-/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
- * vim:expandtab:shiftwidth=8:tabstop=8:
- *
+/*
* GPL HEADER START
*
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* You should have received a copy of the GNU General Public License
* version 2 along with this program; If not, see
- * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf
- *
- * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
- * CA 95054 USA or visit www.sun.com if you need additional information or
- * have any questions.
+ * http://www.gnu.org/licenses/gpl-2.0.html
*
* GPL HEADER END
*/
/*
- * Copyright 2008 Sun Microsystems, Inc. All rights reserved
+ * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
+ *
+ * Copyright (c) 2011, 2016, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
* Lustre is a trademark of Sun Microsystems, Inc.
*/
+/** \defgroup obd_import PtlRPC import definitions
+ * Imports are client-side representation of remote obd target.
+ *
+ * @{
+ */
#ifndef __IMPORT_H
#define __IMPORT_H
-#include <lustre_handles.h>
-#include <lustre/lustre_idl.h>
+/** \defgroup export export
+ *
+ * @{
+ */
+#include <lustre_handles.h>
+#include <uapi/linux/lustre/lustre_idl.h>
-/* Adaptive Timeout stuff */
+/**
+ * Adaptive Timeout stuff
+ *
+ * @{
+ */
#define D_ADAPTTO D_OTHER
#define AT_BINS 4 /* "bin" means "N seconds of history" */
#define AT_FLG_NOHIST 0x1 /* use last reported value only */
struct adaptive_timeout {
- time_t at_binstart; /* bin start time */
- unsigned int at_hist[AT_BINS]; /* timeout history bins */
- unsigned int at_flags;
- unsigned int at_current; /* current timeout value */
- unsigned int at_worst_ever; /* worst-ever timeout value */
- time_t at_worst_time; /* worst-ever timeout timestamp */
- spinlock_t at_lock;
+ time64_t at_binstart; /* bin start time */
+ unsigned int at_hist[AT_BINS]; /* timeout history bins */
+ unsigned int at_flags;
+ unsigned int at_current; /* current timeout value */
+ unsigned int at_worst_ever; /* worst-ever timeout value */
+ time64_t at_worst_time; /* worst-ever timeout timestamp */
+ spinlock_t at_lock;
+};
+
+enum lustre_at_flags {
+ LATF_SKIP = 0x0,
+ LATF_STATS = 0x1,
+};
+
+struct ptlrpc_at_array {
+ struct list_head *paa_reqs_array; /** array to hold requests */
+ __u32 paa_size; /** the size of array */
+ __u32 paa_count; /** the total count of reqs */
+ time64_t paa_deadline; /** the earliest deadline of reqs */
+ __u32 *paa_reqs_count; /** the count of reqs in each entry */
};
+#define IMP_AT_MAX_PORTALS 8
+struct imp_at {
+ int iat_portal[IMP_AT_MAX_PORTALS];
+ struct adaptive_timeout iat_net_latency;
+ struct adaptive_timeout iat_service_estimate[IMP_AT_MAX_PORTALS];
+};
+
+
+/** @} */
+
+/** Possible import states */
enum lustre_imp_state {
LUSTRE_IMP_CLOSED = 1,
LUSTRE_IMP_NEW = 2,
LUSTRE_IMP_EVICTED = 10,
};
+/** Returns test string representation of numeric import state \a state */
static inline char * ptlrpc_import_state_name(enum lustre_imp_state state)
{
static char* import_state_names[] = {
return import_state_names[state];
}
+/**
+ * List of import event types
+ */
enum obd_import_event {
IMP_EVENT_DISCON = 0x808001,
IMP_EVENT_INACTIVE = 0x808002,
IMP_EVENT_INVALIDATE = 0x808003,
IMP_EVENT_ACTIVE = 0x808004,
IMP_EVENT_OCD = 0x808005,
+ IMP_EVENT_DEACTIVATE = 0x808006,
+ IMP_EVENT_ACTIVATE = 0x808007,
};
+/**
+ * Definition of import connection structure
+ */
struct obd_import_conn {
- struct list_head oic_item;
+ /** Item for linking connections together */
+ struct list_head oic_item;
+ /** Pointer to actual PortalRPC connection */
struct ptlrpc_connection *oic_conn;
+ /** uuid of remote side */
struct obd_uuid oic_uuid;
- __u64 oic_last_attempt; /* jiffies, 64-bit */
+ /**
+ * Time (64 bit jiffies) of last connection attempt on this connection
+ */
+ __u64 oic_last_attempt;
};
-#define IMP_AT_MAX_PORTALS 8
-struct imp_at {
- int iat_portal[IMP_AT_MAX_PORTALS];
- struct adaptive_timeout iat_net_latency;
- struct adaptive_timeout iat_service_estimate[IMP_AT_MAX_PORTALS];
+/* state history */
+#define IMP_STATE_HIST_LEN 16
+struct import_state_hist {
+ enum lustre_imp_state ish_state;
+ time64_t ish_time;
};
+/**
+ * Defintion of PortalRPC import structure.
