diff options
Diffstat (limited to 'fs/notify/fanotify/fanotify_user.c')
| -rw-r--r-- | fs/notify/fanotify/fanotify_user.c | 1137 |
1 files changed, 800 insertions, 337 deletions
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 73b1615f9d96..d0b9b984002f 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -1,7 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/fanotify.h> #include <linux/fcntl.h> -#include <linux/fdtable.h> #include <linux/file.h> #include <linux/fs.h> #include <linux/anon_inodes.h> @@ -23,7 +22,7 @@ #include <asm/ioctls.h> -#include "../../mount.h" +#include "../fsnotify.h" #include "../fdinfo.h" #include "fanotify.h" @@ -51,6 +50,7 @@ /* configurable via /proc/sys/fs/fanotify/ */ static int fanotify_max_queued_events __read_mostly; +static int perm_group_timeout __read_mostly; #ifdef CONFIG_SYSCTL @@ -59,7 +59,7 @@ static int fanotify_max_queued_events __read_mostly; static long ft_zero = 0; static long ft_int_max = INT_MAX; -static struct ctl_table fanotify_table[] = { +static const struct ctl_table fanotify_table[] = { { .procname = "max_user_groups", .data = &init_user_ns.ucount_max[UCOUNT_FANOTIFY_GROUPS], @@ -86,7 +86,14 @@ static struct ctl_table fanotify_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO }, - { } + { + .procname = "watchdog_timeout", + .data = &perm_group_timeout, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + }, }; static void __init fanotify_sysctls_init(void) @@ -97,13 +104,97 @@ static void __init fanotify_sysctls_init(void) #define fanotify_sysctls_init() do { } while (0) #endif /* CONFIG_SYSCTL */ +static LIST_HEAD(perm_group_list); +static DEFINE_SPINLOCK(perm_group_lock); +static void perm_group_watchdog(struct work_struct *work); +static DECLARE_DELAYED_WORK(perm_group_work, perm_group_watchdog); + +static void perm_group_watchdog_schedule(void) +{ + schedule_delayed_work(&perm_group_work, secs_to_jiffies(perm_group_timeout)); +} + +static void perm_group_watchdog(struct work_struct *work) +{ + struct fsnotify_group *group; + struct fanotify_perm_event *event; + struct task_struct *task; + pid_t failed_pid = 0; + + guard(spinlock)(&perm_group_lock); + if (list_empty(&perm_group_list)) + return; + + list_for_each_entry(group, &perm_group_list, + fanotify_data.perm_grp_list) { + /* + * Ok to test without lock, racing with an addition is + * fine, will deal with it next round + */ + if (list_empty(&group->fanotify_data.access_list)) + continue; + + spin_lock(&group->notification_lock); + list_for_each_entry(event, &group->fanotify_data.access_list, + fae.fse.list) { + if (likely(event->watchdog_cnt == 0)) { + event->watchdog_cnt = 1; + } else if (event->watchdog_cnt == 1) { + /* Report on event only once */ + event->watchdog_cnt = 2; + + /* Do not report same pid repeatedly */ + if (event->recv_pid == failed_pid) + continue; + + failed_pid = event->recv_pid; + rcu_read_lock(); + task = find_task_by_pid_ns(event->recv_pid, + &init_pid_ns); + pr_warn_ratelimited( + "PID %u (%s) failed to respond to fanotify queue for more than %d seconds\n", + event->recv_pid, + task ? task->comm : NULL, + perm_group_timeout); + rcu_read_unlock(); + } + } + spin_unlock(&group->notification_lock); + } + perm_group_watchdog_schedule(); +} + +static void fanotify_perm_watchdog_group_remove(struct fsnotify_group *group) +{ + if (!list_empty(&group->fanotify_data.perm_grp_list)) { + /* Perm event watchdog can no longer scan this group. */ + spin_lock(&perm_group_lock); + list_del_init(&group->fanotify_data.perm_grp_list); + spin_unlock(&perm_group_lock); + } +} + +static void fanotify_perm_watchdog_group_add(struct fsnotify_group *group) +{ + if (!perm_group_timeout) + return; + + spin_lock(&perm_group_lock); + if (list_empty(&group->fanotify_data.perm_grp_list)) { + /* Add to perm_group_list for monitoring by watchdog. */ + if (list_empty(&perm_group_list)) + perm_group_watchdog_schedule(); + list_add_tail(&group->fanotify_data.perm_grp_list, &perm_group_list); + } + spin_unlock(&perm_group_lock); +} + /* * All flags that may be specified in parameter event_f_flags of fanotify_init. * * Internal and external open flags are stored together in field f_flags of * struct file. Only external open flags shall be allowed in event_f_flags. - * Internal flags like FMODE_NONOTIFY, FMODE_EXEC, FMODE_NOCMTIME shall be - * excluded. + * Internal flags like FMODE_EXEC shall be excluded. */ #define FANOTIFY_INIT_ALL_EVENT_F_BITS ( \ O_ACCMODE | O_APPEND | O_NONBLOCK | \ @@ -112,18 +203,23 @@ static void __init fanotify_sysctls_init(void) extern const struct fsnotify_ops fanotify_fsnotify_ops; -struct kmem_cache *fanotify_mark_cache __read_mostly; -struct kmem_cache *fanotify_fid_event_cachep __read_mostly; -struct kmem_cache *fanotify_path_event_cachep __read_mostly; -struct kmem_cache *fanotify_perm_event_cachep __read_mostly; +struct kmem_cache *fanotify_mark_cache __ro_after_init; +struct kmem_cache *fanotify_fid_event_cachep __ro_after_init; +struct kmem_cache *fanotify_path_event_cachep __ro_after_init; +struct kmem_cache *fanotify_perm_event_cachep __ro_after_init; +struct kmem_cache *fanotify_mnt_event_cachep __ro_after_init; #define FANOTIFY_EVENT_ALIGN 4 #define FANOTIFY_FID_INFO_HDR_LEN \ (sizeof(struct fanotify_event_info_fid) + sizeof(struct file_handle)) -#define FANOTIFY_PIDFD_INFO_HDR_LEN \ +#define FANOTIFY_PIDFD_INFO_LEN \ sizeof(struct fanotify_event_info_pidfd) #define FANOTIFY_ERROR_INFO_LEN \ (sizeof(struct fanotify_event_info_error)) +#define FANOTIFY_RANGE_INFO_LEN \ + (sizeof(struct fanotify_event_info_range)) +#define FANOTIFY_MNT_INFO_LEN \ + (sizeof(struct fanotify_event_info_mnt)) static int fanotify_fid_info_len(int fh_len, int name_len) { @@ -158,18 +254,12 @@ static size_t fanotify_event_len(unsigned int info_mode, struct fanotify_event *event) { size_t event_len = FAN_EVENT_METADATA_LEN; - struct fanotify_info *info; int fh_len; int dot_len = 0; - if (!info_mode) - return event_len; - if (fanotify_is_error_event(event->mask)) event_len += FANOTIFY_ERROR_INFO_LEN; - info = fanotify_event_info(event); - if (fanotify_event_has_any_dir_fh(event)) { event_len += fanotify_dir_name_info_len(event); } else