diff options
Diffstat (limited to 'collectors/ebpf.plugin/ebpf_dcstat.c')
| -rw-r--r-- | collectors/ebpf.plugin/ebpf_dcstat.c | 1225 |
1 files changed, 1225 insertions, 0 deletions
diff --git a/collectors/ebpf.plugin/ebpf_dcstat.c b/collectors/ebpf.plugin/ebpf_dcstat.c new file mode 100644 index 0000000..71169e1 --- /dev/null +++ b/collectors/ebpf.plugin/ebpf_dcstat.c @@ -0,0 +1,1225 @@ +// SPDX-License-Identifier: GPL-3.0-or-later + +#include "ebpf.h" +#include "ebpf_dcstat.h" + +static char *dcstat_counter_dimension_name[NETDATA_DCSTAT_IDX_END] = { "ratio", "reference", "slow", "miss" }; +static netdata_syscall_stat_t dcstat_counter_aggregated_data[NETDATA_DCSTAT_IDX_END]; +static netdata_publish_syscall_t dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_END]; + +netdata_dcstat_pid_t *dcstat_vector = NULL; +netdata_publish_dcstat_t **dcstat_pid = NULL; + +static netdata_idx_t dcstat_hash_values[NETDATA_DCSTAT_IDX_END]; +static netdata_idx_t *dcstat_values = NULL; + +struct config dcstat_config = { .first_section = NULL, + .last_section = NULL, + .mutex = NETDATA_MUTEX_INITIALIZER, + .index = { .avl_tree = { .root = NULL, .compar = appconfig_section_compare }, + .rwlock = AVL_LOCK_INITIALIZER } }; + +struct netdata_static_thread dcstat_threads = {"DCSTAT KERNEL", + .config_section = NULL, + .config_name = NULL, + .env_name = NULL, + .enabled = 1, + .thread = NULL, + .init_routine = NULL, + .start_routine = NULL}; + +ebpf_local_maps_t dcstat_maps[] = {{.name = "dcstat_global", .internal_input = NETDATA_DIRECTORY_CACHE_END, + .user_input = 0, .type = NETDATA_EBPF_MAP_STATIC, + .map_fd = ND_EBPF_MAP_FD_NOT_INITIALIZED}, + {.name = "dcstat_pid", .internal_input = ND_EBPF_DEFAULT_PID_SIZE, + .user_input = 0, + .type = NETDATA_EBPF_MAP_RESIZABLE | NETDATA_EBPF_MAP_PID, + .map_fd = ND_EBPF_MAP_FD_NOT_INITIALIZED}, + {.name = "dcstat_ctrl", .internal_input = NETDATA_CONTROLLER_END, + .user_input = 0, + .type = NETDATA_EBPF_MAP_CONTROLLER, + .map_fd = ND_EBPF_MAP_FD_NOT_INITIALIZED}, + {.name = NULL, .internal_input = 0, .user_input = 0, + .type = NETDATA_EBPF_MAP_CONTROLLER, + .map_fd = ND_EBPF_MAP_FD_NOT_INITIALIZED}}; + +static ebpf_specify_name_t dc_optional_name[] = { {.program_name = "netdata_lookup_fast", + .function_to_attach = "lookup_fast", + .optional = NULL, + .retprobe = CONFIG_BOOLEAN_NO}, + {.program_name = NULL}}; + +netdata_ebpf_targets_t dc_targets[] = { {.name = "lookup_fast", .mode = EBPF_LOAD_TRAMPOLINE}, + {.name = "d_lookup", .mode = EBPF_LOAD_TRAMPOLINE}, + {.name = NULL, .mode = EBPF_LOAD_TRAMPOLINE}}; + +#ifdef LIBBPF_MAJOR_VERSION +#include "includes/dc.skel.h" // BTF code + +static struct dc_bpf *bpf_obj = NULL; + +/** + * Disable probe + * + * Disable all probes to use exclusively another method. + * + * @param obj is the main structure for bpf objects + */ +static inline void ebpf_dc_disable_probes(struct dc_bpf *obj) +{ + bpf_program__set_autoload(obj->progs.netdata_lookup_fast_kprobe, false); + bpf_program__set_autoload(obj->progs.netdata_d_lookup_kretprobe, false); + bpf_program__set_autoload(obj->progs.netdata_dcstat_release_task_kprobe, false); +} + +/* + * Disable trampoline + * + * Disable all trampoline to use exclusively another method. + * + * @param obj is the main structure for bpf objects. + */ +static inline void ebpf_dc_disable_trampoline(struct dc_bpf *obj) +{ + bpf_program__set_autoload(obj->progs.netdata_lookup_fast_fentry, false); + bpf_program__set_autoload(obj->progs.netdata_d_lookup_fexit, false); + bpf_program__set_autoload(obj->progs.netdata_dcstat_release_task_fentry, false); +} + +/** + * Set trampoline target + * + * Set the targets we will monitor. + * + * @param obj is the main structure for bpf objects. + */ +static void ebpf_dc_set_trampoline_target(struct dc_bpf *obj) +{ + bpf_program__set_attach_target(obj->progs.netdata_lookup_fast_fentry, 0, + dc_targets[NETDATA_DC_TARGET_LOOKUP_FAST].name); + + bpf_program__set_attach_target(obj->progs.netdata_d_lookup_fexit, 0, + dc_targets[NETDATA_DC_TARGET_D_LOOKUP].name); + + bpf_program__set_attach_target(obj->progs.netdata_dcstat_release_task_fentry, 0, + EBPF_COMMON_FNCT_CLEAN_UP); +} + +/** + * Mount Attach Probe + * + * Attach probes to target + * + * @param obj is the main structure for bpf objects. + * + * @return It returns 0 on success and -1 otherwise. + */ +static int ebpf_dc_attach_probes(struct dc_bpf *obj) +{ + obj->links.netdata_d_lookup_kretprobe = bpf_program__attach_kprobe(obj->progs.netdata_d_lookup_kretprobe, + true, + dc_targets[NETDATA_DC_TARGET_D_LOOKUP].name); + int ret = libbpf_get_error(obj->links.netdata_d_lookup_kretprobe); + if (ret) + return -1; + + char *lookup_name = (dc_optional_name[NETDATA_DC_TARGET_LOOKUP_FAST].optional) ? + dc_optional_name[NETDATA_DC_TARGET_LOOKUP_FAST].optional : + dc_targets[NETDATA_DC_TARGET_LOOKUP_FAST].name ; + + obj->links.netdata_lookup_fast_kprobe = bpf_program__attach_kprobe(obj->progs.netdata_lookup_fast_kprobe, + false, + lookup_name); + ret = libbpf_get_error(obj->links.netdata_lookup_fast_kprobe); + if (ret) + return -1; + + obj->links.netdata_dcstat_release_task_kprobe = bpf_program__attach_kprobe(obj->progs.netdata_dcstat_release_task_kprobe, + false, + EBPF_COMMON_FNCT_CLEAN_UP); + ret = libbpf_get_error(obj->links.netdata_dcstat_release_task_kprobe); + if (ret) + return -1; + + return 0; +} + +/** + * Adjust Map Size + * + * Resize maps according input from users. + * + * @param obj is the main structure for bpf objects. + * @param em structure with configuration + */ +static void ebpf_dc_adjust_map_size(struct dc_bpf *obj, ebpf_module_t *em) +{ + ebpf_update_map_size(obj->maps.dcstat_pid, &dcstat_maps[NETDATA_DCSTAT_PID_STATS], + em, bpf_map__name(obj->maps.dcstat_pid)); +} + +/** + * Set hash tables + * + * Set the values for maps according the value given by kernel. + * + * @param obj is the main structure for bpf objects. + */ +static void ebpf_dc_set_hash_tables(struct dc_bpf *obj) +{ + dcstat_maps[NETDATA_DCSTAT_GLOBAL_STATS].map_fd = bpf_map__fd(obj->maps.dcstat_global); + dcstat_maps[NETDATA_DCSTAT_PID_STATS].map_fd = bpf_map__fd(obj->maps.dcstat_pid); + dcstat_maps[NETDATA_DCSTAT_CTRL].map_fd = bpf_map__fd(obj->maps.dcstat_ctrl); +} + +/** + * Update Load + * + * For directory cache, some distributions change the function name, and we do not have condition to use + * TRAMPOLINE like other functions. + * + * @param em structure with configuration + * + * @return When then symbols were not modified, it returns TRAMPOLINE, else it returns RETPROBE. + */ +netdata_ebpf_program_loaded_t ebpf_dc_update_load(ebpf_module_t *em) +{ + if (!strcmp(dc_optional_name[NETDATA_DC_TARGET_LOOKUP_FAST].optional, + dc_optional_name[NETDATA_DC_TARGET_LOOKUP_FAST].function_to_attach)) + return EBPF_LOAD_TRAMPOLINE; + + if (em->targets[NETDATA_DC_TARGET_LOOKUP_FAST].mode != EBPF_LOAD_RETPROBE) + info("When your kernel was compiled the symbol %s was modified, instead to use `trampoline`, the plugin will use `probes`.", + dc_optional_name[NETDATA_DC_TARGET_LOOKUP_FAST].function_to_attach); + + return EBPF_LOAD_RETPROBE; +} + +/** + * Disable Release Task + * + * Disable release task when apps is not enabled. + * + * @param obj is the main structure for bpf objects. + */ +static void ebpf_dc_disable_release_task(struct dc_bpf *obj) +{ + bpf_program__set_autoload(obj->progs.netdata_dcstat_release_task_kprobe, false); + bpf_program__set_autoload(obj->progs.netdata_dcstat_release_task_fentry, false); +} + +/** + * Load and attach + * + * Load and attach the eBPF code in kernel. + * + * @param obj is the main structure for bpf objects. + * @param em structure with configuration + * + * @return it returns 0 on succes and -1 otherwise + */ +static inline int ebpf_dc_load_and_attach(struct dc_bpf *obj, ebpf_module_t *em) +{ + netdata_ebpf_program_loaded_t test = ebpf_dc_update_load(em); + if (test == EBPF_LOAD_TRAMPOLINE) { + ebpf_dc_disable_probes(obj); + + ebpf_dc_set_trampoline_target(obj); + } else { + ebpf_dc_disable_trampoline(obj); + } + + ebpf_dc_adjust_map_size(obj, em); + + if (!em->apps_charts && !em->cgroup_charts) + ebpf_dc_disable_release_task(obj); + + int ret = dc_bpf__load(obj); + if (ret) { + return ret; + } + + ret = (test == EBPF_LOAD_TRAMPOLINE) ? dc_bpf__attach(obj) : ebpf_dc_attach_probes(obj); + if (!ret) { + ebpf_dc_set_hash_tables(obj); + + ebpf_update_controller(dcstat_maps[NETDATA_DCSTAT_CTRL].map_fd, em); + } + + return ret; +} +#endif + +/***************************************************************** + * + * COMMON FUNCTIONS + * + *****************************************************************/ + +/** + * Update publish + * + * Update publish values before to write dimension. + * + * @param out structure that will receive data. + * @param cache_access number of access to directory cache. + * @param not_found number of files not found on the file system + */ +void dcstat_update_publish(netdata_publish_dcstat_t *out, uint64_t cache_access, uint64_t not_found) +{ + NETDATA_DOUBLE successful_access = (NETDATA_DOUBLE) (((long long)cache_access) - ((long long)not_found)); + NETDATA_DOUBLE ratio = (cache_access) ? successful_access/(NETDATA_DOUBLE)cache_access : 0; + + out->ratio = (long long )(ratio*100); +} + +/***************************************************************** + * + * FUNCTIONS TO CLOSE THE THREAD + * + *****************************************************************/ + +/** + * Clean names + * + * Clean the optional names allocated during startup. + */ +void ebpf_dcstat_clean_names() +{ + size_t i = 0; + while (dc_optional_name[i].program_name) { + freez(dc_optional_name[i].optional); + i++; + } +} + +/** + * DCstat Free + * + * Cleanup variables after child threads to stop + * + * @param ptr thread data. + */ +static void ebpf_dcstat_free(ebpf_module_t *em ) +{ + pthread_mutex_lock(&ebpf_exit_cleanup); + if (em->thread->enabled == NETDATA_THREAD_EBPF_RUNNING) { + em->thread->enabled = NETDATA_THREAD_EBPF_STOPPING; + pthread_mutex_unlock(&ebpf_exit_cleanup); + return; + } + pthread_mutex_unlock(&ebpf_exit_cleanup); + + freez(dcstat_vector); + freez(dcstat_values); + freez(dcstat_threads.thread); + + ebpf_cleanup_publish_syscall(dcstat_counter_publish_aggregated); + + ebpf_dcstat_clean_names(); + +#ifdef LIBBPF_MAJOR_VERSION + if (bpf_obj) + dc_bpf__destroy(bpf_obj); +#endif + + pthread_mutex_lock(&ebpf_exit_cleanup); + em->thread->enabled = NETDATA_THREAD_EBPF_STOPPED; + pthread_mutex_unlock(&ebpf_exit_cleanup); +} + +/** + * DCstat exit + * + * Cancel child and exit. + * + * @param ptr thread data. + */ +static void ebpf_dcstat_exit(void *ptr) +{ + ebpf_module_t *em = (ebpf_module_t *)ptr; + netdata_thread_cancel(*dcstat_threads.thread); + ebpf_dcstat_free(em); +} + +/** + * Clean up the main thread. + * + * @param ptr thread data. + */ +static void ebpf_dcstat_cleanup(void *ptr) +{ + ebpf_module_t *em = (ebpf_module_t *)ptr; + ebpf_dcstat_free(em); +} + +/***************************************************************** + * + * APPS + * + *****************************************************************/ + +/** + * Create apps charts + * + * Call ebpf_create_chart to create the charts on apps submenu. + * + * @param em a pointer to the structure with the default values. + */ +void ebpf_dcstat_create_apps_charts(struct ebpf_module *em, void *ptr) +{ + struct target *root = ptr; + ebpf_create_charts_on_apps(NETDATA_DC_HIT_CHART, + "Percentage of files inside directory cache", + EBPF_COMMON_DIMENSION_PERCENTAGE, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, + 20100, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + root, em->update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_charts_on_apps(NETDATA_DC_REFERENCE_CHART, + "Count file access", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_STACKED, + 20101, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + root, em->update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_charts_on_apps(NETDATA_DC_REQUEST_NOT_CACHE_CHART, + "Files not present inside directory cache", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_STACKED, + 20102, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + root, em->update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_charts_on_apps(NETDATA_DC_REQUEST_NOT_FOUND_CHART, + "Files not found", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_STACKED, + 20103, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + root, em->update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + em->apps_charts |= NETDATA_EBPF_APPS_FLAG_CHART_CREATED; +} + +/***************************************************************** + * + * MAIN LOOP + * + *****************************************************************/ + +/** + * Apps Accumulator + * + * Sum all values read from kernel and store in the first address. + * + * @param out the vector with read values. + */ +static void dcstat_apps_accumulator(netdata_dcstat_pid_t *out) +{ + int i, end = (running_on_kernel >= NETDATA_KERNEL_V4_15) ? ebpf_nprocs : 1; + netdata_dcstat_pid_t *total = &out[0]; + for (i = 1; i < end; i++) { + netdata_dcstat_pid_t *w = &out[i]; + total->cache_access += w->cache_access; + total->file_system += w->file_system; + total->not_found += w->not_found; + } +} + +/** + * Save PID values + * + * Save the current values inside the structure + * + * @param out vector used to plot charts + * @param publish vector with values read from hash tables. + */ +static inline void dcstat_save_pid_values(netdata_publish_dcstat_t *out, netdata_dcstat_pid_t *publish) +{ + memcpy(&out->curr, &publish[0], sizeof(netdata_dcstat_pid_t)); +} + +/** + * Fill PID + * + * Fill PID structures + * + * @param current_pid pid that we are collecting data + * @param out values read from hash tables; + */ +static void dcstat_fill_pid(uint32_t current_pid, netdata_dcstat_pid_t *publish) +{ + netdata_publish_dcstat_t *curr = dcstat_pid[current_pid]; + if (!curr) { + curr = callocz(1, sizeof(netdata_publish_dcstat_t)); + dcstat_pid[current_pid] = curr; + } + + dcstat_save_pid_values(curr, publish); +} + +/** + * Read APPS table + * + * Read the apps table and store data inside the structure. + */ +static void read_apps_table() +{ + netdata_dcstat_pid_t *cv = dcstat_vector; + uint32_t key; + struct pid_stat *pids = root_of_pids; + int fd = dcstat_maps[NETDATA_DCSTAT_PID_STATS].map_fd; + size_t length = sizeof(netdata_dcstat_pid_t)*ebpf_nprocs; + while (pids) { + key = pids->pid; + + if (bpf_map_lookup_elem(fd, &key, cv)) { + pids = pids->next; + continue; + } + + dcstat_apps_accumulator(cv); + + dcstat_fill_pid(key, cv); + + // We are cleaning to avoid passing data read from one process to other. + memset(cv, 0, length); + + pids = pids->next; + } +} + +/** + * Update cgroup + * + * Update cgroup data based in + */ +static void ebpf_update_dc_cgroup() +{ + netdata_dcstat_pid_t *cv = dcstat_vector; + int fd = dcstat_maps[NETDATA_DCSTAT_PID_STATS].map_fd; + size_t length = sizeof(netdata_dcstat_pid_t)*ebpf_nprocs; + + ebpf_cgroup_target_t *ect; + pthread_mutex_lock(&mutex_cgroup_shm); + for (ect = ebpf_cgroup_pids; ect; ect = ect->next) { + struct pid_on_target2 *pids; + for (pids = ect->pids; pids; pids = pids->next) { + int pid = pids->pid; + netdata_dcstat_pid_t *out = &pids->dc; + if (likely(dcstat_pid) && dcstat_pid[pid]) { + netdata_publish_dcstat_t *in = dcstat_pid[pid]; + + memcpy(out, &in->curr, sizeof(netdata_dcstat_pid_t)); + } else { + memset(cv, 0, length); + if (bpf_map_lookup_elem(fd, &pid, cv)) { + continue; + } + + dcstat_apps_accumulator(cv); + + memcpy(out, cv, sizeof(netdata_dcstat_pid_t)); + } + } + } + pthread_mutex_unlock(&mutex_cgroup_shm); +} + +/** + * Read global table + * + * Read the table with number of calls for all functions + */ +static void read_global_table() +{ + uint32_t idx; + netdata_idx_t *val = dcstat_hash_values; + netdata_idx_t *stored = dcstat_values; + int fd = dcstat_maps[NETDATA_DCSTAT_GLOBAL_STATS].map_fd; + + for (idx = NETDATA_KEY_DC_REFERENCE; idx < NETDATA_DIRECTORY_CACHE_END; idx++) { + if (!bpf_map_lookup_elem(fd, &idx, stored)) { + int i; + int end = ebpf_nprocs; + netdata_idx_t total = 0; + for (i = 0; i < end; i++) + total += stored[i]; + + val[idx] = total; + } + } +} + +/** + * DCstat read hash + * + * This is the thread callback. + * This thread is necessary, because we cannot freeze the whole plugin to read the data. + * + * @param ptr It is a NULL value for this thread. + * + * @return It always returns NULL. + */ +void *ebpf_dcstat_read_hash(void *ptr) +{ + netdata_thread_cleanup_push(ebpf_dcstat_cleanup, ptr); + heartbeat_t hb; + heartbeat_init(&hb); + + ebpf_module_t *em = (ebpf_module_t *)ptr; + + usec_t step = NETDATA_LATENCY_DCSTAT_SLEEP_MS * em->update_every; + while (!ebpf_exit_plugin) { + (void)heartbeat_next(&hb, step); + + read_global_table(); + } + + netdata_thread_cleanup_pop(1); + return NULL; +} + +/** + * Cachestat sum PIDs + * + * Sum values for all PIDs associated to a group + * + * @param publish output structure. + * @param root structure with listed IPs + */ +void ebpf_dcstat_sum_pids(netdata_publish_dcstat_t *publish, struct pid_on_target *root) +{ + memset(&publish->curr, 0, sizeof(netdata_dcstat_pid_t)); + netdata_dcstat_pid_t *dst = &publish->curr; + while (root) { + int32_t pid = root->pid; + netdata_publish_dcstat_t *w = dcstat_pid[pid]; + if (w) { + netdata_dcstat_pid_t *src = &w->curr; + dst->cache_access += src->cache_access; + dst->file_system += src->file_system; + dst->not_found += src->not_found; + } + + root = root->next; + } +} + +/** + * Send data to Netdata calling auxiliary functions. + * + * @param root the target list. +*/ +void ebpf_dcache_send_apps_data(struct target *root) +{ + struct target *w; + collected_number value; + + write_begin_chart(NETDATA_APPS_FAMILY, NETDATA_DC_HIT_CHART); + for (w = root; w; w = w->next) { + if (unlikely(w->exposed && w->processes)) { + ebpf_dcstat_sum_pids(&w->dcstat, w->root_pid); + + uint64_t cache = w->dcstat.curr.cache_access; + uint64_t not_found = w->dcstat.curr.not_found; + + dcstat_update_publish(&w->dcstat, cache, not_found); + value = (collected_number) w->dcstat.ratio; + write_chart_dimension(w->name, value); + } + } + write_end_chart(); + + write_begin_chart(NETDATA_APPS_FAMILY, NETDATA_DC_REFERENCE_CHART); + for (w = root; w; w = w->next) { + if (unlikely(w->exposed && w->processes)) { + if (w->dcstat.curr.cache_access < w->dcstat.prev.cache_access) { + w->dcstat.prev.cache_access = 0; + } + + w->dcstat.cache_access = (long long)w->dcstat.curr.cache_access - (long long)w->dcstat.prev.cache_access; + value = (collected_number) w->dcstat.cache_access; + write_chart_dimension(w->name, value); + w->dcstat.prev.cache_access = w->dcstat.curr.cache_access; + } + } + write_end_chart(); + + write_begin_chart(NETDATA_APPS_FAMILY, NETDATA_DC_REQUEST_NOT_CACHE_CHART); + for (w = root; w; w = w->next) { + if (unlikely(w->exposed && w->processes)) { + if (w->dcstat.curr.file_system < w->dcstat.prev.file_system) { + w->dcstat.prev.file_system = 0; + } + + value = (collected_number) (!w->dcstat.cache_access) ? 