+ * Imports are representing client-side view to remote target.
+ */
struct obd_import {
- struct portals_handle imp_handle;
- atomic_t imp_refcount;
- struct lustre_handle imp_dlm_handle; /* client's ldlm export */
- struct ptlrpc_connection *imp_connection;
+ /** Local handle (== id) for this import. */
+ struct portals_handle imp_handle;
+ /** Reference counter */
+ atomic_t imp_refcount;
+ struct lustre_handle imp_dlm_handle; /* client's ldlm export */
+ /** Currently active connection */
+ struct ptlrpc_connection *imp_connection;
+ /** PortalRPC client structure for this import */
struct ptlrpc_client *imp_client;
- struct list_head imp_pinger_chain;
- struct list_head imp_zombie_chain; /* queue for destruction */
+ /** List element for linking into pinger chain */
+ struct list_head imp_pinger_chain;
+ /** List element for linking into chain for destruction */
+ struct list_head imp_zombie_chain;
- /* Lists of requests that are retained for replay, waiting for a reply,
+ /**
+ * Lists of requests that are retained for replay, waiting for a reply,
* or waiting for recovery to complete, respectively.
+ * @{
+ */
+ struct list_head imp_replay_list;
+ struct list_head imp_sending_list;
+ struct list_head imp_delayed_list;
+ /** @} */
+
+ /**
+ * List of requests that are retained for committed open replay. Once
+ * open is committed, open replay request will be moved from the
+ * imp_replay_list into the imp_committed_list.
+ * The imp_replay_cursor is for accelerating searching during replay.
+ * @{
+ */
+ struct list_head imp_committed_list;
+ struct list_head *imp_replay_cursor;
+ /** @} */
+
+ /** List of not replied requests */
+ struct list_head imp_unreplied_list;
+ /** Known maximal replied XID */
+ __u64 imp_known_replied_xid;
+
+ /** obd device for this import */
+ struct obd_device *imp_obd;
+
+ /**
+ * some seciruty-related fields
+ * @{
*/
- struct list_head imp_replay_list;
- struct list_head imp_sending_list;
- struct list_head imp_delayed_list;
-
- struct obd_device *imp_obd;
- struct ptlrpc_sec *imp_sec;
- struct semaphore imp_sec_mutex;
- cfs_time_t imp_sec_expire;
- cfs_waitq_t imp_recovery_waitq;
-
- atomic_t imp_inflight;
- atomic_t imp_unregistering;
- atomic_t imp_replay_inflight;
- atomic_t imp_inval_count;
+ struct ptlrpc_sec *imp_sec;
+ struct mutex imp_sec_mutex;
+ time64_t imp_sec_expire;
+ pid_t imp_sec_refpid;
+ /** @} */
+
+ /** Wait queue for those who need to wait for recovery completion */
+ wait_queue_head_t imp_recovery_waitq;
+
+ /** Number of requests currently in-flight */
+ atomic_t imp_inflight;
+ /** Number of requests currently unregistering */
+ atomic_t imp_unregistering;
+ /** Number of replay requests inflight */
+ atomic_t imp_replay_inflight;
+ /** Number of currently happening import invalidations */
+ atomic_t imp_inval_count;
+ /** Numbner of request timeouts */
+ atomic_t imp_timeouts;
+ /** Current import state */
enum lustre_imp_state imp_state;
+ /** Last replay state */
+ enum lustre_imp_state imp_replay_state;
+ /** History of import states */
+ struct import_state_hist imp_state_hist[IMP_STATE_HIST_LEN];
+ int imp_state_hist_idx;
+ /** Current import generation. Incremented on every reconnect */
int imp_generation;
