diff options
Diffstat (limited to 'include/linux/ceph/rados.h')
| -rw-r--r-- | include/linux/ceph/rados.h | 343 |
1 files changed, 239 insertions, 104 deletions
diff --git a/include/linux/ceph/rados.h b/include/linux/ceph/rados.h index 68c96a508ac2..73c3efbec36c 100644 --- a/include/linux/ceph/rados.h +++ b/include/linux/ceph/rados.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef CEPH_RADOS_H #define CEPH_RADOS_H @@ -50,7 +51,7 @@ struct ceph_timespec { #define CEPH_PG_LAYOUT_LINEAR 2 #define CEPH_PG_LAYOUT_HYBRID 3 -#define CEPH_PG_MAX_SIZE 16 /* max # osds in a single pg */ +#define CEPH_PG_MAX_SIZE 32 /* max # osds in a single pg */ /* * placement group. @@ -81,8 +82,9 @@ struct ceph_pg_v1 { */ #define CEPH_NOPOOL ((__u64) (-1)) /* pool id not defined */ -#define CEPH_PG_TYPE_REP 1 -#define CEPH_PG_TYPE_RAID4 2 +#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. @@ -113,8 +115,8 @@ struct ceph_object_layout { * compound epoch+version, used by storage layer to serialize mutations */ struct ceph_eversion { - __le32 epoch; __le64 version; + __le32 epoch; } __attribute__ ((packed)); /* @@ -133,12 +135,18 @@ extern const char *ceph_osd_state_name(int s); #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) */ -#define CEPH_OSDMAP_FULL (1<<1) /* no data writes (ENOSPC) */ +#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 */ @@ -148,6 +156,15 @@ extern const char *ceph_osd_state_name(int s); #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 @@ -167,6 +184,7 @@ extern const char *ceph_osd_state_name(int s); #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 @@ -176,100 +194,135 @@ extern const char *ceph_osd_state_name(int s); #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 { - /** data **/ - /* read */ - CEPH_OSD_OP_READ = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 1, - CEPH_OSD_OP_STAT = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 2, - CEPH_OSD_OP_MAPEXT = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 3, - - /* fancy read */ - CEPH_OSD_OP_MASKTRUNC = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 4, - CEPH_OSD_OP_SPARSE_READ = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 5, - - CEPH_OSD_OP_NOTIFY = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 6, - CEPH_OSD_OP_NOTIFY_ACK = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 7, - - /* versioning */ - CEPH_OSD_OP_ASSERT_VER = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 8, - - /* write */ - CEPH_OSD_OP_WRITE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 1, - CEPH_OSD_OP_WRITEFULL = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 2, - CEPH_OSD_OP_TRUNCATE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 3, - CEPH_OSD_OP_ZERO = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 4, - CEPH_OSD_OP_DELETE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 5, - - /* fancy write */ - CEPH_OSD_OP_APPEND = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 6, - CEPH_OSD_OP_STARTSYNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 7, - CEPH_OSD_OP_SETTRUNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 8, - CEPH_OSD_OP_TRIMTRUNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 9, - - CEPH_OSD_OP_TMAPUP = CEPH_OSD_OP_MODE_RMW | CEPH_OSD_OP_TYPE_DATA | 10, - CEPH_OSD_OP_TMAPPUT = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 11, - CEPH_OSD_OP_TMAPGET = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 12, - - CEPH_OSD_OP_CREATE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 13, - CEPH_OSD_OP_ROLLBACK= CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 14, - - CEPH_OSD_OP_WATCH = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 15, - - /* omap */ - CEPH_OSD_OP_OMAPGETKEYS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 17, - CEPH_OSD_OP_OMAPGETVALS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 