0 : + (long long )w->dcstat.curr.file_system - (long long)w->dcstat.prev.file_system; + write_chart_dimension(w->name, value); + w->dcstat.prev.file_system = w->dcstat.curr.file_system; + } + } + write_end_chart(); + + write_begin_chart(NETDATA_APPS_FAMILY, NETDATA_DC_REQUEST_NOT_FOUND_CHART); + for (w = root; w; w = w->next) { + if (unlikely(w->exposed && w->processes)) { + if (w->dcstat.curr.not_found < w->dcstat.prev.not_found) { + w->dcstat.prev.not_found = 0; + } + value = (collected_number) (!w->dcstat.cache_access) ? 0 : + (long long)w->dcstat.curr.not_found - (long long)w->dcstat.prev.not_found; + write_chart_dimension(w->name, value); + w->dcstat.prev.not_found = w->dcstat.curr.not_found; + } + } + write_end_chart(); +} + +/** + * Send global + * + * Send global charts to Netdata + */ +static void dcstat_send_global(netdata_publish_dcstat_t *publish) +{ + dcstat_update_publish(publish, dcstat_hash_values[NETDATA_KEY_DC_REFERENCE], + dcstat_hash_values[NETDATA_KEY_DC_MISS]); + + netdata_publish_syscall_t *ptr = dcstat_counter_publish_aggregated; + netdata_idx_t value = dcstat_hash_values[NETDATA_KEY_DC_REFERENCE]; + if (value != ptr[NETDATA_DCSTAT_IDX_REFERENCE].pcall) { + ptr[NETDATA_DCSTAT_IDX_REFERENCE].ncall = value - ptr[NETDATA_DCSTAT_IDX_REFERENCE].pcall; + ptr[NETDATA_DCSTAT_IDX_REFERENCE].pcall = value; + + value = dcstat_hash_values[NETDATA_KEY_DC_SLOW]; + ptr[NETDATA_DCSTAT_IDX_SLOW].ncall = value - ptr[NETDATA_DCSTAT_IDX_SLOW].pcall; + ptr[NETDATA_DCSTAT_IDX_SLOW].pcall = value; + + value = dcstat_hash_values[NETDATA_KEY_DC_MISS]; + ptr[NETDATA_DCSTAT_IDX_MISS].ncall = value - ptr[NETDATA_DCSTAT_IDX_MISS].pcall; + ptr[NETDATA_DCSTAT_IDX_MISS].pcall = value; + } else { + ptr[NETDATA_DCSTAT_IDX_REFERENCE].ncall = 0; + ptr[NETDATA_DCSTAT_IDX_SLOW].ncall = 0; + ptr[NETDATA_DCSTAT_IDX_MISS].ncall = 0; + } + + ebpf_one_dimension_write_charts(NETDATA_FILESYSTEM_FAMILY, NETDATA_DC_HIT_CHART, + ptr[NETDATA_DCSTAT_IDX_RATIO].dimension, publish->ratio); + + write_count_chart( + NETDATA_DC_REFERENCE_CHART, NETDATA_FILESYSTEM_FAMILY, + &dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_REFERENCE], 3); +} + +/** + * Create specific directory cache charts + * + * Create charts for cgroup/application. + * + * @param type the chart type. + * @param update_every value to overwrite the update frequency set by the server. + */ +static void ebpf_create_specific_dc_charts(char *type, int update_every) +{ + ebpf_create_chart(type, NETDATA_DC_HIT_CHART, "Percentage of files inside directory cache", + EBPF_COMMON_DIMENSION_PERCENTAGE, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_CGROUP_DC_HIT_RATIO_CONTEXT, NETDATA_EBPF_CHART_TYPE_LINE, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5700, + ebpf_create_global_dimension, + dcstat_counter_publish_aggregated, 1, update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_chart(type, NETDATA_DC_REFERENCE_CHART, "Count file access", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_CGROUP_DC_REFERENCE_CONTEXT, NETDATA_EBPF_CHART_TYPE_LINE, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5701, + ebpf_create_global_dimension, + &dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_REFERENCE], 1, + update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_chart(type, NETDATA_DC_REQUEST_NOT_CACHE_CHART, + "Files not present inside directory cache", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_CGROUP_DC_NOT_CACHE_CONTEXT, NETDATA_EBPF_CHART_TYPE_LINE, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5702, + ebpf_create_global_dimension, + &dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_SLOW], 1, + update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_chart(type, NETDATA_DC_REQUEST_NOT_FOUND_CHART, + "Files not found", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_CGROUP_DC_NOT_FOUND_CONTEXT, NETDATA_EBPF_CHART_TYPE_LINE, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5703, + ebpf_create_global_dimension, + &dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_MISS], 