diff options
Diffstat (limited to '')
| -rw-r--r-- | collectors/ebpf.plugin/ebpf.c | 430 |
1 files changed, 357 insertions, 73 deletions
diff --git a/collectors/ebpf.plugin/ebpf.c b/collectors/ebpf.plugin/ebpf.c index 67fe477c..c0764c60 100644 --- a/collectors/ebpf.plugin/ebpf.c +++ b/collectors/ebpf.plugin/ebpf.c @@ -28,11 +28,22 @@ int running_on_kernel = 0; int ebpf_nprocs; int isrh = 0; int main_thread_id = 0; +int process_pid_fd = -1; pthread_mutex_t lock; pthread_mutex_t ebpf_exit_cleanup; pthread_mutex_t collect_data_mutex; -pthread_cond_t collect_data_cond_var; + +struct netdata_static_thread cgroup_integration_thread = { + .name = "EBPF CGROUP INT", + .config_section = NULL, + .config_name = NULL, + .env_name = NULL, + .enabled = 1, + .thread = NULL, + .init_routine = NULL, + .start_routine = NULL +}; ebpf_module_t ebpf_modules[] = { { .thread_name = "process", .config_name = "process", .enabled = 0, .start_routine = ebpf_process_thread, @@ -435,9 +446,6 @@ ebpf_sync_syscalls_t local_syscalls[] = { }; -// Link with apps.plugin -ebpf_process_stat_t *global_process_stat = NULL; - // Link with cgroup.plugin netdata_ebpf_cgroup_shm_t shm_ebpf_cgroup = {NULL, NULL}; int shm_fd_ebpf_cgroup = -1; @@ -449,10 +457,19 @@ ebpf_network_viewer_options_t network_viewer_opt; // Statistic ebpf_plugin_stats_t plugin_statistics = {.core = 0, .legacy = 0, .running = 0, .threads = 0, .tracepoints = 0, - .probes = 0, .retprobes = 0, .trampolines = 0}; + .probes = 0, .retprobes = 0, .trampolines = 0, .memlock_kern = 0, + .hash_tables = 0}; #ifdef LIBBPF_MAJOR_VERSION struct btf *default_btf = NULL; +struct cachestat_bpf *cachestat_bpf_obj = NULL; +struct dc_bpf *dc_bpf_obj = NULL; +struct fd_bpf *fd_bpf_obj = NULL; +struct mount_bpf *mount_bpf_obj = NULL; +struct shm_bpf *shm_bpf_obj = NULL; +struct socket_bpf *socket_bpf_obj = NULL; +struct swap_bpf *bpf_obj = NULL; +struct vfs_bpf *vfs_bpf_obj = NULL; #else void *default_btf = NULL; #endif @@ -460,6 +477,35 @@ char *btf_path = NULL; /***************************************************************** * + * FUNCTIONS USED TO ALLOCATE APPS/CGROUP MEMORIES (ARAL) + * + *****************************************************************/ + +/** + * Allocate PID ARAL + * + * Allocate memory using ARAL functions to speed up processing. + * + * @param name the internal name used for allocated region. + * @param size size of each element inside allocated space + * + * @return It returns the address on success and NULL otherwise. + */ +ARAL *ebpf_allocate_pid_aral(char *name, size_t size) +{ + static size_t max_elements = NETDATA_EBPF_ALLOC_MAX_PID; + if (max_elements < NETDATA_EBPF_ALLOC_MIN_ELEMENTS) { + error("Number of elements given is too small, adjusting it for %d", NETDATA_EBPF_ALLOC_MIN_ELEMENTS); + max_elements = NETDATA_EBPF_ALLOC_MIN_ELEMENTS; + } + + return aral_create(name, size, + 0, max_elements, + NULL, NULL, NULL, false, false); +} + +/***************************************************************** + * * FUNCTIONS USED TO CLEAN MEMORY AND OPERATE SYSTEM FILES * *****************************************************************/ @@ -488,10 +534,12 @@ static void ebpf_exit() #endif printf("DISABLE\n"); + pthread_mutex_lock(&mutex_cgroup_shm); if (shm_ebpf_cgroup.header) { - munmap(shm_ebpf_cgroup.header, shm_ebpf_cgroup.header->body_length); + ebpf_unmap_cgroup_shared_memory(); shm_unlink(NETDATA_SHARED_MEMORY_EBPF_CGROUP_NAME); } + pthread_mutex_unlock(&mutex_cgroup_shm); exit(0); } @@ -518,6 +566,126 @@ static void ebpf_unload_legacy_code(struct bpf_object *objects, struct bpf_link bpf_object__close(objects); } +/** + * Unload