u32 index; if (info->index >= map->max_entries) return NULL; if (*pos == 0) ++*pos; array = container_of(map, struct bpf_array, map); index = info->index & array->index_mask; if (info->percpu_value_buf) return array->pptrs[index]; return array->value + array->elem_size * index;} static void *bpf_array_map_seq_next(struct seq_file *seq, void *v, loff_t *pos){ struct bpf_iter_seq_array_map_info *info = seq->private; struct bpf_map *map = info->map; struct bpf_array *array; u32 index; ++*pos; ++info->index; if (info->index >= map->max_entries) return NULL; array = container_of(map, struct bpf_array, map); index = info->index & array->index_mask; if (info->percpu_value_buf) return array->pptrs[index]; return array->value + array->elem_size * index;} static int __bpf_array_map_seq_show(struct seq_file *seq, void *v){ struct bpf_iter_seq_array_map_info *info = seq->private; struct bpf_iter__bpf_map_elem ctx = {}; struct bpf_map *map = info->map; struct bpf_iter_meta meta; struct bpf_prog *prog; int off = 0, cpu = 0; void __percpu **pptr; u32 size; meta.seq = seq; prog = bpf_iter_get_info(&meta, v == NULL); if (!prog) return 0; ctx.meta = &meta; ctx.map = info->map; if (v) { ctx.key = &info->index; if (!info->percpu_value_buf) { ctx.value = v; } else { pptr = v; size = round_up(map->value_size, 8); for_each_possible_cpu(cpu) { bpf_long_memcpy(info->percpu_value_buf + off, per_cpu_ptr(pptr, cpu), size); off += size; } ctx.value = info->percpu_value_buf; } } return bpf_iter_run_prog(prog, &ctx);} static int bpf_array_map_seq_show(struct seq_file *seq, void *v){ return __bpf_array_map_seq_show(seq, v);} static void bpf_array_map_seq_stop(struct seq_file *seq, void *v){ if (!v) (void)__bpf_array_map_seq_show(seq, NULL);} static int bpf_iter_init_array_map(void *priv_data, struct bpf_iter_aux_info *aux){ struct bpf_iter_seq_array_map_info *seq_info = priv_data; struct bpf_map *map = aux->map; void *value_buf; u32 buf_size; if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { buf_size = round_up(map->value_size, 8) * num_possible_cpus(); value_buf = kmalloc(buf_size, GFP_USER | __GFP_NOWARN); if (!value_buf) return -ENOMEM; seq_info->percpu_value_buf = value_buf; } seq_info->map = map; return 0;} static void bpf_iter_fini_array_map(void *priv_data){ struct bpf_iter_seq_array_map_info *seq_info = priv_data; kfree(seq_info->percpu_value_buf);} static const struct seq_operations bpf_array_map_seq_ops = { .start = bpf_array_map_seq_start, .next = bpf_array_map_seq_next, .stop = bpf_array_map_seq_stop, .show = bpf_array_map_seq_show,}; static const struct bpf_iter_seq_info iter_seq_info = { .seq_ops = &bpf_array_map_seq_ops, .init_seq_private = bpf_iter_init_array_map, .fini_seq_private = bpf_iter_fini_array_map, .seq_priv_size = sizeof(struct bpf_iter_seq_array_map_info),}; static int array_map_btf_id;const struct bpf_map_ops array_map_ops = { .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = array_map_lookup_elem, .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, .map_gen_lookup = array_map_gen_lookup, .map_direct_value_addr = array_map_direct_value_addr, .map_direct_value_meta = array_map_direct_value_meta, .map_mmap = array_map_mmap, .map_seq_show_elem = array_map_seq_show_elem, .map_check_btf = array_map_check_btf, .map_lookup_batch = generic_map_lookup_batch, .map_update_batch = generic_map_update_batch, .map_btf_name = "bpf_array", .map_btf_id = &array_map_btf_id, .iter_seq_info = &iter_seq_info,}; static int percpu_array_map_btf_id;const struct bpf_map_ops percpu_array_map_ops = { .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = percpu_array_map_lookup_elem, .