o2iblnd.h

来自「lustre 1.6.5 source code」· C头文件 代码 · 共 762 行 · 第 1/3 页

        kib_rdma_desc_t   ibpam_rd;             /* sender's sink buffer */
} WIRE_ATTR kib_putack_msg_t;

typedef struct
{
        lnet_hdr_t        ibgm_hdr;             /* portals header */
        __u64             ibgm_cookie;          /* opaque completion cookie */
        kib_rdma_desc_t   ibgm_rd;              /* rdma descriptor */
} WIRE_ATTR kib_get_msg_t;

typedef struct
{
        __u64             ibcm_cookie;          /* opaque completion cookie */
        __s32             ibcm_status;          /* < 0 failure: >= 0 length */
} WIRE_ATTR kib_completion_msg_t;

typedef struct
{
        /* First 2 fields fixed FOR ALL TIME */
        __u32             ibm_magic;            /* I'm an openibnal message */
        __u16             ibm_version;          /* this is my version number */

        __u8              ibm_type;             /* msg type */
        __u8              ibm_credits;          /* returned credits */
        __u32             ibm_nob;              /* # bytes in whole message */
        __u32             ibm_cksum;            /* checksum (0 == no checksum) */
        __u64             ibm_srcnid;           /* sender's NID */
        __u64             ibm_srcstamp;         /* sender's incarnation */
        __u64             ibm_dstnid;           /* destination's NID */
        __u64             ibm_dststamp;         /* destination's incarnation */

        union {
                kib_connparams_t      connparams;
                kib_immediate_msg_t   immediate;
                kib_putreq_msg_t      putreq;
                kib_putack_msg_t      putack;
                kib_get_msg_t         get;
                kib_completion_msg_t  completion;
        } WIRE_ATTR ibm_u;
} WIRE_ATTR kib_msg_t;

#define IBLND_MSG_MAGIC LNET_PROTO_IB_MAGIC	/* unique magic */

#define IBLND_MSG_VERSION           0x11

#define IBLND_MSG_CONNREQ           0xc0        /* connection request */
#define IBLND_MSG_CONNACK           0xc1        /* connection acknowledge */
#define IBLND_MSG_NOOP              0xd0        /* nothing (just credits) */
#define IBLND_MSG_IMMEDIATE         0xd1        /* immediate */
#define IBLND_MSG_PUT_REQ           0xd2        /* putreq (src->sink) */
#define IBLND_MSG_PUT_NAK           0xd3        /* completion (sink->src) */
#define IBLND_MSG_PUT_ACK           0xd4        /* putack (sink->src) */
#define IBLND_MSG_PUT_DONE          0xd5        /* completion (src->sink) */
#define IBLND_MSG_GET_REQ           0xd6        /* getreq (sink->src) */
#define IBLND_MSG_GET_DONE          0xd7        /* completion (src->sink: all OK) */

typedef struct {
        __u32            ibr_magic;             /* sender's magic */
        __u16            ibr_version;           /* sender's version */
        __u8             ibr_why;               /* reject reason */
} WIRE_ATTR kib_rej_t;


/* connection rejection reasons */
#define IBLND_REJECT_CONN_RACE       1          /* You lost connection race */
#define IBLND_REJECT_NO_RESOURCES    2          /* Out of memory/conns etc */
#define IBLND_REJECT_FATAL           3          /* Anything else */

/***********************************************************************/

typedef struct kib_rx                           /* receive message */
{
        struct list_head          rx_list;      /* queue for attention */
        struct kib_conn          *rx_conn;      /* owning conn */
        int                       rx_nob;       /* # bytes received (-1 while posted) */
        enum ib_wc_status         rx_status;    /* completion status */
        kib_msg_t                *rx_msg;       /* message buffer (host vaddr) */
        __u64                     rx_msgaddr;   /* message buffer (I/O addr) */
        DECLARE_PCI_UNMAP_ADDR   (rx_msgunmap); /* for dma_unmap_single() */
        struct ib_recv_wr         rx_wrq;       /* receive work item... */
        struct ib_sge             rx_sge;       /* ...and its memory */
} kib_rx_t;

