* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2011, 2016, Intel Corporation.
+ * Copyright (c) 2011, 2017, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
#define SEQ_DATA_PORTAL 31
#define SEQ_CONTROLLER_PORTAL 32
#define MGS_BULK_PORTAL 33
-
-/* Portal 63 is reserved for the Cray Inc DVS - nic@cray.com, roe@cray.com, n8851@cray.com */
-
-/* packet types */
-#define PTL_RPC_MSG_REQUEST 4711
-#define PTL_RPC_MSG_ERR 4712
-#define PTL_RPC_MSG_REPLY 4713
-
-/* DON'T use swabbed values of MAGIC as magic! */
-#define LUSTRE_MSG_MAGIC_V2 0x0BD00BD3
-#define LUSTRE_MSG_MAGIC_V2_SWABBED 0xD30BD00B
-
-#define LUSTRE_MSG_MAGIC LUSTRE_MSG_MAGIC_V2
-
-#define PTLRPC_MSG_VERSION 0x00000003
-#define LUSTRE_VERSION_MASK 0xffff0000
-#define LUSTRE_OBD_VERSION 0x00010000
-#define LUSTRE_MDS_VERSION 0x00020000
-#define LUSTRE_OST_VERSION 0x00030000
-#define LUSTRE_DLM_VERSION 0x00040000
-#define LUSTRE_LOG_VERSION 0x00050000
-#define LUSTRE_MGS_VERSION 0x00060000
+/* #define DVS_PORTAL 63 */
+/* reserved for Cray DVS - spitzcor@cray.com, roe@cray.com, n8851@cray.com */
/**
* Describes a range of sequence, lsr_start is included but lsr_end is
tgt->cookie = src->cookie;
}
-struct lustre_handle_array {
- unsigned int count;
- struct lustre_handle handles[0];
+/* lustre_msg struct magic. DON'T use swabbed values of MAGIC as magic! */
+enum lustre_msg_magic {
+ LUSTRE_MSG_MAGIC_V2 = 0x0BD00BD3,
+ LUSTRE_MSG_MAGIC_V2_SWABBED = 0xD30BD00B,
+ LUSTRE_MSG_MAGIC = LUSTRE_MSG_MAGIC_V2
};
/* flags for lm_flags */
-#define MSGHDR_AT_SUPPORT 0x1
-#define MSGHDR_CKSUM_INCOMPAT18 0x2
+enum lustre_msghdr {
+ MSGHDR_AT_SUPPORT = 0x1, /* adaptive timeouts, lm_cksum valid
+ * in early reply messages */
+ MSGHDR_CKSUM_INCOMPAT18 = 0x2, /* compat for 1.8, needs to be set well
+ * beyond 2.8.0 for compatibility */
+};
#define lustre_msg lustre_msg_v2
/* we depend on this structure to be 8-byte aligned */
/* this type is only endian-adjusted in lustre_unpack_msg() */
struct lustre_msg_v2 {
- __u32 lm_bufcount;
- __u32 lm_secflvr;
- __u32 lm_magic;
- __u32 lm_repsize;
- __u32 lm_cksum;
- __u32 lm_flags;
- __u32 lm_padding_2;
- __u32 lm_padding_3;
- __u32 lm_buflens[0];
-};
-
-/* without gss, ptlrpc_body is put at the first buffer. */
+ __u32 lm_bufcount; /* number of buffers in lm_buflens[] */
+ __u32 lm_secflvr; /* 0 = no crypto, or sptlrpc security flavour */
+ __u32 lm_magic; /* RPC version magic = LUSTRE_MSG_MAGIC_V2 */
+ __u32 lm_repsize; /* size of preallocated reply buffer */
+ __u32 lm_cksum; /* CRC32 of ptlrpc_body early reply messages */
+ __u32 lm_flags; /* enum lustre_msghdr MSGHDR_* flags */
+ __u32 lm_padding_2; /* unused */
+ __u32 lm_padding_3; /* unused */
+ __u32 lm_buflens[0]; /* length of additional buffers in bytes,
+ * padded to a multiple of 8 bytes. */
+ /*
+ * message buffers are packed after padded lm_buflens[] array,
+ * padded to a multiple of 8 bytes each to align contents.
