* Copyright (c) 2003, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2012, 2016, Intel Corporation.
+ * Copyright (c) 2012, 2017, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
#include <linux/kthread.h>
#include <linux/uio.h>
+#include <linux/semaphore.h>
#include <linux/types.h>
#include <uapi/linux/lnet/lnet-dlc.h>
#include <uapi/linux/lnet/lnetctl.h>
/* Max payload size */
-#ifndef CONFIG_LNET_MAX_PAYLOAD
-# error "CONFIG_LNET_MAX_PAYLOAD must be defined in config.h"
-#endif
+#define LNET_MAX_PAYLOAD LNET_MTU
-#define LNET_MAX_PAYLOAD CONFIG_LNET_MAX_PAYLOAD
-#if (LNET_MAX_PAYLOAD < LNET_MTU)
-# error "LNET_MAX_PAYLOAD too small - error in configure --with-max-payload-mb"
-#elif (LNET_MAX_PAYLOAD > (PAGE_SIZE * LNET_MAX_IOV))
-# error "LNET_MAX_PAYLOAD too large - error in configure --with-max-payload-mb"
-#endif
+#define LNET_MAX_IOV (LNET_MAX_PAYLOAD >> PAGE_SHIFT)
+
+/*
+ * This is the maximum health value.
+ * All local and peer NIs created have their health default to this value.
+ */
+#define LNET_MAX_HEALTH_VALUE 1000
/* forward refs */
struct lnet_libmd;
-typedef struct lnet_msg {
+enum lnet_msg_hstatus {
+ LNET_MSG_STATUS_OK = 0,
+ LNET_MSG_STATUS_LOCAL_INTERRUPT,
+ LNET_MSG_STATUS_LOCAL_DROPPED,
+ LNET_MSG_STATUS_LOCAL_ABORTED,
+ LNET_MSG_STATUS_LOCAL_NO_ROUTE,
+ LNET_MSG_STATUS_LOCAL_ERROR,
+ LNET_MSG_STATUS_LOCAL_TIMEOUT,
+ LNET_MSG_STATUS_REMOTE_ERROR,
+ LNET_MSG_STATUS_REMOTE_DROPPED,
+ LNET_MSG_STATUS_REMOTE_TIMEOUT,
+ LNET_MSG_STATUS_NETWORK_TIMEOUT,
+ LNET_MSG_STATUS_END,
+};
+
+struct lnet_rsp_tracker {
+ /* chain on the waiting list */
+ struct list_head rspt_on_list;
+ /* cpt to lock */
+ int rspt_cpt;
+ /* deadline of the REPLY/ACK */
+ ktime_t rspt_deadline;
+ /* parent MD */
+ struct lnet_handle_md rspt_mdh;
+};
+
+struct lnet_msg {
struct list_head msg_activelist;
struct list_head msg_list; /* Q for credits/MD */
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;
+
+ /*
+ * Deadline for the message after which it will be finalized if it
+ * has not completed.
+ */
+ ktime_t msg_deadline;
+
+ /* The message health status. */
+ enum lnet_msg_hstatus msg_health_status;
+ /* This is a recovery message */
+ bool msg_recovery;
+ /* the number of times a transmission has been retried */
+ int msg_retry_count;
+ /* flag to indicate that we do not want to resend this message */
+ bool msg_no_resend;
+
/* committed for sending */
unsigned int msg_tx_committed:1;
/* CPT # this message committed for sending */
struct lnet_event msg_ev;
struct lnet_hdr msg_hdr;
-} lnet_msg_t;
+};
-typedef struct lnet_libhandle {
+struct lnet_libhandle {
struct list_head lh_hash_chain;
__u64 lh_cookie;
-} lnet_libhandle_t;
+};
#define lh_entry(ptr, type, member) \
((type *)((char *)(ptr)-(char *)(&((type *)0)->member)))
-typedef struct lnet_eq {
+struct lnet_eq {
struct list_head eq_list;
struct lnet_libhandle eq_lh;
unsigned long eq_enq_seq;
lnet_eq_handler_t eq_callback;
struct lnet_event *eq_events;
int **eq_refs; /* percpt refcount for EQ */
-} lnet_eq_t;
+};
-typedef struct lnet_me {
+struct lnet_me {
struct list_head me_list;
struct lnet_libhandle me_lh;
struct lnet_process_id me_match_id;
__u64 me_ignore_bits;
enum lnet_unlink me_unlink;
struct lnet_libmd *me_md;
-} lnet_me_t;
-
-typedef struct lnet_libmd {
- struct list_head md_list;
- struct lnet_libhandle md_lh;
- struct lnet_me *md_me;
- char *md_start;
- unsigned int md_offset;
- unsigned int md_length;
- unsigned int md_max_size;
- int md_threshold;
- int md_refcount;
- unsigned int md_options;
- unsigned int md_flags;
- unsigned int md_niov; /* # frags at end of struct */
