); +struct ceph_entity_addr *ceph_client_addr(struct ceph_client *client); +u64 ceph_client_gid(struct ceph_client *client); +extern void ceph_destroy_client(struct ceph_client *client); +extern void ceph_reset_client_addr(struct ceph_client *client); +extern int __ceph_open_session(struct ceph_client *client, + unsigned long started); +extern int ceph_open_session(struct ceph_client *client); +int ceph_wait_for_latest_osdmap(struct ceph_client *client, + unsigned long timeout); + +/* pagevec.c */ +extern void ceph_release_page_vector(struct page **pages, int num_pages); +extern void ceph_put_page_vector(struct page **pages, int num_pages, + bool dirty); +extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags); +extern int ceph_copy_user_to_page_vector(struct page **pages, + const void __user *data, + loff_t off, size_t len); +extern void ceph_copy_to_page_vector(struct page **pages, + const void *data, + loff_t off, size_t len); +extern void ceph_copy_from_page_vector(struct page **pages, + void *data, + loff_t off, size_t len); +extern void ceph_zero_page_vector_range(int off, int len, struct page **pages); + + +#endif /* _FS_CEPH_SUPER_H */ diff --git a/include/linux/ceph/mdsmap.h b/include/linux/ceph/mdsmap.h new file mode 100644 index 000000000..4c3e0648d --- /dev/null +++ b/include/linux/ceph/mdsmap.h @@ -0,0 +1,72 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_MDSMAP_H +#define _FS_CEPH_MDSMAP_H + +#include <linux/bug.h> +#include <linux/ceph/types.h> + +/* + * mds map - describe servers in the mds cluster. + * + * we limit fields to those the client actually xcares about + */ +struct ceph_mds_info { + u64 global_id; + struct ceph_entity_addr addr; + s32 state; + int num_export_targets; + bool laggy; + u32 *export_targets; +}; + +struct ceph_mdsmap { + u32 m_epoch, m_client_epoch, m_last_failure; + u32 m_root; + u32 m_session_timeout; /* seconds */ + u32 m_session_autoclose; /* seconds */ + u64 m_max_file_size; + u64 m_max_xattr_size; /* maximum size for xattrs blob */ + u32 m_max_mds; /* expected up:active mds number */ + u32 m_num_active_mds; /* actual up:active mds number */ + u32 possible_max_rank; /* possible max rank index */ + struct ceph_mds_info *m_info; + + /* which object pools file data can be stored in */ + int m_num_data_pg_pools; + u64 *m_data_pg_pools; + u64 m_cas_pg_pool; + + bool m_enabled; + bool m_damaged; + int m_num_laggy; +}; + +static inline struct ceph_entity_addr * +ceph_mdsmap_get_addr(struct ceph_mdsmap *m, int w) +{ + if (w >= m->possible_max_rank) + return NULL; + return &m->m_info[w].addr; +} + +static inline int ceph_mdsmap_get_state(struct ceph_mdsmap *m, int w) +{ + BUG_ON(w < 0); + if (w >= m->possible_max_rank) + return CEPH_MDS_STATE_DNE; + return m->m_info[w].state; +} + +static inline bool ceph_mdsmap_is_laggy(struct ceph_mdsmap *m, int w) +{ + if (w >= 0 && w < m->possible_max_rank) + return m->m_info[w].laggy; + return false; +} + +extern int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m); +struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end, bool msgr2); +extern void ceph_mdsmap_destroy(struct ceph_mdsmap *m); +extern bool ceph_mdsmap_is_cluster_available(struct ceph_mdsmap *m); + +#endif diff --git a/include/linux/ceph/messenger.h b/include/linux/ceph/messenger.h new file mode 100644 index 000000000..99c1726be --- /dev/null +++ b/include/linux/ceph/messenger.h @@ -0,0 +1,587 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __FS_CEPH_MESSENGER_H +#define __FS_CEPH_MESSENGER_H + +#include <linux/bvec.h> +#include <linux/crypto.h> +#include <linux/kref.h> +#include <linux/mutex.h> +#include <linux/net.h> +#include <linux/radix-tree.h> +#include <linux/uio.h> +#include <linux/workqueue.h> +#include <net/net_namespace.h> + +#include <linux/ceph/types.h> +#include <linux/ceph/buffer.h> + +struct ceph_msg; +struct ceph_connection; + +/* + * Ceph defines these callbacks for handling connection events. + */ +struct ceph_connection_operations { + struct ceph_connection *(*get)(struct ceph_connection *); + void (*put)(struct ceph_connection *); + + /* handle an incoming message. */ + void (*dispatch) (struct ceph_connection *con, struct ceph_msg *m); + + /* authorize an outgoing connection */ + struct ceph_auth_handshake *(*get_authorizer) ( + struct ceph_connection *con, + int *proto, int force_new); + int (*add_authorizer_challenge)(struct ceph_connection *con, + void *challenge_buf, + int challenge_buf_len); + int (*verify_authorizer_reply) (struct ceph_connection *con); + int (*invalidate_authorizer)(struct ceph_connection *con); + + /* there was some error on the socket (disconnect, whatever) */ + void (*fault) (struct ceph_connection *con); + + /* a remote host as terminated a message exchange session, and messages + * we sent (or they tried to send us) may be lost. */ + void (*peer_reset) (struct ceph_connection *con); + + struct ceph_msg * (*alloc_msg) (struct ceph_connection *con, + struct ceph_msg_header *hdr, + int *skip); + + void (*reencode_message) (struct ceph_msg *msg); + + int (*sign_message) (struct ceph_msg *msg); + int (*check_message_signature) (struct ceph_msg *msg); + + /* msgr2 authentication exchange */ + int (*get_auth_request)(struct ceph_connection *con, + void *buf, int *buf_len, + void **authorizer, int *authorizer_len); + int (*handle_auth_reply_more)(struct ceph_connection *con, + void *reply, int reply_len, + void *buf, int *buf_len, + void **authorizer, int *authorizer_len); + int (*handle_auth_done)(struct ceph_connection *con, + u64 global_id, void *reply, int reply_len, + u8 *session_key, int *session_key_len, + u8 *con_secret, int *con_secret_len); + int (*handle_auth_bad_method)(struct ceph_connection *con, + int used_proto, int result, + const int *allowed_protos, int proto_cnt, + const int *allowed_modes, int mode_cnt); +}; + +/* use format string %s%lld */ +#define ENTITY_NAME(n) ceph_entity_type_name((n).type), le64_to_cpu((n).num) + +struct ceph_messenger { + struct ceph_entity_inst inst; /* my name+address */ + struct ceph_entity_addr my_enc_addr; + + atomic_t stopping; + possible_net_t net; + + /* + * the global_seq counts connections i (attempt to) initiate + * in order to disambiguate certain connect race conditions. + */ + u32 global_seq; + spinlock_t global_seq_lock; +}; + +enum ceph_msg_data_type { + CEPH_MSG_DATA_NONE, /* message contains no data payload */ + CEPH_MSG_DATA_PAGES, /* data source/destination is a page array */ + CEPH_MSG_DATA_PAGELIST, /* data source/destination is a pagelist */ +#ifdef CONFIG_BLOCK + CEPH_MSG_DATA_BIO, /* data source/destination is a bio list */ +#endif /* CONFIG_BLOCK */ + CEPH_MSG_DATA_BVECS, /* data source/destination is a bio_vec array */ +}; + +#ifdef CONFIG_BLOCK + +struct ceph_bio_iter { + struct bio *bio; + struct bvec_iter iter; +}; + +#define __ceph_bio_iter_advance_step(it, n, STEP) do { \ + unsigned int __n = (n), __cur_n; \ + \ + while (__n) { \ + BUG_ON(!(it)->iter.bi_size); \ + __cur_n = min((it)->iter.bi_size, __n); \ + (void)(STEP); \ + bio_advance_iter((it)->bio, &(it)->iter, __cur_n); \ + if (!(it)->iter.bi_size && (it)->bio->bi_next) { \ + dout("__ceph_bio_iter_advance_step next bio\n"); \ + (it)->bio = (it)->bio->bi_next; \ + (it)->iter = (it)->bio->bi_iter; \ + } \ + __n -= __cur_n; \ + } \ +} while (0) + +/* + * Advance @it by @n bytes. + */ +#define ceph_bio_iter_advance(it, n) \ + __ceph_bio_iter_advance_step(it, n, 0) + +/* + * Advance @it by @n bytes, executing BVEC_STEP for each bio_vec. + */ +#define ceph_bio_iter_advance_step(it, n, BVEC_STEP) \ + __ceph_bio_iter_advance_step(it, n, ({ \ + struct bio_vec bv; \ + struct bvec_iter __cur_iter; \ + \ + __cur_iter = (it)->iter; \ + __cur_iter.bi_size = __cur_n; \ + __bio_for_each_segment(bv, (it)->bio, __cur_iter, __cur_iter) \ + (void)(BVEC_STEP); \ + })) + +#endif /* CONFIG_BLOCK */ + +struct ceph_bvec_iter { + struct bio_vec *bvecs; + struct bvec_iter iter; +}; + +#define __ceph_bvec_iter_advance_step(it, n, STEP) do { \ + BUG_ON((n) > (it)->iter.bi_size); \ + (void)(STEP); \ + bvec_iter_advance((it)->bvecs, &(it)->iter, (n)); \ +} while (0) + +/* + * Advance @it by @n bytes. + */ +#define ceph_bvec_iter_advance(it, n) \ + __ceph_bvec_iter_advance_step(it, n, 0) + +/* + * Advance @it by @n bytes, executing BVEC_STEP for each bio_vec. + */ +#define ceph_bvec_iter_advance_step(it, n, BVEC_STEP) \ + __ceph_bvec_iter_advance_step(it, n, ({ \ + struct bio_vec bv; \ + struct bvec_iter __cur_iter; \ + \ + __cur_iter = (it)->iter; \ + __cur_iter.bi_size = (n); \ + for_each_bvec(bv, (it)->bvecs, __cur_iter, __cur_iter) \ + (void)(BVEC_STEP); \ + })) + +#define ceph_bvec_iter_shorten(it, n) do { \ + BUG_ON((n) > (it)->iter.bi_size); \ + (it)->iter.bi_size = (n); \ +} while (0) + +struct ceph_msg_data { + enum ceph_msg_data_type type; + union { +#ifdef CONFIG_BLOCK + struct { + struct ceph_bio_iter bio_pos; + u32 bio_length; + }; +#endif /* CONFIG_BLOCK */ + struct ceph_bvec_iter bvec_pos; + struct { + struct page **pages; + size_t length; /* total # bytes */ + unsigned int alignment; /* first page */ + bool own_pages; + }; + struct ceph_pagelist *pagelist; + }; +}; + +struct ceph_msg_data_cursor { + size_t total_resid; /* across all data items */ + + struct ceph_msg_data *data; /* current data item */ + size_t resid; /* bytes not yet consumed */ + bool need_crc; /* crc update needed */ + union { +#ifdef CONFIG_BLOCK + struct ceph_bio_iter bio_iter; +#endif /* CONFIG_BLOCK */ + struct bvec_iter bvec_iter; + struct { /* pages */ + unsigned int page_offset; /* offset in page */ + unsigned short page_index; /* index in array */ + unsigned short page_count; /* pages in array */ + }; + struct { /* pagelist */ + struct page *page; /* page from list */ + size_t offset; /* bytes from list */ + }; + }; +}; + +/* + * a single message. it contains a header (src, dest, message type, etc.), + * footer (crc values, mainly), a "front" message body, and possibly a + * data payload (stored in some number of pages). + */ +struct ceph_msg { + struct ceph_msg_header hdr; /* header */ + union { + struct ceph_msg_footer footer; /* footer */ + struct ceph_msg_footer_old old_footer; /* old format footer */ + }; + struct kvec front; /* unaligned blobs of message */ + struct ceph_buffer *middle; + + size_t data_length; + struct ceph_msg_data *data; + int num_data_items; + int max_data_items; + struct ceph_msg_data_cursor cursor; + + struct ceph_connection *con; + struct list_head list_head; /* links for connection lists */ + + struct kref kref; + bool more_to_follow; + bool needs_out_seq; + int front_alloc_len; + + struct ceph_msgpool *pool; +}; + +/* + * connection states + */ +#define CEPH_CON_S_CLOSED 1 +#define CEPH_CON_S_PREOPEN 2 +#define CEPH_CON_S_V1_BANNER 3 +#define CEPH_CON_S_V1_CONNECT_MSG 4 +#define CEPH_CON_S_V2_BANNER_PREFIX 5 +#define CEPH_CON_S_V2_BANNER_PAYLOAD 6 +#define CEPH_CON_S_V2_HELLO 7 +#define CEPH_CON_S_V2_AUTH 8 +#define CEPH_CON_S_V2_AUTH_SIGNATURE 9 +#define CEPH_CON_S_V2_SESSION_CONNECT 10 +#define CEPH_CON_S_V2_SESSION_RECONNECT 11 +#define CEPH_CON_S_OPEN 12 +#define CEPH_CON_S_STANDBY 13 + +/* + * ceph_connection flag bits + */ +#define CEPH_CON_F_LOSSYTX 0 /* we can close channel or drop + messages on errors */ +#define CEPH_CON_F_KEEPALIVE_PENDING 1 /* we need to send a keepalive */ +#define CEPH_CON_F_WRITE_PENDING 2 /* we have data ready to send */ +#define CEPH_CON_F_SOCK_CLOSED 3 /* socket state changed to closed */ +#define CEPH_CON_F_BACKOFF 4 /* need to retry queuing delayed + work */ + +/* ceph connection fault delay defaults, for exponential backoff */ +#define BASE_DELAY_INTERVAL (HZ / 4) +#define MAX_DELAY_INTERVAL (15 * HZ) + +struct ceph_connection_v1_info { + struct kvec out_kvec[8], /* sending header/footer data */ + *out_kvec_cur; + int out_kvec_left; /* kvec's left in out_kvec */ + int out_skip; /* skip this many bytes */ + int out_kvec_bytes; /* total bytes left */ + bool out_more; /* there is more data after the kvecs */ + bool out_msg_done; + + struct ceph_auth_handshake *auth; + int auth_retry; /* true if we need a newer authorizer */ + + /* connection negotiation temps */ + u8 in_banner[CEPH_BANNER_MAX_LEN]; + struct ceph_entity_addr actual_peer_addr; + struct ceph_entity_addr peer_addr_for_me; + struct ceph_msg_connect out_connect; + struct ceph_msg_connect_reply in_reply; + + int in_base_pos; /* bytes read */ + + /* message in temps */ + u8 in_tag; /* protocol control byte */ + struct ceph_msg_header in_hdr; + __le64 in_temp_ack; /* for reading an ack */ + + /* message out temps */ + struct ceph_msg_header out_hdr; + __le64 out_temp_ack; /* for writing an ack */ + struct ceph_timespec out_temp_keepalive2; /* for writing keepalive2 + stamp */ + + u32 connect_seq; /* identify the most recent connection + attempt for this session */ + u32 peer_global_seq; /* peer's global seq for this connection */ +}; + +#define CEPH_CRC_LEN 4 +#define CEPH_GCM_KEY_LEN 16 +#define CEPH_GCM_IV_LEN sizeof(struct ceph_gcm_nonce) +#define CEPH_GCM_BLOCK_LEN 16 +#define CEPH_GCM_TAG_LEN 16 + +#define CEPH_PREAMBLE_LEN 32 +#define CEPH_PREAMBLE_INLINE_LEN 48 +#define CEPH_PREAMBLE_PLAIN_LEN CEPH_PREAMBLE_LEN +#define CEPH_PREAMBLE_SECURE_LEN (CEPH_PREAMBLE_LEN + \ + CEPH_PREAMBLE_INLINE_LEN + \ + CEPH_GCM_TAG_LEN) +#define CEPH_EPILOGUE_PLAIN_LEN (1 + 3 * CEPH_CRC_LEN) +#define CEPH_EPILOGUE_SECURE_LEN (CEPH_GCM_BLOCK_LEN + CEPH_GCM_TAG_LEN) + +#define CEPH_FRAME_MAX_SEGMENT_COUNT 4 + +struct ceph_frame_desc { + int fd_tag; /* FRAME_TAG_* */ + int fd_seg_cnt; + int fd_lens[CEPH_FRAME_MAX_SEGMENT_COUNT]; /* logical */ + int fd_aligns[CEPH_FRAME_MAX_SEGMENT_COUNT]; +}; + +struct ceph_gcm_nonce { + __le32 fixed; + __le64 counter __packed; +}; + +struct ceph_connection_v2_info { + struct iov_iter in_iter; + struct kvec in_kvecs[5]; /* recvmsg */ + struct bio_vec in_bvec; /* recvmsg (in_cursor) */ + int in_kvec_cnt; + int in_state; /* IN_S_* */ + + struct iov_iter out_iter; + struct kvec out_kvecs[8]; /* sendmsg */ + struct bio_vec out_bvec; /* sendpage (out_cursor, out_zero), + sendmsg (out_enc_pages) */ + int out_kvec_cnt; + int out_state; /* OUT_S_* */ + + int out_zero; /* # of zero bytes to send */ + bool out_iter_sendpage; /* use sendpage if possible */ + + struct ceph_frame_desc in_desc; + struct ceph_msg_data_cursor in_cursor; + struct ceph_msg_data_cursor out_cursor; + + struct crypto_shash *hmac_tfm; /* post-auth signature */ + struct crypto_aead *gcm_tfm; /* on-wire encryption */ + struct aead_request *gcm_req; + struct crypto_wait gcm_wait; + struct ceph_gcm_nonce in_gcm_nonce; + struct ceph_gcm_nonce out_gcm_nonce; + + struct page **in_enc_pages; + int in_enc_page_cnt; + int in_enc_resid; + int in_enc_i; + struct page **out_enc_pages; + int out_enc_page_cnt; + int out_enc_resid; + int out_enc_i; + + int con_mode; /* CEPH_CON_MODE_* */ + + void *conn_bufs[16]; + int conn_buf_cnt; + + struct kvec in_sign_kvecs[8]; + struct kvec out_sign_kvecs[8]; + int in_sign_kvec_cnt; + int out_sign_kvec_cnt; + + u64 client_cookie; + u64 server_cookie; + u64 global_seq; + u64 connect_seq; + u64 peer_global_seq; + + u8 in_buf[CEPH_PREAMBLE_SECURE_LEN]; + u8 out_buf[CEPH_PREAMBLE_SECURE_LEN]; + struct { + u8 late_status; /* FRAME_LATE_STATUS_* */ + union { + struct { + u32 front_crc; + u32 middle_crc; + u32 data_crc; + } __packed; + u8 pad[CEPH_GCM_BLOCK_LEN - 1]; + }; + } out_epil; +}; + +/* + * A single connection with another host. + * + * We maintain a queue of outgoing messages, and some session state to + * ensure that we can preserve the lossless, ordered delivery of + * messages in the case of a TCP disconnect. + */ +struct ceph_connection { + void *private; + + const struct ceph_connection_operations *ops; + + struct ceph_messenger *msgr; + + int state; /* CEPH_CON_S_* */ + atomic_t sock_state; + struct socket *sock; + + unsigned long flags; /* CEPH_CON_F_* */ + const char *error_msg; /* error message, if any */ + + struct ceph_entity_name peer_name; /* peer name */ + struct ceph_entity_addr peer_addr; /* peer address */ + u64 peer_features; + + struct mutex mutex; + + /* out queue */ + struct list_head out_queue; + struct list_head out_sent; /* sending or sent but unacked */ + u64 out_seq; /* last message queued for send */ + + u64 in_seq, in_seq_acked; /* last message received, acked */ + + struct ceph_msg *in_msg; + struct ceph_msg *out_msg; /* sending message (== tail of + out_sent) */ + + struct page *bounce_page; + u32 in_front_crc, in_middle_crc, in_data_crc; /* calculated crc */ + + struct timespec64 last_keepalive_ack; /* keepalive2 ack stamp */ + + struct delayed_work work; /* send|recv work */ + unsigned long delay; /* current delay interval */ + + union { + struct ceph_connection_v1_info v1; + struct ceph_connection_v2_info v2; + }; +}; + +extern struct page *ceph_zero_page; + +void ceph_con_flag_clear(struct ceph_connection *con, unsigned long con_flag); +void ceph_con_flag_set(struct ceph_connection *con, unsigned long con_flag); +bool ceph_con_flag_test(struct ceph_connection *con, unsigned long con_flag); +bool ceph_con_flag_test_and_clear(struct ceph_connection *con, + unsigned long con_flag); +bool ceph_con_flag_test_and_set(struct ceph_connection *con, + unsigned long con_flag); + +void ceph_encode_my_addr(struct ceph_messenger *msgr); + +int ceph_tcp_connect(struct ceph_connection *con); +int ceph_con_close_socket(struct ceph_connection *con); +void ceph_con_reset_session(struct ceph_connection *con); + +u32 ceph_get_global_seq(struct ceph_messenger *msgr, u32 gt); +void ceph_con_discard_sent(struct