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, .map_seq_show_elem = percpu_array_map_seq_show_elem, .map_check_btf = array_map_check_btf, .map_btf_name = "bpf_array", .map_btf_id = &percpu_array_map_btf_id, .iter_seq_info = &iter_seq_info,}; static int fd_array_map_alloc_check(union bpf_attr *attr){ /* only file descriptors can be stored in this type of map */ if (attr->value_size != sizeof(u32)) return -EINVAL; /* Program read-only/write-only not supported for special maps yet. */ if (attr->map_flags & (BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG)) return -EINVAL; return array_map_alloc_check(attr);} static void fd_array_map_free(struct bpf_map *map){ struct bpf_array *array = container_of(map, struct bpf_array, map); int i; /* make sure it's empty */ for (i = 0; i < array->map.max_entries; i++) BUG_ON(array->ptrs[i] != NULL); bpf_map_area_free(array);} static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key){ return ERR_PTR(-EOPNOTSUPP);} /* only called from syscall */int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value){ void **elem, *ptr; int ret = 0; if (!map->ops->map_fd_sys_lookup_elem) return -ENOTSUPP; rcu_read_lock(); elem = array_map_lookup_elem(map, key); if (elem && (ptr = READ_ONCE(*elem))) *value = map->ops->map_fd_sys_lookup_elem(ptr); else ret = -ENOENT; rcu_read_unlock(); return ret;} /* only called from syscall */int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags){ struct bpf_array *array = container_of(map, struct bpf_array, map); void *new_ptr, *old_ptr; u32 index = *(u32 *)key, ufd; if (map_flags != BPF_ANY) return -EINVAL; if (index >= array->map.max_entries) return -E2BIG; ufd = *(u32 *)value; new_ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); if (IS_ERR(new_ptr)) return PTR_ERR(new_ptr); if (map->ops->map_poke_run) { mutex_lock(&array->aux->poke_mutex); old_ptr = xchg(array->ptrs + index, new_ptr); map->ops->map_poke_run(map, index, old_ptr, new_ptr); mutex_unlock(&array->aux->poke_mutex); } else { old_ptr = xchg(array->ptrs + index, new_ptr); } if (old_ptr) map->ops->map_fd_put_ptr(old_ptr); return 0;} static int fd_array_map_delete_elem(struct bpf_map *map, void *key){ struct bpf_array *array = container_of(map, struct bpf_array, map); void *old_ptr; u32 index = *(u32 *)key; if (index >= array->map.max_entries) return -E2BIG; if (map->ops->map_poke_run) { mutex_lock(&array->aux->poke_mutex); old_ptr = xchg(array->ptrs + index, NULL); map->ops->map_poke_run(map, index, old_ptr, NULL); mutex_unlock(&array->aux->poke_mutex); } else { old_ptr = xchg(array->ptrs + index, NULL); } if (old_ptr) { map->ops->map_fd_put_ptr(old_ptr); return 0; } else { return -ENOENT; }} static void *prog_fd_array_get_ptr(struct bpf_map *map, struct file *map_file, int fd){ struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_prog *prog = bpf_prog_get(fd); if (IS_ERR(prog)) return prog; if (!bpf_prog_array_compatible(array, prog)) { bpf_prog_put(prog); return ERR_PTR(-EINVAL); } return prog;} static void prog_fd_array_put_ptr(void *ptr){ bpf_prog_put(ptr);} static u32 prog_fd_array_sys_lookup_elem(void *ptr){ return ((struct bpf_prog *)ptr)->aux->id;} /* decrement refcnt of all bpf_progs that are stored in this map */static void bpf_fd_array_map_clear(struct bpf_map *map){ struct bpf_array *array = container_of(map, struct bpf_array, map); int i; for (i = 0; i < array->map.max_entries; i++) fd_array_map_delete_elem(map, &i);} static void prog_array_map_seq_show_elem(struct bpf_map *map, void *key, struct seq_file *m){ void **elem, *ptr; u32 prog_id; rcu_read_lock(); elem = array_map_lookup_elem(map, key); if (elem) { ptr = READ_ONCE(*elem); if (ptr) { seq_printf(m, "%u: ", *(u32 *)key); prog_id = prog_fd_array_sys_lookup_elem(ptr); btf_type_seq_show(map->btf, map->btf_value_type_id, &prog_id, m); seq_puts(m, "\n"); } } rcu_read_unlock();} struct prog_poke_elem { struct list_head list; struct bpf_prog_aux *aux;}; static int prog_array_map_poke_track(struct bpf_map *map, struct bpf_prog_aux *prog_aux){ struct prog_poke_elem *elem; struct bpf_array_aux *aux; int ret = 0; aux = container_of(map, struct bpf_array, map)->aux; mutex_lock(&aux->poke_mutex); list_for_each_entry(elem, &aux->poke_progs, list) { if (elem->aux == prog_aux) goto out; } elem = kmalloc(sizeof(*elem), GFP_KERNEL); if (!elem) { ret = -ENOMEM; goto out; } INIT_LIST_HEAD(&elem->list); /* We must track the program's aux info at this point in time * since the program pointer itself may not be stable yet, see * also comment in prog_array_map_poke_run(). */ elem->aux = prog_aux; list_add_tail(&elem->list, &aux->poke_progs);out: mutex_unlock(&aux->poke_mutex); return ret;} static void prog_array_map_poke_untrack(struct bpf_map *map, struct bpf_prog_aux *prog_aux){ struct prog_poke_elem *elem, *tmp; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; mutex_lock(&aux->poke_mutex); list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) { if (elem->aux == prog_aux) { list_del_init(&elem->list); kfree(elem); break; } } mutex_unlock(&aux->poke_mutex);} static void prog_array_map_poke_run(struct bpf_map *map, u32 key, struct bpf_prog *old, struct bpf_prog *new){ struct prog_poke_elem *elem; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; WARN_ON_ONCE(!mutex_is_locked(&aux->poke_mutex)); list_for_each_entry(elem, &aux->poke_progs, list) { struct bpf_jit_poke_descriptor *poke; int i, ret; for (i = 0; i < elem->aux->size_poke_tab; i++) { poke = &elem->aux->poke_tab[i]; /* Few things to be aware of: * * 1) We can only ever access aux in this context, but * not aux->prog since it might not be stable yet and * there could be danger of use after free otherwise. * 2) Initially when we start tracking aux, the program * is not JITed yet and also does not have a kallsyms * entry. We skip these as poke->ip_stable is not * active yet. The JIT will do the final fixup before * setting it stable. The various poke->ip_stable are * successively activated, so tail call updates can * arrive from here while JIT is still finishing its * final fixup for non-activated poke entries. * 3) On program teardown, the program's kallsym entry gets * removed out of RCU callback, but we can only untrack * from sleepable context, therefore bpf_arch_text_poke() * might not see that this is in BPF text section and * bails out with -EINVAL. As these are unreachable since * RCU grace period already passed, we simply skip them. * 4) Also programs reaching refcount of zero while patching * is in progress is okay since we're protected under * poke_mutex and untrack the programs before the JIT * buffer is freed. When we're still in the middle of * patching and suddenly kallsyms entry of the program * gets evicted, we just skip the rest which is fine due * to point 3). * 5) Any other error happening below from bpf_arch_text_poke() * is a unexpected bug. */ if (!READ_ONCE(poke->ip_stable)) continue; if (poke->reason != BPF_POKE_REASON_TAIL_CALL) continue; if (poke->tail_call.map != map || poke->tail_call.key != key) continue; ret = bpf_arch_text_poke(poke->ip, BPF_MOD_JUMP, old ? (u8 *)old->bpf_func + poke->adj_off : NULL, new ? (u8 *)new->bpf_func + poke->adj_off : NULL); BUG_ON(ret < 0 && ret != -EINVAL); } }} static void prog_array_map_clear_deferred(struct work_struct *work){ struct bpf_map *map = container_of(work, struct bpf_array_aux, work)->map; bpf_fd_array_map_clear(map); bpf_map_put(map);} static void prog_array_map_clear(struct bpf_map *map){ struct bpf_array_aux *aux = container_of(map, struct bpf_array, map)->aux; bpf_map_inc(map); schedule_work(&aux->work);} static struct bpf_map *prog_array_map_alloc(union bpf_attr *attr){ struct bpf_array_aux *aux; struct bpf_map *map; aux = kzalloc(sizeof(*aux), GFP_KERNEL); if (!aux) return ERR_PTR(-ENOMEM); INIT_WORK(&aux->work, prog_array_map_clear_deferred); INIT_LIST_HEAD(&aux->poke_progs); mutex_init(&aux->poke_mutex); map = array_map_alloc(attr); if (IS_ERR(map)) { kfree(aux); return map; } container_of(map, struct bpf_array, map)->aux = aux; aux->map = map; return map;} static void prog_array_map_free(struct bpf_map *map){ struct prog_poke_elem *elem, *tmp; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) { list_del_init(&elem->list); kfree(elem);Lines 501–1000 of 1239. Previous lines Load more