Unique maps + * + * This function unload all BPF maps from threads using one unique BPF object. + */ +static void ebpf_unload_unique_maps() +{ + int i; + for (i = 0; ebpf_modules[i].thread_name; i++) { + if (ebpf_modules[i].enabled != NETDATA_THREAD_EBPF_STOPPED) { + if (ebpf_modules[i].enabled != NETDATA_THREAD_EBPF_NOT_RUNNING) + error("Cannot unload maps for thread %s, because it is not stopped.", ebpf_modules[i].thread_name); + + continue; + } + + ebpf_unload_legacy_code(ebpf_modules[i].objects, ebpf_modules[i].probe_links); + switch (i) { + case EBPF_MODULE_CACHESTAT_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (cachestat_bpf_obj) + cachestat_bpf__destroy(cachestat_bpf_obj); +#endif + break; + } + case EBPF_MODULE_DCSTAT_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (dc_bpf_obj) + dc_bpf__destroy(dc_bpf_obj); +#endif + break; + } + case EBPF_MODULE_FD_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (fd_bpf_obj) + fd_bpf__destroy(fd_bpf_obj); +#endif + break; + } + case EBPF_MODULE_MOUNT_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (mount_bpf_obj) + mount_bpf__destroy(mount_bpf_obj); +#endif + break; + } + case EBPF_MODULE_SHM_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (shm_bpf_obj) + shm_bpf__destroy(shm_bpf_obj); +#endif + break; + } + case EBPF_MODULE_SOCKET_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (socket_bpf_obj) + socket_bpf__destroy(socket_bpf_obj); +#endif + break; + } + case EBPF_MODULE_SWAP_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (bpf_obj) + swap_bpf__destroy(bpf_obj); +#endif + break; + } + case EBPF_MODULE_VFS_IDX: { +#ifdef LIBBPF_MAJOR_VERSION + if (vfs_bpf_obj) + vfs_bpf__destroy(vfs_bpf_obj); +#endif + break; + } + case EBPF_MODULE_PROCESS_IDX: + case EBPF_MODULE_DISK_IDX: + case EBPF_MODULE_HARDIRQ_IDX: + case EBPF_MODULE_SOFTIRQ_IDX: + case EBPF_MODULE_OOMKILL_IDX: + case EBPF_MODULE_MDFLUSH_IDX: + default: + continue; + } + } +} + +/** + * Unload filesystem maps + * + * This function unload all BPF maps from filesystem thread. + */ +static void ebpf_unload_filesystems() +{ + if (ebpf_modules[EBPF_MODULE_FILESYSTEM_IDX].enabled == NETDATA_THREAD_EBPF_NOT_RUNNING || + ebpf_modules[EBPF_MODULE_SYNC_IDX].enabled == NETDATA_THREAD_EBPF_RUNNING) + return; + + int i; + for (i = 0; localfs[i].filesystem != NULL; i++) { + ebpf_unload_legacy_code(localfs[i].objects, localfs[i].probe_links); + } +} + +/** + * Unload sync maps + * + * This function unload all BPF maps from sync thread. + */ +static void ebpf_unload_sync() +{ + if (ebpf_modules[EBPF_MODULE_SYNC_IDX].enabled == NETDATA_THREAD_EBPF_NOT_RUNNING || + ebpf_modules[EBPF_MODULE_SYNC_IDX].enabled == NETDATA_THREAD_EBPF_RUNNING) + return; + + int i; + for (i = 0; local_syscalls[i].syscall != NULL; i++) { + ebpf_unload_legacy_code(local_syscalls[i].objects, local_syscalls[i].probe_links); + } +} + int ebpf_exit_plugin = 0; /** * Close the collector gracefully @@ -529,7 +697,6 @@ static void ebpf_stop_threads(int sig) UNUSED(sig); static int only_one = 0; - int i; // Child thread should be closed by itself. pthread_mutex_lock(&ebpf_exit_cleanup); if (main_thread_id != gettid() || only_one) { @@ -537,13 +704,26 @@ static void ebpf_stop_threads(int sig) return; } only_one = 1; - for (i = 0; ebpf_threads[i].name != NULL; i++) { - if (ebpf_threads[i].enabled != NETDATA_THREAD_EBPF_STOPPED) - netdata_thread_cancel(*ebpf_threads[i].thread); + int i; + for (i = 0; ebpf_modules[i].thread_name != NULL; i++) { + if (ebpf_modules[i].enabled == NETDATA_THREAD_EBPF_RUNNING) { + netdata_thread_cancel(*ebpf_modules[i].thread->thread); +#ifdef NETDATA_DEV_MODE + info("Sending cancel for thread %s", ebpf_modules[i].thread_name); +#endif + } } pthread_mutex_unlock(&ebpf_exit_cleanup); + pthread_mutex_lock(&mutex_cgroup_shm); + netdata_thread_cancel(*cgroup_integration_thread.thread); +#ifdef NETDATA_DEV_MODE + info("Sending cancel for thread %s", cgroup_integration_thread.name); +#endif + pthread_mutex_unlock(&mutex_cgroup_shm); + ebpf_exit_plugin = 1; + usec_t max = USEC_PER_SEC, step = 100000; while (i && max) { max -= step; @@ -551,42 +731,18 @@ static void ebpf_stop_threads(int sig) i = 0; int j; pthread_mutex_lock(&ebpf_exit_cleanup); - for (j = 0; ebpf_threads[j].name != NULL; j++) { - if (ebpf_threads[j].enabled != NETDATA_THREAD_EBPF_STOPPED) + for (j = 0; ebpf_modules[j].thread_name != NULL; j++) { + if (ebpf_modules[j].enabled == NETDATA_THREAD_EBPF_RUNNING) i++; } pthread_mutex_unlock(&ebpf_exit_cleanup); } - if (!i) { - //Unload threads(except sync and filesystem) - pthread_mutex_lock(&ebpf_exit_cleanup); - for (i = 0; ebpf_threads[i].name != NULL; i++) { - if (ebpf_threads[i].enabled == NETDATA_THREAD_EBPF_STOPPED && i != EBPF_MODULE_FILESYSTEM_IDX && - i != EBPF_MODULE_SYNC_IDX) - ebpf_unload_legacy_code(ebpf_modules[i].objects, ebpf_modules[i].probe_links); - } - pthread_mutex_unlock(&ebpf_exit_cleanup); - - //Unload filesystem - pthread_mutex_lock(&ebpf_exit_cleanup); - if (ebpf_threads[EBPF_MODULE_FILESYSTEM_IDX].enabled == NETDATA_THREAD_EBPF_STOPPED) { - for (i = 0; localfs[i].filesystem != NULL; i++) { - ebpf_unload_legacy_code(localfs[i].objects, localfs[i].probe_links); - } - } - pthread_mutex_unlock(&ebpf_exit_cleanup); - - //Unload Sync - pthread_mutex_lock(&ebpf_exit_cleanup); - if (ebpf_threads[EBPF_MODULE_SYNC_IDX].enabled == NETDATA_THREAD_EBPF_STOPPED) { - for (i = 0; local_syscalls[i].syscall != NULL; i++) { - ebpf_unload_legacy_code(local_syscalls[i].objects, local_syscalls[i].probe_links); - } - } - pthread_mutex_unlock(&ebpf_exit_cleanup); - - } + pthread_mutex_lock(&ebpf_exit_cleanup); + ebpf_unload_unique_maps(); + ebpf_unload_filesystems(); + ebpf_unload_sync(); + pthread_mutex_unlock(&ebpf_exit_cleanup); ebpf_exit(); } @@ -598,6 +754,58 @@ static void ebpf_stop_threads(int sig) *****************************************************************/ /** + * Create apps charts + * + * Call ebpf_create_chart to create the charts on apps submenu. + * + * @param root a pointer for the targets. + */ +static void ebpf_create_apps_charts(struct ebpf_target *root) +{ + if (unlikely(!ebpf_all_pids)) + return; + + struct ebpf_target *w; + int newly_added = 0; + + for (w = root; w; w = w->next) { + if (w->target) + continue; + + if (unlikely(w->processes && (debug_enabled || w->debug_enabled))) { + struct ebpf_pid_on_target *pid_on_target; + + fprintf( + stderr, "ebpf.plugin: target '%s' has aggregated %u process%s:", w->name, w->processes, + (w->processes == 1) ? "" : "es"); + + for (pid_on_target = w->root_pid; pid_on_target; pid_on_target = pid_on_target->next) { + fprintf(stderr, " %d", pid_on_target->pid); + } + + fputc('\n', stderr); + } + + if (!w->exposed && w->processes) { + newly_added++; + w->exposed = 1; + if (debug_enabled || w->debug_enabled) + debug_log_int("%s just added - regenerating charts.", w->name); + } + } + + if (!newly_added) + return; + + int counter; + for (counter = 0; ebpf_modules[counter].thread_name; counter++) { + ebpf_module_t *current = &ebpf_modules[counter]; + if (current->enabled == NETDATA_THREAD_EBPF_RUNNING && current->apps_charts && current->apps_routine) + current->apps_routine(current, root); + } +} + +/** * Get a value from a structure. * * @param basis it is the first