--- /dev/null
+/*
+ * Copyright (C) 2011 EfficiOS Inc.
+ *
+ * SPDX-License-Identifier: GPL-2.0-only
+ *
+ */
+
+#ifndef _LTT_SESSION_H
+#define _LTT_SESSION_H
+
+#include <limits.h>
+#include <stdbool.h>
+#include <urcu/list.h>
+
+#include <common/hashtable/hashtable.hpp>
+#include <common/dynamic-array.hpp>
+#include <lttng/rotation.h>
+#include <lttng/location.h>
+#include <lttng/lttng-error.h>
+
+#include "snapshot.hpp"
+#include "trace-kernel.hpp"
+#include "consumer.hpp"
+
+struct ltt_ust_session;
+
+typedef void (*ltt_session_destroy_notifier)(const struct ltt_session *session,
+ void *user_data);
+typedef void (*ltt_session_clear_notifier)(const struct ltt_session *session,
+ void *user_data);
+
+/*
+ * Tracing session list
+ *
+ * Statically declared in session.c and can be accessed by using
+ * session_get_list() function that returns the pointer to the list.
+ */
+struct ltt_session_list {
+ /*
+ * This lock protects any read/write access to the list and
+ * next_uuid. All public functions in session.c acquire this
+ * lock and release it before returning. If none of those
+ * functions are used, the lock MUST be acquired in order to
+ * iterate or/and do any actions on that list.
+ */
+ pthread_mutex_t lock;
+ /*
+ * This condition variable is signaled on every removal from
+ * the session list.
+ */
+ pthread_cond_t removal_cond;
+
+ /*
+ * Session unique ID generator. The session list lock MUST be
+ * upon update and read of this counter.
+ */
+ uint64_t next_uuid;
+
+ /* Linked list head */
+ struct cds_list_head head;
+};
+
+/*
+ * This data structure contains information needed to identify a tracing
+ * session for both LTTng and UST.
+ */
+struct ltt_session {
+ char name[NAME_MAX];
+ bool has_auto_generated_name;
+ bool name_contains_creation_time;
+ char hostname[LTTNG_HOST_NAME_MAX]; /* Local hostname. */
+ /* Path of the last closed chunk. */
+ char last_chunk_path[LTTNG_PATH_MAX];
+ time_t creation_time;
+ struct ltt_kernel_session *kernel_session;
+ struct ltt_ust_session *ust_session;
+ struct urcu_ref ref;
+ /*
+ * Protect any read/write on this session data structure. This lock must be
+ * acquired *before* using any public functions declared below. Use
+ * session_lock() and session_unlock() for that.
+ */
+ pthread_mutex_t lock;
+ struct cds_list_head list;
+ uint64_t id; /* session unique identifier */
+ /* Indicates if the session has been added to the session list and ht.*/
+ bool published;
+ /* Indicates if a destroy command has been applied to this session. */
+ bool destroyed;
+ /* UID/GID of the user owning the session */
+ uid_t uid;
+ gid_t gid;
+ /*
+ * Network session handle. A value of 0 means that there is no remote
+ * session established.
+ */
+ uint64_t net_handle;
+ /*
+ * This consumer is only set when the create_session_uri call is made.
+ * This contains the temporary information for a consumer output. Upon
+ * creation of the UST or kernel session, this consumer, if available, is
+ * copied into those sessions.
+ */
+ struct consumer_output *consumer;
+ /*
+ * Indicates whether or not the user has specified an output directory
+ * or if it was configured using the default configuration.
+ */
+ bool has_user_specified_directory;
+ /* Did at least ONE start command has been triggered?. */
+ unsigned int has_been_started:1;
+ /*
+ * Is the session active? Start trace command sets this to 1 and the stop
+ * command reset it to 0.
+ */
+ unsigned int active:1;
+
+ /* Snapshot representation in a session. */
+ struct snapshot snapshot;
+ /* Indicate if the session has to output the traces or not. */
+ unsigned int output_traces;
+ /*
+ * This session is in snapshot mode. This means that channels enabled
+ * will be set in overwrite mode by default and must be in mmap
+ * output mode. Note that snapshots can be taken on a session that
+ * is not in "snapshot_mode". This parameter only affects channel
+ * creation defaults.
+ */
+ unsigned int snapshot_mode;
+ /*
+ * A session that has channels that don't use 'mmap' output can't be
+ * used to capture snapshots. This is set to true whenever a
+ * 'splice' kernel channel is enabled.
+ */
+ bool has_non_mmap_channel;
+ /*
+ * Timer set when the session is created for live reading.
+ */
+ unsigned int live_timer;
+ /*
+ * Path where to keep the shared memory files.
+ */
+ char shm_path[PATH_MAX];
+ /*
+ * Node in ltt_sessions_ht_by_id.
+ */
+ struct lttng_ht_node_u64 node;
+ /*
+ * Node in ltt_sessions_ht_by_name.
+ */
+ struct lttng_ht_node_str node_by_name;
+ /*
+ * Timer to check periodically if a relay and/or consumer has completed
+ * the last rotation.
+ */
+ bool rotation_pending_check_timer_enabled;
+ timer_t rotation_pending_check_timer;
+ /* Timer to periodically rotate a session. */
+ bool rotation_schedule_timer_enabled;
+ timer_t rotation_schedule_timer;
+ /* Value for periodic rotations, 0 if disabled. */
+ uint64_t rotate_timer_period;
+ /* Value for size-based rotations, 0 if disabled. */
+ uint64_t rotate_size;
+ /*
+ * Keep a state if this session was rotated after the last stop command.