u32 index; if (info->index >= map->max_entries) return NULL; if (*pos == 0) ++*pos; array = container_of(map, struct bpf_array, map); index = info->index & array->index_mask; if (info->percpu_value_buf) return array->pptrs[index]; return array->value + array->elem_size * index; } static void *bpf_array_map_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct bpf_iter_seq_array_map_info *info = seq->private; struct bpf_map *map = info->map; struct bpf_array *array; u32 index; ++*pos; ++info->index; if (info->index >= map->max_entries) return NULL; array = container_of(map, struct bpf_array, map); index = info->index & array->index_mask; if (info->percpu_value_buf) return array->pptrs[index]; return array->value + array->elem_size * index; } static int __bpf_array_map_seq_show(struct seq_file *seq, void *v) { struct bpf_iter_seq_array_map_info *info = seq->private; struct bpf_iter__bpf_map_elem ctx = {}; struct bpf_map *map = info->map; struct bpf_iter_meta meta; struct bpf_prog *prog; int off = 0, cpu = 0; void __percpu **pptr; u32 size; meta.seq = seq; prog = bpf_iter_get_info(&meta, v == NULL); if (!prog) return 0; ctx.meta = &meta; ctx.map = info->map; if (v) { ctx.key = &info->index; if (!info->percpu_value_buf) { ctx.value = v; } else { pptr = v; size = round_up(map->value_size, 8); for_each_possible_cpu(cpu) { bpf_long_memcpy(info->percpu_value_buf + off, per_cpu_ptr(pptr, cpu), size); off += size; } ctx.value = info->percpu_value_buf; } } return bpf_iter_run_prog(prog, &ctx); } static int bpf_array_map_seq_show(struct seq_file *seq, void *v) { return __bpf_array_map_seq_show(seq, v); } static void bpf_array_map_seq_stop(struct seq_file *seq, void *v) { if (!v) (void)__bpf_array_map_seq_show(seq, NULL); } static int bpf_iter_init_array_map(void *priv_data, struct bpf_iter_aux_info *aux) { struct bpf_iter_seq_array_map_info *seq_info = priv_data; struct bpf_map *map = aux->map; void *value_buf; u32 buf_size; if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { buf_size = round_up(map->value_size, 8) * num_possible_cpus(); value_buf = kmalloc(buf_size, GFP_USER | __GFP_NOWARN); if (!value_buf) return -ENOMEM; seq_info->percpu_value_buf = value_buf; } seq_info->map = map; return 0; } static void bpf_iter_fini_array_map(void *priv_data) { struct bpf_iter_seq_array_map_info *seq_info = priv_data; kfree(seq_info->percpu_value_buf); } static const struct seq_operations bpf_array_map_seq_ops = { .start = bpf_array_map_seq_start, .next = bpf_array_map_seq_next, .stop = bpf_array_map_seq_stop, .show = bpf_array_map_seq_show, }; static const struct bpf_iter_seq_info iter_seq_info = { .seq_ops = &bpf_array_map_seq_ops, .init_seq_private = bpf_iter_init_array_map, .fini_seq_private = bpf_iter_fini_array_map, .seq_priv_size = sizeof(struct bpf_iter_seq_array_map_info), }; static int array_map_btf_id; const struct bpf_map_ops array_map_ops = { .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = array_map_lookup_elem, .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, .map_gen_lookup = array_map_gen_lookup, .map_direct_value_addr = array_map_direct_value_addr, .map_direct_value_meta = array_map_direct_value_meta, .map_mmap = array_map_mmap, .map_seq_show_elem = array_map_seq_show_elem, .map_check_btf = array_map_check_btf, .map_lookup_batch = generic_map_lookup_batch, .map_update_batch = generic_map_update_batch, .map_btf_name = "bpf_array", .map_btf_id = &array_map_btf_id, .iter_seq_info = &iter_seq_info, }; static int percpu_array_map_btf_id; const struct bpf_map_ops percpu_array_map_ops = { .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = percpu_array_map_lookup_elem, .