+ */
+};
+
+/* ptlrpc_body packet pb_types */
+#define PTL_RPC_MSG_REQUEST 4711 /* normal RPC request message */
+#define PTL_RPC_MSG_ERR 4712 /* error reply if request unprocessed */
+#define PTL_RPC_MSG_REPLY 4713 /* normal RPC reply message */
+
+/* ptlrpc_body pb_version ((target_version << 16) | rpc_version) */
+enum lustre_msg_version {
+ PTLRPC_MSG_VERSION = 0x00000003,
+ LUSTRE_VERSION_MASK = 0xffff0000,
+ LUSTRE_OBD_VERSION = 0x00010000,
+ LUSTRE_MDS_VERSION = 0x00020000,
+ LUSTRE_OST_VERSION = 0x00030000,
+ LUSTRE_DLM_VERSION = 0x00040000,
+ LUSTRE_LOG_VERSION = 0x00050000,
+ LUSTRE_MGS_VERSION = 0x00060000,
+};
+
+/* pb_flags that apply to all request messages */
+/* #define MSG_LAST_REPLAY 0x0001 obsolete 2.0 => {REQ,LOCK}_REPLAY_DONE */
+#define MSG_RESENT 0x0002 /* was previously sent, no reply seen */
+#define MSG_REPLAY 0x0004 /* was processed, got reply, recovery */
+/* #define MSG_AT_SUPPORT 0x0008 obsolete since 1.5, AT always enabled */
+/* #define MSG_DELAY_REPLAY 0x0010 obsolete since 2.0 */
+/* #define MSG_VERSION_REPLAY 0x0020 obsolete since 1.8.2, VBR always on */
+#define MSG_REQ_REPLAY_DONE 0x0040 /* request replay over, locks next */
+#define MSG_LOCK_REPLAY_DONE 0x0080 /* lock replay over, client done */
+
+/* pb_op_flags for connect opcodes: MDS_CONNECT, OST_CONNECT, MGS_CONNECT */
+#define MSG_CONNECT_RECOVERING 0x00000001 /* target is in recovery */
+#define MSG_CONNECT_RECONNECT 0x00000002 /* tgt already has client import */
+#define MSG_CONNECT_REPLAYABLE 0x00000004 /* target supports RPC replay */
+/* #define MSG_CONNECT_PEER 0x00000008 obsolete since 1.2, removed in 1.5 */
+#define MSG_CONNECT_LIBCLIENT 0x00000010 /* obsolete since 2.3, removed 2.6 */
+#define MSG_CONNECT_INITIAL 0x00000020 /* first client connection attempt */
+/* #define MSG_CONNECT_ASYNC 0x00000040 obsolete since 1.5 */
+#define MSG_CONNECT_NEXT_VER 0x00000080 /* use next version of lustre_msg */
+#define MSG_CONNECT_TRANSNO 0x00000100 /* client sent transno in replay */
+
+/* number of previous object versions in pb_pre_versions[] */
#define PTLRPC_NUM_VERSIONS 4
+/* without gss, ptlrpc_body is put at the first buffer. */
struct ptlrpc_body_v3 {
struct lustre_handle pb_handle;
- __u32 pb_type;
- __u32 pb_version;
- __u32 pb_opc;
- __u32 pb_status;
- __u64 pb_last_xid; /* highest replied XID without lower unreplied XID */
- __u16 pb_tag; /* virtual slot idx for multiple modifying RPCs */
+ __u32 pb_type; /* request/reply/err type: PTL_RPC_MSG_* */
+ __u32 pb_version; /* LUSTRE_*_VERSION | PTLRPC_MSG_VERSION */
+ __u32 pb_opc; /* RPC opcodes: MDS_*, OST_*, LDLM_, ... */
+ __u32 pb_status; /* negative Linux x86 error number */
+ __u64 pb_last_xid; /* highest replied XID w/o lower unreplied XID*/
+ __u16 pb_tag; /* multiple modifying RPCs virtual slot index */
__u16 pb_padding0;
__u32 pb_padding1;
- __u64 pb_last_committed;
- __u64 pb_transno;
- __u32 pb_flags;
- __u32 pb_op_flags;
- __u32 pb_conn_cnt;
- __u32 pb_timeout; /* for req, the deadline, for rep, the service est */
- __u32 pb_service_time; /* for rep, actual service time */
- __u32 pb_limit;
- __u64 pb_slv;
- /* VBR: pre-versions */
+ __u64 pb_last_committed;/* rep: highest pb_transno committed to disk */
+ __u64 pb_transno; /* server-assigned transno for modifying RPCs */
+ __u32 pb_flags; /* req: MSG_* flags */
+ __u32 pb_op_flags; /* req: MSG_CONNECT_* flags */
+ __u32 pb_conn_cnt; /* connect instance of this client on server */
+ __u32 pb_timeout; /* req: max wait time; rep: service estimate */
+ __u32 pb_service_time; /* rep: server arrival to reply in seconds */
+ __u32 pb_limit; /* rep: dynamic DLM LRU lock count limit */
+ __u64 pb_slv; /* rep: dynamic DLM LRU server lock volume */
+ /* VBR: rep: previous pb_version(s) of objects modified by this RPC */
__u64 pb_pre_versions[PTLRPC_NUM_VERSIONS];
__u64 pb_mbits; /**< match bits for bulk request */
- /* padding for future needs */
+ /* padding for future needs - fix lustre_swab_ptlrpc_body() also */