+ /** Incremented every time we send reconnection request */
__u32 imp_conn_cnt;
+ /**
+ * \see ptlrpc_free_committed remembers imp_generation value here
+ * after a check to save on unnecessary replay list iterations
+ */
int imp_last_generation_checked;
+ /** Last tranno we replayed */
__u64 imp_last_replay_transno;
+ /** Last transno committed on remote side */
__u64 imp_peer_committed_transno;
+ /**
+ * \see ptlrpc_free_committed remembers last_transno since its last
+ * check here and if last_transno did not change since last run of
+ * ptlrpc_free_committed and import generation is the same, we can
+ * skip looking for requests to remove from replay list as optimisation
+ */
__u64 imp_last_transno_checked;
+ /**
+ * Remote export handle. This is how remote side knows what export
+ * we are talking to. Filled from response to connect request
+ */
struct lustre_handle imp_remote_handle;
- cfs_time_t imp_next_ping; /* jiffies */
- __u64 imp_last_success_conn; /* jiffies, 64-bit */
+ /** When to perform next ping. time in jiffies. */
+ cfs_time_t imp_next_ping;
+ /** When we last successfully connected. time in 64bit jiffies */
+ __u64 imp_last_success_conn;
- /* all available obd_import_conn linked here */
- struct list_head imp_conn_list;
+ /** List of all possible connection for import. */
+ struct list_head imp_conn_list;
+ /**
+ * Current connection. \a imp_connection is imp_conn_current->oic_conn
+ */
struct obd_import_conn *imp_conn_current;
- /* Protects flags, level, generation, conn_cnt, *_list */
- spinlock_t imp_lock;
-
- /* flags */
- unsigned long imp_no_timeout:1, /* timeouts are disabled */
- imp_invalid:1, /* evicted */
- imp_deactive:1, /* administratively disabled */
- imp_replayable:1, /* try to recover the import */
- imp_dlm_fake:1, /* don't run recovery (timeout instead) */
- imp_server_timeout:1, /* use 1/2 timeout on MDS' OSCs */
- imp_initial_recov:1, /* retry the initial connection */
- imp_initial_recov_bk:1, /* turn off init_recov after trying all failover nids */
- imp_force_verify:1, /* force an immidiate ping */
- imp_pingable:1, /* pingable */
- imp_resend_replay:1, /* resend for replay */
- imp_recon_bk:1, /* turn off reconnect if all failovers fail */
- imp_last_recon:1; /* internally used by above */
- __u32 imp_connect_op;
- struct obd_connect_data imp_connect_data;
- __u64 imp_connect_flags_orig;
- int imp_connect_error;
-
- __u32 imp_msg_magic;
- __u32 imp_msghdr_flags; /* adjusted based on server capability */
-
- struct ptlrpc_request_pool *imp_rq_pool; /* emergency request pool */
-
- struct imp_at imp_at; /* adaptive timeout data */
- time_t imp_last_reply_time; /* for health check */
-};
+ /** Protects flags, level, generation, conn_cnt, *_list */
+ spinlock_t imp_lock;
-typedef void (*obd_import_callback)(struct obd_import *imp, void *closure,
- int event, void *event_arg, void *cb_data);
+ /* flags */
+ unsigned long imp_no_timeout:1, /* timeouts are disabled */
+ imp_invalid:1, /* evicted */
+ /* administratively disabled */
+ imp_deactive:1,
+ /* try to recover the import */
+ imp_replayable:1,
+ /* don't run recovery (timeout instead) */
+ imp_dlm_fake:1,
+ /* use 1/2 timeout on MDS' OSCs */
+ imp_server_timeout:1,
+ /* VBR: imp in delayed recovery */