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, .map_seq_show_elem = percpu_array_map_seq_show_elem, .map_check_btf = array_map_check_btf, .map_btf_name = "bpf_array", .map_btf_id = &percpu_array_map_btf_id, .iter_seq_info = &iter_seq_info, }; static int fd_array_map_alloc_check(union bpf_attr *attr) { /* only file descriptors can be stored in this type of map */ if (attr->value_size != sizeof(u32)) return -EINVAL; /* Program read-only/write-only not supported for special maps yet. */ if (attr->map_flags & (BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG)) return -EINVAL; return array_map_alloc_check(attr); } static void fd_array_map_free(struct bpf_map *map) { struct bpf_array *array = container_of(map, struct bpf_array, map); int i; /* make sure it's empty */ for (i = 0; i < array->map.max_entries; i++) BUG_ON(array->ptrs[i] != NULL); bpf_map_area_free(array); } static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key) { return ERR_PTR(-EOPNOTSUPP); } /* only called from syscall */ int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value) { void **elem, *ptr; int ret = 0; if (!map->ops->map_fd_sys_lookup_elem) return -ENOTSUPP; rcu_read_lock(); elem = array_map_lookup_elem(map, key); if (elem && (ptr = READ_ONCE(*elem))) *value = map->ops->map_fd_sys_lookup_elem(ptr); else ret = -ENOENT; rcu_read_unlock(); return ret; } /* only called from syscall */ int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags) { struct bpf_array *array = container_of(map, struct bpf_array, map); void *new_ptr, *old_ptr; u32 index = *(u32 *)key, ufd; if (map_flags != BPF_ANY) return -EINVAL; if (index >= array->map.max_entries) return -E2BIG; ufd = *(u32 *)value; new_ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); if (IS_ERR(new_ptr)) return PTR_ERR(new_ptr); if (map->ops->map_poke_run) { mutex_lock(&array->aux->poke_mutex); old_ptr = xchg(array->ptrs + index, new_ptr); map->ops->map_poke_run(map, index, old_ptr, new_ptr); mutex_unlock(&array->aux->poke_mutex); } else { old_ptr = xchg(array->ptrs + index, new_ptr); } if (old_ptr) map->ops->map_fd_put_ptr(old_ptr); return 0; } static int fd_array_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_array *array = container_of(map, struct bpf_array, map); void *old_ptr; u32 index = *(u32 *)key; if (index >= array->map.max_entries) return -E2BIG; if (map->ops->map_poke_run) { mutex_lock(&array->aux->poke_mutex); old_ptr = xchg(array->ptrs + index, NULL); map->ops->map_poke_run(map, index, old_ptr, NULL); mutex_unlock(&array->aux->poke_mutex); } else { old_ptr = xchg(array->ptrs + index, NULL); } if (old_ptr) { map->ops->map_fd_put_ptr(old_ptr); return 0; } else { return -ENOENT; } } static void *prog_fd_array_get_ptr(struct bpf_map *map, struct file *map_file, int fd) { struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_prog *prog = bpf_prog_get(fd); if (IS_ERR(prog)) return prog; if (!bpf_prog_array_compatible(array, prog)) { bpf_prog_put(prog); return ERR_PTR(-EINVAL); } return prog; } static void prog_fd_array_put_ptr(void *ptr) { bpf_prog_put(ptr); } static u32 prog_fd_array_sys_lookup_elem(void *ptr) { return ((struct bpf_prog *)ptr)->aux->id; } /* decrement refcnt of all bpf_progs that are stored in this map */ static void bpf_fd_array_map_clear(struct bpf_map *map) { struct bpf_array *array = container_of(map, struct bpf_array, map); int i; for (i = 0; i < array->map.max_entries; i++) fd_array_map_delete_elem(map, &i); } static void prog_array_map_seq_show_elem(struct bpf_map *map, void *key, struct seq_file *m) { void **elem, *ptr; u32 prog_id; rcu_read_lock(); elem = array_map_lookup_elem(map, key); if (elem) { ptr = READ_ONCE(*elem); if (ptr) { seq_printf(m, "%u: ", *(u32 *)key); prog_id = prog_fd_array_sys_lookup_elem(ptr); btf_type_seq_show(map->btf, map->btf_value_type_id, &prog_id, m); seq_puts(m, "\n"); } } rcu_read_unlock(); } struct prog_poke_elem { struct list_head list; struct bpf_prog_aux *aux; }; static int prog_array_map_poke_track(struct bpf_map *map, struct bpf_prog_aux *prog_aux) { struct prog_poke_elem *elem; struct bpf_array_aux *aux; int ret = 0; aux = container_of(map, struct bpf_array, map)->aux; mutex_lock(&aux->poke_mutex); list_for_each_entry(elem, &aux->poke_progs, list) { if (elem->aux == prog_aux) goto out; } elem = kmalloc(sizeof(*elem), GFP_KERNEL); if (!elem) { ret = -ENOMEM; goto