__u64 pb_padding64_0;
__u64 pb_padding64_1;
__u64 pb_padding64_2;
- char pb_jobid[LUSTRE_JOBID_SIZE];
+ char pb_jobid[LUSTRE_JOBID_SIZE]; /* req: ASCII jobid from env + NUL */
};
#define ptlrpc_body ptlrpc_body_v3
/** only use in req->rq_{req,rep}_swab_mask */
#define MSG_PTLRPC_HEADER_OFF 31
-/* Flags that are operation-specific go in the top 16 bits. */
-#define MSG_OP_FLAG_MASK 0xffff0000
-#define MSG_OP_FLAG_SHIFT 16
-
-/* Flags that apply to all requests are in the bottom 16 bits */
-#define MSG_GEN_FLAG_MASK 0x0000ffff
-#define MSG_LAST_REPLAY 0x0001
-#define MSG_RESENT 0x0002
-#define MSG_REPLAY 0x0004
-/* #define MSG_AT_SUPPORT 0x0008
- * This was used in early prototypes of adaptive timeouts, and while there
- * shouldn't be any users of that code there also isn't a need for using this
- * bits. Defer usage until at least 1.10 to avoid potential conflict. */
-#define MSG_DELAY_REPLAY 0x0010
-#define MSG_VERSION_REPLAY 0x0020
-#define MSG_REQ_REPLAY_DONE 0x0040
-#define MSG_LOCK_REPLAY_DONE 0x0080
-
-/*
- * Flags for all connect opcodes (MDS_CONNECT, OST_CONNECT)
- */
-
-#define MSG_CONNECT_RECOVERING 0x00000001
-#define MSG_CONNECT_RECONNECT 0x00000002
-#define MSG_CONNECT_REPLAYABLE 0x00000004
-/* #define MSG_CONNECT_PEER 0x00000008 removed 1.5 */
-#define MSG_CONNECT_LIBCLIENT 0x00000010
-#define MSG_CONNECT_INITIAL 0x00000020
-#define MSG_CONNECT_ASYNC 0x00000040
-#define MSG_CONNECT_NEXT_VER 0x00000080 /* use next version of lustre_msg */
-#define MSG_CONNECT_TRANSNO 0x00000100 /* report transno */
-
/* Connect flags */
#define OBD_CONNECT_RDONLY 0x1ULL /*client has read-only access*/
#define OBD_CONNECT_INDEX 0x2ULL /*connect specific LOV idx */
#define OBD_CONNECT_OBDOPACK 0x4000000000000000ULL /* compact OUT obdo */
#define OBD_CONNECT_FLAGS2 0x8000000000000000ULL /* second flags word */
/* ocd_connect_flags2 flags */
-#define OBD_CONNECT2_FILE_SECCTX 0x1ULL /* set file security context at create */
-#define OBD_CONNECT2_LOCKAHEAD 0x2ULL /* ladvise lockahead v2 */
+#define OBD_CONNECT2_FILE_SECCTX 0x1ULL /* set file security context at create */
+#define OBD_CONNECT2_LOCKAHEAD 0x2ULL /* ladvise lockahead v2 */
+#define OBD_CONNECT2_DIR_MIGRATE 0x4ULL /* migrate striped dir */
+#define OBD_CONNECT2_FLR 0x20ULL /* FLR support */
+#define OBD_CONNECT2_WBC_INTENTS 0x40ULL /* create/unlink/... intents for wbc, also operations under client-held parent locks */
/* XXX README XXX:
* Please DO NOT add flag values here before first ensuring that this same
OBD_CONNECT_MULTIMODRPCS | \
OBD_CONNECT_SUBTREE | OBD_CONNECT_LARGE_ACL | \
OBD_CONNECT_GRANT_PARAM | \
- OBD_CONNECT_FLAGS2)
+ OBD_CONNECT_SHORTIO | OBD_CONNECT_FLAGS2)
-#define MDT_CONNECT_SUPPORTED2 OBD_CONNECT2_FILE_SECCTX
+#define MDT_CONNECT_SUPPORTED2 (OBD_CONNECT2_FILE_SECCTX | OBD_CONNECT2_FLR)
#define OST_CONNECT_SUPPORTED (OBD_CONNECT_SRVLOCK | OBD_CONNECT_GRANT | \
OBD_CONNECT_REQPORTAL | OBD_CONNECT_VERSION | \
OBD_CONNECT_LAYOUTLOCK | OBD_CONNECT_FID | \
OBD_CONNECT_PINGLESS | OBD_CONNECT_LFSCK | \
OBD_CONNECT_BULK_MBITS | \
- OBD_CONNECT_GRANT_PARAM | OBD_CONNECT_FLAGS2)
+ OBD_CONNECT_GRANT_PARAM | \
+ OBD_CONNECT_SHORTIO | OBD_CONNECT_FLAGS2)
#define OST_CONNECT_SUPPORTED2 OBD_CONNECT2_LOCKAHEAD
-#define ECHO_CONNECT_SUPPORTED 0
+#define ECHO_CONNECT_SUPPORTED (OBD_CONNECT_FID)
#define ECHO_CONNECT_SUPPORTED2 0
#define MGS_CONNECT_SUPPORTED (OBD_CONNECT_VERSION | OBD_CONNECT_AT | \
* Please update DECLARE_CKSUM_NAME/OBD_CKSUM_ALL in obd.h when adding a new
* algorithm and also the OBD_FL_CKSUM* flags.
*/
-typedef enum cksum_types {
+enum cksum_types {
OBD_CKSUM_CRC32 = 0x00000001,
OBD_CKSUM_ADLER = 0x00000002,
OBD_CKSUM_CRC32C= 0x00000004,
-} cksum_type_t;
+};
/*
* OST requests: OBDO & OBD request records
*/
/* opcodes */
-typedef enum {
+enum ost_cmd {
OST_REPLY = 0, /* reply ? */
OST_GETATTR = 1,
OST_SETATTR = 2,
OST_QUOTA_ADJUST_QUNIT = 20, /* not used since 2.4 */
OST_LADVISE = 21,
OST_LAST_OPC /* must be < 33 to avoid MDS_GETATTR */
-} ost_cmd_t;
+};
#define OST_FIRST_OPC OST_REPLY
enum obdo_flags {
#define XATTR_TRUSTED_PREFIX "trusted."