if ((info_mode & FAN_REPORT_NAME) && @@ -181,13 +271,18 @@ static size_t fanotify_event_len(unsigned int info_mode, dot_len = 1; } - if (info_mode & FAN_REPORT_PIDFD) - event_len += FANOTIFY_PIDFD_INFO_HDR_LEN; - if (fanotify_event_has_object_fh(event)) { fh_len = fanotify_event_object_fh_len(event); event_len += fanotify_fid_info_len(fh_len, dot_len); } + if (fanotify_is_mnt_event(event->mask)) + event_len += FANOTIFY_MNT_INFO_LEN; + + if (info_mode & FAN_REPORT_PIDFD) + event_len += FANOTIFY_PIDFD_INFO_LEN; + + if (fanotify_event_has_access_range(event)) + event_len += FANOTIFY_RANGE_INFO_LEN; return event_len; } @@ -252,7 +347,7 @@ out: return event; } -static int create_fd(struct fsnotify_group *group, struct path *path, +static int create_fd(struct fsnotify_group *group, const struct path *path, struct file **file) { int client_fd; @@ -263,20 +358,12 @@ static int create_fd(struct fsnotify_group *group, struct path *path, return client_fd; /* - * we need a new file handle for the userspace program so it can read even if it was - * originally opened O_WRONLY. + * We provide an fd for the userspace program, so it could access the + * file without generating fanotify events itself. */ - new_file = dentry_open(path, - group->fanotify_data.f_flags | FMODE_NONOTIFY, - current_cred()); + new_file = dentry_open_nonotify(path, group->fanotify_data.f_flags, + current_cred()); if (IS_ERR(new_file)) { - /* - * we still send an event even if we can't open the file. this - * can happen when say tasks are gone and we try to open their - * /proc files or we try to open a WRONLY file like in sysfs - * we just send the errno to userspace since there isn't much - * else we can do. - */ put_unused_fd(client_fd); client_fd = PTR_ERR(new_file); } else { @@ -286,19 +373,42 @@ static int create_fd(struct fsnotify_group *group, struct path *path, return client_fd; } +static int process_access_response_info(const char __user *info, + size_t info_len, + struct fanotify_response_info_audit_rule *friar) +{ + if (info_len != sizeof(*friar)) + return -EINVAL; + + if (copy_from_user(friar, info, sizeof(*friar))) + return -EFAULT; + + if (friar->hdr.type != FAN_RESPONSE_INFO_AUDIT_RULE) + return -EINVAL; + if (friar->hdr.pad != 0) + return -EINVAL; + if (friar->hdr.len != sizeof(*friar)) + return -EINVAL; + + return info_len; +} + /* * Finish processing of permission event by setting it to ANSWERED state and * drop group->notification_lock. */ static void finish_permission_event(struct fsnotify_group *group, - struct fanotify_perm_event *event, - unsigned int response) + struct fanotify_perm_event *event, u32 response, + struct fanotify_response_info_audit_rule *friar) __releases(&group->notification_lock) { bool destroy = false; assert_spin_locked(&group->notification_lock); - event->response = response; + event->response = response & ~FAN_INFO; + if (response & FAN_INFO) + memcpy(&event->audit_rule, friar, sizeof(*friar)); + if (event->state == FAN_EVENT_CANCELED) destroy = true; else @@ -309,31 +419,73 @@ static void finish_permission_event(struct fsnotify_group *group, } static int process_access_response(struct fsnotify_group *group, - struct fanotify_response *response_struct) + struct fanotify_response *response_struct, + const char __user *info, + size_t info_len) { struct fanotify_perm_event *event; int fd = response_struct->fd; - int response = response_struct->response; + u32 response = response_struct->response; + int errno = fanotify_get_response_errno(response); + int ret = info_len; + struct fanotify_response_info_audit_rule friar; - pr_debug("%s: group=%p fd=%d response=%d\n", __func__, group, - fd, response); + pr_debug("%s: group=%p fd=%d response=%x errno=%d buf=%p size=%zu\n", + __func__, group, fd, response, errno, info, info_len); /* * make sure the response is valid, if invalid we do nothing and either * userspace can send a valid response or we will clean it up after the * timeout */ - switch (response & ~FAN_AUDIT) { + if (response & ~FANOTIFY_RESPONSE_VALID_MASK) + return -EINVAL; + + switch (response & FANOTIFY_RESPONSE_ACCESS) { case FAN_ALLOW: + if (errno) + return -EINVAL; + break; case FAN_DENY: + /* Custom errno is supported only for pre-content groups */ + if (errno && group->priority != FSNOTIFY_PRIO_PRE_CONTENT) + return -EINVAL; + + /* + * Limit errno to values expected on open(2)/read(2)/write(2) + * of regular files. + */ + switch (errno) { + case 0: + case EIO: + case EPERM: + case EBUSY: + case ETXTBSY: + case EAGAIN: + case ENOSPC: + case EDQUOT: + break; + default: + return -EINVAL; + } break; default: return -EINVAL; } - if (fd < 0) + if ((response & FAN_AUDIT) && !FAN_GROUP_FLAG(group, FAN_ENABLE_AUDIT)) return -EINVAL; - if ((response & FAN_AUDIT) && !FAN_GROUP_FLAG(group, FAN_ENABLE_AUDIT)) + if (response & FAN_INFO) { + ret = process_access_response_info(info, info_len, &friar); + if (ret < 0) + return ret; + if (fd == FAN_NOFD) + return ret; + } else { + ret = 0; + } + + if (fd < 0) return -EINVAL; spin_lock(&group->notification_lock); @@ -343,15 +495,34 @@ static int process_access_response(struct fsnotify_group *group, continue; list_del_init(&event->fae.fse.list); - finish_permission_event(group, event, response); + finish_permission_event(group, event, response, &friar); wake_up(&group->fanotify_data.access_waitq); - return 0; + return ret; } spin_unlock(&group->notification_lock); return -ENOENT; } +static size_t copy_mnt_info_to_user(struct fanotify_event *event, + char __user *buf, int count) +{ + struct fanotify_event_info_mnt info = { }; + + info.hdr.info_type = FAN_EVENT_INFO_TYPE_MNT; + info.hdr.len = FANOTIFY_MNT_INFO_LEN; + + if (WARN_ON(count < info.hdr.len)) + return -EFAULT; + + info.mnt_id = FANOTIFY_ME(event)->mnt_id; + + if (copy_to_user(buf, &info, sizeof(info))) + return -EFAULT; + + return info.hdr.len; +} + static size_t copy_error_info_to_user(struct fanotify_event *event, char __user *buf, int count) { @@ -478,7 +649,7 @@ static int copy_pidfd_info_to_user(int pidfd, size_t count) { struct fanotify_event_info_pidfd info = { }; - size_t info_len = FANOTIFY_PIDFD_INFO_HDR_LEN; + size_t