#define IBLND_POSTRX_DONT_POST    0             /* don't post */
#define IBLND_POSTRX_NO_CREDIT    1             /* post: no credits */
#define IBLND_POSTRX_PEER_CREDIT  2             /* post: give peer back 1 credit */
#define IBLND_POSTRX_RSRVD_CREDIT 3             /* post: give myself back 1 reserved credit */

typedef struct kib_tx                           /* transmit message */
{
        struct list_head          tx_list;      /* queue on idle_txs ibc_tx_queue etc. */
        struct kib_conn          *tx_conn;      /* owning conn */
        int                       tx_sending;   /* # tx callbacks outstanding */
        int                       tx_queued;    /* queued for sending */
        int                       tx_waiting;   /* waiting for peer */
        int                       tx_status;    /* LNET completion status */
        unsigned long             tx_deadline;  /* completion deadline */
        __u64                     tx_cookie;    /* completion cookie */
        lnet_msg_t               *tx_lntmsg[2]; /* lnet msgs to finalize on completion */
        kib_msg_t                *tx_msg;       /* message buffer (host vaddr) */
        __u64                     tx_msgaddr;   /* message buffer (I/O addr) */
        DECLARE_PCI_UNMAP_ADDR   (tx_msgunmap); /* for dma_unmap_single() */
        int                       tx_nwrq;      /* # send work items */
#if IBLND_MAP_ON_DEMAND
        struct ib_send_wr         tx_wrq[2];    /* send work items... */
        struct ib_sge             tx_sge[2];    /* ...and their memory */
        kib_rdma_desc_t           tx_rd[1];     /* rdma descriptor */
        __u64                    *tx_pages;     /* rdma phys page addrs */
        struct ib_pool_fmr       *tx_fmr;       /* rdma mapping (mapped if != NULL) */
#else
        struct ib_send_wr        *tx_wrq;       /* send work items... */
        struct ib_sge            *tx_sge;       /* ...and their memory */
        kib_rdma_desc_t          *tx_rd;        /* rdma descriptor */
        int                       tx_nfrags;    /* # entries in... */
        struct scatterlist       *tx_frags;     /* dma_map_sg descriptor */
        int                       tx_dmadir;    /* dma direction */
#endif        
} kib_tx_t;

typedef struct kib_connvars
{
        /* connection-in-progress variables */
        kib_msg_t                 cv_msg;
} kib_connvars_t;

typedef struct kib_conn
{
        struct kib_peer    *ibc_peer;           /* owning peer */
        struct list_head    ibc_list;           /* stash on peer's conn list */
        struct list_head    ibc_sched_list;     /* schedule for attention */
        __u64               ibc_incarnation;    /* which instance of the peer */
        atomic_t            ibc_refcount;       /* # users */
        int                 ibc_state;          /* what's happening */
        int                 ibc_nsends_posted;  /* # uncompleted sends */
        int                 ibc_credits;        /* # credits I have */
        int                 ibc_outstanding_credits; /* # credits to return */
        int                 ibc_reserved_credits;/* # ACK/DONE msg credits */
        int                 ibc_comms_error;    /* set on comms error */
        int                 ibc_nrx:8;          /* receive buffers owned */
        int                 ibc_scheduled:1;    /* scheduled for attention */
        int                 ibc_ready:1;        /* CQ callback fired */
        unsigned long       ibc_last_send;      /* time of last send */
        struct list_head    ibc_early_rxs;      /* rxs completed before ESTABLISHED */
        struct list_head    ibc_tx_noops;       /* IBLND_MSG_NOOPs */
        struct list_head    ibc_tx_queue;       /* sends that need a credit */
        struct list_head    ibc_tx_queue_nocred;/* sends that don't need a credit */
        struct list_head    ibc_tx_queue_rsrvd; /* sends that need to reserve an ACK/DONE msg */
        struct list_head    ibc_active_txs;     /* active tx awaiting completion */
        spinlock_t          ibc_lock;           /* serialise */
        kib_rx_t           *ibc_rxs;            /* the rx descs */
        kib_pages_t        *ibc_rx_pages;       /* premapped rx msg pages */

        struct rdma_cm_id  *ibc_cmid;           /* CM id */
        struct ib_cq       *ibc_cq;             /* completion queue */

        kib_connvars_t     *ibc_connvars;       /* in-progress connection state */
} kib_conn_t;