#define XATTR_SECURITY_PREFIX "security."
+#define XATTR_NAME_SOM "trusted.som"
#define XATTR_NAME_LOV "trusted.lov"
#define XATTR_NAME_LMA "trusted.lma"
#define XATTR_NAME_LMV "trusted.lmv"
#define OBD_MD_DOM_SIZE (0X00001000ULL) /* Data-on-MDT component size */
#define OBD_MD_FLNLINK (0x00002000ULL) /* link count */
#define OBD_MD_FLGENER (0x00004000ULL) /* generation number */
-/*#define OBD_MD_FLINLINE (0x00008000ULL) inline data. used until 1.6.5 */
+#define OBD_MD_LAYOUT_VERSION (0x00008000ULL) /* layout version for
+ * OST objects */
#define OBD_MD_FLRDEV (0x00010000ULL) /* device number */
#define OBD_MD_FLEASIZE (0x00020000ULL) /* extended attribute data */
#define OBD_MD_LINKNAME (0x00040000ULL) /* symbolic link target */
/* OBD_MD_FLRMTPERM (0x0000010000000000ULL) remote perm, obsolete */
#define OBD_MD_FLMDSCAPA (0x0000020000000000ULL) /* MDS capability */
#define OBD_MD_FLOSSCAPA (0x0000040000000000ULL) /* OSS capability */
-#define OBD_MD_FLCKSPLIT (0x0000080000000000ULL) /* Check split on server */
+/* OBD_MD_FLCKSPLIT (0x0000080000000000ULL) obsolete 2.3.58*/
#define OBD_MD_FLCROSSREF (0x0000100000000000ULL) /* Cross-ref case */
#define OBD_MD_FLGETATTRLOCK (0x0000200000000000ULL) /* Get IOEpoch attributes
* under lock; for xattr
#define OBD_BRW_READ 0x01
#define OBD_BRW_WRITE 0x02
#define OBD_BRW_RWMASK (OBD_BRW_READ | OBD_BRW_WRITE)
+#define OBD_BRW_NDELAY 0x04 /* Non-delay RPC should be issued for
+ * this page. Non-delay RPCs have bit
+ * rq_no_delay set. */
#define OBD_BRW_SYNC 0x08 /* this page is a part of synchronous
* transfer and is not accounted in
* the grant. */
#define OBD_BRW_LOCALS (OBD_BRW_LOCAL1)
+#define OBD_MAX_GRANT 0x7fffffffUL /* Max grant allowed to one client: 2 GiB */
+
#define OBD_OBJECT_EOF LUSTRE_EOF
#define OST_MIN_PRECREATE 32
#define lvb_glb_ver lvb_id_may_rel /* current version of the global index */
/* op codes */
-typedef enum {
+enum quota_cmd {
QUOTA_DQACQ = 601,
QUOTA_DQREL = 602,
QUOTA_LAST_OPC
-} quota_cmd_t;
+};
#define QUOTA_FIRST_OPC QUOTA_DQACQ
/*
*/
/* opcodes */
-typedef enum {
+enum mds_cmd {
MDS_GETATTR = 33,
MDS_GETATTR_NAME = 34,
MDS_CLOSE = 35,
MDS_HSM_CT_UNREGISTER = 60,
MDS_SWAP_LAYOUTS = 61,
MDS_LAST_OPC
-} mds_cmd_t;
+};
#define MDS_FIRST_OPC MDS_GETATTR
/* opcodes for object update */
-typedef enum {
+enum update_cmd {
OUT_UPDATE = 1000,
OUT_UPDATE_LAST_OPC
-} update_cmd_t;
+};
#define OUT_UPDATE_FIRST_OPC OUT_UPDATE
REINT_SETXATTR = 7,
REINT_RMENTRY = 8,
REINT_MIGRATE = 9,
- REINT_MAX
+ REINT_RESYNC = 10,
+ REINT_MAX
};
/* the disposition of the intent outlines what was executed */
* 2. If these flags needs to be stored into inode, they will be
* stored in LMA. see LMAI_XXXX */
LUSTRE_ORPHAN_FL = 0x00002000,
+ LUSTRE_SET_SYNC_FL = 0x00040000, /* Synchronous setattr on OSTs */
LUSTRE_LMA_FL_MASKS = LUSTRE_ORPHAN_FL,
};
__s64 mbo_atime;
__s64 mbo_ctime;
__u64 mbo_blocks; /* XID, in the case of MDS_READPAGE */
- __u64 mbo_ioepoch;
+ __u64 mbo_version; /* was mbo_ioepoch before 2.11 */
__u64 mbo_t_state; /* transient file state defined in
* enum md_transient_state
* was "ino" until 2.4.0 */
__u32 mbo_flags; /* LUSTRE_*_FL file attributes */
__u32 mbo_rdev;
__u32 mbo_nlink; /* #bytes to read in the case of MDS_READPAGE */
- __u32 mbo_unused2; /* was "generation" until 2.4.0 */
+ __u32 mbo_layout_gen; /* was "generation" until 2.4.0 */
__u32 mbo_suppgid;
__u32 mbo_eadatasize;
__u32 mbo_aclsize;
#define MDS_ATTR_BLOCKS 0x8000ULL /* = 32768 */
#define MDS_ATTR_PROJID 0x10000ULL /* = 65536 */
-#ifndef FMODE_READ
-#define FMODE_READ 00000001
-#define FMODE_WRITE 00000002
-#endif
-
-#define MDS_FMODE_CLOSED 00000000