ceph_connection *con, u64 ack_seq); +void ceph_con_discard_requeued(struct ceph_connection *con, u64 reconnect_seq); + +void ceph_msg_data_cursor_init(struct ceph_msg_data_cursor *cursor, + struct ceph_msg *msg, size_t length); +struct page *ceph_msg_data_next(struct ceph_msg_data_cursor *cursor, + size_t *page_offset, size_t *length); +void ceph_msg_data_advance(struct ceph_msg_data_cursor *cursor, size_t bytes); + +u32 ceph_crc32c_page(u32 crc, struct page *page, unsigned int page_offset, + unsigned int length); + +bool ceph_addr_is_blank(const struct ceph_entity_addr *addr); +int ceph_addr_port(const struct ceph_entity_addr *addr); +void ceph_addr_set_port(struct ceph_entity_addr *addr, int p); + +void ceph_con_process_message(struct ceph_connection *con); +int ceph_con_in_msg_alloc(struct ceph_connection *con, + struct ceph_msg_header *hdr, int *skip); +void ceph_con_get_out_msg(struct ceph_connection *con); + +/* messenger_v1.c */ +int ceph_con_v1_try_read(struct ceph_connection *con); +int ceph_con_v1_try_write(struct ceph_connection *con); +void ceph_con_v1_revoke(struct ceph_connection *con); +void ceph_con_v1_revoke_incoming(struct ceph_connection *con); +bool ceph_con_v1_opened(struct ceph_connection *con); +void ceph_con_v1_reset_session(struct ceph_connection *con); +void ceph_con_v1_reset_protocol(struct ceph_connection *con); + +/* messenger_v2.c */ +int ceph_con_v2_try_read(struct ceph_connection *con); +int ceph_con_v2_try_write(struct ceph_connection *con); +void ceph_con_v2_revoke(struct ceph_connection *con); +void ceph_con_v2_revoke_incoming(struct ceph_connection *con); +bool ceph_con_v2_opened(struct ceph_connection *con); +void ceph_con_v2_reset_session(struct ceph_connection *con); +void ceph_con_v2_reset_protocol(struct ceph_connection *con); + + +extern const char *ceph_pr_addr(const struct ceph_entity_addr *addr); + +extern int ceph_parse_ips(const char *c, const char *end, + struct ceph_entity_addr *addr, + int max_count, int *count, char delim); + +extern int ceph_msgr_init(void); +extern void ceph_msgr_exit(void); +extern void ceph_msgr_flush(void); + +extern void ceph_messenger_init(struct ceph_messenger *msgr, + struct ceph_entity_addr *myaddr); +extern void ceph_messenger_fini(struct ceph_messenger *msgr); +extern void ceph_messenger_reset_nonce(struct ceph_messenger *msgr); + +extern void ceph_con_init(struct ceph_connection *con, void *private, + const struct ceph_connection_operations *ops, + struct ceph_messenger *msgr); +extern void ceph_con_open(struct ceph_connection *con, + __u8 entity_type, __u64 entity_num, + struct ceph_entity_addr *addr); +extern bool ceph_con_opened(struct ceph_connection *con); +extern void ceph_con_close(struct ceph_connection *con); +extern void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg); + +extern void ceph_msg_revoke(struct ceph_msg *msg); +extern void ceph_msg_revoke_incoming(struct ceph_msg *msg); + +extern void ceph_con_keepalive(struct ceph_connection *con); +extern bool ceph_con_keepalive_expired(struct ceph_connection *con, + unsigned long interval); + +void ceph_msg_data_add_pages(struct ceph_msg *msg, struct page **pages, + size_t length, size_t alignment, bool own_pages); +extern void ceph_msg_data_add_pagelist(struct ceph_msg *msg, + struct ceph_pagelist *pagelist); +#ifdef CONFIG_BLOCK +void ceph_msg_data_add_bio(struct ceph_msg *msg, struct ceph_bio_iter *bio_pos, + u32 length); +#endif /* CONFIG_BLOCK */ +void ceph_msg_data_add_bvecs(struct ceph_msg *msg, + struct ceph_bvec_iter *bvec_pos); + +struct ceph_msg *ceph_msg_new2(int type, int front_len, int max_data_items, + gfp_t flags, bool can_fail); +extern struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags, + bool can_fail); + +extern struct ceph_msg *ceph_msg_get(struct ceph_msg *msg); +extern void ceph_msg_put(struct ceph_msg *msg); + +extern void ceph_msg_dump(struct ceph_msg *msg); + +#endif diff --git a/include/linux/ceph/mon_client.h b/include/linux/ceph/mon_client.h new file mode 100644 index 000000000..b65896115 --- /dev/null +++ b/include/linux/ceph/mon_client.h @@ -0,0 +1,152 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_MON_CLIENT_H +#define _FS_CEPH_MON_CLIENT_H + +#include <linux/completion.h> +#include <linux/kref.h> +#include <linux/rbtree.h> + +#include <linux/ceph/messenger.h> + +struct ceph_client; +struct ceph_mount_args; +struct ceph_auth_client; + +/* + * The monitor map enumerates the set of all monitors. + */ +struct ceph_monmap { + struct ceph_fsid fsid; + u32 epoch; + u32 num_mon; + struct ceph_entity_inst mon_inst[]; +}; + +struct ceph_mon_client; +struct ceph_mon_generic_request; + + +/* + * Generic mechanism for resending monitor requests. + */ +typedef void (*ceph_monc_request_func_t)(struct ceph_mon_client *monc, + int newmon); + +/* a pending monitor request */ +struct ceph_mon_request { + struct ceph_mon_client *monc; + struct delayed_work delayed_work; + unsigned long delay; + ceph_monc_request_func_t do_request; +}; + +typedef void (*ceph_monc_callback_t)(struct ceph_mon_generic_request *); + +/* + * ceph_mon_generic_request is being used for the statfs and + * mon_get_version requests which are being done a bit differently + * because we need to get data back to the caller + */ +struct ceph_mon_generic_request { + struct ceph_mon_client *monc; + struct kref kref; + u64 tid; + struct rb_node node; + int result; + + struct completion completion; + ceph_monc_callback_t complete_cb; + u64 private_data; /* r_tid/linger_id */ + + struct ceph_msg *request; /* original request */ + struct ceph_msg *reply; /* and reply */ + + union { + struct ceph_statfs *st; + u64 newest; + } u; +}; + +struct ceph_mon_client { + struct ceph_client *client; + struct ceph_monmap *monmap; + + struct mutex mutex; + struct delayed_work delayed_work; + + struct ceph_auth_client *auth; + struct ceph_msg *m_auth, *m_auth_reply, *m_subscribe, *m_subscribe_ack; + int pending_auth; + + bool hunting; + int cur_mon; /* last monitor i contacted */ + unsigned long sub_renew_after; + unsigned long sub_renew_sent; + struct ceph_connection con; + + bool had_a_connection; + int hunt_mult; /* [1..CEPH_MONC_HUNT_MAX_MULT] */ + + /* pending generic requests */ + struct rb_root generic_request_tree; + u64 last_tid; + + /* subs, indexed with CEPH_SUB_* */ + struct { + struct ceph_mon_subscribe_item item; + bool want; + u32 have; /* epoch */ + } subs[4]; + int fs_cluster_id; /* "mdsmap.<id>" sub */ + +#ifdef CONFIG_DEBUG_FS + struct dentry *debugfs_file; +#endif +}; + +extern int ceph_monmap_contains(struct ceph_monmap *m, + struct ceph_entity_addr *addr); + +extern int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl); +extern void ceph_monc_stop(struct ceph_mon_client *monc); +extern void ceph_monc_reopen_session(struct ceph_mon_client *monc); + +enum { + CEPH_SUB_MONMAP = 0, + CEPH_SUB_OSDMAP, + CEPH_SUB_FSMAP, + CEPH_SUB_MDSMAP, +}; + +extern const char *ceph_sub_str[]; + +/* + * The model here is to indicate that we need a new map of at least + * epoch @epoch, and also call in when we receive a map. We will + * periodically rerequest the map from the monitor cluster until we + * get what we want. + */ +bool ceph_monc_want_map(struct ceph_mon_client *monc, int sub, u32 epoch, + bool continuous); +void ceph_monc_got_map(struct ceph_mon_client *monc, int sub, u32 epoch); +void ceph_monc_renew_subs(struct ceph_mon_client *monc); + +extern int ceph_monc_wait_osdmap(struct ceph_mon_client *monc, u32 epoch, + unsigned long timeout); + +int ceph_monc_do_statfs(struct ceph_mon_client *monc, u64 data_pool, + struct ceph_statfs *buf); + +int ceph_monc_get_version(struct ceph_mon_client *monc, const char *what, + u64 *newest); +int ceph_monc_get_version_async(struct ceph_mon_client *monc, const char *what, + ceph_monc_callback_t cb, u64 private_data); + +int ceph_monc_blocklist_add(struct ceph_mon_client *monc, + struct ceph_entity_addr *client_addr); + +extern int ceph_monc_open_session(struct ceph_mon_client *monc); + +extern int ceph_monc_validate_auth(struct ceph_mon_client *monc); + +#endif diff --git a/include/linux/ceph/msgpool.h b/include/linux/ceph/msgpool.h new file mode 100644 index 000000000..729cdf700 --- /dev/null +++ b/include/linux/ceph/msgpool.