- void *md_user_ptr;
- struct lnet_eq *md_eq;
- struct lnet_handle_md md_bulk_handle;
+};
+
+struct lnet_libmd {
+ struct list_head md_list;
+ struct lnet_libhandle md_lh;
+ struct lnet_me *md_me;
+ char *md_start;
+ unsigned int md_offset;
+ unsigned int md_length;
+ unsigned int md_max_size;
+ int md_threshold;
+ int md_refcount;
+ unsigned int md_options;
+ unsigned int md_flags;
+ unsigned int md_niov; /* # frags at end of struct */
+ void *md_user_ptr;
+ struct lnet_rsp_tracker *md_rspt_ptr;
+ 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];
+ struct kvec iov[LNET_MAX_IOV];
+ lnet_kiov_t kiov[LNET_MAX_IOV];
} md_iov;
-} lnet_libmd_t;
+};
#define LNET_MD_FLAG_ZOMBIE (1 << 0)
#define LNET_MD_FLAG_AUTO_UNLINK (1 << 1)
#define LNET_MD_FLAG_ABORTED (1 << 2)
-typedef struct lnet_test_peer {
+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 */
unsigned int tp_threshold; /* # failures to simulate */
-} lnet_test_peer_t;
+};
#define LNET_COOKIE_TYPE_MD 1
#define LNET_COOKIE_TYPE_ME 2
struct lnet_ni; /* forward ref */
struct socket;
-typedef struct lnet_lnd {
+struct lnet_lnd {
/* fields managed by portals */
struct list_head lnd_list; /* stash in the LND table */
int lnd_refcount; /* # active instances */
void (*lnd_notify)(struct lnet_ni *ni, lnet_nid_t peer, int alive);
/* query of peer aliveness */
- void (*lnd_query)(struct lnet_ni *ni, lnet_nid_t peer, cfs_time_t *when);
+ void (*lnd_query)(struct lnet_ni *ni, lnet_nid_t peer, time64_t *when);
/* accept a new connection */
int (*lnd_accept)(struct lnet_ni *ni, struct socket *sock);
-} lnd_t;
+};
struct lnet_tx_queue {
int tq_credits; /* # tx credits free */
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
+#define LNET_NI_STATE_INIT (1 << 0)
+#define LNET_NI_STATE_ACTIVE (1 << 1)
+#define LNET_NI_STATE_FAILED (1 << 2)
+#define LNET_NI_STATE_RECOVERY_PENDING (1 << 3)
+#define LNET_NI_STATE_DELETING (1 << 4)
+
+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 {
- atomic_t send_count;
- atomic_t recv_count;
- atomic_t drop_count;
+ struct lnet_comm_count el_send_stats;
+ struct lnet_comm_count el_recv_stats;
+ struct lnet_comm_count el_drop_stats;
+};
+
+struct lnet_health_local_stats {
+ atomic_t hlt_local_interrupt;
+ atomic_t hlt_local_dropped;
+ atomic_t hlt_local_aborted;
+ atomic_t hlt_local_no_route;
+ atomic_t hlt_local_timeout;
+ atomic_t hlt_local_error;
+};
+
+struct lnet_health_remote_stats {
+ atomic_t hlt_remote_dropped;
+ atomic_t hlt_remote_timeout;
+ atomic_t hlt_remote_error;
+ atomic_t hlt_network_timeout;
};
struct lnet_net {
enum lnet_net_state net_state;
};
-typedef struct lnet_ni {
+struct lnet_ni {
/* chain on the lnet_net structure */
struct list_head ni_netlist;
/* chain on net_ni_cpt */
struct list_head ni_cptlist;
+ /* chain on the recovery queue */
+ struct list_head ni_recovery;
+
+ /* MD handle for recovery ping */
+ struct lnet_handle_md ni_ping_mdh;
+
spinlock_t ni_lock;
/* number of CPTs */
int **ni_refs;
/* when I was last alive */
- long ni_last_alive;
+ time64_t ni_last_alive;
/* pointer to parent network */
struct lnet_net *ni_net;
struct lnet_ni_status *ni_status;
/* NI FSM */
- enum lnet_ni_state ni_state;
+ __u32 ni_state;
/* per NI LND tunables */
struct lnet_lnd_tunables ni_lnd_tunables;
/* NI statistics */
struct lnet_element_stats ni_stats;
+ struct lnet_health_local_stats ni_hstats;
/* physical device CPT */
int ni_dev_cpt;
__u32 ni_seq;
/*
+ * health value
+ * initialized to LNET_MAX_HEALTH_VALUE
+ * Value is decremented every time we fail to send a message over
+ * this NI because of a NI specific failure.