-#define MDS_FMODE_EXEC 00000004
-/* MDS_FMODE_EPOCH 01000000 obsolete since 2.8.0 */
-/* MDS_FMODE_TRUNC 02000000 obsolete since 2.8.0 */
-/* MDS_FMODE_SOM 04000000 obsolete since 2.8.0 */
-
-#define MDS_OPEN_CREATED 00000010
-#define MDS_OPEN_CROSS 00000020
-
-#define MDS_OPEN_CREAT 00000100
-#define MDS_OPEN_EXCL 00000200
-#define MDS_OPEN_TRUNC 00001000
-#define MDS_OPEN_APPEND 00002000
-#define MDS_OPEN_SYNC 00010000
-#define MDS_OPEN_DIRECTORY 00200000
-
-#define MDS_OPEN_BY_FID 040000000 /* open_by_fid for known object */
-#define MDS_OPEN_DELAY_CREATE 0100000000 /* delay initial object create */
-#define MDS_OPEN_OWNEROVERRIDE 0200000000 /* NFSD rw-reopen ro file for owner */
-#define MDS_OPEN_JOIN_FILE 0400000000 /* open for join file.
- * We do not support JOIN FILE
- * anymore, reserve this flags
- * just for preventing such bit
- * to be reused. */
-
-#define MDS_OPEN_LOCK 04000000000 /* This open requires open lock */
-#define MDS_OPEN_HAS_EA 010000000000 /* specify object create pattern */
-#define MDS_OPEN_HAS_OBJS 020000000000 /* Just set the EA the obj exist */
-#define MDS_OPEN_NORESTORE 0100000000000ULL /* Do not restore file at open */
-#define MDS_OPEN_NEWSTRIPE 0200000000000ULL /* New stripe needed (restripe or
- * hsm restore) */
-#define MDS_OPEN_VOLATILE 0400000000000ULL /* File is volatile = created
- unlinked */
-#define MDS_OPEN_LEASE 01000000000000ULL /* Open the file and grant lease
- * delegation, succeed if it's not
- * being opened with conflict mode.
- */
-#define MDS_OPEN_RELEASE 02000000000000ULL /* Open the file for HSM release */
-
-/* lustre internal open flags, which should not be set from user space */
-#define MDS_OPEN_FL_INTERNAL (MDS_OPEN_HAS_EA | MDS_OPEN_HAS_OBJS | \
- MDS_OPEN_OWNEROVERRIDE | MDS_OPEN_LOCK | \
- MDS_OPEN_BY_FID | MDS_OPEN_LEASE | \
- MDS_OPEN_RELEASE)
-
enum mds_op_bias {
- MDS_CHECK_SPLIT = 1 << 0,
+/* MDS_CHECK_SPLIT = 1 << 0, obsolete before 2.3.58 */
MDS_CROSS_REF = 1 << 1,
- MDS_VTX_BYPASS = 1 << 2,
+/* MDS_VTX_BYPASS = 1 << 2, obsolete since 2.3.54 */
MDS_PERM_BYPASS = 1 << 3,
/* MDS_SOM = 1 << 4, obsolete since 2.8.0 */
MDS_QUOTA_IGNORE = 1 << 5,
- /* Was MDS_CLOSE_CLEANUP (1 << 6), No more used */
+/* MDS_CLOSE_CLEANUP = 1 << 6, obsolete since 2.3.51 */
MDS_KEEP_ORPHAN = 1 << 7,
MDS_RECOV_OPEN = 1 << 8,
MDS_DATA_MODIFIED = 1 << 9,
MDS_HSM_RELEASE = 1 << 12,
MDS_RENAME_MIGRATE = 1 << 13,
MDS_CLOSE_LAYOUT_SWAP = 1 << 14,
+ MDS_CLOSE_LAYOUT_MERGE = 1 << 15,
+ MDS_CLOSE_RESYNC_DONE = 1 << 16,
+ MDS_CLOSE_LAYOUT_SPLIT = 1 << 17,
};
+#define MDS_CLOSE_INTENT (MDS_HSM_RELEASE | MDS_CLOSE_LAYOUT_SWAP | \
+ MDS_CLOSE_LAYOUT_MERGE | MDS_CLOSE_LAYOUT_SPLIT | \
+ MDS_CLOSE_RESYNC_DONE)
+
/* instance of mdt_reint_rec */
struct mdt_rec_create {
__u32 cr_opcode;
__u32 sx_padding_11; /* rr_padding_4 */
};
+/* instance of mdt_reint_rec
+ * FLR: for file resync MDS_REINT_RESYNC RPC. */
+struct mdt_rec_resync {
+ __u32 rs_opcode;
+ __u32 rs_cap;
+ __u32 rs_fsuid;
+ __u32 rs_fsuid_h;
+ __u32 rs_fsgid;
+ __u32 rs_fsgid_h;
+ __u32 rs_suppgid1;
+ __u32 rs_suppgid1_h;
+ __u32 rs_suppgid2;
+ __u32 rs_suppgid2_h;
+ struct lu_fid rs_fid;
+ __u8 rs_padding0[sizeof(struct lu_fid)];
+ struct lustre_handle rs_handle; /* rr_mtime */
+ __s64 rs_padding1; /* rr_atime */
+ __s64 rs_padding2; /* rr_ctime */
+ __u64 rs_padding3; /* rr_size */
+ __u64 rs_padding4; /* rr_blocks */
+ __u32 rs_bias;
+ __u32 rs_padding5; /* rr_mode */
+ __u32 rs_padding6; /* rr_flags */
+ __u32 rs_padding7; /* rr_flags_h */
+ __u32 rs_padding8; /* rr_umask */
+ __u32 rs_padding9; /* rr_padding_4 */
+};
+
/*
* mdt_rec_reint is the template for all mdt_reint_xxx structures.