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_MSGPOOL +#define _FS_CEPH_MSGPOOL + +#include <linux/mempool.h> + +/* + * we use memory pools for preallocating messages we may receive, to + * avoid unexpected OOM conditions. + */ +struct ceph_msgpool { + const char *name; + mempool_t *pool; + int type; /* preallocated message type */ + int front_len; /* preallocated payload size */ + int max_data_items; +}; + +int ceph_msgpool_init(struct ceph_msgpool *pool, int type, + int front_len, int max_data_items, int size, + const char *name); +extern void ceph_msgpool_destroy(struct ceph_msgpool *pool); +struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool, int front_len, + int max_data_items); +extern void ceph_msgpool_put(struct ceph_msgpool *, struct ceph_msg *); + +#endif diff --git a/include/linux/ceph/msgr.h b/include/linux/ceph/msgr.h new file mode 100644 index 000000000..3989dcb94 --- /dev/null +++ b/include/linux/ceph/msgr.h @@ -0,0 +1,234 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef CEPH_MSGR_H +#define CEPH_MSGR_H + +/* + * Data types for message passing layer used by Ceph. + */ + +#define CEPH_MON_PORT 6789 /* default monitor port */ + +/* + * tcp connection banner. include a protocol version. and adjust + * whenever the wire protocol changes. try to keep this string length + * constant. + */ +#define CEPH_BANNER "ceph v027" +#define CEPH_BANNER_LEN 9 +#define CEPH_BANNER_MAX_LEN 30 + + +/* + * messenger V2 connection banner prefix. + * The full banner string should have the form: "ceph v2\n<le16>" + * the 2 bytes are the length of the remaining banner. + */ +#define CEPH_BANNER_V2 "ceph v2\n" +#define CEPH_BANNER_V2_LEN 8 +#define CEPH_BANNER_V2_PREFIX_LEN (CEPH_BANNER_V2_LEN + sizeof(__le16)) + +/* + * messenger V2 features + */ +#define CEPH_MSGR2_INCARNATION_1 (0ull) + +#define DEFINE_MSGR2_FEATURE(bit, incarnation, name) \ + static const uint64_t __maybe_unused CEPH_MSGR2_FEATURE_##name = (1ULL << bit); \ + static const uint64_t __maybe_unused CEPH_MSGR2_FEATUREMASK_##name = \ + (1ULL << bit | CEPH_MSGR2_INCARNATION_##incarnation); + +#define HAVE_MSGR2_FEATURE(x, name) \ + (((x) & (CEPH_MSGR2_FEATUREMASK_##name)) == (CEPH_MSGR2_FEATUREMASK_##name)) + +DEFINE_MSGR2_FEATURE( 0, 1, REVISION_1) // msgr2.1 + +#define CEPH_MSGR2_SUPPORTED_FEATURES (CEPH_MSGR2_FEATURE_REVISION_1) + +#define CEPH_MSGR2_REQUIRED_FEATURES (CEPH_MSGR2_FEATURE_REVISION_1) + + +/* + * Rollover-safe type and comparator for 32-bit sequence numbers. + * Comparator returns -1, 0, or 1. + */ +typedef __u32 ceph_seq_t; + +static inline __s32 ceph_seq_cmp(__u32 a, __u32 b) +{ + return (__s32)a - (__s32)b; +} + + +/* + * entity_name -- logical name for a process participating in the + * network, e.g. 'mds0' or 'osd3'. + */ +struct ceph_entity_name { + __u8 type; /* CEPH_ENTITY_TYPE_* */ + __le64 num; +} __attribute__ ((packed)); + +#define CEPH_ENTITY_TYPE_MON 0x01 +#define CEPH_ENTITY_TYPE_MDS 0x02 +#define CEPH_ENTITY_TYPE_OSD 0x04 +#define CEPH_ENTITY_TYPE_CLIENT 0x08 +#define CEPH_ENTITY_TYPE_AUTH 0x20 + +#define CEPH_ENTITY_TYPE_ANY 0xFF + +extern const char *ceph_entity_type_name(int type); + +/* + * entity_addr -- network address + */ +struct ceph_entity_addr { + __le32 type; /* CEPH_ENTITY_ADDR_TYPE_* */ + __le32 nonce; /* unique id for process (e.g. pid) */ + struct sockaddr_storage in_addr; +} __attribute__ ((packed)); + +static inline bool ceph_addr_equal_no_type(const struct ceph_entity_addr *lhs, + const struct ceph_entity_addr *rhs) +{ + return !memcmp(&lhs->in_addr, &rhs->in_addr, sizeof(lhs->in_addr)) && + lhs->nonce == rhs->nonce; +} + +struct ceph_entity_inst { + struct ceph_entity_name name; + struct ceph_entity_addr addr; +} __attribute__ ((packed)); + + +/* used by message exchange protocol */ +#define CEPH_MSGR_TAG_READY 1 /* server->client: ready for messages */ +#define CEPH_MSGR_TAG_RESETSESSION 2 /* server->client: reset, try again */ +#define CEPH_MSGR_TAG_WAIT 3 /* server->client: wait for racing + incoming connection */ +#define CEPH_MSGR_TAG_RETRY_SESSION 4 /* server->client + cseq: try again + with higher cseq */ +#define CEPH_MSGR_TAG_RETRY_GLOBAL 5 /* server->client + gseq: try again + with higher gseq */ +#define CEPH_MSGR_TAG_CLOSE 6 /* closing pipe */ +#define CEPH_MSGR_TAG_MSG 7 /* message */ +#define CEPH_MSGR_TAG_ACK 8 /* message ack */ +#define CEPH_MSGR_TAG_KEEPALIVE 9 /* just a keepalive byte! */ +#define CEPH_MSGR_TAG_BADPROTOVER 10 /* bad protocol version */ +#define CEPH_MSGR_TAG_BADAUTHORIZER 11 /* bad authorizer */ +#define CEPH_MSGR_TAG_FEATURES 12 /* insufficient features */ +#define CEPH_MSGR_TAG_SEQ 13 /* 64-bit int follows with seen seq number */ +#define CEPH_MSGR_TAG_KEEPALIVE2 14 /* keepalive2 byte + ceph_timespec */ +#define CEPH_MSGR_TAG_KEEPALIVE2_ACK 15 /* keepalive2 reply */ +#define CEPH_MSGR_TAG_CHALLENGE_AUTHORIZER 16 /* cephx v2 doing server challenge */ + +/* + * connection negotiation + */ +struct ceph_msg_connect { + __le64 features; /* supported feature bits */ + __le32 host_type; /* CEPH_ENTITY_TYPE_* */ + __le32 global_seq; /* count connections initiated by this host */ + __le32 connect_seq; /* count connections initiated in this session */ + __le32 protocol_version; + __le32 authorizer_protocol; + __le32 authorizer_len; + __u8 flags; /* CEPH_MSG_CONNECT_* */ +} __attribute__ ((packed)); + +struct ceph_msg_connect_reply { + __u8 tag; + __le64 features; /* feature bits for this session */ + __le32 global_seq; + __le32 connect_seq; + __le32 protocol_version; + __le32 authorizer_len; + __u8 flags; +} __attribute__ ((packed)); + +#define CEPH_MSG_CONNECT_LOSSY 1 /* messages i send may be safely dropped */ + + +/* + * message header + */ +struct ceph_msg_header_old { + __le64 seq; /* message seq# for this session */ + __le64 tid; /* transaction id */ + __le16 type; /* message type */ + __le16 priority; /* priority. higher value == higher priority */ + __le16 version; /* version of message encoding */ + + __le32 front_len; /* bytes in main payload */ + __le32 middle_len;/* bytes in middle payload */ + __le32 data_len; /* bytes of data payload */ + __le16 data_off; /* sender: include full offset; + receiver: mask against ~PAGE_MASK */ + + struct ceph_entity_inst src, orig_src; + __le32 reserved; + __le32 crc; /* header crc32c */ +} __attribute__ ((packed)); + +struct ceph_msg_header { + __le64 seq; /* message seq# for this session */ + __le64 tid; /* transaction id */ + __le16 type; /* message type */ + __le16 priority; /* priority. higher value == higher priority */ + __le16 version; /* version of message encoding */ + + __le32 front_len; /* bytes in main payload */ + __le32 middle_len;/* bytes in middle payload */ + __le32 data_len; /* bytes of data payload */ + __le16 data_off; /* sender: include full offset; + receiver: mask against ~PAGE_MASK */ + + struct ceph_entity_name src; + __le16 compat_version; + __le16 reserved; + __le32 crc; /* header crc32c */ +} __attribute__ ((packed)); + +struct ceph_msg_header2 { + __le64 seq; /* message seq# for this session */ + __le64 tid; /* transaction id */ + __le16 type; /* message type */ + __le16 priority; /* priority. higher value == higher priority */ + __le16 version; /* version of message encoding */ + + __le32 data_pre_padding_len; + __le16 data_off; /* sender: include full offset; + receiver: mask against ~PAGE_MASK */ + + __le64 ack_seq; + __u8 flags; + /* oldest code we think can decode this. unknown if zero. */ + __le16 compat_version; + __le16 reserved; +} __attribute__ ((packed)); + +#define CEPH_MSG_PRIO_LOW 64 +#define CEPH_MSG_PRIO_DEFAULT 127 +#define CEPH_MSG_PRIO_HIGH 196 +#define CEPH_MSG_PRIO_HIGHEST 255 + +/* + * follows data payload + */ +struct ceph_msg_footer_old { + __le32 front_crc, middle_crc, data_crc; + __u8 flags; +} __attribute__ ((packed)); + +struct ceph_msg_footer { + __le32 front_crc, middle_crc, data_crc; + // sig holds the 64 bits of the digital signature for the message PLR + __le64 sig; + __u8 flags; +} __attribute__ ((packed)); + +#define CEPH_MSG_FOOTER_COMPLETE (1<<0) /* msg wasn't aborted */ +#define CEPH_MSG_FOOTER_NOCRC (1<<1) /* no data crc */ +#define CEPH_MSG_FOOTER_SIGNED (1<<2) /* msg was signed */ + + +#endif diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h new file mode 100644 index 000000000..fb6be7210 --- /dev/null +++ b/include/linux/ceph/osd_client.h @@ -0,0 +1,562 @@ +/* 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 given osd we're communicating with */ +struct ceph_osd { + refcount_t o_ref; + struct ceph_osd_client *o_osdc; + int o_osd; + int o_incarnation; + struct rb_node o_node; + struct ceph_connection o_con; + 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; +}; + +#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, +}; + +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 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; + 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 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; + 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 */ + bool r_linger; /* don't resend on failure */ + + /* internal */ + 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[]; +}; + +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_reply(struct ceph_osd_client *osdc, + struct ceph_msg *msg); +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); +extern void osd_req_op_extent_osd_data_pagelist(struct ceph_osd_request *, + unsigned int which, + struct ceph_pagelist *pagelist); +#ifdef CONFIG_BLOCK +void 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); + +extern void osd_req_op_cls_request_data_pagelist(struct ceph_osd_request *, + unsigned int which, + struct ceph_pagelist *pagelist); +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); + +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_watch_check(struct ceph_osd_client *osdc, + struct ceph_osd_linger_request *lreq); +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); +#endif + diff --git a/include/linux/ceph/osdmap.h b/include/linux/ceph/osdmap.h new file mode 100644 index 000000000..5553019c3 --- /dev/null +++ b/include/linux/ceph/osdmap.h @@ -0,0 +1,339 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_OSDMAP_H +#define _FS_CEPH_OSDMAP_H + +#include <linux/rbtree.h> +#include <linux/ceph/types.h> +#include <linux/ceph/decode.h> +#include <linux/crush/crush.h> + +/* + * The osd map describes the current membership of the osd cluster and + * specifies the mapping of objects to placement groups and placement + * groups to (sets of) osds. That is, it completely specifies the + * (desired) distribution of all data objects in the system at some + * point in time. + * + * Each map version is identified by an epoch, which increases monotonically. + * + * The map can be updated either via an incremental map (diff) describing + * the change between two successive epochs, or as a fully encoded map. + */ +struct ceph_pg { + uint64_t pool; + uint32_t seed; +}; + +#define CEPH_SPG_NOSHARD -1 + +struct ceph_spg { + struct ceph_pg pgid; + s8 shard; +}; + +int ceph_pg_compare(const struct ceph_pg *lhs, const struct ceph_pg *rhs); +int ceph_spg_compare(const struct ceph_spg *lhs, const struct ceph_spg *rhs); + +#define CEPH_POOL_FLAG_HASHPSPOOL (1ULL << 0) /* hash pg seed and pool id + together */ +#define CEPH_POOL_FLAG_FULL (1ULL << 1) /* pool is full */ +#define CEPH_POOL_FLAG_FULL_QUOTA (1ULL << 10) /* pool ran out of quota, + will set FULL too */ +#define CEPH_POOL_FLAG_NEARFULL (1ULL << 11) /* pool is nearfull */ + +struct ceph_pg_pool_info { + struct rb_node node; + s64 id; + u8 type; /* CEPH_POOL_TYPE_* */ + u8 size; + u8 min_size; + u8 crush_ruleset; + u8 object_hash; + u32 last_force_request_resend; + u32 pg_num, pgp_num; + int pg_num_mask, pgp_num_mask; + s64 read_tier; + s64 write_tier; /* wins for read+write ops */ + u64 flags; /* CEPH_POOL_FLAG_* */ + char *name; + + bool was_full; /* for handle_one_map() */ +}; + +static inline bool ceph_can_shift_osds(struct ceph_pg_pool_info *pool) +{ + switch (pool->type) { + case CEPH_POOL_TYPE_REP: + return true; + case CEPH_POOL_TYPE_EC: + return false; + default: + BUG(); + } +} + +struct ceph_object_locator { + s64 pool; + struct ceph_string *pool_ns; +}; + +static inline void ceph_oloc_init(struct ceph_object_locator *oloc) +{ + oloc->pool = -1; + oloc->pool_ns = NULL; +} + +static inline bool ceph_oloc_empty(const struct ceph_object_locator *oloc) +{ + return oloc->pool == -1; +} + +void ceph_oloc_copy(struct ceph_object_locator *dest, + const struct ceph_object_locator *src); +void ceph_oloc_destroy(struct ceph_object_locator *oloc); + +/* + * 51-char inline_name is long enough for all cephfs and all but one + * rbd requests: <imgname> in "<imgname>.rbd"/"rbd_id.<imgname>" can be + * arbitrarily long (~PAGE_SIZE). It's done once during rbd map; all + * other rbd requests fit into inline_name. + * + * Makes ceph_object_id 64 bytes on 64-bit. + */ +#define CEPH_OID_INLINE_LEN 52 + +/* + * Both inline and external buffers have space for a NUL-terminator, + * which is carried around. It's not required though - RADOS object + * names don't have to be NUL-terminated and may contain NULs. + */ +struct ceph_object_id { + char *name; + char inline_name[CEPH_OID_INLINE_LEN]; + int name_len; +}; + +#define __CEPH_OID_INITIALIZER(oid) { .name = (oid).inline_name } + +#define CEPH_DEFINE_OID_ONSTACK(oid) \ + struct ceph_object_id oid = __CEPH_OID_INITIALIZER(oid) + +static inline void ceph_oid_init(struct ceph_object_id *oid) +{ + *oid = (struct ceph_object_id) __CEPH_OID_INITIALIZER(*oid); +} + +static inline bool ceph_oid_empty(const struct ceph_object_id *oid) +{ + return oid->name == oid->inline_name && !oid->name_len; +} + +void ceph_oid_copy(struct ceph_object_id *dest, + const struct ceph_object_id *src); +__printf(2, 3) +void ceph_oid_printf(struct ceph_object_id *oid, const char *fmt, ...); +__printf(3, 4) +int ceph_oid_aprintf(struct ceph_object_id *oid, gfp_t gfp, + const char *fmt, ...); +void ceph_oid_destroy(struct ceph_object_id *oid); + +struct workspace_manager { + struct list_head idle_ws; + spinlock_t ws_lock; + /* Number of free workspaces */ + int free_ws; + /* Total number of allocated workspaces */ + atomic_t total_ws; + /* Waiters for a free workspace */ + wait_queue_head_t ws_wait; +}; + +struct ceph_pg_mapping { + struct rb_node node; + struct ceph_pg pgid; + + union { + struct { + int len; + int osds[]; + } pg_temp, pg_upmap; + struct { + int osd; + } primary_temp; + struct { + int len; + int from_to[][2]; + } pg_upmap_items; + }; +}; + +struct ceph_osdmap { + struct ceph_fsid fsid; + u32 epoch; + struct ceph_timespec created, modified; + + u32 flags; /* CEPH_OSDMAP_* */ + + u32 max_osd; /* size of osd_state, _offload, _addr arrays */ + u32 *osd_state; /* CEPH_OSD_* */ + u32 *osd_weight; /* 0 = failed, 0x10000 = 100% normal */ + struct ceph_entity_addr *osd_addr; + + struct rb_root pg_temp; + struct rb_root primary_temp; + + /* remap (post-CRUSH, pre-up) */ + struct rb_root pg_upmap; /* PG := raw set */ + struct rb_root pg_upmap_items; /* from -> to within raw set */ + + u32 *osd_primary_affinity; + + struct rb_root pg_pools; + u32 pool_max; + + /* the CRUSH map specifies the mapping of placement groups to + * the list of osds that store+replicate them. */ + struct crush_map *crush; + + struct workspace_manager crush_wsm; +}; + +static inline bool ceph_osd_exists(struct ceph_osdmap *map, int osd) +{ + return osd >= 0 && osd < map->max_osd && + (map->osd_state[osd] & CEPH_OSD_EXISTS); +} + +static inline bool ceph_osd_is_up(struct ceph_osdmap *map, int osd) +{ + return ceph_osd_exists(map, osd) && + (map->osd_state[osd] & CEPH_OSD_UP); +} + +static inline bool ceph_osd_is_down(struct ceph_osdmap *map, int osd) +{ + return !ceph_osd_is_up(map, osd); +} + +char *ceph_osdmap_state_str(char *str, int len, u32 state); +extern u32 ceph_get_primary_affinity(struct ceph_osdmap *map, int osd); + +static inline struct ceph_entity_addr *ceph_osd_addr(struct ceph_osdmap *map, + int osd) +{ + if (osd >= map->max_osd) + return NULL; + return &map->osd_addr[osd]; +} + +#define CEPH_PGID_ENCODING_LEN (1 + 8 + 4 + 4) + +static inline int ceph_decode_pgid(void **p, void *end, struct ceph_pg *pgid) +{ + __u8 version; + + if (!ceph_has_room(p, end, CEPH_PGID_ENCODING_LEN)) { + pr_warn("incomplete pg encoding\n"); + return -EINVAL; + } + version = ceph_decode_8(p); + if (version > 1) { + pr_warn("do not understand pg encoding %d > 1\n", + (int)version); + return -EINVAL; + } + + pgid->pool = ceph_decode_64(p); + pgid->seed = ceph_decode_32(p); + *p += 4; /* skip deprecated preferred value */ + + return 0; +} + +struct ceph_osdmap *ceph_osdmap_alloc(void); +struct ceph_osdmap *ceph_osdmap_decode(void **p, void *end, bool msgr2); +struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, bool msgr2, + struct ceph_osdmap *map); +extern void ceph_osdmap_destroy(struct ceph_osdmap *map); + +struct ceph_osds { + int osds[CEPH_PG_MAX_SIZE]; + int size; + int primary; /* id, NOT index */ +}; + +static inline void ceph_osds_init(struct ceph_osds *set) +{ + set->size = 0; + set->primary = -1; +} + +void ceph_osds_copy(struct ceph_osds *dest, const struct ceph_osds *src); + +bool ceph_pg_is_split(const struct ceph_pg *pgid, u32 old_pg_num, + u32 new_pg_num); +bool ceph_is_new_interval(const struct ceph_osds *old_acting, + const struct ceph_osds *new_acting, + const struct ceph_osds *old_up, + const struct ceph_osds *new_up, + int old_size, + int new_size, + int old_min_size, + int new_min_size, + u32 old_pg_num, + u32 new_pg_num, + bool old_sort_bitwise, + bool new_sort_bitwise, + bool old_recovery_deletes, + bool new_recovery_deletes, + const struct ceph_pg *pgid); +bool ceph_osds_changed(const struct ceph_osds *old_acting, + const struct ceph_osds *new_acting, + bool any_change); + +void __ceph_object_locator_to_pg(struct ceph_pg_pool_info *pi, + const struct ceph_object_id *oid, + const struct ceph_object_locator *oloc, + struct ceph_pg *raw_pgid); +int ceph_object_locator_to_pg(struct ceph_osdmap *osdmap, + const struct ceph_object_id *oid, + const struct ceph_object_locator *oloc, + struct ceph_pg *raw_pgid); + +void ceph_pg_to_up_acting_osds(struct ceph_osdmap *osdmap, + struct ceph_pg_pool_info *pi, + const struct ceph_pg *raw_pgid, + struct ceph_osds *up, + struct ceph_osds *acting); +bool ceph_pg_to_primary_shard(struct ceph_osdmap *osdmap, + struct ceph_pg_pool_info *pi, + const struct ceph_pg *raw_pgid, + struct ceph_spg *spgid); +int ceph_pg_to_acting_primary(struct ceph_osdmap *osdmap, + const struct ceph_pg *raw_pgid); + +struct crush_loc { + char *cl_type_name; + char *cl_name; +}; + +struct crush_loc_node { + struct rb_node cl_node; + struct crush_loc cl_loc; /* pointers into cl_data */ + char cl_data[]; +}; + +int ceph_parse_crush_location(char *crush_location, struct rb_root *locs); +int ceph_compare_crush_locs(struct rb_root *locs1, struct rb_root *locs2); +void ceph_clear_crush_locs(struct rb_root *locs); + +int ceph_get_crush_locality(struct ceph_osdmap *osdmap, int id, + struct rb_root *locs); + +extern struct ceph_pg_pool_info *ceph_pg_pool_by_id(struct ceph_osdmap *map, + u64 id); +extern const char *ceph_pg_pool_name_by_id(struct ceph_osdmap *map, u64 id); +extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name); +u64 ceph_pg_pool_flags(struct ceph_osdmap *map, u64 id); + +#endif diff --git a/include/linux/ceph/pagelist.h b/include/linux/ceph/pagelist.h new file mode 100644 index 000000000..5dead8486 --- /dev/null +++ b/include/linux/ceph/pagelist.h @@ -0,0 +1,72 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __FS_CEPH_PAGELIST_H +#define __FS_CEPH_PAGELIST_H + +#include <asm/byteorder.h> +#include <linux/refcount.h> +#include <linux/list.h> +#include <linux/types.h> + +struct ceph_pagelist { + struct list_head head; + void *mapped_tail; + size_t length; + size_t room; + struct list_head free_list; + size_t num_pages_free; + refcount_t refcnt; +}; + +struct ceph_pagelist_cursor { + struct ceph_pagelist *pl; /* pagelist, for error checking */ + struct list_head *page_lru; /* page in list */ + size_t room; /* room remaining to reset to */ +}; + +struct ceph_pagelist *ceph_pagelist_alloc(gfp_t gfp_flags); + +extern void ceph_pagelist_release(struct ceph_pagelist *pl); + +extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l); + +extern int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space); + +extern int ceph_pagelist_free_reserve(struct ceph_pagelist *pl); + +extern void ceph_pagelist_set_cursor(struct ceph_pagelist *pl, + struct ceph_pagelist_cursor *c); + +extern int ceph_pagelist_truncate(struct ceph_pagelist *pl, + struct ceph_pagelist_cursor *c); + +static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v) +{ + __le64 ev = cpu_to_le64(v); + return ceph_pagelist_append(pl, &ev, sizeof(ev)); +} +static inline int ceph_pagelist_encode_32(struct ceph_pagelist *pl, u32 v) +{ + __le32 ev = cpu_to_le32(v); + return ceph_pagelist_append(pl, &ev, sizeof(ev)); +} +static inline int ceph_pagelist_encode_16(struct ceph_pagelist *pl, u16 v) +{ + __le16 ev = cpu_to_le16(v); + return ceph_pagelist_append(pl, &ev, sizeof(ev)); +} +static inline int ceph_pagelist_encode_8(struct ceph_pagelist *pl, u8 v) +{ + return ceph_pagelist_append(pl, &v, 1); +} +static inline int ceph_pagelist_encode_string(struct ceph_pagelist *pl, + char *s, u32 len) +{ + int ret = ceph_pagelist_encode_32(pl, len); + if (ret) + return ret; + if (len) + return ceph_pagelist_append(pl, s, len); + return 0; +} + +#endif diff --git a/include/linux/ceph/rados.h b/include/linux/ceph/rados.h new file mode 100644 index 000000000..43a7a1573 --- /dev/null +++ b/include/linux/ceph/rados.h @@ -0,0 +1,551 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef CEPH_RADOS_H +#define CEPH_RADOS_H + +/* + * Data types for the Ceph distributed object storage layer RADOS + * (Reliable Autonomic Distributed Object Store). + */ + +#include <linux/ceph/msgr.h> + +/* + * fs id + */ +struct ceph_fsid { + unsigned char fsid[16]; +}; + +static inline int ceph_fsid_compare(const struct ceph_fsid *a, + const struct ceph_fsid *b) +{ + return memcmp(a, b, sizeof(*a)); +} + +/* + * ino, object, etc. + */ +typedef __le64 ceph_snapid_t; +#define CEPH_SNAPDIR ((__u64)(-1)) /* reserved for hidden .snap dir */ +#define CEPH_NOSNAP ((__u64)(-2)) /* "head", "live" revision */ +#define CEPH_MAXSNAP ((__u64)(-3)) /* largest valid snapid */ + +struct ceph_timespec { + __le32 tv_sec; + __le32 tv_nsec; +} __attribute__ ((packed)); + + +/* + * object layout - how objects are mapped into PGs + */ +#define CEPH_OBJECT_LAYOUT_HASH 1 +#define CEPH_OBJECT_LAYOUT_LINEAR 2 +#define CEPH_OBJECT_LAYOUT_HASHINO 3 + +/* + * pg layout -- how PGs are mapped onto (sets of) OSDs + */ +#define CEPH_PG_LAYOUT_CRUSH 0 +#define CEPH_PG_LAYOUT_HASH 1 +#define CEPH_PG_LAYOUT_LINEAR 2 +#define CEPH_PG_LAYOUT_HYBRID 3 + +#define CEPH_PG_MAX_SIZE 32 /* max # osds in a single pg */ + +/* + * placement group. + * we encode this into one __le64. + */ +struct ceph_pg_v1 { + __le16 preferred; /* preferred primary osd */ + __le16 ps; /* placement seed */ + __le32 pool; /* object pool */ +} __attribute__ ((packed)); + +/* + * pg_pool is a set of pgs storing a pool of objects + * + * pg_num -- base number of pseudorandomly placed pgs + * + * pgp_num -- effective number when calculating pg placement. this + * is used for pg_num increases. new pgs result in data being "split" + * into new pgs. for this to proceed smoothly, new pgs are intiially + * colocated with their parents; that is, pgp_num doesn't increase + * until the new pgs have successfully split. only _then_ are the new + * pgs placed independently. + * + * lpg_num -- localized pg count (per device). replicas are randomly + * selected. + * + * lpgp_num -- as above. + */ +#define CEPH_NOPOOL ((__u64) (-1)) /* pool id not defined */ + +#define CEPH_POOL_TYPE_REP 1 +#define CEPH_POOL_TYPE_RAID4 2 /* never implemented */ +#define CEPH_POOL_TYPE_EC 3 + +/* + * stable_mod func is used to control number of placement groups. + * similar to straight-up modulo, but produces a stable mapping as b + * increases over time. b is the number of bins, and bmask is the + * containing power of 2 minus 1. + * + * b <= bmask and bmask=(2**n)-1 + * e.g., b=12 -> bmask=15, b=123 -> bmask=127 + */ +static inline int ceph_stable_mod(int x, int b, int bmask) +{ + if ((x & bmask) < b) + return x & bmask; + else + return x & (bmask >> 1); +} + +/* + * object layout - how a given object should be stored. + */ +struct ceph_object_layout { + struct ceph_pg_v1 ol_pgid; /* raw pg, with _full_ ps precision. */ + __le32 ol_stripe_unit; /* for per-object parity, if any */ +} __attribute__ ((packed)); + +/* + * compound epoch+version, used by storage layer to serialize mutations + */ +struct ceph_eversion { + __le64 version; + __le32 epoch; +} __attribute__ ((packed)); + +/* + * osd map bits + */ + +/* status bits */ +#define CEPH_OSD_EXISTS (1<<0) +#define CEPH_OSD_UP (1<<1) +#define CEPH_OSD_AUTOOUT (1<<2) /* osd was automatically marked out */ +#define CEPH_OSD_NEW (1<<3) /* osd is new, never marked in */ + +extern const char *ceph_osd_state_name(int s); + +/* osd weights. fixed point value: 0x10000 == 1.0 ("in"), 0 == "out" */ +#define CEPH_OSD_IN 0x10000 +#define CEPH_OSD_OUT 0 + +/* osd primary-affinity. fixed point value: 0x10000 == baseline */ +#define CEPH_OSD_MAX_PRIMARY_AFFINITY 0x10000 +#define CEPH_OSD_DEFAULT_PRIMARY_AFFINITY 0x10000 + + +/* + * osd map flag bits + */ +#define CEPH_OSDMAP_NEARFULL (1<<0) /* sync writes (near ENOSPC), + not set since ~luminous */ +#define CEPH_OSDMAP_FULL (1<<1) /* no data writes (ENOSPC), + not set since ~luminous */ +#define CEPH_OSDMAP_PAUSERD (1<<2) /* pause all reads */ +#define CEPH_OSDMAP_PAUSEWR (1<<3) /* pause all writes */ +#define CEPH_OSDMAP_PAUSEREC (1<<4) /* pause recovery */ +#define CEPH_OSDMAP_NOUP (1<<5) /* block osd boot */ +#define CEPH_OSDMAP_NODOWN (1<<6) /* block osd mark-down/failure */ +#define CEPH_OSDMAP_NOOUT (1<<7) /* block osd auto mark-out */ +#define CEPH_OSDMAP_NOIN (1<<8) /* block osd auto mark-in */ +#define CEPH_OSDMAP_NOBACKFILL (1<<9) /* block osd backfill */ +#define CEPH_OSDMAP_NORECOVER (1<<10) /* block osd recovery and backfill */ +#define CEPH_OSDMAP_NOSCRUB (1<<11) /* block periodic scrub */ +#define CEPH_OSDMAP_NODEEP_SCRUB (1<<12) /* block periodic deep-scrub */ +#define CEPH_OSDMAP_NOTIERAGENT (1<<13) /* disable tiering agent */ +#define CEPH_OSDMAP_NOREBALANCE (1<<14) /* block osd backfill unless pg is degraded */ +#define CEPH_OSDMAP_SORTBITWISE (1<<15) /* use bitwise hobject_t sort */ +#define CEPH_OSDMAP_REQUIRE_JEWEL (1<<16) /* require jewel for booting osds */ +#define CEPH_OSDMAP_REQUIRE_KRAKEN (1<<17) /* require kraken for booting osds */ +#define