out; } INIT_LIST_HEAD(&elem->list); /* We must track the program's aux info at this point in time * since the program pointer itself may not be stable yet, see * also comment in prog_array_map_poke_run(). */ elem->aux = prog_aux; list_add_tail(&elem->list, &aux->poke_progs); out: mutex_unlock(&aux->poke_mutex); return ret; } static void prog_array_map_poke_untrack(struct bpf_map *map, struct bpf_prog_aux *prog_aux) { struct prog_poke_elem *elem, *tmp; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; mutex_lock(&aux->poke_mutex); list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) { if (elem->aux == prog_aux) { list_del_init(&elem->list); kfree(elem); break; } } mutex_unlock(&aux->poke_mutex); } static void prog_array_map_poke_run(struct bpf_map *map, u32 key, struct bpf_prog *old, struct bpf_prog *new) { struct prog_poke_elem *elem; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; WARN_ON_ONCE(!mutex_is_locked(&aux->poke_mutex)); list_for_each_entry(elem, &aux->poke_progs, list) { struct bpf_jit_poke_descriptor *poke; int i, ret; for (i = 0; i < elem->aux->size_poke_tab; i++) { poke = &elem->aux->poke_tab[i]; /* Few things to be aware of: * * 1) We can only ever access aux in this context, but * not aux->prog since it might not be stable yet and * there could be danger of use after free otherwise. * 2) Initially when we start tracking aux, the program * is not JITed yet and also does not have a kallsyms * entry. We skip these as poke->ip_stable is not * active yet. The JIT will do the final fixup before * setting it stable. The various poke->ip_stable are * successively activated, so tail call updates can * arrive from here while JIT is still finishing its * final fixup for non-activated poke entries. * 3) On program teardown, the program's kallsym entry gets * removed out of RCU callback, but we can only untrack * from sleepable context, therefore bpf_arch_text_poke() * might not see that this is in BPF text section and * bails out with -EINVAL. As these are unreachable since * RCU grace period already passed, we simply skip them. * 4) Also programs reaching refcount of zero while patching * is in progress is okay since we're protected under * poke_mutex and untrack the programs before the JIT * buffer is freed. When we're still in the middle of * patching and suddenly kallsyms entry of the program * gets evicted, we just skip the rest which is fine due * to point 3). * 5) Any other error happening below from bpf_arch_text_poke() * is a unexpected bug. */ if (!READ_ONCE(poke->ip_stable)) continue; if (poke->reason != BPF_POKE_REASON_TAIL_CALL) continue; if (poke->tail_call.map != map || poke->tail_call.key != key) continue; ret = bpf_arch_text_poke(poke->ip, BPF_MOD_JUMP, old ? (u8 *)old->bpf_func + poke->adj_off : NULL, new ? (u8 *)new->bpf_func + poke->adj_off : NULL); BUG_ON(ret < 0 && ret != -EINVAL); } } } static void prog_array_map_clear_deferred(struct work_struct *work) { struct bpf_map *map = container_of(work, struct bpf_array_aux, work)->map; bpf_fd_array_map_clear(map); bpf_map_put(map); } static void prog_array_map_clear(struct bpf_map *map) { struct bpf_array_aux *aux = container_of(map, struct bpf_array, map)->aux; bpf_map_inc(map); schedule_work(&aux->work); } static struct bpf_map *prog_array_map_alloc(union bpf_attr *attr) { struct bpf_array_aux *aux; struct bpf_map *map; aux = kzalloc(sizeof(*aux), GFP_KERNEL); if (!aux) return ERR_PTR(-ENOMEM); INIT_WORK(&aux->work, prog_array_map_clear_deferred); INIT_LIST_HEAD(&aux->poke_progs); mutex_init(&aux->poke_mutex); map = array_map_alloc(attr); if (IS_ERR(map)) { kfree(aux); return map; } container_of(map, struct bpf_array, map)->aux = aux; aux->map = map; return map; } static void prog_array_map_free(struct bpf_map *map) { struct prog_poke_elem *elem, *tmp; struct bpf_array_aux *aux; aux = container_of(map, struct bpf_array, map)->aux; list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) { list_del_init(&elem->list); kfree(elem);