* Do NOT change the size of various members, otherwise the value
};
/* LFSCK opcodes */
-typedef enum {
+enum lfsck_cmd {
LFSCK_NOTIFY = 1101,
LFSCK_QUERY = 1102,
LFSCK_LAST_OPC,
- LFSCK_FIRST_OPC = LFSCK_NOTIFY
-} lfsck_cmd_t;
+ LFSCK_FIRST_OPC = LFSCK_NOTIFY
+};
/*
* LOV data structures
* LDLM requests:
*/
/* opcodes -- MUST be distinct from OST/MDS opcodes */
-typedef enum {
+enum ldlm_cmd {
LDLM_ENQUEUE = 101,
LDLM_CONVERT = 102,
LDLM_CANCEL = 103,
LDLM_GL_CALLBACK = 106,
LDLM_SET_INFO = 107,
LDLM_LAST_OPC
-} ldlm_cmd_t;
+};
#define LDLM_FIRST_OPC LDLM_ENQUEUE
#define RES_NAME_SIZE 4
(unsigned long long)(res)->lr_name.name[3]
/* lock types */
-typedef enum ldlm_mode {
+enum ldlm_mode {
LCK_MINMODE = 0,
LCK_EX = 1,
LCK_PW = 2,
LCK_GROUP = 64,
LCK_COS = 128,
LCK_MAXMODE
-} ldlm_mode_t;
+};
#define LCK_MODE_NUM 8
-typedef enum ldlm_type {
+enum ldlm_type {
LDLM_PLAIN = 10,
LDLM_EXTENT = 11,
LDLM_FLOCK = 12,
LDLM_IBITS = 13,
LDLM_MAX_TYPE
-} ldlm_type_t;
+};
#define LDLM_MIN_TYPE LDLM_PLAIN
}
struct ldlm_inodebits {
- __u64 bits;
- __u64 try_bits; /* optional bits to try */
+ __u64 bits;
+ union {
+ __u64 try_bits; /* optional bits to try */
+ __u64 cancel_bits; /* for lock convert */
+ };
};
struct ldlm_flock_wire {
* this ever changes we will need to swab the union differently based
* on the resource type. */
-typedef union ldlm_wire_policy_data {
+union ldlm_wire_policy_data {
struct ldlm_extent l_extent;
struct ldlm_flock_wire l_flock;
struct ldlm_inodebits l_inodebits;
-} ldlm_wire_policy_data_t;
+};
struct barrier_lvb {
__u32 lvb_status;
#define LDLM_ENQUEUE_CANCEL_OFF 1
struct ldlm_request {
- __u32 lock_flags;
- __u32 lock_count;
- struct ldlm_lock_desc lock_desc;
- struct lustre_handle lock_handle[LDLM_LOCKREQ_HANDLES];
+ __u32 lock_flags; /* LDLM_FL_*, see lustre_dlm_flags.h */
+ __u32 lock_count; /* number of locks in lock_handle[] */
+ struct ldlm_lock_desc lock_desc;/* lock descriptor */
+ struct lustre_handle lock_handle[LDLM_LOCKREQ_HANDLES];
};
struct ldlm_reply {
/*
* Opcodes for mountconf (mgs and mgc)
*/
-typedef enum {
+enum mgs_cmd {
MGS_CONNECT = 250,
MGS_DISCONNECT,
MGS_EXCEPTION, /* node died, etc. */
MGS_SET_INFO,
MGS_CONFIG_READ,
MGS_LAST_OPC
-} mgs_cmd_t;
+};
#define MGS_FIRST_OPC MGS_CONNECT
#if LUSTRE_VERSION_CODE < OBD_OCD_VERSION(2, 13, 53, 0)
#define MTI_PARAM_MAXLEN 4096
#define MTI_NIDS_MAX 32
struct mgs_target_info {
- __u32 mti_lustre_ver;
- __u32 mti_stripe_index;
- __u32 mti_config_ver;
- __u32 mti_flags;
- __u32 mti_nid_count;
- __u32 mti_instance; /* Running instance of target */
- char mti_fsname[MTI_NAME_MAXLEN];
- char mti_svname[MTI_NAME_MAXLEN];
- char mti_uuid[sizeof(struct obd_uuid)];
- __u64 mti_nids[MTI_NIDS_MAX]; /* host nids (lnet_nid_t)*/
- char mti_params[MTI_PARAM_MAXLEN];
+ __u32 mti_lustre_ver;
+ __u32 mti_stripe_index;
+ __u32 mti_config_ver;
+ __u32 mti_flags; /* LDD_F_* */
+ __u32 mti_nid_count;
+ __u32 mti_instance; /* Running instance of target */
+ char mti_fsname[MTI_NAME_MAXLEN];
+ char mti_svname[MTI_NAME_MAXLEN];
+ char mti_uuid[sizeof(struct obd_uuid)];
+ __u64 mti_nids[MTI_NIDS_MAX]; /* host nids (lnet_nid_t) */
+ char mti_params[MTI_PARAM_MAXLEN];
};
struct mgs_nidtbl_entry {
/*
* Opcodes for multiple servers.