1, + update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); +} + +/** + * Obsolete specific directory cache charts + * + * Obsolete charts for cgroup/application. + * + * @param type the chart type. + * @param update_every value to overwrite the update frequency set by the server. + */ +static void ebpf_obsolete_specific_dc_charts(char *type, int update_every) +{ + ebpf_write_chart_obsolete(type, NETDATA_DC_HIT_CHART, + "Percentage of files inside directory cache", + EBPF_COMMON_DIMENSION_PERCENTAGE, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, NETDATA_CGROUP_DC_HIT_RATIO_CONTEXT, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5700, update_every); + + ebpf_write_chart_obsolete(type, NETDATA_DC_REFERENCE_CHART, + "Count file access", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, NETDATA_CGROUP_DC_REFERENCE_CONTEXT, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5701, update_every); + + ebpf_write_chart_obsolete(type, NETDATA_DC_REQUEST_NOT_CACHE_CHART, + "Files not present inside directory cache", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, NETDATA_CGROUP_DC_NOT_CACHE_CONTEXT, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5702, update_every); + + ebpf_write_chart_obsolete(type, NETDATA_DC_REQUEST_NOT_FOUND_CHART, + "Files not found", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, NETDATA_CGROUP_DC_NOT_FOUND_CONTEXT, + NETDATA_CHART_PRIO_CGROUPS_CONTAINERS + 5703, update_every); +} + +/** + * Cachestat sum PIDs + * + * Sum values for all PIDs associated to a group + * + * @param publish output structure. + * @param root structure with listed IPs + */ +void ebpf_dc_sum_cgroup_pids(netdata_publish_dcstat_t *publish, struct pid_on_target2 *root) +{ + memset(&publish->curr, 0, sizeof(netdata_dcstat_pid_t)); + netdata_dcstat_pid_t *dst = &publish->curr; + while (root) { + netdata_dcstat_pid_t *src = &root->dc; + + dst->cache_access += src->cache_access; + dst->file_system += src->file_system; + dst->not_found += src->not_found; + + root = root->next; + } +} + +/** + * Calc chart values + * + * Do necessary math to plot charts. + */ +void ebpf_dc_calc_chart_values() +{ + ebpf_cgroup_target_t *ect; + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + ebpf_dc_sum_cgroup_pids(&ect->publish_dc, ect->pids); + uint64_t cache = ect->publish_dc.curr.cache_access; + uint64_t not_found = ect->publish_dc.curr.not_found; + + dcstat_update_publish(&ect->publish_dc, cache, not_found); + + ect->publish_dc.cache_access = (long long)ect->publish_dc.curr.cache_access - + (long long)ect->publish_dc.prev.cache_access; + ect->publish_dc.prev.cache_access = ect->publish_dc.curr.cache_access; + + if (ect->publish_dc.curr.not_found < ect->publish_dc.prev.not_found) { + ect->publish_dc.prev.not_found = 0; + } + } +} + +/** + * Create Systemd directory cache Charts + * + * Create charts when systemd is enabled + * + * @param update_every value to overwrite the update frequency set by the server. + **/ +static void ebpf_create_systemd_dc_charts(int update_every) +{ + ebpf_create_charts_on_systemd(NETDATA_DC_HIT_CHART, + "Percentage of files inside directory cache", + EBPF_COMMON_DIMENSION_PERCENTAGE, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, + 21200, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + NETDATA_SYSTEMD_DC_HIT_RATIO_CONTEXT, NETDATA_EBPF_MODULE_NAME_DCSTAT, + update_every); + + ebpf_create_charts_on_systemd(NETDATA_DC_REFERENCE_CHART, + "Count file access", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, + 21201, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + NETDATA_SYSTEMD_DC_REFERENCE_CONTEXT, NETDATA_EBPF_MODULE_NAME_DCSTAT, + update_every); + + ebpf_create_charts_on_systemd(NETDATA_DC_REQUEST_NOT_CACHE_CHART, + "Files not present inside directory cache", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, + 21202, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + NETDATA_SYSTEMD_DC_NOT_CACHE_CONTEXT, NETDATA_EBPF_MODULE_NAME_DCSTAT, + update_every); + + ebpf_create_charts_on_systemd(NETDATA_DC_REQUEST_NOT_FOUND_CHART, + "Files not found", + EBPF_COMMON_DIMENSION_FILES, + NETDATA_DIRECTORY_CACHE_SUBMENU, + NETDATA_EBPF_CHART_TYPE_LINE, + 21202, + ebpf_algorithms[NETDATA_EBPF_ABSOLUTE_IDX], + NETDATA_SYSTEMD_DC_NOT_FOUND_CONTEXT, NETDATA_EBPF_MODULE_NAME_DCSTAT, + update_every); +} + +/** + * Send Directory Cache charts + * + * Send collected data to Netdata. + */ +static void ebpf_send_systemd_dc_charts() +{ + collected_number value; + ebpf_cgroup_target_t *ect; + write_begin_chart(NETDATA_SERVICE_FAMILY, NETDATA_DC_HIT_CHART); + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + if (unlikely(ect->systemd) && unlikely(ect->updated)) { + write_chart_dimension(ect->name, (long long) ect->publish_dc.ratio); + } + } + write_end_chart(); + + write_begin_chart(NETDATA_SERVICE_FAMILY, NETDATA_DC_REFERENCE_CHART); + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + if (unlikely(ect->systemd) && unlikely(ect->updated)) { + write_chart_dimension(ect->name, (long long) ect->publish_dc.cache_access); + } + } + write_end_chart(); + + write_begin_chart(NETDATA_SERVICE_FAMILY, NETDATA_DC_REQUEST_NOT_CACHE_CHART); + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + if (unlikely(ect->systemd) && unlikely(ect->updated)) { + value = (collected_number) (!ect->publish_dc.cache_access) ? 0 : + (long long )ect->publish_dc.curr.file_system - (long long)ect->publish_dc.prev.file_system; + ect->publish_dc.prev.file_system = ect->publish_dc.curr.file_system; + + write_chart_dimension(ect->name, (long long) value); + } + } + write_end_chart(); + + write_begin_chart(NETDATA_SERVICE_FAMILY, NETDATA_DC_REQUEST_NOT_FOUND_CHART); + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + if (unlikely(ect->systemd) && unlikely(ect->updated)) { + value = (collected_number) (!ect->publish_dc.cache_access) ? 0 : + (long long)ect->publish_dc.curr.not_found - (long long)ect->publish_dc.prev.not_found; + + ect->publish_dc.prev.not_found = ect->publish_dc.curr.not_found; + + write_chart_dimension(ect->name, (long long) value); + } + } + write_end_chart(); +} + +/** + * Send Directory Cache charts + * + * Send collected data to Netdata. + * + */ +static void ebpf_send_specific_dc_data(char *type, netdata_publish_dcstat_t *pdc) +{ + collected_number value; + write_begin_chart(type, NETDATA_DC_HIT_CHART); + write_chart_dimension(dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_RATIO].name, + (long long) pdc->ratio); + write_end_chart(); + + write_begin_chart(type, NETDATA_DC_REFERENCE_CHART); + write_chart_dimension(dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_REFERENCE].name, + (long long) pdc->cache_access); + write_end_chart(); + + value = (collected_number) (!pdc->cache_access) ? 0 : + (long long )pdc->curr.file_system - (long long)pdc->prev.file_system; + pdc->prev.file_system = pdc->curr.file_system; + + write_begin_chart(type, NETDATA_DC_REQUEST_NOT_CACHE_CHART); + write_chart_dimension(dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_SLOW].name, (long long) value); + write_end_chart(); + + value = (collected_number) (!pdc->cache_access) ? 0 : + (long long)pdc->curr.not_found - (long long)pdc->prev.not_found; + pdc->prev.not_found = pdc->curr.not_found; + + write_begin_chart(type, NETDATA_DC_REQUEST_NOT_FOUND_CHART); + write_chart_dimension(dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_MISS].name, (long long) value); + write_end_chart(); +} + +/** + * Send data to Netdata calling auxiliary functions. + * + * @param update_every value to overwrite the update frequency set by the server. +*/ +void ebpf_dc_send_cgroup_data(int update_every) +{ + if (!ebpf_cgroup_pids) + return; + + pthread_mutex_lock(&mutex_cgroup_shm); + ebpf_cgroup_target_t *ect; + ebpf_dc_calc_chart_values(); + + int has_systemd = shm_ebpf_cgroup.header->systemd_enabled; + if (has_systemd) { + if (send_cgroup_chart) { + ebpf_create_systemd_dc_charts(update_every); + } + + ebpf_send_systemd_dc_charts(); + } + + for (ect = ebpf_cgroup_pids; ect ; ect = ect->next) { + if (ect->systemd) + continue; + + if (!(ect->flags & NETDATA_EBPF_CGROUP_HAS_DC_CHART) && ect->updated) { + ebpf_create_specific_dc_charts(ect->name, update_every); + ect->flags |= NETDATA_EBPF_CGROUP_HAS_DC_CHART; + } + + if (ect->flags & NETDATA_EBPF_CGROUP_HAS_DC_CHART) { + if (ect->updated) { + ebpf_send_specific_dc_data(ect->name, &ect->publish_dc); + } else { + ebpf_obsolete_specific_dc_charts(ect->name, update_every); + ect->flags &= ~NETDATA_EBPF_CGROUP_HAS_DC_CHART; + } + } + } + + pthread_mutex_unlock(&mutex_cgroup_shm); +} + +/** +* Main loop for this collector. +*/ +static void dcstat_collector(ebpf_module_t *em) +{ + dcstat_threads.thread = mallocz(sizeof(netdata_thread_t)); + dcstat_threads.start_routine = ebpf_dcstat_read_hash; + + netdata_thread_create(dcstat_threads.thread, dcstat_threads.name, NETDATA_THREAD_OPTION_DEFAULT, + ebpf_dcstat_read_hash, em); + + netdata_publish_dcstat_t publish; + memset(&publish, 0, sizeof(publish)); + int cgroups = em->cgroup_charts; + int update_every = em->update_every; + heartbeat_t hb; + heartbeat_init(&hb); + usec_t step = update_every * USEC_PER_SEC; + while (!ebpf_exit_plugin) { + (void)heartbeat_next(&hb, step); + if (ebpf_exit_plugin) + break; + + netdata_apps_integration_flags_t