info_len = FANOTIFY_PIDFD_INFO_LEN; if (WARN_ON_ONCE(info_len > count)) return -EFAULT; @@ -493,6 +664,30 @@ static int copy_pidfd_info_to_user(int pidfd, return info_len; } +static size_t copy_range_info_to_user(struct fanotify_event *event, + char __user *buf, int count) +{ + struct fanotify_perm_event *pevent = FANOTIFY_PERM(event); + struct fanotify_event_info_range info = { }; + size_t info_len = FANOTIFY_RANGE_INFO_LEN; + + if (WARN_ON_ONCE(info_len > count)) + return -EFAULT; + + if (WARN_ON_ONCE(!pevent->ppos)) + return -EINVAL; + + info.hdr.info_type = FAN_EVENT_INFO_TYPE_RANGE; + info.hdr.len = info_len; + info.offset = *(pevent->ppos); + info.count = pevent->count; + + if (copy_to_user(buf, &info, info_len)) + return -EFAULT; + + return info_len; +} + static int copy_info_records_to_user(struct fanotify_event *event, struct fanotify_info *info, unsigned int info_mode, int pidfd, @@ -614,6 +809,24 @@ static int copy_info_records_to_user(struct fanotify_event *event, total_bytes += ret; } + if (fanotify_event_has_access_range(event)) { + ret = copy_range_info_to_user(event, buf, count); + if (ret < 0) + return ret; + buf += ret; + count -= ret; + total_bytes += ret; + } + + if (fanotify_is_mnt_event(event->mask)) { + ret = copy_mnt_info_to_user(event, buf, count); + if (ret < 0) + return ret; + buf += ret; + count -= ret; + total_bytes += ret; + } + return total_bytes; } @@ -622,12 +835,12 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, char __user *buf, size_t count) { struct fanotify_event_metadata metadata; - struct path *path = fanotify_event_path(event); + const struct path *path = fanotify_event_path(event); struct fanotify_info *info = fanotify_event_info(event); unsigned int info_mode = FAN_GROUP_FLAG(group, FANOTIFY_INFO_MODES); unsigned int pidfd_mode = info_mode & FAN_REPORT_PIDFD; - struct file *f = NULL; - int ret, pidfd = FAN_NOPIDFD, fd = FAN_NOFD; + struct file *f = NULL, *pidfd_file = NULL; + int ret, pidfd = -ESRCH, fd = -EBADF; pr_debug("%s: group=%p event=%p\n", __func__, group, event); @@ -655,10 +868,39 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, if (!FAN_GROUP_FLAG(group, FANOTIFY_UNPRIV) && path && path->mnt && path->dentry) { fd = create_fd(group, path, &f); - if (fd < 0) - return fd; + /* + * Opening an fd from dentry can fail for several reasons. + * For example, when tasks are gone and we try to open their + * /proc files or we try to open a WRONLY file like in sysfs + * or when trying to open a file that was deleted on the + * remote network server. + * + * For a group with FAN_REPORT_FD_ERROR, we will send the + * event with the error instead of the open fd, otherwise + * Userspace may not get the error at all. + * In any case, userspace will not know which file failed to + * open, so add a debug print for further investigation. + */ + if (fd < 0) { + pr_debug("fanotify: create_fd(%pd2) failed err=%d\n", + path->dentry, fd); + if (!FAN_GROUP_FLAG(group, FAN_REPORT_FD_ERROR)) { + /* + * Historically, we've handled EOPENSTALE in a + * special way and silently dropped such + * events. Now we have to keep it to maintain + * backward compatibility... + */ + if (fd == -EOPENSTALE) + fd = 0; + return fd; + } + } } - metadata.fd = fd; + if (FAN_GROUP_FLAG(group, FAN_REPORT_FD_ERROR)) + metadata.fd = fd; + else + metadata.fd = fd >= 0 ? fd : FAN_NOFD; if (pidfd_mode) { /* @@ -673,18 +915,16 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, * The PIDTYPE_TGID check for an event->pid is performed * preemptively in an attempt to catch out cases where the event * listener reads events after the event generating process has - * already terminated. Report FAN_NOPIDFD to the event listener - * in those cases, with all other pidfd creation errors being - * reported as FAN_EPIDFD. + * already terminated. Depending on flag FAN_REPORT_FD_ERROR, + * report either -ESRCH or FAN_NOPIDFD to the event listener in + * those cases with all other pidfd creation errors reported as + * the error code itself or as FAN_EPIDFD. */ - if (metadata.pid == 0 || - !pid_has_task(event->pid, PIDTYPE_TGID)) { - pidfd = FAN_NOPIDFD; - } else { - pidfd = pidfd_create(event->pid, 0); - if (pidfd < 0) - pidfd = FAN_EPIDFD; - } + if (metadata.pid && pid_has_task(event->pid, PIDTYPE_TGID)) + pidfd = pidfd_prepare(event->pid, 0, &pidfd_file); + + if (!FAN_GROUP_FLAG(group, FAN_REPORT_FD_ERROR) && pidfd < 0) + pidfd = pidfd == -ESRCH ? FAN_NOPIDFD : FAN_EPIDFD; } ret = -EFAULT; @@ -701,29 +941,32 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, buf += FAN_EVENT_METADATA_LEN; count -= FAN_EVENT_METADATA_LEN; - if (fanotify_is_perm_event(event->mask)) - FANOTIFY_PERM(event)->fd = fd; + ret = copy_info_records_to_user(event, info, info_mode, pidfd, + buf, count); + if (ret < 0) + goto out_close_fd; if (f) fd_install(fd, f); - if (info_mode) { - ret = copy_info_records_to_user(event, info, info_mode, pidfd, - buf, count); - if (ret < 0) - goto out_close_fd; - } + if (pidfd_file) + fd_install(pidfd, pidfd_file); + + if (fanotify_is_perm_event(event->mask)) + FANOTIFY_PERM(event)->fd = fd; return metadata.event_len; out_close_fd: - if (fd != FAN_NOFD) { + if (f) { put_unused_fd(fd); fput(f); } - if (pidfd >= 0) - close_fd(pidfd); + if (pidfd_file) { + put_unused_fd(pidfd); + fput(pidfd_file); + } return ret; } @@ -787,15 +1030,6 @@ static ssize_t fanotify_read(struct file *file, char __user *buf, } ret = copy_event_to_user(group, event, buf, count); - if (unlikely(ret == -EOPENSTALE)) { - /* - * We cannot report events with stale fd so drop it. - * Setting ret to 0 will continue the event loop and - * do the right thing if there are no more events to - * read (i.e. return bytes read, -EAGAIN or wait). - */ - ret = 0; - } /* * Permission events get queued to wait for response. Other @@ -804,15 +1038,16 @@ static ssize_t fanotify_read(struct file *file, char __user *buf, if (!fanotify_is_perm_event(event->mask)) { fsnotify_destroy_event(group, &event->fse); } else { - if (ret <= 0) { + if (ret <= 0 || FANOTIFY_PERM(event)->fd < 0) { spin_lock(&group->notification_lock); finish_permission_event(group, - FANOTIFY_PERM(event), FAN_DENY); + FANOTIFY_PERM(event), FAN_DENY, NULL); wake_up(&group->fanotify_data.access_waitq); } else { spin_lock(&group->notification_lock); list_add_tail(&event->fse.list, &group->fanotify_data.access_list); + FANOTIFY_PERM(event)->recv_pid = current->pid; spin_unlock(&group->notification_lock); } } @@ -830,28 +1065,32 @@ static ssize_t fanotify_read(struct file *file, char __user *buf, static ssize_t fanotify_write(struct file *file, const char __user *buf, size_t count, loff_t *pos) { - struct fanotify_response response = { .fd = -1, .response = -1 }; + struct fanotify_response response; struct fsnotify_group *group; int ret; + const char __user *info_buf = buf + sizeof(struct fanotify_response); + size_t info_len; if (!IS_ENABLED(CONFIG_FANOTIFY_ACCESS_PERMISSIONS)) return -EINVAL; group = file->private_data; + pr_debug("%s: group=%p count=%zu\n", __func__, group, count); + if (count < sizeof(response)) return -EINVAL; - count = sizeof(response); - - pr_debug("%s: group=%p count=%zu\n", __func__, group, count); - - if (copy_from_user(&response, buf, count)) + if (copy_from_user(&response, buf, sizeof(response))) return -EFAULT; - ret = process_access_response(group, &response); + info_len = count - sizeof(response); + + ret = process_access_response(group, &response, info_buf, info_len); if (ret < 0) count = ret; + else + count = sizeof(response) + ret; return count; } @@ -868,6 +1107,8 @@ static int fanotify_release(struct inode *ignored, struct file *file) */ fsnotify_group_stop_queueing(group); + fanotify_perm_watchdog_group_remove(group); + /* * Process all permission events on access_list and notification queue * and simulate reply from userspace. @@ -879,7 +1120,7 @@ static int fanotify_release(struct inode *ignored, struct file *file) event = list_first_entry(&group->fanotify_data.access_list, struct fanotify_perm_event, fae.fse.list); list_del_init(&event->fae.fse.list); - finish_permission_event(group, event, FAN_ALLOW); + finish_permission_event(group, event, FAN_ALLOW, NULL); spin_lock(&group->notification_lock); } @@ -896,7 +1137,7 @@ static int fanotify_release(struct inode *ignored, struct file *file) fsnotify_destroy_event(group, fsn_event); } else { finish_permission_event(group, FANOTIFY_PERM(event), - FAN_ALLOW); + FAN_ALLOW, NULL); } spin_lock(&group->notification_lock); } @@ -958,22 +1199,17 @@ static int fanotify_find_path(int dfd, const char __user *filename, dfd, filename, flags); if (filename == NULL) { - struct fd f = fdget(dfd); + CLASS(fd, f)(dfd); - ret = -EBADF; - if (!f.file) - goto out; + if (fd_empty(f)) + return -EBADF; - ret = -ENOTDIR; if ((flags & FAN_MARK_ONLYDIR) && - !(S_ISDIR(file_inode(f.file)->i_mode))) { - fdput(f); - goto out; - } + !(S_ISDIR(file_inode(fd_file(f))->i_mode))) + return -ENOTDIR; - *path = f.file->f_path; + *path = fd_file(f)->f_path; path_get(path); - fdput(f); } else { unsigned int lookup_flags = 0; @@ -1006,41 +1242,42 @@ static __u32 fanotify_mark_remove_from_mask(struct fsnotify_mark *fsn_mark, __u32 mask, unsigned int flags, __u32 umask, int *destroy) { - __u32 oldmask = 0; + __u32 oldmask, newmask; /* umask bits cannot be removed by user */ mask &= ~umask; spin_lock(&fsn_mark->lock); - if (!(flags & FAN_MARK_IGNORED_MASK)) { - oldmask = fsn_mark->mask; + oldmask = fsnotify_calc_mask(fsn_mark); + if (!(flags & FANOTIFY_MARK_IGNORE_BITS)) { fsn_mark->mask &= ~mask; } else { - fsn_mark->ignored_mask &= ~mask; + fsn_mark->ignore_mask &= ~mask; } + newmask = fsnotify_calc_mask(fsn_mark); /* * We need to keep the mark around even if remaining mask cannot * result in any events (e.g. mask == FAN_ONDIR) to support incremenal * changes to the mask. * Destroy mark when only umask bits remain. */ - *destroy = !((fsn_mark->mask | fsn_mark->ignored_mask) & ~umask); + *destroy = !((fsn_mark->mask | fsn_mark->ignore_mask) & ~umask); spin_unlock(&fsn_mark->lock); - return mask & oldmask; + return oldmask & ~newmask; } static int fanotify_remove_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, __u32 mask, + void *obj, unsigned int obj_type, __u32 mask, unsigned int flags, __u32 umask) { struct fsnotify_mark *fsn_mark = NULL; __u32 removed; int destroy_mark; - mutex_lock(&group->mark_mutex); - fsn_mark = fsnotify_find_mark(connp, group); + fsnotify_group_lock(group); + fsn_mark = fsnotify_find_mark(obj, obj_type, group); if (!fsn_mark) { - mutex_unlock(&group->mark_mutex); + fsnotify_group_unlock(group); return -ENOENT; } @@ -1050,7 +1287,7 @@ static int fanotify_remove_mark(struct fsnotify_group *group, fsnotify_recalc_mask(fsn_mark->connector); if (destroy_mark) fsnotify_detach_mark(fsn_mark); - mutex_unlock(&group->mark_mutex); + fsnotify_group_unlock(group); if (destroy_mark) fsnotify_free_mark(fsn_mark); @@ -1059,56 +1296,133 @@ static int fanotify_remove_mark(struct fsnotify_group *group, return 0; } -static int fanotify_remove_vfsmount_mark(struct fsnotify_group *group, - struct vfsmount *mnt, __u32 mask, - unsigned int flags, __u32 umask) +static bool fanotify_mark_update_flags(struct fsnotify_mark *fsn_mark, + unsigned int fan_flags) { - return fanotify_remove_mark(group, &real_mount(mnt)->mnt_fsnotify_marks, - mask, flags, umask); -} + bool want_iref = !(fan_flags & FAN_MARK_EVICTABLE); + unsigned int ignore = fan_flags & FANOTIFY_MARK_IGNORE_BITS; + bool recalc = false; -static int fanotify_remove_sb_mark(struct fsnotify_group *group, - struct super_block *sb, __u32 mask, - unsigned int flags, __u32 umask) -{ - return fanotify_remove_mark(group, &sb->s_fsnotify_marks, mask, - flags, umask); -} + /* + * When using FAN_MARK_IGNORE for the first time, mark starts using + * independent event flags in ignore mask. After that, trying to + * update the ignore mask with the old FAN_MARK_IGNORED_MASK API + * will result in EEXIST error. + */ + if (ignore == FAN_MARK_IGNORE) + fsn_mark->flags |= FSNOTIFY_MARK_FLAG_HAS_IGNORE_FLAGS; -static int fanotify_remove_inode_mark(struct fsnotify_group *group, - struct inode *inode, __u32 mask, - unsigned int flags, __u32 umask) -{ - return fanotify_remove_mark(group, &inode->i_fsnotify_marks, mask, - flags, umask); + /* + * Setting FAN_MARK_IGNORED_SURV_MODIFY for the first time may lead to + * the removal of the FS_MODIFY bit in calculated mask if it was set + * because of an ignore mask that is now going to survive FS_MODIFY. + */ + if (ignore && (fan_flags & FAN_MARK_IGNORED_SURV_MODIFY) && + !