address of the structure @@ -876,9 +1084,9 @@ void ebpf_create_chart(char *type, * @param module chart module name, this is the eBPF thread. */ void ebpf_create_charts_on_apps(char *id, char *title, char *units, char *family, char *charttype, int order, - char *algorithm, struct target *root, int update_every, char *module) + char *algorithm, struct ebpf_target *root, int update_every, char *module) { - struct target *w; + struct ebpf_target *w; ebpf_write_chart_cmd(NETDATA_APPS_FAMILY, id, title, units, family, charttype, NULL, order, update_every, module); @@ -913,6 +1121,79 @@ void write_histogram_chart(char *family, char *name, const netdata_idx_t *hist, fflush(stdout); } +/** + * ARAL Charts + * + * Add chart to monitor ARAL usage + * Caller must call this function with mutex locked. + * + * @param name the name used to create aral + * @param em a pointer to the structure with the default values. + */ +void ebpf_statistic_create_aral_chart(char *name, ebpf_module_t *em) +{ + static int priority = 140100; + char *mem = { NETDATA_EBPF_STAT_DIMENSION_MEMORY }; + char *aral = { NETDATA_EBPF_STAT_DIMENSION_ARAL }; + + snprintfz(em->memory_usage, NETDATA_EBPF_CHART_MEM_LENGTH -1, "aral_%s_size", name); + snprintfz(em->memory_allocations, NETDATA_EBPF_CHART_MEM_LENGTH -1, "aral_%s_alloc", name); + + ebpf_write_chart_cmd(NETDATA_MONITORING_FAMILY, + em->memory_usage, + "Bytes allocated for ARAL.", + "bytes", + NETDATA_EBPF_FAMILY, + NETDATA_EBPF_CHART_TYPE_STACKED, + "netdata.ebpf_aral_stat_size", + priority++, + em->update_every, + NETDATA_EBPF_MODULE_NAME_PROCESS); + + ebpf_write_global_dimension(mem, + mem, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX]); + + ebpf_write_chart_cmd(NETDATA_MONITORING_FAMILY, + em->memory_allocations, + "Calls to allocate memory.", + "calls", + NETDATA_EBPF_FAMILY, + NETDATA_EBPF_CHART_TYPE_STACKED, + "netdata.ebpf_aral_stat_alloc", + priority++, + em->update_every, + NETDATA_EBPF_MODULE_NAME_PROCESS); + + ebpf_write_global_dimension(aral, + aral, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX]); +} + +/** + * Send data from aral chart + * + * Send data for eBPF plugin + * + * @param memory a pointer to the allocated address + * @param em a pointer to the structure with the default values. + */ +void ebpf_send_data_aral_chart(ARAL *memory, ebpf_module_t *em) +{ + char *mem = { NETDATA_EBPF_STAT_DIMENSION_MEMORY }; + char *aral = { NETDATA_EBPF_STAT_DIMENSION_ARAL }; + + struct aral_statistics *stats = aral_statistics(memory); + + write_begin_chart(NETDATA_MONITORING_FAMILY, em->memory_usage); + write_chart_dimension(mem, (long long)stats->structures.allocated_bytes); + write_end_chart(); + + write_begin_chart(NETDATA_MONITORING_FAMILY, em->memory_allocations); + write_chart_dimension(aral, (long long)stats->structures.allocations); + write_end_chart(); +} + /***************************************************************** * * FUNCTIONS TO DEFINE OPTIONS @@ -944,7 +1225,7 @@ void ebpf_global_labels(netdata_syscall_stat_t *is, netdata_publish_syscall_t *p pio[i].dimension = dim[i]; pio[i].name = name[i]; - pio[i].algorithm = strdupz(ebpf_algorithms[algorithm[i]]); + pio[i].algorithm = ebpf_algorithms[algorithm[i]]; if (publish_prev) { publish_prev->next = &pio[i]; } @@ -1342,21 +1623,13 @@ static void read_local_addresses() * Start Pthread Variable * * This function starts all pthread variables. - * - * @return It returns 0 on success and -1. */ -int ebpf_start_pthread_variables() +void ebpf_start_pthread_variables() { pthread_mutex_init(&lock, NULL); pthread_mutex_init(&ebpf_exit_cleanup, NULL); pthread_mutex_init(&collect_data_mutex, NULL); - - if (pthread_cond_init(&collect_data_cond_var, NULL)) { - error("Cannot start conditional variable to control Apps charts."); - return -1; - } - - return 0; + pthread_mutex_init(&mutex_cgroup_shm, NULL); } /** @@ -1386,8 +1659,8 @@ static void ebpf_allocate_common_vectors() return; } - all_pids = callocz((size_t)pid_max, sizeof(struct pid_stat *)); - global_process_stat = callocz((size_t)ebpf_nprocs, sizeof(ebpf_process_stat_t)); + ebpf_all_pids = callocz((size_t)pid_max, sizeof(struct ebpf_pid_stat *)); + ebpf_aral_init(); } /** @@ -1720,8 +1993,9 @@ void set_global_variables() ebpf_configured_log_dir = LOG_DIR; ebpf_nprocs = (int)sysconf(_SC_NPROCESSORS_ONLN); - if (ebpf_nprocs > NETDATA_MAX_PROCESSOR) { + if (ebpf_nprocs < 0) { ebpf_nprocs = NETDATA_MAX_PROCESSOR; + error("Cannot identify number of process, using default value %d", ebpf_nprocs); } isrh = get_redhat_release(); @@ -2088,7 +2362,7 @@ static pid_t ebpf_read_previous_pid(char *filename) length = 63; buffer[length] = '\0'; - old_pid = (pid_t)str2uint32_t(buffer); + old_pid = (pid_t) str2uint32_t(buffer, NULL); } fclose(fp); @@ -2219,10 +2493,7 @@ int main(int argc, char **argv) signal(SIGTERM, ebpf_stop_threads); signal(SIGPIPE, ebpf_stop_threads); - if (ebpf_start_pthread_variables()) { - error("Cannot start mutex to control overall charts."); - ebpf_exit(); - } + ebpf_start_pthread_variables(); netdata_configured_host_prefix = getenv("NETDATA_HOST_PREFIX"); if(verify_netdata_host_prefix() == -1) ebpf_exit(6); @@ -2241,6 +2512,12 @@ int main(int argc, char **argv) ebpf_set_static_routine(); + cgroup_integration_thread.thread = mallocz(sizeof(netdata_thread_t)); + cgroup_integration_thread.start_routine = ebpf_cgroup_integration; + + netdata_thread_create(cgroup_integration_thread.thread, cgroup_integration_thread.name, + NETDATA_THREAD_OPTION_DEFAULT, ebpf_cgroup_integration, NULL); + int i; for (i = 0; ebpf_threads[i].name != NULL; i++) { struct netdata_static_thread *st = &ebpf_threads[i]; @@ -2251,30 +2528,37 @@ int main(int argc, char **argv) if (em->enabled || !i) { st->thread = mallocz(sizeof(netdata_thread_t)); em->thread_id = i; - st->enabled = NETDATA_THREAD_EBPF_RUNNING; + em->enabled = NETDATA_THREAD_EBPF_RUNNING; netdata_thread_create(st->thread, st->name, NETDATA_THREAD_OPTION_DEFAULT, st->start_routine, em); } else { - st->enabled = NETDATA_THREAD_EBPF_STOPPED; + em->enabled = NETDATA_THREAD_EBPF_NOT_RUNNING; } } usec_t step = USEC_PER_SEC; - int counter = NETDATA_EBPF_CGROUP_UPDATE - 1; heartbeat_t hb; heartbeat_init(&hb); + int update_apps_every = (int) EBPF_CFG_UPDATE_APPS_EVERY_DEFAULT; + int update_apps_list = update_apps_every - 1; //Plugin will be killed when it receives a signal while (!ebpf_exit_plugin) { (void)heartbeat_next(&hb, step); - // We are using a small heartbeat time to wake up thread, - // but we should not update so frequently the shared memory data - if (++counter >= NETDATA_EBPF_CGROUP_UPDATE) { - counter = 0; - if (!shm_ebpf_cgroup.header) - ebpf_map_cgroup_shared_memory(); - - ebpf_parse_cgroup_shm_data(); + pthread_mutex_lock(&ebpf_exit_cleanup); + if (ebpf_modules[i].enabled == NETDATA_THREAD_EBPF_RUNNING && process_pid_fd != -1) { + pthread_mutex_lock(&collect_data_mutex); + if (++update_apps_list == update_apps_every) { + update_apps_list = 0; + cleanup_exited_pids(); + collect_data_for_all_processes(process_pid_fd); + + pthread_mutex_lock(&lock); + ebpf_create_apps_charts(apps_groups_root_target); + pthread_mutex_unlock(&lock); + } + pthread_mutex_unlock(&collect_data_mutex); } + pthread_mutex_unlock(&ebpf_exit_cleanup); } ebpf_stop_threads(0); |