18, - CEPH_OSD_OP_OMAPGETHEADER = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 19, - CEPH_OSD_OP_OMAPGETVALSBYKEYS = - CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 20, - CEPH_OSD_OP_OMAPSETVALS = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 21, - CEPH_OSD_OP_OMAPSETHEADER = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 22, - CEPH_OSD_OP_OMAPCLEAR = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 23, - CEPH_OSD_OP_OMAPRMKEYS = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 24, - CEPH_OSD_OP_OMAP_CMP = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 25, - - /** multi **/ - CEPH_OSD_OP_CLONERANGE = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_MULTI | 1, - CEPH_OSD_OP_ASSERT_SRC_VERSION = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_MULTI | 2, - CEPH_OSD_OP_SRC_CMPXATTR = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_MULTI | 3, - - /** attrs **/ - /* read */ - CEPH_OSD_OP_GETXATTR = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 1, - CEPH_OSD_OP_GETXATTRS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 2, - CEPH_OSD_OP_CMPXATTR = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 3, - - /* write */ - CEPH_OSD_OP_SETXATTR = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 1, - CEPH_OSD_OP_SETXATTRS = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 2, - CEPH_OSD_OP_RESETXATTRS = CEPH_OSD_OP_MODE_WR|CEPH_OSD_OP_TYPE_ATTR | 3, - CEPH_OSD_OP_RMXATTR = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 4, - - /** subop **/ - CEPH_OSD_OP_PULL = CEPH_OSD_OP_MODE_SUB | 1, - CEPH_OSD_OP_PUSH = CEPH_OSD_OP_MODE_SUB | 2, - CEPH_OSD_OP_BALANCEREADS = CEPH_OSD_OP_MODE_SUB | 3, - CEPH_OSD_OP_UNBALANCEREADS = CEPH_OSD_OP_MODE_SUB | 4, - CEPH_OSD_OP_SCRUB = CEPH_OSD_OP_MODE_SUB | 5, - CEPH_OSD_OP_SCRUB_RESERVE = CEPH_OSD_OP_MODE_SUB | 6, - CEPH_OSD_OP_SCRUB_UNRESERVE = CEPH_OSD_OP_MODE_SUB | 7, - CEPH_OSD_OP_SCRUB_STOP = CEPH_OSD_OP_MODE_SUB | 8, - CEPH_OSD_OP_SCRUB_MAP = CEPH_OSD_OP_MODE_SUB | 9, - - /** lock **/ - CEPH_OSD_OP_WRLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 1, - CEPH_OSD_OP_WRUNLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 2, - CEPH_OSD_OP_RDLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 3, - CEPH_OSD_OP_RDUNLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 4, - CEPH_OSD_OP_UPLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 5, - CEPH_OSD_OP_DNLOCK = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 6, - - /** exec **/ - /* note: the RD bit here is wrong; see special-case below in helper */ - CEPH_OSD_OP_CALL = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_EXEC | 1, - - /** pg **/ - CEPH_OSD_OP_PGLS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_PG | 1, - CEPH_OSD_OP_PGLS_FILTER = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_PG | 2, +#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) @@ -344,15 +397,34 @@ enum { 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 EBLACKLISTED ESHUTDOWN /* blacklisted */ +#define EBLOCKLISTED ESHUTDOWN /* blocklisted */ /* xattr comparison */ enum { @@ -370,7 +442,46 @@ enum { CEPH_OSD_CMPXATTR_MODE_U64 = 2 }; -#define RADOS_NOTIFY_VER 1 +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 @@ -378,7 +489,7 @@ enum { */ struct ceph_osd_op { __le16 op; /* CEPH_OSD_OP_* */ - __le32 flags; /* CEPH_OSD_FLAG_* */ + __le32 flags; /* CEPH_OSD_OP_FLAG_* */ union { struct { __le64 offset, length; @@ -405,13 +516,37 @@ struct ceph_osd_op { } __attribute__ ((packed)) snap; struct { __le64 cookie; - __le64 ver; - __u8 flag; /* 0 = unwatch, 1 = watch */ + __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 unused; + __le64 ver; + } __attribute__ ((packed)) assert_ver; + 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)); |