apps = em->apps_charts; + pthread_mutex_lock(&collect_data_mutex); + if (apps) + read_apps_table(); + + if (cgroups) + ebpf_update_dc_cgroup(); + + pthread_mutex_lock(&lock); + + dcstat_send_global(&publish); + + if (apps & NETDATA_EBPF_APPS_FLAG_CHART_CREATED) + ebpf_dcache_send_apps_data(apps_groups_root_target); + + if (cgroups) + ebpf_dc_send_cgroup_data(update_every); + + pthread_mutex_unlock(&lock); + pthread_mutex_unlock(&collect_data_mutex); + } +} + +/***************************************************************** + * + * INITIALIZE THREAD + * + *****************************************************************/ + +/** + * Create filesystem charts + * + * Call ebpf_create_chart to create the charts for the collector. + * + * @param update_every value to overwrite the update frequency set by the server. + */ +static void ebpf_create_filesystem_charts(int update_every) +{ + ebpf_create_chart(NETDATA_FILESYSTEM_FAMILY, NETDATA_DC_HIT_CHART, + "Percentage of files inside directory cache", + EBPF_COMMON_DIMENSION_PERCENTAGE, NETDATA_DIRECTORY_CACHE_SUBMENU, + NULL, + NETDATA_EBPF_CHART_TYPE_LINE, + 21200, + ebpf_create_global_dimension, + dcstat_counter_publish_aggregated, 1, update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + ebpf_create_chart(NETDATA_FILESYSTEM_FAMILY, NETDATA_DC_REFERENCE_CHART, + "Variables used to calculate hit ratio.", + EBPF_COMMON_DIMENSION_FILES, NETDATA_DIRECTORY_CACHE_SUBMENU, + NULL, + NETDATA_EBPF_CHART_TYPE_LINE, + 21201, + ebpf_create_global_dimension, + &dcstat_counter_publish_aggregated[NETDATA_DCSTAT_IDX_REFERENCE], 3, + update_every, NETDATA_EBPF_MODULE_NAME_DCSTAT); + + fflush(stdout); +} + +/** + * Allocate vectors used with this thread. + * + * We are not testing the return, because callocz does this and shutdown the software + * case it was not possible to allocate. + * + * @param apps is apps enabled? + */ +static void ebpf_dcstat_allocate_global_vectors(int apps) +{ + if (apps) + dcstat_pid = callocz((size_t)pid_max, sizeof(netdata_publish_dcstat_t *)); + + dcstat_vector = callocz((size_t)ebpf_nprocs, sizeof(netdata_dcstat_pid_t)); + dcstat_values = callocz((size_t)ebpf_nprocs, sizeof(netdata_idx_t)); + + memset(dcstat_counter_aggregated_data, 0, NETDATA_DCSTAT_IDX_END * sizeof(netdata_syscall_stat_t)); + memset(dcstat_counter_publish_aggregated, 0, NETDATA_DCSTAT_IDX_END * sizeof(netdata_publish_syscall_t)); +} + +/***************************************************************** + * + * MAIN THREAD + * + *****************************************************************/ + +/* + * Load BPF + * + * Load BPF files. + * + * @param em the structure with configuration + */ +static int ebpf_dcstat_load_bpf(ebpf_module_t *em) +{ + int ret = 0; + ebpf_adjust_apps_cgroup(em, em->targets[NETDATA_DC_TARGET_LOOKUP_FAST].mode); + if (em->load & EBPF_LOAD_LEGACY) { + em->probe_links = ebpf_load_program(ebpf_plugin_dir, em, running_on_kernel, isrh, &em->objects); + if (!em->probe_links) { + ret = -1; + } + } +#ifdef LIBBPF_MAJOR_VERSION + else { + bpf_obj = dc_bpf__open(); + if (!bpf_obj) + ret = -1; + else + ret = ebpf_dc_load_and_attach(bpf_obj, em); + } +#endif + + if (ret) + error("%s %s", EBPF_DEFAULT_ERROR_MSG, em->thread_name); + + return ret; +} + +/** + * Directory Cache thread + * + * Thread used to make dcstat thread + * + * @param ptr a pointer to `struct ebpf_module` + * + * @return It always returns NULL + */ +void *ebpf_dcstat_thread(void *ptr) +{ + netdata_thread_cleanup_push(ebpf_dcstat_exit, ptr); + + ebpf_module_t *em = (ebpf_module_t *)ptr; + em->maps = dcstat_maps; + + ebpf_update_pid_table(&dcstat_maps[NETDATA_DCSTAT_PID_STATS], em); + + ebpf_update_names(dc_optional_name, em); + +#ifdef LIBBPF_MAJOR_VERSION + ebpf_adjust_thread_load(em, default_btf); +#endif + if (ebpf_dcstat_load_bpf(em)) { + em->thread->enabled = NETDATA_THREAD_EBPF_STOPPED; + goto enddcstat; + } + + ebpf_dcstat_allocate_global_vectors(em->apps_charts); + + int algorithms[NETDATA_DCSTAT_IDX_END] = { + NETDATA_EBPF_ABSOLUTE_IDX, NETDATA_EBPF_ABSOLUTE_IDX, NETDATA_EBPF_ABSOLUTE_IDX, + NETDATA_EBPF_ABSOLUTE_IDX + }; + + ebpf_global_labels(dcstat_counter_aggregated_data, dcstat_counter_publish_aggregated, + dcstat_counter_dimension_name, dcstat_counter_dimension_name, + algorithms, NETDATA_DCSTAT_IDX_END); + + pthread_mutex_lock(&lock); + ebpf_create_filesystem_charts(em->update_every); + ebpf_update_stats(&plugin_statistics, em); + pthread_mutex_unlock(&lock); + + dcstat_collector(em); + +enddcstat: + ebpf_update_disabled_plugin_stats(em); + + netdata_thread_cleanup_pop(1); + return NULL; +} |