+ imp_delayed_recovery:1,
+ /* VBR: if gap was found then no lock replays
+ */
+ imp_no_lock_replay:1,
+ /* recovery by versions was failed */
+ imp_vbr_failed:1,
+ /* force an immidiate ping */
+ imp_force_verify:1,
+ /* force a scheduled ping */
+ imp_force_next_verify:1,
+ /* pingable */
+ imp_pingable:1,
+ /* resend for replay */
+ imp_resend_replay:1,
+ /* disable normal recovery, for test only. */
+ imp_no_pinger_recover:1,
+#if LUSTRE_VERSION_CODE < OBD_OCD_VERSION(3, 0, 53, 0)
+ /* need IR MNE swab */
+ imp_need_mne_swab:1,
+#endif
+ /* import must be reconnected instead of
+ * chouse new connection */
+ imp_force_reconnect:1,
+ /* import has tried to connect with server */
+ imp_connect_tried:1,
+ /* connected but not FULL yet */
+ imp_connected:1;
+ __u32 imp_connect_op;
+ struct obd_connect_data imp_connect_data;
+ __u64 imp_connect_flags_orig;
+ __u64 imp_connect_flags2_orig;
+ int imp_connect_error;
-struct obd_import_observer {
- struct list_head oio_chain;
- obd_import_callback oio_cb;
- void *oio_cb_data;
-};
+ __u32 imp_msg_magic;
+ /* adjusted based on server capability */
+ __u32 imp_msghdr_flags;
-void class_observe_import(struct obd_import *imp, obd_import_callback cb,
- void *cb_data);
-void class_unobserve_import(struct obd_import *imp, obd_import_callback cb,
- void *cb_data);
-void class_notify_import_observers(struct obd_import *imp, int event,
- void *event_arg);
+ /* adaptive timeout data */
+ struct imp_at imp_at;
+ time64_t imp_last_reply_time; /* for health check */
+};
/* import.c */
static inline unsigned int at_est2timeout(unsigned int val)
return (max((val << 2) / 5, 5U) - 4);
}
-static inline void at_init(struct adaptive_timeout *at, int val, int flags) {
- memset(at, 0, sizeof(*at));
+static inline void at_reset_nolock(struct adaptive_timeout *at, int val)
+{
at->at_current = val;
at->at_worst_ever = val;
- at->at_worst_time = cfs_time_current_sec();
- at->at_flags = flags;
- spin_lock_init(&at->at_lock);
+ at->at_worst_time = ktime_get_real_seconds();
+}
+
+static inline void at_reset(struct adaptive_timeout *at, int val)
+{
+ spin_lock(&at->at_lock);
+ at_reset_nolock(at, val);
+ spin_unlock(&at->at_lock);
}
+
+static inline void at_init(struct adaptive_timeout *at, int val, int flags) {
+ memset(at, 0, sizeof(*at));
+ spin_lock_init(&at->at_lock);
+ at->at_flags = flags;
+ at_reset(at, val);
+}
+
+static inline void at_reinit(struct adaptive_timeout *at, int val, int flags)
+{
+ spin_lock(&at->at_lock);
+ at->at_binstart = 0;
+ memset(at->at_hist, 0, sizeof(at->at_hist));
+ at->at_flags = flags;
+ at_reset_nolock(at, val);
+ spin_unlock(&at->at_lock);
+}
+
+extern unsigned int at_min;
static inline int at_get(struct adaptive_timeout *at) {
- return at->at_current;
+ return (at->at_current > at_min) ? at->at_current : at_min;
}
-int at_add(struct adaptive_timeout *at, unsigned int val);
+int at_measured(struct adaptive_timeout *at, unsigned int val);
int import_at_get_index(struct obd_import *imp, int portal);
extern unsigned int at_max;
#define AT_OFF (at_max == 0)
extern struct obd_import *class_exp2cliimp(struct obd_export *);
extern struct obd_import *class_conn2cliimp(struct lustre_handle *);
+/** @} import */
+
#endif /* __IMPORT_H */
+
+/** @} obd_import */