*/
-
-typedef enum {
- OBD_PING = 400,
- OBD_LOG_CANCEL,
+enum obd_cmd {
+ OBD_PING = 400,
+ OBD_LOG_CANCEL, /* Obsolete since 1.5. */
OBD_QC_CALLBACK, /* not used since 2.4 */
OBD_IDX_READ,
OBD_LAST_OPC
-} obd_cmd_t;
+};
#define OBD_FIRST_OPC OBD_PING
/**
#define LLOG_OP_MAGIC 0x10600000
#define LLOG_OP_MASK 0xfff00000
-typedef enum {
+enum llog_op_type {
LLOG_PAD_MAGIC = LLOG_OP_MAGIC | 0x00000,
OST_SZ_REC = LLOG_OP_MAGIC | 0x00f00,
/* OST_RAID1_REC = LLOG_OP_MAGIC | 0x01000, never used */
UPDATE_REC = LLOG_OP_MAGIC | 0xa0000,
LLOG_HDR_MAGIC = LLOG_OP_MAGIC | 0x45539,
LLOG_LOGID_MAGIC = LLOG_OP_MAGIC | 0x4553b,
-} llog_op_type;
+};
#define LLOG_REC_HDR_NEEDS_SWABBING(r) \
(((r)->lrh_type & __swab32(LLOG_OP_MASK)) == __swab32(LLOG_OP_MAGIC))
__u32 lsr_gid_h;
__u64 lsr_valid;
__u32 lsr_projid;
- __u32 lsr_padding1;
+ __u32 lsr_layout_version;
__u64 lsr_padding2;
__u64 lsr_padding3;
struct llog_rec_tail lsr_tail;
#define CHANGELOG_ALLMASK 0XFFFFFFFF
/** default \a changelog_rec_type mask. Allow all of them, except
* CL_ATIME since it can really be time consuming, and not necessary
- * under normal use. */
-#define CHANGELOG_DEFMASK (CHANGELOG_ALLMASK & ~(1 << CL_ATIME))
+ * under normal use.
+ * Remove also CL_OPEN, CL_GETXATTR and CL_DN_OPEN from default list as it can
+ * be costly and only necessary for audit purpose.
+ */
+#define CHANGELOG_DEFMASK (CHANGELOG_ALLMASK & \
+ ~(1 << CL_ATIME | 1 << CL_OPEN | 1 << CL_GETXATTR | \
+ 1 << CL_DN_OPEN))
/* changelog llog name, needed by client replicators */
#define CHANGELOG_CATALOG "changelog_catalog"
#define CHANGELOG_USER_PREFIX "cl"
struct llog_changelog_user_rec {
- struct llog_rec_hdr cur_hdr;
- __u32 cur_id;
- __u32 cur_padding;
- __u64 cur_endrec;
- struct llog_rec_tail cur_tail;
+ struct llog_rec_hdr cur_hdr;
+ __u32 cur_id;
+ /* only intended to be used in relative time comparisons to
+ * detect idle users */
+ __u32 cur_time;
+ __u64 cur_endrec;
+ struct llog_rec_tail cur_tail;
} __attribute__((packed));
enum agent_req_status {
* agent_req_status */
__u32 arr_archive_id; /**< backend archive number */
__u64 arr_flags; /**< req flags */
- __u64 arr_compound_id; /**< compound cookie */
+ __u64 arr_compound_id; /**< compound cookie, ignored */
__u64 arr_req_create; /**< req. creation time */
__u64 arr_req_change; /**< req. status change time */
struct hsm_action_item arr_hai; /**< req. to the agent */
LLOG_F_EXT_JOBID = 0x8,
LLOG_F_IS_FIXSIZE = 0x10,
LLOG_F_EXT_EXTRA_FLAGS = 0x20,
+ LLOG_F_EXT_X_UIDGID = 0x40,
+ LLOG_F_EXT_X_NID = 0x80,
+ LLOG_F_EXT_X_OMODE = 0x100,
+ LLOG_F_EXT_X_XATTR = 0x200,
/* Note: Flags covered by LLOG_F_EXT_MASK will be inherited from
* catlog to plain log, so do not add LLOG_F_IS_FIXSIZE here,
* because the catlog record is usually fixed size, but its plain
* log record can be variable */
- LLOG_F_EXT_MASK = LLOG_F_EXT_JOBID | LLOG_F_EXT_EXTRA_FLAGS,
+ LLOG_F_EXT_MASK = LLOG_F_EXT_JOBID | LLOG_F_EXT_EXTRA_FLAGS |