CEPH_OSDMAP_REQUIRE_LUMINOUS (1<<18) /* require l for booting osds */ +#define CEPH_OSDMAP_RECOVERY_DELETES (1<<19) /* deletes performed during recovery instead of peering */ + +/* + * The error code to return when an OSD can't handle a write + * because it is too large. + */ +#define OSD_WRITETOOBIG EMSGSIZE + +/* + * osd ops + * + * WARNING: do not use these op codes directly. Use the helpers + * defined below instead. In certain cases, op code behavior was + * redefined, resulting in special-cases in the helpers. + */ +#define CEPH_OSD_OP_MODE 0xf000 +#define CEPH_OSD_OP_MODE_RD 0x1000 +#define CEPH_OSD_OP_MODE_WR 0x2000 +#define CEPH_OSD_OP_MODE_RMW 0x3000 +#define CEPH_OSD_OP_MODE_SUB 0x4000 +#define CEPH_OSD_OP_MODE_CACHE 0x8000 + +#define CEPH_OSD_OP_TYPE 0x0f00 +#define CEPH_OSD_OP_TYPE_LOCK 0x0100 +#define CEPH_OSD_OP_TYPE_DATA 0x0200 +#define CEPH_OSD_OP_TYPE_ATTR 0x0300 +#define CEPH_OSD_OP_TYPE_EXEC 0x0400 +#define CEPH_OSD_OP_TYPE_PG 0x0500 +#define CEPH_OSD_OP_TYPE_MULTI 0x0600 /* multiobject */ + +#define __CEPH_OSD_OP1(mode, nr) \ + (CEPH_OSD_OP_MODE_##mode | (nr)) + +#define __CEPH_OSD_OP(mode, type, nr) \ + (CEPH_OSD_OP_MODE_##mode | CEPH_OSD_OP_TYPE_##type | (nr)) + +#define __CEPH_FORALL_OSD_OPS(f) \ + /** data **/ \ + /* read */ \ + f(READ, __CEPH_OSD_OP(RD, DATA, 1), "read") \ + f(STAT, __CEPH_OSD_OP(RD, DATA, 2), "stat") \ + f(MAPEXT, __CEPH_OSD_OP(RD, DATA, 3), "mapext") \ + \ + /* fancy read */ \ + f(MASKTRUNC, __CEPH_OSD_OP(RD, DATA, 4), "masktrunc") \ + f(SPARSE_READ, __CEPH_OSD_OP(RD, DATA, 5), "sparse-read") \ + \ + f(NOTIFY, __CEPH_OSD_OP(RD, DATA, 6), "notify") \ + f(NOTIFY_ACK, __CEPH_OSD_OP(RD, DATA, 7), "notify-ack") \ + \ + /* versioning */ \ + f(ASSERT_VER, __CEPH_OSD_OP(RD, DATA, 8), "assert-version") \ + \ + f(LIST_WATCHERS, __CEPH_OSD_OP(RD, DATA, 9), "list-watchers") \ + \ + f(LIST_SNAPS, __CEPH_OSD_OP(RD, DATA, 10), "list-snaps") \ + \ + /* sync */ \ + f(SYNC_READ, __CEPH_OSD_OP(RD, DATA, 11), "sync_read") \ + \ + /* write */ \ + f(WRITE, __CEPH_OSD_OP(WR, DATA, 1), "write") \ + f(WRITEFULL, __CEPH_OSD_OP(WR, DATA, 2), "writefull") \ + f(TRUNCATE, __CEPH_OSD_OP(WR, DATA, 3), "truncate") \ + f(ZERO, __CEPH_OSD_OP(WR, DATA, 4), "zero") \ + f(DELETE, __CEPH_OSD_OP(WR, DATA, 5), "delete") \ + \ + /* fancy write */ \ + f(APPEND, __CEPH_OSD_OP(WR, DATA, 6), "append") \ + f(SETTRUNC, __CEPH_OSD_OP(WR, DATA, 8), "settrunc") \ + f(TRIMTRUNC, __CEPH_OSD_OP(WR, DATA, 9), "trimtrunc") \ + \ + f(TMAPUP, __CEPH_OSD_OP(RMW, DATA, 10), "tmapup") \ + f(TMAPPUT, __CEPH_OSD_OP(WR, DATA, 11), "tmapput") \ + f(TMAPGET, __CEPH_OSD_OP(RD, DATA, 12), "tmapget") \ + \ + f(CREATE, __CEPH_OSD_OP(WR, DATA, 13), "create") \ + f(ROLLBACK, __CEPH_OSD_OP(WR, DATA, 14), "rollback") \ + \ + f(WATCH, __CEPH_OSD_OP(WR, DATA, 15), "watch") \ + \ + /* omap */ \ + f(OMAPGETKEYS, __CEPH_OSD_OP(RD, DATA, 17), "omap-get-keys") \ + f(OMAPGETVALS, __CEPH_OSD_OP(RD, DATA, 18), "omap-get-vals") \ + f(OMAPGETHEADER, __CEPH_OSD_OP(RD, DATA, 19), "omap-get-header") \ + f(OMAPGETVALSBYKEYS, __CEPH_OSD_OP(RD, DATA, 20), "omap-get-vals-by-keys") \ + f(OMAPSETVALS, __CEPH_OSD_OP(WR, DATA, 21), "omap-set-vals") \ + f(OMAPSETHEADER, __CEPH_OSD_OP(WR, DATA, 22), "omap-set-header") \ + f(OMAPCLEAR, __CEPH_OSD_OP(WR, DATA, 23), "omap-clear") \ + f(OMAPRMKEYS, __CEPH_OSD_OP(WR, DATA, 24), "omap-rm-keys") \ + f(OMAP_CMP, __CEPH_OSD_OP(RD, DATA, 25), "omap-cmp") \ + \ + /* tiering */ \ + f(COPY_FROM, __CEPH_OSD_OP(WR, DATA, 26), "copy-from") \ + f(COPY_FROM2, __CEPH_OSD_OP(WR, DATA, 45), "copy-from2") \ + f(COPY_GET_CLASSIC, __CEPH_OSD_OP(RD, DATA, 27), "copy-get-classic") \ + f(UNDIRTY, __CEPH_OSD_OP(WR, DATA, 28), "undirty") \ + f(ISDIRTY, __CEPH_OSD_OP(RD, DATA, 29), "isdirty") \ + f(COPY_GET, __CEPH_OSD_OP(RD, DATA, 30), "copy-get") \ + f(CACHE_FLUSH, __CEPH_OSD_OP(CACHE, DATA, 31), "cache-flush") \ + f(CACHE_EVICT, __CEPH_OSD_OP(CACHE, DATA, 32), "cache-evict") \ + f(CACHE_TRY_FLUSH, __CEPH_OSD_OP(CACHE, DATA, 33), "cache-try-flush") \ + \ + /* convert tmap to omap */ \ + f(TMAP2OMAP, __CEPH_OSD_OP(RMW, DATA, 34), "tmap2omap") \ + \ + /* hints */ \ + f(SETALLOCHINT, __CEPH_OSD_OP(WR, DATA, 35), "set-alloc-hint") \ + \ + /** multi **/ \ + f(CLONERANGE, __CEPH_OSD_OP(WR, MULTI, 1), "clonerange") \ + f(ASSERT_SRC_VERSION, __CEPH_OSD_OP(RD, MULTI, 2), "assert-src-version") \ + f(SRC_CMPXATTR, __CEPH_OSD_OP(RD, MULTI, 3), "src-cmpxattr") \ + \ + /** attrs **/ \ + /* read */ \ + f(GETXATTR, __CEPH_OSD_OP(RD, ATTR, 1), "getxattr") \ + f(GETXATTRS, __CEPH_OSD_OP(RD, ATTR, 2), "getxattrs") \ + f(CMPXATTR, __CEPH_OSD_OP(RD, ATTR, 3), "cmpxattr") \ + \ + /* write */ \ + f(SETXATTR, __CEPH_OSD_OP(WR, ATTR, 1), "setxattr") \ + f(SETXATTRS, __CEPH_OSD_OP(WR, ATTR, 2), "setxattrs") \ + f(RESETXATTRS, __CEPH_OSD_OP(WR, ATTR, 3), "resetxattrs") \ + f(RMXATTR, __CEPH_OSD_OP(WR, ATTR, 4), "rmxattr") \ + \ + /** subop **/ \ + f(PULL, __CEPH_OSD_OP1(SUB, 1), "pull") \ + f(PUSH, __CEPH_OSD_OP1(SUB, 2), "push") \ + f(BALANCEREADS, __CEPH_OSD_OP1(SUB, 3), "balance-reads") \ + f(UNBALANCEREADS, __CEPH_OSD_OP1(SUB, 4), "unbalance-reads") \ + f(SCRUB, __CEPH_OSD_OP1(SUB, 5), "scrub") \ + f(SCRUB_RESERVE, __CEPH_OSD_OP1(SUB, 6), "scrub-reserve") \ + f(SCRUB_UNRESERVE, __CEPH_OSD_OP1(SUB, 7), "scrub-unreserve") \ + f(SCRUB_STOP, __CEPH_OSD_OP1(SUB, 8), "scrub-stop") \ + f(SCRUB_MAP, __CEPH_OSD_OP1(SUB, 9), "scrub-map") \ + \ + /** lock **/ \ + f(WRLOCK, __CEPH_OSD_OP(WR, LOCK, 1), "wrlock") \ + f(WRUNLOCK, __CEPH_OSD_OP(WR, LOCK, 2), "wrunlock") \ + f(RDLOCK, __CEPH_OSD_OP(WR, LOCK, 3), "rdlock") \ + f(RDUNLOCK, __CEPH_OSD_OP(WR, LOCK, 4), "rdunlock") \ + f(UPLOCK, __CEPH_OSD_OP(WR, LOCK, 5), "uplock") \ + f(DNLOCK, __CEPH_OSD_OP(WR, LOCK, 6), "dnlock") \ + \ + /** exec **/ \ + /* note: the RD bit here is wrong; see special-case below in helper */ \ + f(CALL, __CEPH_OSD_OP(RD, EXEC, 1), "call") \ + \ + /** pg **/ \ + f(PGLS, __CEPH_OSD_OP(RD, PG, 1), "pgls") \ + f(PGLS_FILTER, __CEPH_OSD_OP(RD, PG, 2), "pgls-filter") \ + f(PG_HITSET_LS, __CEPH_OSD_OP(RD, PG, 3), "pg-hitset-ls") \ + f(PG_HITSET_GET, __CEPH_OSD_OP(RD, PG, 4), "pg-hitset-get") + +enum { +#define GENERATE_ENUM_ENTRY(op, opcode, str) CEPH_OSD_OP_##op = (opcode), +__CEPH_FORALL_OSD_OPS(GENERATE_ENUM_ENTRY) +#undef GENERATE_ENUM_ENTRY +}; + +static inline int ceph_osd_op_type_lock(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_LOCK; +} +static inline int ceph_osd_op_type_data(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_DATA; +} +static inline int ceph_osd_op_type_attr(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_ATTR; +} +static inline int ceph_osd_op_type_exec(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_EXEC; +} +static inline int ceph_osd_op_type_pg(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_PG; +} +static inline int ceph_osd_op_type_multi(int op) +{ + return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_MULTI; +} + +static inline int ceph_osd_op_mode_subop(int op) +{ + return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_SUB; +} +static inline int ceph_osd_op_mode_read(int op) +{ + return (op & CEPH_OSD_OP_MODE_RD) && + op != CEPH_OSD_OP_CALL; +} +static inline int ceph_osd_op_mode_modify(int op) +{ + return op & CEPH_OSD_OP_MODE_WR; +} + +/* + * note that the following tmap stuff is also defined in the ceph librados.h + * any modification here needs to be updated there + */ +#define CEPH_OSD_TMAP_HDR 'h' +#define CEPH_OSD_TMAP_SET 's' +#define CEPH_OSD_TMAP_CREATE 'c' /* create key */ +#define CEPH_OSD_TMAP_RM 'r' +#define CEPH_OSD_TMAP_RMSLOPPY 'R' + +extern const char *ceph_osd_op_name(int op); + +/* + * osd op flags + * + * An op may be READ, WRITE, or READ|WRITE. + */ +enum { + CEPH_OSD_FLAG_ACK = 0x0001, /* want (or is) "ack" ack */ + CEPH_OSD_FLAG_ONNVRAM = 0x0002, /* want (or is) "onnvram" ack */ + CEPH_OSD_FLAG_ONDISK = 0x0004, /* want (or is) "ondisk" ack */ + CEPH_OSD_FLAG_RETRY = 0x0008, /* resend attempt */ + CEPH_OSD_FLAG_READ = 0x0010, /* op may read */ + CEPH_OSD_FLAG_WRITE = 0x0020, /* op may write */ + CEPH_OSD_FLAG_ORDERSNAP = 0x0040, /* EOLDSNAP if snapc is out of order */ + CEPH_OSD_FLAG_PEERSTAT_OLD = 0x0080, /* DEPRECATED msg includes osd_peer_stat */ + CEPH_OSD_FLAG_BALANCE_READS = 0x0100, + CEPH_OSD_FLAG_PARALLELEXEC = 0x0200, /* execute op in parallel */ + CEPH_OSD_FLAG_PGOP = 0x0400, /* pg op, no object */ + CEPH_OSD_FLAG_EXEC = 0x0800, /* op may exec */ + CEPH_OSD_FLAG_EXEC_PUBLIC = 0x1000, /* DEPRECATED op may exec (public) */ + CEPH_OSD_FLAG_LOCALIZE_READS = 0x2000, /* read from nearby replica, if any */ + CEPH_OSD_FLAG_RWORDERED = 0x4000, /* order wrt concurrent reads */ + CEPH_OSD_FLAG_IGNORE_CACHE = 0x8000, /* ignore cache logic */ + CEPH_OSD_FLAG_SKIPRWLOCKS = 0x10000, /* skip rw locks */ + CEPH_OSD_FLAG_IGNORE_OVERLAY = 0x20000, /* ignore pool overlay */ + CEPH_OSD_FLAG_FLUSH = 0x40000, /* this is part of flush */ + CEPH_OSD_FLAG_MAP_SNAP_CLONE = 0x80000, /* map snap direct to clone id */ + CEPH_OSD_FLAG_ENFORCE_SNAPC = 0x100000, /* use snapc provided even if + pool uses pool snaps */ + CEPH_OSD_FLAG_REDIRECTED = 0x200000, /* op has been redirected */ + CEPH_OSD_FLAG_KNOWN_REDIR = 0x400000, /* redirect bit is authoritative */ + CEPH_OSD_FLAG_FULL_TRY = 0x800000, /* try op despite full flag */ + CEPH_OSD_FLAG_FULL_FORCE = 0x1000000, /* force op despite full flag */ +}; + +enum { + CEPH_OSD_OP_FLAG_EXCL = 1, /* EXCL object create */ + CEPH_OSD_OP_FLAG_FAILOK = 2, /* continue despite failure */ + CEPH_OSD_OP_FLAG_FADVISE_RANDOM = 0x4, /* the op is random */ + CEPH_OSD_OP_FLAG_FADVISE_SEQUENTIAL = 0x8, /* the op is sequential */ + CEPH_OSD_OP_FLAG_FADVISE_WILLNEED = 0x10,/* data will be accessed in + the near future */ + CEPH_OSD_OP_FLAG_FADVISE_DONTNEED = 0x20,/* data will not be accessed + in the near future */ + CEPH_OSD_OP_FLAG_FADVISE_NOCACHE = 0x40,/* data will be accessed only + once by this client */ +}; + +#define EOLDSNAPC ERESTART /* ORDERSNAP flag set; writer has old snapc*/ +#define EBLOCKLISTED ESHUTDOWN /* blocklisted */ + +/* xattr comparison */ +enum { + CEPH_OSD_CMPXATTR_OP_NOP = 0, + CEPH_OSD_CMPXATTR_OP_EQ = 1, + CEPH_OSD_CMPXATTR_OP_NE = 2, + CEPH_OSD_CMPXATTR_OP_GT = 3, + CEPH_OSD_CMPXATTR_OP_GTE = 4, + CEPH_OSD_CMPXATTR_OP_LT = 5, + CEPH_OSD_CMPXATTR_OP_LTE = 6 +}; + +enum { + CEPH_OSD_CMPXATTR_MODE_STRING = 1, + CEPH_OSD_CMPXATTR_MODE_U64 = 2 +}; + +enum { + CEPH_OSD_COPY_FROM_FLAG_FLUSH = 1, /* part of a flush operation */ + CEPH_OSD_COPY_FROM_FLAG_IGNORE_OVERLAY = 2, /* ignore pool overlay */ + CEPH_OSD_COPY_FROM_FLAG_IGNORE_CACHE = 4, /* ignore osd cache logic */ + CEPH_OSD_COPY_FROM_FLAG_MAP_SNAP_CLONE = 8, /* map snap direct to + * cloneid */ + CEPH_OSD_COPY_FROM_FLAG_RWORDERED = 16, /* order with write */ + CEPH_OSD_COPY_FROM_FLAG_TRUNCATE_SEQ = 32, /* send truncate_{seq,size} */ +}; + +enum { + CEPH_OSD_WATCH_OP_UNWATCH = 0, + CEPH_OSD_WATCH_OP_LEGACY_WATCH = 1, + /* note: use only ODD ids to prevent pre-giant code from + interpreting the op as UNWATCH */ + CEPH_OSD_WATCH_OP_WATCH = 3, + CEPH_OSD_WATCH_OP_RECONNECT = 5, + CEPH_OSD_WATCH_OP_PING = 7, +}; + +const char *ceph_osd_watch_op_name(int o); + +enum { + CEPH_OSD_ALLOC_HINT_FLAG_SEQUENTIAL_WRITE = 1, + CEPH_OSD_ALLOC_HINT_FLAG_RANDOM_WRITE = 2, + CEPH_OSD_ALLOC_HINT_FLAG_SEQUENTIAL_READ = 4, + CEPH_OSD_ALLOC_HINT_FLAG_RANDOM_READ = 8, + CEPH_OSD_ALLOC_HINT_FLAG_APPEND_ONLY = 16, + CEPH_OSD_ALLOC_HINT_FLAG_IMMUTABLE = 32, + CEPH_OSD_ALLOC_HINT_FLAG_SHORTLIVED = 64, + CEPH_OSD_ALLOC_HINT_FLAG_LONGLIVED = 128, + CEPH_OSD_ALLOC_HINT_FLAG_COMPRESSIBLE = 256, + CEPH_OSD_ALLOC_HINT_FLAG_INCOMPRESSIBLE = 512, +}; + +enum { + CEPH_OSD_BACKOFF_OP_BLOCK = 1, + CEPH_OSD_BACKOFF_OP_ACK_BLOCK = 2, + CEPH_OSD_BACKOFF_OP_UNBLOCK = 3, +}; + +/* + * an individual object operation. each may be accompanied by some data + * payload + */ +struct ceph_osd_op { + __le16 op; /* CEPH_OSD_OP_* */ + __le32 flags; /* CEPH_OSD_OP_FLAG_* */ + union { + struct { + __le64 offset, length; + __le64 truncate_size; + __le32 truncate_seq; + } __attribute__ ((packed)) extent; + struct { + __le32 name_len; + __le32 value_len; + __u8 cmp_op; /* CEPH_OSD_CMPXATTR_OP_* */ + __u8 cmp_mode; /* CEPH_OSD_CMPXATTR_MODE_* */ + } __attribute__ ((packed)) xattr; + struct { + __u8 class_len; + __u8 method_len; + __u8 argc; + __le32 indata_len; + } __attribute__ ((packed)) cls; + struct { + __le64 cookie, count; + } __attribute__ ((packed)) pgls; + struct { + __le64 snapid; + } __attribute__ ((packed)) snap; + struct { + __le64 cookie; + __le64 ver; /* no longer used */ + __u8 op; /* CEPH_OSD_WATCH_OP_* */ + __le32 gen; /* registration generation */ + } __attribute__ ((packed)) watch; + struct { + __le64 cookie; + } __attribute__ ((packed)) notify; + struct { + __le64 offset, length; + __le64 src_offset; + } __attribute__ ((packed)) clonerange; + struct { + __le64 expected_object_size; + __le64 expected_write_size; + __le32 flags; /* CEPH_OSD_OP_ALLOC_HINT_FLAG_* */ + } __attribute__ ((packed)) alloc_hint; + struct { + __le64 snapid; + __le64 src_version; + __u8 flags; /* CEPH_OSD_COPY_FROM_FLAG_* */ + /* + * CEPH_OSD_OP_FLAG_FADVISE_*: fadvise flags + * for src object, flags for dest object are in + * ceph_osd_op::flags. + */ + __le32 src_fadvise_flags; + } __attribute__ ((packed)) copy_from; + }; + __le32 payload_len; +} __attribute__ ((packed)); + + +#endif diff --git a/include/linux/ceph/string_table.h b/include/linux/ceph/string_table.h new file mode 100644 index 000000000..a4a9962d1 --- /dev/null +++ b/include/linux/ceph/string_table.h @@ -0,0 +1,63 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_STRING_TABLE_H +#define _FS_CEPH_STRING_TABLE_H + +#include <linux/types.h> +#include <linux/kref.h> +#include <linux/rbtree.h> +#include <linux/rcupdate.h> + +struct ceph_string { + struct kref kref; + union { + struct rb_node node; + struct rcu_head rcu; + }; + size_t len; + char str[]; +}; + +extern void ceph_release_string(struct kref *ref); +extern struct ceph_string *ceph_find_or_create_string(const char *str, + size_t len); +extern bool ceph_strings_empty(void); + +static inline struct ceph_string *ceph_get_string(struct ceph_string *str) +{ + kref_get(&str->kref); + return str; +} + +static inline void ceph_put_string(struct ceph_string *str) +{ + if (!str) + return; + kref_put(&str->kref, ceph_release_string); +} + +static inline int ceph_compare_string(struct ceph_string *cs, + const char* str, size_t len) +{ + size_t cs_len = cs ? cs->len : 0; + if (cs_len != len) + return cs_len - len; + if (len == 0) + return 0; + return strncmp(cs->str, str, len); +} + +#define ceph_try_get_string(x) \ +({ \ + struct ceph_string *___str; \ + rcu_read_lock(); \ + for (;;) { \ + ___str = rcu_dereference(x); \ + if (!___str || \ + kref_get_unless_zero(&___str->kref)) \ + break; \ + } \ + rcu_read_unlock(); \ + (___str); \ +}) + +#endif diff --git a/include/linux/ceph/striper.h b/include/linux/ceph/striper.h new file mode 100644 index 000000000..3486636c0 --- /dev/null +++ b/include/linux/ceph/striper.h @@ -0,0 +1,71 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_CEPH_STRIPER_H +#define _LINUX_CEPH_STRIPER_H + +#include <linux/list.h> +#include <linux/types.h> + +struct ceph_file_layout; + +void ceph_calc_file_object_mapping(struct ceph_file_layout *l, + u64 off, u64 len, + u64 *objno, u64 *objoff, u32 *xlen); + +struct ceph_object_extent { + struct list_head oe_item; + u64 oe_objno; + u64 oe_off; + u64 oe_len; +}; + +static inline void ceph_object_extent_init(struct ceph_object_extent *ex) +{ + INIT_LIST_HEAD(&ex->oe_item); +} + +/* + * Called for each mapped stripe unit. + * + * @bytes: number of bytes mapped, i.e. the minimum of the full length + * requested (file extent length) or the remainder of the stripe + * unit within an object + */ +typedef void (*ceph_object_extent_fn_t)(struct ceph_object_extent *ex, + u32 bytes, void *arg); + +int ceph_file_to_extents(struct ceph_file_layout *l, u64 off, u64 len, + struct list_head *object_extents, + struct ceph_object_extent *alloc_fn(void *arg), + void *alloc_arg, + ceph_object_extent_fn_t action_fn, + void *action_arg); +int ceph_iterate_extents(struct ceph_file_layout *l, u64 off, u64 len, + struct list_head *object_extents, + ceph_object_extent_fn_t action_fn, + void *action_arg); + +struct ceph_file_extent { + u64 fe_off; + u64 fe_len; +}; + +static inline u64 ceph_file_extents_bytes(struct ceph_file_extent *file_extents, + u32 num_file_extents) +{ + u64 bytes = 0; + u32 i; + + for (i = 0; i < num_file_extents; i++) + bytes += file_extents[i].fe_len; + + return bytes; +} + +int ceph_extent_to_file(struct ceph_file_layout *l, + u64 objno, u64 objoff, u64 objlen, + struct ceph_file_extent **file_extents, + u32 *num_file_extents); + +u64 ceph_get_num_objects(struct ceph_file_layout *l, u64 size); + +#endif diff --git a/include/linux/ceph/types.h b/include/linux/ceph/types.h new file mode 100644 index 000000000..bd3d53290 --- /dev/null +++ b/include/linux/ceph/types.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _FS_CEPH_TYPES_H +#define _FS_CEPH_TYPES_H + +/* needed before including ceph_fs.h */ +#include <linux/in.h> +#include <linux/types.h> +#include <linux/fcntl.h> +#include <linux/string.h> + +#include <linux/ceph/ceph_fs.h> +#include <linux/ceph/ceph_frag.h> +#include <linux/ceph/ceph_hash.h> + +/* + * Identify inodes by both their ino AND snapshot id (a u64). + */ +struct ceph_vino { + u64 ino; + u64 snap; +}; + + +/* context for the caps reservation mechanism */ +struct ceph_cap_reservation { + int count; + int used; +}; + + +#endif |