(fsn_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY)) { + fsn_mark->flags |= FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY; + if (!(fsn_mark->mask & FS_MODIFY)) + recalc = true; + } + + if (fsn_mark->connector->type != FSNOTIFY_OBJ_TYPE_INODE || + want_iref == !(fsn_mark->flags & FSNOTIFY_MARK_FLAG_NO_IREF)) + return recalc; + + /* + * NO_IREF may be removed from a mark, but not added. + * When removed, fsnotify_recalc_mask() will take the inode ref. + */ + WARN_ON_ONCE(!want_iref); + fsn_mark->flags &= ~FSNOTIFY_MARK_FLAG_NO_IREF; + + return true; } -static __u32 fanotify_mark_add_to_mask(struct fsnotify_mark *fsn_mark, - __u32 mask, - unsigned int flags) +static bool fanotify_mark_add_to_mask(struct fsnotify_mark *fsn_mark, + __u32 mask, unsigned int fan_flags) { - __u32 oldmask = -1; + bool recalc; spin_lock(&fsn_mark->lock); - if (!(flags & FAN_MARK_IGNORED_MASK)) { - oldmask = fsn_mark->mask; + if (!(fan_flags & FANOTIFY_MARK_IGNORE_BITS)) fsn_mark->mask |= mask; - } else { - fsn_mark->ignored_mask |= mask; - if (flags & FAN_MARK_IGNORED_SURV_MODIFY) - fsn_mark->flags |= FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY; - } + else + fsn_mark->ignore_mask |= mask; + + recalc = fsnotify_calc_mask(fsn_mark) & + ~fsnotify_conn_mask(fsn_mark->connector); + + recalc |= fanotify_mark_update_flags(fsn_mark, fan_flags); spin_unlock(&fsn_mark->lock); - return mask & ~oldmask; + return recalc; +} + +struct fan_fsid { + struct super_block *sb; + __kernel_fsid_t id; + bool weak; +}; + +static int fanotify_set_mark_fsid(struct fsnotify_group *group, + struct fsnotify_mark *mark, + struct fan_fsid *fsid) +{ + struct fsnotify_mark_connector *conn; + struct fsnotify_mark *old; + struct super_block *old_sb = NULL; + + FANOTIFY_MARK(mark)->fsid = fsid->id; + mark->flags |= FSNOTIFY_MARK_FLAG_HAS_FSID; + if (fsid->weak) + mark->flags |= FSNOTIFY_MARK_FLAG_WEAK_FSID; + + /* First mark added will determine if group is single or multi fsid */ + if (list_empty(&group->marks_list)) + return 0; + + /* Find sb of an existing mark */ + list_for_each_entry(old, &group->marks_list, g_list) { + conn = READ_ONCE(old->connector); + if (!conn) + continue; + old_sb = fsnotify_connector_sb(conn); + if (old_sb) + break; + } + + /* Only detached marks left? */ + if (!old_sb) + return 0; + + /* Do not allow mixing of marks with weak and strong fsid */ + if ((mark->flags ^ old->flags) & FSNOTIFY_MARK_FLAG_WEAK_FSID) + return -EXDEV; + + /* Allow mixing of marks with strong fsid from different fs */ + if (!fsid->weak) + return 0; + + /* Do not allow mixing marks with weak fsid from different fs */ + if (old_sb != fsid->sb) + return -EXDEV; + + /* Do not allow mixing marks from different btrfs sub-volumes */ + if (!fanotify_fsid_equal(&FANOTIFY_MARK(old)->fsid, + &FANOTIFY_MARK(mark)->fsid)) + return -EXDEV; + + return 0; } static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, + void *obj, unsigned int obj_type, - __kernel_fsid_t *fsid) + unsigned int fan_flags, + struct fan_fsid *fsid) { struct ucounts *ucounts = group->fanotify_data.ucounts; + struct fanotify_mark *fan_mark; struct fsnotify_mark *mark; int ret; @@ -1117,25 +1431,39 @@ static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group, * A group with FAN_UNLIMITED_MARKS does not contribute to mark count * in the limited groups account. */ + BUILD_BUG_ON(!(FANOTIFY_ADMIN_INIT_FLAGS & FAN_UNLIMITED_MARKS)); if (!FAN_GROUP_FLAG(group, FAN_UNLIMITED_MARKS) && !inc_ucount(ucounts->ns, ucounts->uid, UCOUNT_FANOTIFY_MARKS)) return ERR_PTR(-ENOSPC); - mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL); - if (!mark) { + fan_mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL); + if (!fan_mark) { ret = -ENOMEM; goto out_dec_ucounts; } + mark = &fan_mark->fsn_mark; fsnotify_init_mark(mark, group); - ret = fsnotify_add_mark_locked(mark, connp, obj_type, 0, fsid); - if (ret) { - fsnotify_put_mark(mark); - goto out_dec_ucounts; + if (fan_flags & FAN_MARK_EVICTABLE) + mark->flags |= FSNOTIFY_MARK_FLAG_NO_IREF; + + /* Cache fsid of filesystem containing the marked object */ + if (fsid) { + ret = fanotify_set_mark_fsid(group, mark, fsid); + if (ret) + goto out_put_mark; + } else { + fan_mark->fsid.val[0] = fan_mark->fsid.val[1] = 0; } + ret = fsnotify_add_mark_locked(mark, obj, obj_type, 0); + if (ret) + goto out_put_mark; + return mark; +out_put_mark: + fsnotify_put_mark(mark); out_dec_ucounts: if (!FAN_GROUP_FLAG(group, FAN_UNLIMITED_MARKS)) dec_ucount(ucounts, UCOUNT_FANOTIFY_MARKS); @@ -1152,80 +1480,93 @@ static int fanotify_group_init_error_pool(struct fsnotify_group *group) sizeof(struct fanotify_error_event)); } +static int fanotify_may_update_existing_mark(struct fsnotify_mark *fsn_mark, + __u32 mask, unsigned int fan_flags) +{ + /* + * Non evictable mark cannot be downgraded to evictable mark. + */ + if (fan_flags & FAN_MARK_EVICTABLE && + !(fsn_mark->flags & FSNOTIFY_MARK_FLAG_NO_IREF)) + return -EEXIST; + + /* + * New ignore mask semantics cannot be downgraded to old semantics. + */ + if (fan_flags & FAN_MARK_IGNORED_MASK && + fsn_mark->flags & FSNOTIFY_MARK_FLAG_HAS_IGNORE_FLAGS) + return -EEXIST; + + /* + * An ignore mask that survives modify could never be downgraded to not + * survive modify. With new FAN_MARK_IGNORE semantics we make that rule + * explicit and return an error when trying to update the ignore mask + * without the original FAN_MARK_IGNORED_SURV_MODIFY value. + */ + if (fan_flags & FAN_MARK_IGNORE && + !