X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=lnet%2Finclude%2Flnet%2Flib-types.h;h=7c7a9bd7c56f0d5acf51e7407ec0a23da10f0337;hb=ed052504713d1db49531454a87055b2ee54399f0;hp=3ff7b337c1a4ae3ba870154acb562899fae41c57;hpb=07ff616a18cc602460724e6693e142782cad0fcc;p=fs%2Flustre-release.git diff --git a/lnet/include/lnet/lib-types.h b/lnet/include/lnet/lib-types.h index 3ff7b33..7c7a9bd 100644 --- a/lnet/include/lnet/lib-types.h +++ b/lnet/include/lnet/lib-types.h @@ -46,7 +46,8 @@ #include #include -#include +#include +#include /* Max payload size */ #ifndef CONFIG_LNET_MAX_PAYLOAD @@ -67,11 +68,20 @@ typedef struct lnet_msg { struct list_head msg_activelist; struct list_head msg_list; /* Q for credits/MD */ - lnet_process_id_t msg_target; + struct lnet_process_id msg_target; + /* Primary NID of the source. */ + lnet_nid_t msg_initiator; /* where is it from, it's only for building event */ lnet_nid_t msg_from; __u32 msg_type; + /* + * hold parameters in case message is with held due + * to discovery + */ + lnet_nid_t msg_src_nid_param; + lnet_nid_t msg_rtr_nid_param; + /* committed for sending */ unsigned int msg_tx_committed:1; /* CPT # this message committed for sending */ @@ -87,37 +97,39 @@ typedef struct lnet_msg { /* ready for pending on RX delay list */ unsigned int msg_rx_ready_delay:1; - unsigned int msg_vmflush:1; /* VM trying to free memory */ - unsigned int msg_target_is_router:1; /* sending to a router */ - unsigned int msg_routing:1; /* being forwarded */ - unsigned int msg_ack:1; /* ack on finalize (PUT) */ - unsigned int msg_sending:1; /* outgoing message */ - unsigned int msg_receiving:1; /* being received */ - unsigned int msg_txcredit:1; /* taken an NI send credit */ - unsigned int msg_peertxcredit:1; /* taken a peer send credit */ - unsigned int msg_rtrcredit:1; /* taken a globel router credit */ - unsigned int msg_peerrtrcredit:1; /* taken a peer router credit */ - unsigned int msg_onactivelist:1; /* on the activelist */ + unsigned int msg_vmflush:1; /* VM trying to free memory */ + unsigned int msg_target_is_router:1; /* sending to a router */ + unsigned int msg_routing:1; /* being forwarded */ + unsigned int msg_ack:1; /* ack on finalize (PUT) */ + unsigned int msg_sending:1; /* outgoing message */ + unsigned int msg_receiving:1; /* being received */ + unsigned int msg_txcredit:1; /* taken an NI send credit */ + unsigned int msg_peertxcredit:1; /* taken a peer send credit */ + unsigned int msg_rtrcredit:1; /* taken a globel router credit */ + unsigned int msg_peerrtrcredit:1; /* taken a peer router credit */ + unsigned int msg_onactivelist:1; /* on the activelist */ unsigned int msg_rdma_get:1; - struct lnet_peer *msg_txpeer; /* peer I'm sending to */ - struct lnet_peer *msg_rxpeer; /* peer I received from */ + struct lnet_peer_ni *msg_txpeer; /* peer I'm sending to */ + struct lnet_peer_ni *msg_rxpeer; /* peer I received from */ - void *msg_private; + void *msg_private; struct lnet_libmd *msg_md; - - unsigned int msg_len; - unsigned int msg_wanted; - unsigned int msg_offset; - unsigned int msg_niov; + /* the NI the message was sent or received over */ + struct lnet_ni *msg_txni; + struct lnet_ni *msg_rxni; + + unsigned int msg_len; + unsigned int msg_wanted; + unsigned int msg_offset; + unsigned int msg_niov; struct kvec *msg_iov; - lnet_kiov_t *msg_kiov; + lnet_kiov_t *msg_kiov; - lnet_event_t msg_ev; - lnet_hdr_t msg_hdr; + struct lnet_event msg_ev; + struct lnet_hdr msg_hdr; } lnet_msg_t; - typedef struct lnet_libhandle { struct list_head lh_hash_chain; __u64 lh_cookie; @@ -128,31 +140,31 @@ typedef struct lnet_libhandle { typedef struct lnet_eq { struct list_head eq_list; - lnet_libhandle_t eq_lh; - lnet_seq_t eq_enq_seq; - lnet_seq_t eq_deq_seq; + struct lnet_libhandle eq_lh; + unsigned long eq_enq_seq; + unsigned long eq_deq_seq; unsigned int eq_size; lnet_eq_handler_t eq_callback; - lnet_event_t *eq_events; + struct lnet_event *eq_events; int **eq_refs; /* percpt refcount for EQ */ } lnet_eq_t; typedef struct lnet_me { struct list_head me_list; - lnet_libhandle_t me_lh; - lnet_process_id_t me_match_id; + struct lnet_libhandle me_lh; + struct lnet_process_id me_match_id; unsigned int me_portal; unsigned int me_pos; /* hash offset in mt_hash */ __u64 me_match_bits; __u64 me_ignore_bits; - lnet_unlink_t me_unlink; + enum lnet_unlink me_unlink; struct lnet_libmd *me_md; } lnet_me_t; typedef struct lnet_libmd { struct list_head md_list; - lnet_libhandle_t md_lh; - lnet_me_t *md_me; + struct lnet_libhandle md_lh; + struct lnet_me *md_me; char *md_start; unsigned int md_offset; unsigned int md_length; @@ -163,7 +175,8 @@ typedef struct lnet_libmd { unsigned int md_flags; unsigned int md_niov; /* # frags at end of struct */ void *md_user_ptr; - lnet_eq_t *md_eq; + struct lnet_eq *md_eq; + struct lnet_handle_md md_bulk_handle; union { struct kvec iov[LNET_MAX_IOV]; lnet_kiov_t kiov[LNET_MAX_IOV]; @@ -174,7 +187,7 @@ typedef struct lnet_libmd { #define LNET_MD_FLAG_AUTO_UNLINK (1 << 1) #define LNET_MD_FLAG_ABORTED (1 << 2) -typedef struct { +typedef struct lnet_test_peer { /* info about peers we are trying to fail */ struct list_head tp_list; /* ln_test_peers */ lnet_nid_t tp_nid; /* matching nid */ @@ -190,8 +203,7 @@ typedef struct { struct lnet_ni; /* forward ref */ struct socket; -typedef struct lnet_lnd -{ +typedef struct lnet_lnd { /* fields managed by portals */ struct list_head lnd_list; /* stash in the LND table */ int lnd_refcount; /* # active instances */ @@ -219,14 +231,15 @@ typedef struct lnet_lnd * and 'private' is the 'private' passed to lnet_parse(). Return * non-zero for immediate failure, otherwise complete later with * lnet_finalize() */ - int (*lnd_send)(struct lnet_ni *ni, void *private, lnet_msg_t *msg); + int (*lnd_send)(struct lnet_ni *ni, void *private, + struct lnet_msg *msg); /* Start receiving 'mlen' bytes of payload data, skipping the following * 'rlen' - 'mlen' bytes. 'private' is the 'private' passed to * lnet_parse(). Return non-zero for immedaite failure, otherwise * complete later with lnet_finalize(). This also gives back a receive * credit if the LND does flow control. */ - int (*lnd_recv)(struct lnet_ni *ni, void *private, lnet_msg_t *msg, + int (*lnd_recv)(struct lnet_ni *ni, void *private, struct lnet_msg *msg, int delayed, unsigned int niov, struct kvec *iov, lnet_kiov_t *kiov, unsigned int offset, unsigned int mlen, unsigned int rlen); @@ -237,8 +250,8 @@ typedef struct lnet_lnd * for success and do NOT give back a receive credit; that has to wait * until lnd_recv() gets called. On failure return < 0 and * release resources; lnd_recv() will not be called. */ - int (*lnd_eager_recv)(struct lnet_ni *ni, void *private, lnet_msg_t *msg, - void **new_privatep); + int (*lnd_eager_recv)(struct lnet_ni *ni, void *private, + struct lnet_msg *msg, void **new_privatep); /* notification of peer health */ void (*lnd_notify)(struct lnet_ni *ni, lnet_nid_t peer, int alive); @@ -250,12 +263,6 @@ typedef struct lnet_lnd int (*lnd_accept)(struct lnet_ni *ni, struct socket *sock); } lnd_t; -typedef struct { - lnet_nid_t ns_nid; - __u32 ns_status; - __u32 ns_unused; -} WIRE_ATTR lnet_ni_status_t; - struct lnet_tx_queue { int tq_credits; /* # tx credits free */ int tq_credits_min; /* lowest it's been */ @@ -263,141 +270,452 @@ struct lnet_tx_queue { struct list_head tq_delayed; /* delayed TXs */ }; +enum lnet_net_state { + /* set when net block is allocated */ + LNET_NET_STATE_INIT = 0, + /* set when NIs in net are started successfully */ + LNET_NET_STATE_ACTIVE, + /* set if all NIs in net are in FAILED state */ + LNET_NET_STATE_INACTIVE, + /* set when shutting down a NET */ + LNET_NET_STATE_DELETING +}; + +enum lnet_ni_state { + /* set when NI block is allocated */ + LNET_NI_STATE_INIT = 0, + /* set when NI is started successfully */ + LNET_NI_STATE_ACTIVE, + /* set when LND notifies NI failed */ + LNET_NI_STATE_FAILED, + /* set when LND notifies NI degraded */ + LNET_NI_STATE_DEGRADED, + /* set when shuttding down NI */ + LNET_NI_STATE_DELETING +}; + +enum lnet_stats_type { + LNET_STATS_TYPE_SEND = 0, + LNET_STATS_TYPE_RECV, + LNET_STATS_TYPE_DROP +}; + +struct lnet_comm_count { + atomic_t co_get_count; + atomic_t co_put_count; + atomic_t co_reply_count; + atomic_t co_ack_count; + atomic_t co_hello_count; +}; + +struct lnet_element_stats { + struct lnet_comm_count el_send_stats; + struct lnet_comm_count el_recv_stats; + struct lnet_comm_count el_drop_stats; +}; + +struct lnet_net { + /* chain on the ln_nets */ + struct list_head net_list; + + /* net ID, which is composed of + * (net_type << 16) | net_num. + * net_type can be one of the enumerated types defined in + * lnet/include/lnet/nidstr.h */ + __u32 net_id; + + /* priority of the network */ + __u32 net_prio; + + /* total number of CPTs in the array */ + __u32 net_ncpts; + + /* cumulative CPTs of all NIs in this net */ + __u32 *net_cpts; + + /* network tunables */ + struct lnet_ioctl_config_lnd_cmn_tunables net_tunables; + + /* + * boolean to indicate that the tunables have been set and + * shouldn't be reset + */ + bool net_tunables_set; + + /* procedural interface */ + struct lnet_lnd *net_lnd; + + /* list of NIs on this net */ + struct list_head net_ni_list; + + /* list of NIs being added, but not started yet */ + struct list_head net_ni_added; + + /* dying LND instances */ + struct list_head net_ni_zombie; + + /* network state */ + enum lnet_net_state net_state; +}; + typedef struct lnet_ni { + /* chain on the lnet_net structure */ + struct list_head ni_netlist; + + /* chain on net_ni_cpt */ + struct list_head ni_cptlist; + spinlock_t ni_lock; - struct list_head ni_list; /* chain on ln_nis */ - struct list_head ni_cptlist; /* chain on ln_nis_cpt */ - int ni_maxtxcredits; /* # tx credits */ - /* # per-peer send credits */ - int ni_peertxcredits; - /* # per-peer router buffer credits */ - int ni_peerrtrcredits; - /* seconds to consider peer dead */ - int ni_peertimeout; - int ni_ncpts; /* number of CPTs */ - __u32 *ni_cpts; /* bond NI on some CPTs */ - lnet_nid_t ni_nid; /* interface's NID */ - void *ni_data; /* instance-specific data */ - lnd_t *ni_lnd; /* procedural interface */ - struct lnet_tx_queue **ni_tx_queues; /* percpt TX queues */ - int **ni_refs; /* percpt reference count */ - time64_t ni_last_alive; /* when I was last alive */ - lnet_ni_status_t *ni_status; /* my health status */ + + /* number of CPTs */ + int ni_ncpts; + + /* bond NI on some CPTs */ + __u32 *ni_cpts; + + /* interface's NID */ + lnet_nid_t ni_nid; + + /* instance-specific data */ + void *ni_data; + + /* per ni credits */ + atomic_t ni_tx_credits; + + /* percpt TX queues */ + struct lnet_tx_queue **ni_tx_queues; + + /* percpt reference count */ + int **ni_refs; + + /* when I was last alive */ + long ni_last_alive; + + /* pointer to parent network */ + struct lnet_net *ni_net; + + /* my health status */ + struct lnet_ni_status *ni_status; + + /* NI FSM */ + enum lnet_ni_state ni_state; + /* per NI LND tunables */ - struct lnet_ioctl_config_lnd_tunables *ni_lnd_tunables; - /* equivalent interfaces to use */ - char *ni_interfaces[LNET_MAX_INTERFACES]; + struct lnet_lnd_tunables ni_lnd_tunables; + + /* lnd tunables set explicitly */ + bool ni_lnd_tunables_set; + + /* NI statistics */ + struct lnet_element_stats ni_stats; + + /* physical device CPT */ + int ni_dev_cpt; + + /* sequence number used to round robin over nis within a net */ + __u32 ni_seq; + + /* + * equivalent interfaces to use + * This is an array because socklnd bonding can still be configured + */ + char *ni_interfaces[LNET_INTERFACES_NUM]; struct net *ni_net_ns; /* original net namespace */ } lnet_ni_t; #define LNET_PROTO_PING_MATCHBITS 0x8000000000000000LL -/* NB: value of these features equal to LNET_PROTO_PING_VERSION_x - * of old LNet, so there shouldn't be any compatibility issue */ -#define LNET_PING_FEAT_INVAL (0) /* no feature */ -#define LNET_PING_FEAT_BASE (1 << 0) /* just a ping */ -#define LNET_PING_FEAT_NI_STATUS (1 << 1) /* return NI status */ -#define LNET_PING_FEAT_RTE_DISABLED (1 << 2) /* Routing enabled */ +/* + * Descriptor of a ping info buffer: keep a separate indicator of the + * size and a reference count. The type is used both as a source and + * sink of data, so we need to keep some information outside of the + * area that may be overwritten by network data. + */ +struct lnet_ping_buffer { + int pb_nnis; + atomic_t pb_refcnt; + struct lnet_ping_info pb_info; +}; -#define LNET_PING_FEAT_MASK (LNET_PING_FEAT_BASE | \ - LNET_PING_FEAT_NI_STATUS) +#define LNET_PING_BUFFER_SIZE(NNIDS) \ + offsetof(struct lnet_ping_buffer, pb_info.pi_ni[NNIDS]) +#define LNET_PING_BUFFER_LONI(PBUF) ((PBUF)->pb_info.pi_ni[0].ns_nid) +#define LNET_PING_BUFFER_SEQNO(PBUF) ((PBUF)->pb_info.pi_ni[0].ns_status) -typedef struct { - __u32 pi_magic; - __u32 pi_features; - lnet_pid_t pi_pid; - __u32 pi_nnis; - lnet_ni_status_t pi_ni[0]; -} WIRE_ATTR lnet_ping_info_t; +#define LNET_PING_INFO_TO_BUFFER(PINFO) \ + container_of((PINFO), struct lnet_ping_buffer, pb_info) /* router checker data, per router */ -#define LNET_MAX_RTR_NIS 16 -#define LNET_PINGINFO_SIZE offsetof(lnet_ping_info_t, pi_ni[LNET_MAX_RTR_NIS]) -typedef struct { +typedef struct lnet_rc_data { /* chain on the_lnet.ln_zombie_rcd or ln_deathrow_rcd */ struct list_head rcd_list; - lnet_handle_md_t rcd_mdh; /* ping buffer MD */ - struct lnet_peer *rcd_gateway; /* reference to gateway */ - lnet_ping_info_t *rcd_pinginfo; /* ping buffer */ + struct lnet_handle_md rcd_mdh; /* ping buffer MD */ + struct lnet_peer_ni *rcd_gateway; /* reference to gateway */ + struct lnet_ping_buffer *rcd_pingbuffer;/* ping buffer */ + int rcd_nnis; /* desired size of buffer */ } lnet_rc_data_t; -typedef struct lnet_peer { +struct lnet_peer_ni { + /* chain on lpn_peer_nis */ + struct list_head lpni_peer_nis; + /* chain on remote peer list */ + struct list_head lpni_on_remote_peer_ni_list; /* chain on peer hash */ - struct list_head lp_hashlist; + struct list_head lpni_hashlist; /* messages blocking for tx credits */ - struct list_head lp_txq; + struct list_head lpni_txq; /* messages blocking for router credits */ - struct list_head lp_rtrq; + struct list_head lpni_rtrq; /* chain on router list */ - struct list_head lp_rtr_list; + struct list_head lpni_rtr_list; + /* pointer to peer net I'm part of */ + struct lnet_peer_net *lpni_peer_net; + /* statistics kept on each peer NI */ + struct lnet_element_stats lpni_stats; + /* spin