#define IBLND_CONN_INIT               0         /* being intialised */
#define IBLND_CONN_ACTIVE_CONNECT     1         /* active sending req */
#define IBLND_CONN_PASSIVE_WAIT       2         /* passive waiting for rtu */
#define IBLND_CONN_ESTABLISHED        3         /* connection established */
#define IBLND_CONN_CLOSING            4         /* being closed */
#define IBLND_CONN_DISCONNECTED       5         /* disconnected */

typedef struct kib_peer
{
        struct list_head    ibp_list;           /* stash on global peer list */
        lnet_nid_t          ibp_nid;            /* who's on the other end(s) */
        lnet_ni_t          *ibp_ni;             /* LNet interface */
        atomic_t            ibp_refcount;       /* # users */
        struct list_head    ibp_conns;          /* all active connections */
        struct list_head    ibp_tx_queue;       /* msgs waiting for a conn */
        int                 ibp_connecting;     /* current active connection attempts */
        int                 ibp_accepting;      /* current passive connection attempts */
        int                 ibp_error;          /* errno on closing this peer */
        cfs_time_t          ibp_last_alive;     /* when (in jiffies) I was last alive */
} kib_peer_t;


extern kib_data_t      kiblnd_data;
extern kib_tunables_t  kiblnd_tunables;

#define kiblnd_conn_addref(conn)                                \
do {                                                            \
        CDEBUG(D_NET, "conn[%p] (%d)++\n",                      \
               (conn), atomic_read(&(conn)->ibc_refcount));     \
        LASSERT(atomic_read(&(conn)->ibc_refcount) > 0);        \
        atomic_inc(&(conn)->ibc_refcount);                      \
} while (0)

#define kiblnd_conn_decref(conn)                                              \
do {                                                                          \
        unsigned long   flags;                                                \
                                                                              \
        CDEBUG(D_NET, "conn[%p] (%d)--\n",                                    \
               (conn), atomic_read(&(conn)->ibc_refcount));                   \
        LASSERT(atomic_read(&(conn)->ibc_refcount) > 0);                      \
        if (atomic_dec_and_test(&(conn)->ibc_refcount)) {                     \
                spin_lock_irqsave(&kiblnd_data.kib_connd_lock, flags);        \
                list_add_tail(&(conn)->ibc_list,                              \
                              &kiblnd_data.kib_connd_zombies);                \
                wake_up(&kiblnd_data.kib_connd_waitq);                        \
                spin_unlock_irqrestore(&kiblnd_data.kib_connd_lock, flags);   \
        }                                                                     \
} while (0)

#define kiblnd_peer_addref(peer)                                \
do {                                                            \
        CDEBUG(D_NET, "peer[%p] -> %s (%d)++\n",                \
               (peer), libcfs_nid2str((peer)->ibp_nid),         \
               atomic_read (&(peer)->ibp_refcount));            \
        LASSERT(atomic_read(&(peer)->ibp_refcount) > 0);        \
        atomic_inc(&(peer)->ibp_refcount);                      \
} while (0)

#define kiblnd_peer_decref(peer)                                \
do {                                                            \
        CDEBUG(D_NET, "peer[%p] -> %s (%d)--\n",                \
               (peer), libcfs_nid2str((peer)->ibp_nid),         \
               atomic_read (&(peer)->ibp_refcount));            \
        LASSERT(atomic_read(&(peer)->ibp_refcount) > 0);        \
        if (atomic_dec_and_test(&(peer)->ibp_refcount))         \
                kiblnd_destroy_peer(peer);                      \
} while (0)

static inline struct list_head *
kiblnd_nid2peerlist (lnet_nid_t nid)
{
        unsigned int hash = ((unsigned int)nid) % kiblnd_data.kib_peer_hash_size;

        return (&kiblnd_data.kib_peers [hash]);
}

static inline int
kiblnd_peer_active (kib_peer_t *peer)
{
        /* Am I in the peer hash table? */
        return (!list_empty(&peer->ibp_list));
}

static inline kib_conn_t *
kiblnd_get_conn_locked (kib_peer_t *peer)
{
        LASSERT (!list_empty(&peer->ibp_conns));
        
        /* just return the first connection */
        return list_entry(peer->ibp_conns.next, kib_conn_t, ibc_list);
}

static inline int
kiblnd_send_keepalive(kib_conn_t *conn) 
{
        return (*kiblnd_tunables.kib_keepalive > 0) &&