(fan_flags & FAN_MARK_IGNORED_SURV_MODIFY) && + fsn_mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY) + return -EEXIST; + + /* For now pre-content events are not generated for directories */ + mask |= fsn_mark->mask; + if (mask & FANOTIFY_PRE_CONTENT_EVENTS && mask & FAN_ONDIR) + return -EEXIST; + + return 0; +} + static int fanotify_add_mark(struct fsnotify_group *group, - fsnotify_connp_t *connp, unsigned int obj_type, - __u32 mask, unsigned int flags, - __kernel_fsid_t *fsid) + void *obj, unsigned int obj_type, + __u32 mask, unsigned int fan_flags, + struct fan_fsid *fsid) { struct fsnotify_mark *fsn_mark; - __u32 added; + bool recalc; int ret = 0; - mutex_lock(&group->mark_mutex); - fsn_mark = fsnotify_find_mark(connp, group); + fsnotify_group_lock(group); + fsn_mark = fsnotify_find_mark(obj, obj_type, group); if (!fsn_mark) { - fsn_mark = fanotify_add_new_mark(group, connp, obj_type, fsid); + fsn_mark = fanotify_add_new_mark(group, obj, obj_type, + fan_flags, fsid); if (IS_ERR(fsn_mark)) { - mutex_unlock(&group->mark_mutex); + fsnotify_group_unlock(group); return PTR_ERR(fsn_mark); } } /* + * Check if requested mark flags conflict with an existing mark flags. + */ + ret = fanotify_may_update_existing_mark(fsn_mark, mask, fan_flags); + if (ret) + goto out; + + /* * Error events are pre-allocated per group, only if strictly * needed (i.e. FAN_FS_ERROR was requested). */ - if (!(flags & FAN_MARK_IGNORED_MASK) && (mask & FAN_FS_ERROR)) { + if (!(fan_flags & FANOTIFY_MARK_IGNORE_BITS) && + (mask & FAN_FS_ERROR)) { ret = fanotify_group_init_error_pool(group); if (ret) goto out; } - added = fanotify_mark_add_to_mask(fsn_mark, mask, flags); - if (added & ~fsnotify_conn_mask(fsn_mark->connector)) + recalc = fanotify_mark_add_to_mask(fsn_mark, mask, fan_flags); + if (recalc) fsnotify_recalc_mask(fsn_mark->connector); out: - mutex_unlock(&group->mark_mutex); + fsnotify_group_unlock(group); fsnotify_put_mark(fsn_mark); - return ret; -} - -static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, - struct vfsmount *mnt, __u32 mask, - unsigned int flags, __kernel_fsid_t *fsid) -{ - return fanotify_add_mark(group, &real_mount(mnt)->mnt_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_VFSMOUNT, mask, flags, fsid); -} - -static int fanotify_add_sb_mark(struct fsnotify_group *group, - struct super_block *sb, __u32 mask, - unsigned int flags, __kernel_fsid_t *fsid) -{ - return fanotify_add_mark(group, &sb->s_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_SB, mask, flags, fsid); -} -static int fanotify_add_inode_mark(struct fsnotify_group *group, - struct inode *inode, __u32 mask, - unsigned int flags, __kernel_fsid_t *fsid) -{ - pr_debug("%s: group=%p inode=%p\n", __func__, group, inode); + if (!ret && (mask & FANOTIFY_PERM_EVENTS)) + fanotify_perm_watchdog_group_add(group); - /* - * If some other task has this inode open for write we should not add - * an ignored mark, unless that ignored mark is supposed to survive - * modification changes anyway. - */ - if ((flags & FAN_MARK_IGNORED_MASK) && - !(flags & FAN_MARK_IGNORED_SURV_MODIFY) && - inode_is_open_for_write(inode)) - return 0; - - return fanotify_add_mark(group, &inode->i_fsnotify_marks, - FSNOTIFY_OBJ_TYPE_INODE, mask, flags, fsid); + return ret; } static struct fsnotify_event *fanotify_alloc_overflow_event(void) @@ -1256,10 +1597,16 @@ static struct hlist_head *fanotify_alloc_merge_hash(void) return hash; } +DEFINE_CLASS(fsnotify_group, + struct fsnotify_group *, + if (!IS_ERR_OR_NULL(_T)) fsnotify_destroy_group(_T), + fsnotify_alloc_group(ops, flags), + const struct fsnotify_ops *ops, int flags) + /* fanotify syscalls */ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) { - struct fsnotify_group *group; + struct user_namespace *user_ns = current_user_ns(); int f_flags, fd; unsigned int fid_mode = flags & FANOTIFY_FID_BITS; unsigned int class = flags & FANOTIFY_CLASS_BITS; @@ -1272,10 +1619,11 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) /* * An unprivileged user can setup an fanotify group with * limited functionality - an unprivileged group is limited to - * notification events with file handles and it cannot use - * unlimited queue/marks. + * notification events with file handles or mount ids and it + * cannot use unlimited queue/marks. */ - if ((flags & FANOTIFY_ADMIN_INIT_FLAGS) || !fid_mode) + if ((flags & FANOTIFY_ADMIN_INIT_FLAGS) || + !(flags & (FANOTIFY_FID_BITS | FAN_REPORT_MNT))) return -EPERM; /* @@ -1301,6 +1649,14 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) if ((flags & FAN_REPORT_PIDFD) && (flags & FAN_REPORT_TID)) return -EINVAL; + /* Don't allow mixing mnt events with inode events for now */ + if (flags & FAN_REPORT_MNT) { + if (class != FAN_CLASS_NOTIF) + return -EINVAL; + if (flags & (FANOTIFY_FID_BITS | FAN_REPORT_FD_ERROR)) + return -EINVAL; + } + if (event_f_flags & ~FANOTIFY_INIT_ALL_EVENT_F_BITS) return -EINVAL; @@ -1332,108 +1688,95 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) (!(fid_mode & FAN_REPORT_NAME) || !(fid_mode & FAN_REPORT_FID))) return -EINVAL; - f_flags = O_RDWR | FMODE_NONOTIFY; + f_flags = O_RDWR; if (flags & FAN_CLOEXEC) f_flags |= O_CLOEXEC; if (flags & FAN_NONBLOCK) f_flags |= O_NONBLOCK; + CLASS(fsnotify_group, group)(&fanotify_fsnotify_ops, + FSNOTIFY_GROUP_USER); /* fsnotify_alloc_group takes a ref. Dropped in fanotify_release */ - group = fsnotify_alloc_user_group(&fanotify_fsnotify_ops); - if (IS_ERR(group)) { + if (IS_ERR(group)) return PTR_ERR(group); - } /* Enforce groups limits per user in all containing user ns */ - group->fanotify_data.ucounts = inc_ucount(current_user_ns(), - current_euid(), + group->fanotify_data.ucounts = inc_ucount(user_ns, current_euid(), UCOUNT_FANOTIFY_GROUPS); - if (!group->fanotify_data.ucounts) { - fd = -EMFILE; - goto out_destroy_group; - } + if (!group->fanotify_data.ucounts) + return -EMFILE; group->fanotify_data.flags = flags | internal_flags; group->memcg = get_mem_cgroup_from_mm(current->mm); + group->user_ns = get_user_ns(user_ns); group->fanotify_data.merge_hash = fanotify_alloc_merge_hash(); - if (!group->fanotify_data.merge_hash) { - fd = -ENOMEM; - goto out_destroy_group; - } + if (!group->fanotify_data.merge_hash) + return -ENOMEM; group->overflow_event = fanotify_alloc_overflow_event(); - if (unlikely(!group->overflow_event)) { - fd = -ENOMEM; - goto out_destroy_group; - } + if (unlikely(!group->overflow_event)) + return -ENOMEM; if (force_o_largefile()) event_f_flags |= O_LARGEFILE; group->fanotify_data.f_flags = event_f_flags; init_waitqueue_head(&group->fanotify_data.access_waitq); INIT_LIST_HEAD(&group->fanotify_data.access_list); + INIT_LIST_HEAD(&group->fanotify_data.perm_grp_list); switch (class) { case FAN_CLASS_NOTIF: - group->priority = FS_PRIO_0; + group->priority = FSNOTIFY_PRIO_NORMAL; break; case FAN_CLASS_CONTENT: - group->priority = FS_PRIO_1; + group->priority = FSNOTIFY_PRIO_CONTENT; break; case FAN_CLASS_PRE_CONTENT: - group->priority = FS_PRIO_2; + group->priority = FSNOTIFY_PRIO_PRE_CONTENT; break; default: - fd = -EINVAL; - goto out_destroy_group; + return -EINVAL; } + BUILD_BUG_ON(!