+ * We only allow one rotation after a stop. At destroy, we also need to
+ * know if a rotation occurred since the last stop to rename the current
+ * chunk. After a stop followed by rotate, all subsequent clear
+ * (without prior start) will succeed, but will be effect-less.
+ */
+ bool rotated_after_last_stop;
+ /*
+ * Track whether the session was cleared after last stop. All subsequent
+ * clear (without prior start) will succeed, but will be effect-less. A
+ * subsequent rotate (without prior start) will return an error.
+ */
+ bool cleared_after_last_stop;
+ /*
+ * True if the session has had an explicit non-quiet rotation.
+ */
+ bool rotated;
+ /*
+ * Condition and trigger for size-based rotations.
+ */
+ struct lttng_condition *rotate_condition;
+ struct lttng_trigger *rotate_trigger;
+ LTTNG_OPTIONAL(uint64_t) most_recent_chunk_id;
+ struct lttng_trace_chunk *current_trace_chunk;
+ struct lttng_trace_chunk *chunk_being_archived;
+ /* Current state of a rotation. */
+ enum lttng_rotation_state rotation_state;
+ bool quiet_rotation;
+ char *last_archived_chunk_name;
+ LTTNG_OPTIONAL(uint64_t) last_archived_chunk_id;
+ struct lttng_dynamic_array destroy_notifiers;
+ struct lttng_dynamic_array clear_notifiers;
+ /* Session base path override. Set non-null. */
+ char *base_path;
+};
+
+enum lttng_error_code session_create(const char *name, uid_t uid, gid_t gid,
+ struct ltt_session **out_session);
+void session_lock(struct ltt_session *session);
+void session_unlock(struct ltt_session *session);
+
+/*
+ * The session list lock covers more ground than its name implies. While
+ * it does protect against concurent mutations of the session list, it is
+ * also used as a multi-session lock when synchronizing newly-registered
+ * 'user space tracer' and 'agent' applications.
+ *
+ * In other words, it prevents tracer configurations from changing while they
+ * are being transmitted to the various applications.
+ */
+void session_lock_list(void);
+int session_trylock_list(void);
+void session_unlock_list(void);
+
+void session_destroy(struct ltt_session *session);
+int session_add_destroy_notifier(struct ltt_session *session,
+ ltt_session_destroy_notifier notifier, void *user_data);
+
+int session_add_clear_notifier(struct ltt_session *session,
+ ltt_session_clear_notifier notifier, void *user_data);
+void session_notify_clear(struct ltt_session *session);
+
+bool session_get(struct ltt_session *session);
+void session_put(struct ltt_session *session);
+
+enum consumer_dst_type session_get_consumer_destination_type(
+ const struct ltt_session *session);
+const char *session_get_net_consumer_hostname(
+ const struct ltt_session *session);
+void session_get_net_consumer_ports(
+ const struct ltt_session *session,
+ uint16_t *control_port, uint16_t *data_port);
+struct lttng_trace_archive_location *session_get_trace_archive_location(
+ const struct ltt_session *session);
+
+struct ltt_session *session_find_by_name(const char *name);
+struct ltt_session *session_find_by_id(uint64_t id);
+
+struct ltt_session_list *session_get_list(void);
+void session_list_wait_empty(void);
+
+bool session_access_ok(struct ltt_session *session, uid_t uid);
+
+int session_reset_rotation_state(struct ltt_session *session,
+ enum lttng_rotation_state result);
+
+/* Create a new trace chunk object from the session's configuration. */
+struct lttng_trace_chunk *session_create_new_trace_chunk(
+ const struct ltt_session *session,
+ const struct consumer_output *consumer_output_override,
+ const char *session_base_path_override,
+ const char *chunk_name_override);
+
+/*
+ * Set `new_trace_chunk` as the session's current trace chunk. A reference
+ * to `new_trace_chunk` is acquired by the session. The chunk is created
+ * on remote peers (consumer and relay daemons).
+ *
+ * A reference to the session's current trace chunk is returned through
+ * `current_session_trace_chunk` on success.
+ */
+int session_set_trace_chunk(struct ltt_session *session,
+ struct lttng_trace_chunk *new_trace_chunk,
+ struct lttng_trace_chunk **current_session_trace_chunk);
+
+/*
+ * Close a chunk on the remote peers of a session. Has no effect on the
+ * ltt_session itself.
+ */
+int session_close_trace_chunk(struct ltt_session *session,
+ struct lttng_trace_chunk *trace_chunk,
+ enum lttng_trace_chunk_command_type close_command,
+ char *path);
+
+/* Open a packet in all channels of a given session. */
+enum lttng_error_code session_open_packets(struct ltt_session *session);
+
+bool session_output_supports_trace_chunks(const struct ltt_session *session);
+
+/*
+ * Sample the id of a session looked up via its name.
+ * Here the term "sampling" hint the caller that this return the id at a given
+ * point in time with no guarantee that the session for which the id was
+ * sampled still exist at that point.
+ *
+ * Return 0 when the session is not found,
+ * Return 1 when the session is found and set `id`.
+ */
+bool sample_session_id_by_name(const char *name, uint64_t *id);
+
+#endif /* _LTT_SESSION_H */
projects / lttng-tools.git / blobdiff