+ LLOG_F_EXT_X_UIDGID | LLOG_F_EXT_X_NID |
+ LLOG_F_EXT_X_OMODE | LLOG_F_EXT_X_XATTR,
};
/* On-disk header structure of each log object, stored in little endian order */
llh->llh_hdr.lrh_len - \
sizeof(llh->llh_tail)))
-/** log cookies are used to reference a specific log file and a record therein */
+/** log cookies are used to reference a specific log file and a record therein,
+ and pass record offset from llog_process_thread to llog_write */
struct llog_cookie {
- struct llog_logid lgc_lgl;
+ union {
+ struct llog_logid lgc_lgl;
+ __u64 lgc_offset;
+ };
__u32 lgc_subsys;
__u32 lgc_index;
__u32 lgc_padding;
/** llog protocol */
enum llogd_rpc_ops {
- LLOG_ORIGIN_HANDLE_CREATE = 501,
- LLOG_ORIGIN_HANDLE_NEXT_BLOCK = 502,
- LLOG_ORIGIN_HANDLE_READ_HEADER = 503,
- LLOG_ORIGIN_HANDLE_WRITE_REC = 504,
- LLOG_ORIGIN_HANDLE_CLOSE = 505,
- LLOG_ORIGIN_CONNECT = 506,
- LLOG_CATINFO = 507, /* deprecated */
- LLOG_ORIGIN_HANDLE_PREV_BLOCK = 508,
- LLOG_ORIGIN_HANDLE_DESTROY = 509, /* for destroy llog object*/
- LLOG_LAST_OPC,
- LLOG_FIRST_OPC = LLOG_ORIGIN_HANDLE_CREATE
+ LLOG_ORIGIN_HANDLE_CREATE = 501,
+ LLOG_ORIGIN_HANDLE_NEXT_BLOCK = 502,
+ LLOG_ORIGIN_HANDLE_READ_HEADER = 503,
+ LLOG_ORIGIN_HANDLE_WRITE_REC = 504, /* Obsolete by 2.1. */
+ LLOG_ORIGIN_HANDLE_CLOSE = 505, /* Obsolete by 1.8. */
+ LLOG_ORIGIN_CONNECT = 506, /* Obsolete by 2.4. */
+ LLOG_CATINFO = 507, /* Obsolete by 2.3. */
+ LLOG_ORIGIN_HANDLE_PREV_BLOCK = 508,
+ LLOG_ORIGIN_HANDLE_DESTROY = 509, /* Obsolete. */
+ LLOG_LAST_OPC,
+ LLOG_FIRST_OPC = LLOG_ORIGIN_HANDLE_CREATE
};
struct llogd_body {
*
* sizeof(ost_layout) + sieof(__u32) == sizeof(llog_cookie). */
struct ost_layout o_layout;
- __u32 o_padding_3;
+ __u32 o_layout_version;
__u32 o_uid_h;
__u32 o_gid_h;
};
/* security opcodes */
-typedef enum {
+enum sec_cmd {
SEC_CTX_INIT = 801,
SEC_CTX_INIT_CONT = 802,
SEC_CTX_FINI = 803,
SEC_LAST_OPC,
SEC_FIRST_OPC = SEC_CTX_INIT
-} sec_cmd_t;
+};
/*
* capa related definitions
char gp_name[0]; /**< zero-terminated link name */
} __attribute__((packed));
-enum {
+enum layout_intent_opc {
LAYOUT_INTENT_ACCESS = 0, /** generic access */
LAYOUT_INTENT_READ = 1, /** not used */
LAYOUT_INTENT_WRITE = 2, /** write file, for comp layout */
struct layout_intent {
__u32 li_opc; /* intent operation for enqueue, read, write etc */
__u32 li_flags;
- __u64 li_start;
- __u64 li_end;
+ struct lu_extent li_extent;
} __attribute__((packed));
/**
OUT_PUNCH = 14,
OUT_READ = 15,
OUT_NOOP = 16,
+ OUT_XATTR_LIST = 17,
OUT_LAST
};
__u64 msl_flags;
} __attribute__((packed));
+#define INLINE_RESYNC_ARRAY_SIZE 15
+struct close_data_resync_done {
+ __u32 resync_count;
+ __u32 resync_ids_inline[INLINE_RESYNC_ARRAY_SIZE];
+};
+
struct close_data {
struct lustre_handle cd_handle;
struct lu_fid cd_fid;
__u64 cd_data_version;
- __u64 cd_reserved[8];
+ union {
+ __u64 cd_reserved[8];
+ struct close_data_resync_done cd_resync;
+ /* split close */
+ __u16 cd_mirror_id;
+ };
};
/* Update llog format */