+ * Value is incremented if we successfully send a message.
+ */
+ atomic_t ni_healthv;
+
+ /*
+ * Set to 1 by the LND when it receives an event telling it the device
+ * has gone into a fatal state. Set to 0 when the LND receives an
+ * even telling it the device is back online.
+ */
+ atomic_t ni_fatal_error_on;
+
+ /*
* 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
#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)
+#define LNET_PING_INFO_TO_BUFFER(PINFO) \
+ container_of((PINFO), struct lnet_ping_buffer, pb_info)
/* router checker data, per router */
-typedef struct lnet_rc_data {
+struct lnet_rc_data {
/* chain on the_lnet.ln_zombie_rcd or ln_deathrow_rcd */
struct list_head rcd_list;
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;
+};
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 recovery queue */
+ struct list_head lpni_recovery;
/* chain on peer hash */
struct list_head lpni_hashlist;
/* messages blocking for tx credits */
struct lnet_peer_net *lpni_peer_net;
/* statistics kept on each peer NI */
struct lnet_element_stats lpni_stats;
+ struct lnet_health_remote_stats lpni_hstats;
/* spin lock protecting credits and lpni_txq / lpni_rtrq */
spinlock_t lpni_lock;
/* # tx credits available */
/* # times router went dead<->alive. Protected with lpni_lock */
int lpni_alive_count;
/* time of last aliveness news */
- cfs_time_t lpni_timestamp;
+ time64_t lpni_timestamp;
/* time of last ping attempt */
- cfs_time_t lpni_ping_timestamp;
+ time64_t lpni_ping_timestamp;
/* != 0 if ping reply expected */
- cfs_time_t lpni_ping_deadline;
+ time64_t lpni_ping_deadline;
/* when I was last alive */
- cfs_time_t lpni_last_alive;
+ time64_t lpni_last_alive;
/* when lpni_ni was queried last time */
- cfs_time_t lpni_last_query;
+ time64_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;
+ /* health value for the peer */
+ atomic_t lpni_healthv;
+ /* recovery ping mdh */
+ struct lnet_handle_md lpni_recovery_ping_mdh;
/* CPT this peer attached on */
int lpni_cpt;
/* state flags -- protected by lpni_lock */
/* Preferred path added due to traffic on non-MR peer_ni */
#define LNET_PEER_NI_NON_MR_PREF (1 << 0)
+/* peer is being recovered. */
+#define LNET_PEER_NI_RECOVERY_PENDING (1 << 1)
+/* peer is being deleted */
+#define LNET_PEER_NI_DELETING (1 << 2)
struct lnet_peer {
/* chain on pt_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;
/* buffer for data pushed by peer */
struct lnet_ping_buffer *lp_data;
+ /* MD handle for ping in progress */
+ struct lnet_handle_md lp_ping_mdh;
+
+ /* MD handle for push in progress */
+ struct lnet_handle_md 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 peer knows */
+ /* 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;
* pt_hash[...]