lock protecting credits and lpni_txq / lpni_rtrq */ + spinlock_t lpni_lock; /* # tx credits available */ - int lp_txcredits; + int lpni_txcredits; /* low water mark */ - int lp_mintxcredits; + int lpni_mintxcredits; /* # router credits */ - int lp_rtrcredits; + int lpni_rtrcredits; /* low water mark */ - int lp_minrtrcredits; + int lpni_minrtrcredits; + /* bytes queued for sending */ + long lpni_txqnob; /* alive/dead? */ - unsigned int lp_alive:1; + bool lpni_alive; /* notification outstanding? */ - unsigned int lp_notify:1; + bool lpni_notify; /* outstanding notification for LND? */ - unsigned int lp_notifylnd:1; + bool lpni_notifylnd; /* some thread is handling notification */ - unsigned int lp_notifying:1; + bool lpni_notifying; /* SEND event outstanding from ping */ - unsigned int lp_ping_notsent; - /* # times router went dead<->alive */ - int lp_alive_count; - /* bytes queued for sending */ - long lp_txqnob; + bool lpni_ping_notsent; + /* # times router went dead<->alive. Protected with lpni_lock */ + int lpni_alive_count; /* time of last aliveness news */ - cfs_time_t lp_timestamp; + cfs_time_t lpni_timestamp; /* time of last ping attempt */ - cfs_time_t lp_ping_timestamp; + cfs_time_t lpni_ping_timestamp; /* != 0 if ping reply expected */ - cfs_time_t lp_ping_deadline; + cfs_time_t lpni_ping_deadline; /* when I was last alive */ - cfs_time_t lp_last_alive; - /* when lp_ni was queried last time */ - cfs_time_t lp_last_query; - /* interface peer is on */ - lnet_ni_t *lp_ni; - lnet_nid_t lp_nid; /* peer's NID */ - int lp_refcount; /* # refs */ - int lp_cpt; /* CPT this peer attached on */ + cfs_time_t lpni_last_alive; + /* when lpni_ni was queried last time */ + cfs_time_t lpni_last_query; + /* network peer is on */ + struct lnet_net *lpni_net; + /* peer's NID */ + lnet_nid_t lpni_nid; + /* # refs */ + atomic_t lpni_refcount; + /* CPT this peer attached on */ + int lpni_cpt; + /* state flags -- protected by lpni_lock */ + unsigned lpni_state; /* # refs from lnet_route_t::lr_gateway */ - int lp_rtr_refcount; - /* returned RC ping features */ - unsigned int lp_ping_feats; - struct list_head lp_routes; /* routers on this peer */ - lnet_rc_data_t *lp_rcd; /* router checker state */ -} lnet_peer_t; + int lpni_rtr_refcount; + /* sequence number used to round robin over peer nis within a net */ + __u32 lpni_seq; + /* sequence number used to round robin over gateways */ + __u32 lpni_gw_seq; + /* health flag */ + bool lpni_healthy; + /* returned RC ping features. Protected with lpni_lock */ + unsigned int lpni_ping_feats; + /* routes on this peer */ + struct list_head lpni_routes; + /* preferred local nids: if only one, use lpni_pref.nid */ + union lpni_pref { + lnet_nid_t nid; + lnet_nid_t *nids; + } lpni_pref; + /* number of preferred NIDs in lnpi_pref_nids */ + __u32 lpni_pref_nnids; + /* router checker state */ + struct lnet_rc_data *lpni_rcd; +}; + +/* Preferred path added due to traffic on non-MR peer_ni */ +#define LNET_PEER_NI_NON_MR_PREF (1 << 0) + +struct lnet_peer { + /* chain on pt_peer_list */ + struct list_head lp_peer_list; + + /* list of peer nets */ + struct list_head lp_peer_nets; + + /* list of messages pending discovery*/ + struct list_head lp_dc_pendq; + + /* primary NID of the peer */ + lnet_nid_t lp_primary_nid; + + /* CPT of peer_table */ + int lp_cpt; + + /* number of NIDs on this peer */ + int lp_nnis; + + /* reference count */ + atomic_t lp_refcount; + + /* lock protecting peer state flags */ + spinlock_t lp_lock; + + /* peer state flags */ + unsigned lp_state; + + /* buffer for data pushed by peer */ + struct lnet_ping_buffer *lp_data; + + /* MD handle for ping in progress */ + lnet_handle_md_t lp_ping_mdh; + + /* MD handle for push in progress */ + lnet_handle_md_t lp_push_mdh; + + /* number of NIDs for sizing push data */ + int lp_data_nnis; + + /* NI config sequence number of peer */ + __u32 lp_peer_seqno; + + /* Local NI config sequence number acked by peer */ + __u32 lp_node_seqno; + + /* Local NI config sequence number sent to peer */ + __u32 lp_node_seqno_sent; + + /* Ping error encountered during discovery. */ + int lp_ping_error; + + /* Push error encountered during discovery. */ + int lp_push_error; + + /* Error encountered during discovery. */ + int lp_dc_error; + + /* time it was put on the ln_dc_working queue */ + time64_t lp_last_queued; + + /* link on discovery-related lists */ + struct list_head lp_dc_list; + + /* tasks waiting on discovery of this peer */ + wait_queue_head_t lp_dc_waitq; +}; + +/* + * The status flags in lp_state. Their semantics have chosen so that + * lp_state can be zero-initialized. + * + * A peer is marked MULTI_RAIL in two cases: it was configured using DLC + * as multi-rail aware, or the LNET_PING_FEAT_MULTI_RAIL bit was set. + * + * A peer is marked NO_DISCOVERY if the LNET_PING_FEAT_DISCOVERY bit was + * NOT set when the peer was pinged by discovery. + */ +#define LNET_PEER_MULTI_RAIL (1 << 0) /* Multi-rail aware */ +#define LNET_PEER_NO_DISCOVERY (1 << 1) /* Peer disabled discovery */ +/* + * A peer is marked CONFIGURED if it was configured by DLC. + * + * In addition, a peer is marked DISCOVERED if it has fully passed + * through Peer Discovery. + * + * When Peer Discovery is disabled, the discovery thread will mark + * peers REDISCOVER to indicate that they should be re-examined if + * discovery is (re)enabled on the node. + * + * A peer that was created as the result of inbound traffic will not + * be marked at all. + */ +#define LNET_PEER_CONFIGURED (1 << 2) /* Configured via DLC */ +#define LNET_PEER_DISCOVERED (1 << 3) /* Peer was discovered */ +#define LNET_PEER_REDISCOVER (1 << 4) /* Discovery was disabled */ +/* + * A peer is marked DISCOVERING when discovery is in progress. + * The other flags below correspond to stages of discovery. + */ +#define LNET_PEER_DISCOVERING (1 << 5) /* Discovering */ +#define LNET_PEER_DATA_PRESENT (1 << 6) /* Remote peer data present */ +#define LNET_PEER_NIDS_UPTODATE (1 << 7) /* Remote peer info uptodate */ +#define LNET_PEER_PING_SENT (1 << 8) /* Waiting for REPLY to Ping */ +#define LNET_PEER_PUSH_SENT (1 << 9) /* Waiting for ACK of Push */ +#define LNET_PEER_PING_FAILED (1 << 10) /* Ping send failure */ +#define LNET_PEER_PUSH_FAILED (1 << 11) /* Push send failure */ +/* + * A ping can be forced as a way to fix up state, or as a manual + * intervention by an admin. + * A push can be forced in circumstances that would normally not + * allow for one to happen. + */ +#define LNET_PEER_FORCE_PING (1 << 12) /* Forced Ping */ +#define LNET_PEER_FORCE_PUSH (1 << 13) /* Forced Push */ + +struct lnet_peer_net { + /* chain on lp_peer_nets */ + struct list_head lpn_peer_nets; + + /* list of peer_nis on this network */ + struct list_head lpn_peer_nis; + + /* pointer to the peer I'm part of */ + struct lnet_peer *lpn_peer; + + /* Net ID */ + __u32 lpn_net_id; + + /* reference count */ + atomic_t lpn_refcount; +}; /* peer hash size */ #define LNET_PEER_HASH_BITS 9 #define LNET_PEER_HASH_SIZE (1 << LNET_PEER_HASH_BITS) -/* peer hash table */ +/* + * peer hash table - one per CPT + * + * protected by lnet_net_lock/EX for update + * pt_version + * pt_number + * pt_hash[...] + * pt_peer_list + * pt_peers + * protected by pt_zombie_lock: + * pt_zombie_list + * pt_zombies + * + * pt_zombie lock nests inside lnet_net_lock + */ struct lnet_peer_table { int pt_version; /* /proc validity stamp */ - int pt_number; /* # peers extant */ - int pt_zombies; /* # zombies to go to deathrow - * (and not there yet) */ - struct list_head pt_deathrow; /* zombie peers */ + int pt_number; /* # peers_ni extant */ struct list_head *pt_hash; /* NID->peer hash */ + struct list_head pt_peer_list; /* peers */ + int pt_peers; /* # peers */ + struct list_head pt_zombie_list; /* zombie peer_ni */ + int pt_zombies; /* # zombie peers_ni */ + spinlock_t pt_zombie_lock; /* protect list and count */ }; /* peer aliveness is enabled only on routers for peers in a network where the - * lnet_ni_t::ni_peertimeout has been set to a positive value */ + * struct lnet_ni::ni_peertimeout has been set to a positive value + */ #define lnet_peer_aliveness_enabled(lp) (the_lnet.ln_routing != 0 && \ - (lp)->lp_ni->ni_peertimeout > 0) + ((lp)->lpni_net) && \ + (lp)->lpni_net->net_tunables.lct_peer_timeout > 0) -typedef struct { +typedef struct lnet_route { struct list_head lr_list; /* chain on net */ struct list_head lr_gwlist; /* chain on gateway */ - lnet_peer_t *lr_gateway; /* router node */ + struct lnet_peer_ni *lr_gateway; /* router node */ __u32 lr_net; /* remote network number */ int lr_seq; /* sequence for round-robin */ unsigned int lr_downis; /* number of down NIs */ @@ -409,7 +727,7 @@ typedef struct { #define LNET_REMOTE_NETS_HASH_MAX (1U << 16) #define LNET_REMOTE_NETS_HASH_SIZE (1 << the_lnet.ln_remote_nets_hbits) -typedef struct { +typedef struct lnet_remotenet { /* chain on ln_remote_nets_hash */ struct list_head lrn_list; /* routes to me */ @@ -422,8 +740,10 @@ typedef struct { #define LNET_CREDIT_OK 0 /** lnet message is waiting for credit */ #define LNET_CREDIT_WAIT 1 +/** lnet message is waiting for discovery */ +#define LNET_DC_WAIT 2 -typedef struct { +typedef struct lnet_rtrbufpool { /* my free buffer pool */ struct list_head rbp_bufs; /* messages blocking for a buffer */ @@ -440,15 +760,15 @@ typedef struct { int rbp_mincredits; } lnet_rtrbufpool_t; -typedef struct { +typedef struct lnet_rtrbuf { struct list_head rb_list; /* chain on rbp_bufs */ - lnet_rtrbufpool_t *rb_pool; /* owning pool */ + struct lnet_rtrbufpool *rb_pool; /* owning pool */ lnet_kiov_t rb_kiov[0]; /* the buffer space */ } lnet_rtrbuf_t; #define LNET_PEER_HASHSIZE 503 /* prime! */ -enum { +enum lnet_match_flags { /* Didn't match anything */ LNET_MATCHMD_NONE = (1 << 0), /* Matched OK */ @@ -461,7 +781,7 @@ enum { LNET_MATCHMD_FINISH = (LNET_MATCHMD_OK | LNET_MATCHMD_DROP), }; -/* Options for lnet_portal_t::ptl_options */ +/* Options for struct lnet_portal::ptl_options */ #define LNET_PTL_LAZY (1 << 0) #define LNET_PTL_MATCH_UNIQUE (1 << 1) /* unique match, for RDMA */ #define LNET_PTL_MATCH_WILDCARD (1 << 2) /* wildcard match, request portal */ @@ -469,7 +789,8 @@ enum { /* parameter for matching operations (GET, PUT) */ struct lnet_match_info { __u64 mi_mbits; - lnet_process_id_t mi_id; + struct lnet_process_id mi_id; + unsigned int mi_cpt; unsigned int mi_opc; unsigned int mi_portal; unsigned int mi_rlength; @@ -556,13 +877,22 @@ struct lnet_msg_container { void **msc_finalizers; }; +/* Peer Discovery states */ +#define LNET_DC_STATE_SHUTDOWN 0 /* not started */ +#define LNET_DC_STATE_RUNNING 1 /* started up OK */ +#define LNET_DC_STATE_STOPPING 2 /* telling thread to stop */ + /* Router Checker states */ #define LNET_RC_STATE_SHUTDOWN 0 /* not started */ #define LNET_RC_STATE_RUNNING 1 /* started up OK */ #define LNET_RC_STATE_STOPPING 2 /* telling thread to stop */ -typedef struct -{ +/* LNet states */ +#define LNET_STATE_SHUTDOWN 0 /* not started */ +#define LNET_STATE_RUNNING 1 /* started up OK */ +#define LNET_STATE_STOPPING 2 /* telling thread to stop */ + +typedef struct lnet { /* CPU partition table of LNet */ struct cfs_cpt_table *ln_cpt_table; /* number of CPTs in ln_cpt_table */ @@ -574,7 +904,7 @@ typedef struct /* # portals */ int ln_nportals; /* the vector of portals */ - lnet_portal_t **ln_portals; + struct lnet_portal **ln_portals; /* percpt ME containers */ struct lnet_res_container **ln_me_containers; /* percpt MD container */ @@ -591,19 +921,20 @@ typedef struct struct cfs_percpt_lock *ln_net_lock; /* percpt message containers for active/finalizing/freed message */ struct lnet_msg_container **ln_msg_containers; - lnet_counters_t **ln_counters; + struct lnet_counters **ln_counters; struct lnet_peer_table **ln_peer_tables; + /* list of peer nis not on a local network */ + struct list_head ln_remote_peer_ni_list; /* failure simulation */ struct list_head ln_test_peers; struct list_head ln_drop_rules; struct list_head ln_delay_rules; - - struct list_head ln_nis; /* LND instances */ - /* NIs bond on specific CPT(s) */ - struct list_head ln_nis_cpt; - /* dying LND instances */ - struct list_head ln_nis_zombie; - lnet_ni_t *ln_loni; /* the loopback NI */ + /* LND instances */ + struct list_head ln_nets; + /* the loopback NI */ + struct lnet_ni *ln_loni; + /* network zombie list */ + struct list_head ln_net_zombie; /* remote networks with routes to them */ struct list_head *ln_remote_nets_hash; @@ -614,16 +945,50 @@ typedef struct /* validity stamp */ __u64 ln_routers_version; /* percpt router buffer pools */ - lnet_rtrbufpool_t **ln_rtrpools; - - lnet_handle_md_t ln_ping_target_md; - lnet_handle_eq_t ln_ping_target_eq; - lnet_ping_info_t *ln_ping_info; + struct lnet_rtrbufpool **ln_rtrpools; + + /* + * Ping target / Push source + * + * The ping target and push source share a single buffer. The + * ln_ping_target is protected against concurrent updates by + * ln_api_mutex. + */ + struct lnet_handle_md ln_ping_target_md; + struct lnet_handle_eq ln_ping_target_eq; + struct lnet_ping_buffer *ln_ping_target; + atomic_t ln_ping_target_seqno; + + /* + * Push Target + * + * ln_push_nnis contains the desired size of the push target. + * The lnet_net_lock is used to handle update races. The old + * buffer may linger a while after it has been unlinked, in + * which case the event handler cleans up. + */ + lnet_handle_eq_t ln_push_target_eq; + lnet_handle_md_t ln_push_target_md; + struct lnet_ping_buffer *ln_push_target; + int ln_push_target_nnis; + + /* discovery event queue handle */ + lnet_handle_eq_t ln_dc_eqh; + /* discovery requests */ + struct list_head ln_dc_request; + /* discovery working list */ + struct list_head ln_dc_working; + /* discovery expired list */ + struct list_head ln_dc_expired; + /* discovery thread wait queue */ + wait_queue_head_t ln_dc_waitq; + /* discovery startup/shutdown state */ + int ln_dc_state; /* router checker startup/shutdown state */ int ln_rc_state; /* router checker's event queue */ - lnet_handle_eq_t ln_rc_eqh; + struct lnet_handle_eq ln_rc_eqh; /* rcd still pending on net */ struct list_head ln_rcd_deathrow; /* rcd ready for free */ @@ -637,8 +1002,8 @@ typedef struct int ln_niinit_self; /* LNetNIInit/LNetNIFini counter */ int ln_refcount; - /* shutdown in progress */ - int ln_shutdown; + /* SHUTDOWN/RUNNING/STOPPING */ + int ln_state; int ln_routing; /* am I a router? */ lnet_pid_t ln_pid; /* requested pid */