* pt_peer_list
* pt_peers
- * pt_peer_nnids
* protected by pt_zombie_lock:
* pt_zombie_list
* pt_zombies
struct list_head *pt_hash; /* NID->peer hash */
struct list_head pt_peer_list; /* peers */
int pt_peers; /* # peers */
- int pt_peer_nnids; /* # NIDS on listed 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 */
((lp)->lpni_net) && \
(lp)->lpni_net->net_tunables.lct_peer_timeout > 0)
-typedef struct lnet_route {
+struct lnet_route {
struct list_head lr_list; /* chain on net */
struct list_head lr_gwlist; /* chain on gateway */
struct lnet_peer_ni *lr_gateway; /* router node */
unsigned int lr_downis; /* number of down NIs */
__u32 lr_hops; /* how far I am */
unsigned int lr_priority; /* route priority */
-} lnet_route_t;
+};
#define LNET_REMOTE_NETS_HASH_DEFAULT (1U << 7)
#define LNET_REMOTE_NETS_HASH_MAX (1U << 16)
#define LNET_REMOTE_NETS_HASH_SIZE (1 << the_lnet.ln_remote_nets_hbits)
-typedef struct lnet_remotenet {
+struct lnet_remotenet {
/* chain on ln_remote_nets_hash */
struct list_head lrn_list;
/* routes to me */
struct list_head lrn_routes;
/* my net number */
__u32 lrn_net;
-} lnet_remotenet_t;
+};
/** lnet message has credit and can be submitted to lnd for send/receive */
#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 lnet_rtrbufpool {
+struct lnet_rtrbufpool {
/* my free buffer pool */
struct list_head rbp_bufs;
/* messages blocking for a buffer */
int rbp_credits;
/* low water mark */
int rbp_mincredits;
-} lnet_rtrbufpool_t;
+};
-typedef struct lnet_rtrbuf {
+struct lnet_rtrbuf {
struct list_head rb_list; /* chain on rbp_bufs */
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! */
/* dispatch routed PUT message by hashing source NID for wildcard portals */
#define LNET_PTL_ROTOR_HASH_RT 3
-typedef struct lnet_portal {
+struct lnet_portal {
spinlock_t ptl_lock;
unsigned int ptl_index; /* portal ID, reserved */
/* flags on this portal: lazy, unique... */
int ptl_mt_nmaps;
/* array of active entries' cpu-partition-id */
int ptl_mt_maps[0];
-} lnet_portal_t;
+};
#define LNET_LH_HASH_BITS 12
#define LNET_LH_HASH_SIZE (1ULL << LNET_LH_HASH_BITS)
#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 */
+#define LNET_MT_STATE_SHUTDOWN 0 /* not started */
+#define LNET_MT_STATE_RUNNING 1 /* started up OK */
+#define LNET_MT_STATE_STOPPING 2 /* telling thread to stop */
/* 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 {
+struct lnet {
/* CPU partition table of LNet */
struct cfs_cpt_table *ln_cpt_table;
/* number of CPTs in ln_cpt_table */
struct lnet_ni *ln_loni;
/* network zombie list */
struct list_head ln_net_zombie;
+ /* resend messages list */
+ struct list_head ln_msg_resend;
+ /* spin lock to protect the msg resend list */
+ spinlock_t ln_msg_resend_lock;
/* remote networks with routes to them */
struct list_head *ln_remote_nets_hash;
* 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_handle_eq ln_push_target_eq;
+ struct lnet_handle_md 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;
+ struct lnet_handle_eq 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;
+ /* monitor thread startup/shutdown state */
+ int ln_mt_state;
/* router checker's event queue */
struct lnet_handle_eq ln_rc_eqh;
/* rcd still pending on net */
/* rcd ready for free */
struct list_head ln_rcd_zombie;
/* serialise startup/shutdown */
- struct semaphore ln_rc_signal;
+ struct semaphore ln_mt_signal;
struct mutex ln_api_mutex;
struct mutex ln_lnd_mutex;
*/
bool ln_nis_from_mod_params;
- /* waitq for router checker. As long as there are no routes in
- * the list, the router checker will sleep on this queue. when
- * routes are added the thread will wake up */
- wait_queue_head_t ln_rc_waitq;
-} lnet_t;
+ /*
+ * waitq for the monitor thread. The monitor thread takes care of
+ * checking routes, timedout messages and resending messages.
+ */
+ wait_queue_head_t ln_mt_waitq;
+
+ /* per-cpt resend queues */
+ struct list_head **ln_mt_resendqs;
+ /* local NIs to recover */
+ struct list_head ln_mt_localNIRecovq;
+ /* local NIs to recover */
+ struct list_head ln_mt_peerNIRecovq;
+ /*
+ * An array of queues for GET/PUT waiting for REPLY/ACK respectively.
+ * There are CPT number of queues. Since response trackers will be
+ * added on the fast path we can't afford to grab the exclusive
+ * net lock to protect these queues. The CPT will be calculated
+ * based on the mdh cookie.
+ */
+ struct list_head **ln_mt_rstq;
+ /* recovery eq handler */
+ struct lnet_handle_eq ln_mt_eqh;
+
+};
#endif
git://git.whamcloud.com / fs / lustre-release.git / blobdiff