(FANOTIFY_ADMIN_INIT_FLAGS & FAN_UNLIMITED_QUEUE)); if (flags & FAN_UNLIMITED_QUEUE) { - fd = -EPERM; - if (!capable(CAP_SYS_ADMIN)) - goto out_destroy_group; group->max_events = UINT_MAX; } else { group->max_events = fanotify_max_queued_events; } - if (flags & FAN_UNLIMITED_MARKS) { - fd = -EPERM; - if (!capable(CAP_SYS_ADMIN)) - goto out_destroy_group; - } - if (flags & FAN_ENABLE_AUDIT) { - fd = -EPERM; if (!capable(CAP_AUDIT_WRITE)) - goto out_destroy_group; + return -EPERM; } - fd = anon_inode_getfd("[fanotify]", &fanotify_fops, group, f_flags); - if (fd < 0) - goto out_destroy_group; - - return fd; - -out_destroy_group: - fsnotify_destroy_group(group); + fd = FD_ADD(f_flags, + anon_inode_getfile_fmode("[fanotify]", &fanotify_fops, + group, f_flags, FMODE_NONOTIFY)); + if (fd >= 0) + retain_and_null_ptr(group); return fd; } -static int fanotify_test_fsid(struct dentry *dentry, __kernel_fsid_t *fsid) +static int fanotify_test_fsid(struct dentry *dentry, unsigned int flags, + struct fan_fsid *fsid) { + unsigned int mark_type = flags & FANOTIFY_MARK_TYPE_BITS; __kernel_fsid_t root_fsid; int err; /* * Make sure dentry is not of a filesystem with zero fsid (e.g. fuse). */ - err = vfs_get_fsid(dentry, fsid); + err = vfs_get_fsid(dentry, &fsid->id); if (err) return err; - if (!fsid->val[0] && !fsid->val[1]) - return -ENODEV; + fsid->sb = dentry->d_sb; + if (!fsid->id.val[0] && !fsid->id.val[1]) { + err = -ENODEV; + goto weak; + } /* * Make sure dentry is not of a filesystem subvolume (e.g. btrfs) @@ -1443,32 +1786,67 @@ static int fanotify_test_fsid(struct dentry *dentry, __kernel_fsid_t *fsid) if (err) return err; - if (root_fsid.val[0] != fsid->val[0] || - root_fsid.val[1] != fsid->val[1]) - return -EXDEV; + if (!fanotify_fsid_equal(&root_fsid, &fsid->id)) { + err = -EXDEV; + goto weak; + } + fsid->weak = false; return 0; + +weak: + /* Allow weak fsid when marking inodes */ + fsid->weak = true; + return (mark_type == FAN_MARK_INODE) ? 0 : err; } /* Check if filesystem can encode a unique fid */ -static int fanotify_test_fid(struct dentry *dentry) +static int fanotify_test_fid(struct dentry *dentry, unsigned int flags) { + unsigned int mark_type = flags & FANOTIFY_MARK_TYPE_BITS; + const struct export_operations *nop = dentry->d_sb->s_export_op; + + /* + * We need to make sure that the filesystem supports encoding of + * file handles so user can use name_to_handle_at() to compare fids + * reported with events to the file handle of watched objects. + */ + if (!exportfs_can_encode_fid(nop)) + return -EOPNOTSUPP; + /* - * We need to make sure that the file system supports at least - * encoding a file handle so user can use name_to_handle_at() to - * compare fid returned with event to the file handle of watched - * objects. However, name_to_handle_at() requires that the - * filesystem also supports decoding file handles. + * For sb/mount mark, we also need to make sure that the filesystem + * supports decoding file handles, so user has a way to map back the + * reported fids to filesystem objects. */ - if (!dentry->d_sb->s_export_op || - !dentry->d_sb->s_export_op->fh_to_dentry) + if (mark_type != FAN_MARK_INODE && !exportfs_can_decode_fh(nop)) return -EOPNOTSUPP; return 0; } -static int fanotify_events_supported(struct path *path, __u64 mask) +static int fanotify_events_supported(struct fsnotify_group *group, + const struct path *path, __u64 mask, + unsigned int flags) { + unsigned int mark_type = flags & FANOTIFY_MARK_TYPE_BITS; + bool is_dir = d_is_dir(path->dentry); + /* Strict validation of events in non-dir inode mask with v5.17+ APIs */ + bool strict_dir_events = FAN_GROUP_FLAG(group, FAN_REPORT_TARGET_FID) || + (mask & FAN_RENAME) || + (flags & FAN_MARK_IGNORE); + + /* + * Filesystems need to opt-into pre-content evnets (a.k.a HSM) + * and they are only supported on regular files and directories. + */ + if (mask & FANOTIFY_PRE_CONTENT_EVENTS) { + if (!(path->mnt->mnt_sb->s_iflags & SB_I_ALLOW_HSM)) + return -EOPNOTSUPP; + if (!is_dir && !d_is_reg(path->dentry)) + return -EINVAL; + } + /* * Some filesystems such as 'proc' acquire unusual locks when opening * files. For them fanotify permission events have high chances of @@ -1480,6 +1858,30 @@ static int fanotify_events_supported(struct path *path, __u64 mask) if (mask & FANOTIFY_PERM_EVENTS && path->mnt->mnt_sb->s_type->fs_flags & FS_DISALLOW_NOTIFY_PERM) return -EINVAL; + + /* + * mount and sb marks are not allowed on kernel internal pseudo fs, + * like pipe_mnt, because that would subscribe to events on all the + * anonynous pipes in the system. + * + * SB_NOUSER covers all of the internal pseudo fs whose objects are not + * exposed to user's mount namespace, but there are other SB_KERNMOUNT + * fs, like nsfs, debugfs, for which the value of allowing sb and mount + * mark is questionable. For now we leave them alone. + */ + if (mark_type != FAN_MARK_INODE && + path->mnt->mnt_sb->s_flags & SB_NOUSER) + return -EINVAL; + + /* + * We shouldn't have allowed setting dirent events and the directory + * flags FAN_ONDIR and FAN_EVENT_ON_CHILD in mask of non-dir inode, + * but because we always allowed it, error only when using new APIs. + */ + if (strict_dir_events && mark_type == FAN_MARK_INODE && + !is_dir && (mask & FANOTIFY_DIRONLY_EVENT_BITS)) + return -ENOTDIR; + return 0; } @@ -1487,15 +1889,17 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, int dfd, const char __user *pathname) { struct inode *inode = NULL; - struct vfsmount *mnt = NULL; struct fsnotify_group *group; - struct fd f; struct path path; - __kernel_fsid_t __fsid, *fsid = NULL; + struct fan_fsid __fsid, *fsid = NULL; + struct user_namespace *user_ns = NULL; + struct mnt_namespace *mntns; u32 valid_mask = FANOTIFY_EVENTS | FANOTIFY_EVENT_FLAGS; unsigned int mark_type = flags & FANOTIFY_MARK_TYPE_BITS; - bool ignored = flags & FAN_MARK_IGNORED_MASK; + unsigned int mark_cmd = flags & FANOTIFY_MARK_CMD_BITS; + unsigned int ignore = flags & FANOTIFY_MARK_IGNORE_BITS; unsigned int obj_type, fid_mode; + void *obj = NULL; u32 umask = 0; int ret; @@ -1519,11 +1923,14 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, case FAN_MARK_FILESYSTEM: obj_type = FSNOTIFY_OBJ_TYPE_SB; break; + case FAN_MARK_MNTNS: + obj_type = FSNOTIFY_OBJ_TYPE_MNTNS; + break; default: return -EINVAL; } - switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) { + switch (mark_cmd) { case FAN_MARK_ADD: case FAN_MARK_REMOVE: if (!mask) @@ -1543,43 +1950,72 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, if (mask & ~valid_mask) return -EINVAL; - /* Event flags (ONDIR, ON_CHILD) are meaningless in ignored mask */ - if (ignored) + + /* We don't allow FAN_MARK_IGNORE & FAN_MARK_IGNORED_MASK together */ + if (ignore == (FAN_MARK_IGNORE | FAN_MARK_IGNORED_MASK)) + return -EINVAL; + + /* + * Event flags (FAN_ONDIR, FAN_EVENT_ON_CHILD) have no effect with + * FAN_MARK_IGNORED_MASK. + */ + if (ignore == FAN_MARK_IGNORED_MASK) { mask &= ~FANOTIFY_EVENT_FLAGS; + umask = FANOTIFY_EVENT_FLAGS; + } - f = fdget(fanotify_fd); - if (unlikely(!f.file)) + CLASS(fd, f)(fanotify_fd); + if (fd_empty(f)) return -EBADF; /* verify that this is indeed an fanotify instance */ - ret = -EINVAL; - if (unlikely(f.file->f_op != &fanotify_fops)) - goto fput_and_out; - group = f.file->private_data; + if (unlikely(fd_file(f)->f_op != &fanotify_fops)) + return -EINVAL; + group = fd_file(f)->private_data; + + /* Only report mount events on mnt namespace */ + if (FAN_GROUP_FLAG(group, FAN_REPORT_MNT)) { + if (mask & ~FANOTIFY_MOUNT_EVENTS) + return -EINVAL; + if (mark_type != FAN_MARK_MNTNS) + return -EINVAL; + } else { + if (mask & FANOTIFY_MOUNT_EVENTS) + return -EINVAL; + if (mark_type == FAN_MARK_MNTNS) + return -EINVAL; + } /* - * An unprivileged user is not allowed to setup mount nor filesystem - * marks. This also includes setting up such marks by a group that - * was initialized by an unprivileged user. + * A user is allowed to setup sb/mount/mntns marks only if it is + * capable in the user ns where the group was created. */ - ret = -EPERM; - if ((!capable(CAP_SYS_ADMIN) || - FAN_GROUP_FLAG(group, FANOTIFY_UNPRIV)) && + if (!ns_capable(group->user_ns, CAP_SYS_ADMIN) && mark_type != FAN_MARK_INODE) - goto fput_and_out; + return -EPERM; /* - * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF. These are not - * allowed to set permissions events. + * Permission events are not allowed for FAN_CLASS_NOTIF. + * Pre-content permission events are not allowed for FAN_CLASS_CONTENT. */ - ret = -EINVAL; if (mask & FANOTIFY_PERM_EVENTS && - group->priority == FS_PRIO_0) - goto fput_and_out; + group->priority == FSNOTIFY_PRIO_NORMAL) + return -EINVAL; + else if (mask & FANOTIFY_PRE_CONTENT_EVENTS && + group->priority == FSNOTIFY_PRIO_CONTENT) + return -EINVAL; if (mask & FAN_FS_ERROR && mark_type != FAN_MARK_FILESYSTEM) - goto fput_and_out; + return -EINVAL; + + /* + * Evictable is only relevant for inode marks, because only inode object + * can be evicted on memory pressure. + */ + if (flags & FAN_MARK_EVICTABLE && + mark_type != FAN_MARK_INODE) + return -EINVAL; /* * Events that do not carry enough information to report @@ -1589,9 +2025,9 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, * point. */ fid_mode = FAN_GROUP_FLAG(group, FANOTIFY_FID_BITS); - if (mask & ~(FANOTIFY_FD_EVENTS|FANOTIFY_EVENT_FLAGS) && + if (mask & ~(FANOTIFY_FD_EVENTS|FANOTIFY_MOUNT_EVENTS|FANOTIFY_EVENT_FLAGS) && (!fid_mode || mark_type == FAN_MARK_MOUNT)) - goto fput_and_out; + return -EINVAL; /* * FAN_RENAME uses special info type records to report the old and @@ -1599,50 +2035,92 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, * useful and was not implemented. */ if (mask & FAN_RENAME && !(fid_mode & FAN_REPORT_NAME)) - goto fput_and_out; + return -EINVAL; - if (flags & FAN_MARK_FLUSH) { - ret = 0; - if (mark_type == FAN_MARK_MOUNT) - fsnotify_clear_vfsmount_marks_by_group(group); - else if (mark_type == FAN_MARK_FILESYSTEM) - fsnotify_clear_sb_marks_by_group(group); - else - fsnotify_clear_inode_marks_by_group(group); - goto fput_and_out; + /* Pre-content events are not currently generated for directories. */ + if (mask & FANOTIFY_PRE_CONTENT_EVENTS && mask & FAN_ONDIR) + return -EINVAL; + + if (mark_cmd == FAN_MARK_FLUSH) { + fsnotify_clear_marks_by_group(group, obj_type); + return 0; } ret = fanotify_find_path(dfd, pathname, &path, flags, (mask & ALL_FSNOTIFY_EVENTS), obj_type); if (ret) - goto fput_and_out; + return ret; - if (flags & FAN_MARK_ADD) { - ret = fanotify_events_supported(&path, mask); + if (mark_cmd == FAN_MARK_ADD) { + ret = fanotify_events_supported(group, &path, mask, flags); if (ret) goto path_put_and_out; } if (fid_mode) { - ret = fanotify_test_fsid(path.dentry, &__fsid); + ret = fanotify_test_fsid(path.dentry, flags, &__fsid); if (ret) goto path_put_and_out; - ret = fanotify_test_fid(path.dentry); + ret = fanotify_test_fid(path.dentry, flags); if (ret) goto path_put_and_out; fsid = &__fsid; } - /* inode held in place by reference to path; group by fget on fd */ - if (mark_type == FAN_MARK_INODE) + /* + * In addition to being capable in the user ns where group was created, + * the user also needs to be capable in the user ns associated with + * the filesystem or in the user ns associated with the mntns + * (when marking mntns). + */ + if (obj_type == FSNOTIFY_OBJ_TYPE_INODE) { inode = path.dentry->d_inode; - else - mnt = path.mnt; + obj = inode; + } else if (obj_type == FSNOTIFY_OBJ_TYPE_VFSMOUNT) { + user_ns = path.mnt->mnt_sb->s_user_ns; + obj = path.mnt; + } else if (obj_type == FSNOTIFY_OBJ_TYPE_SB) { + user_ns = path.mnt->mnt_sb->s_user_ns; + obj = path.mnt->mnt_sb; + } else if (obj_type == FSNOTIFY_OBJ_TYPE_MNTNS) { + ret = -EINVAL; + mntns = mnt_ns_from_dentry(path.dentry); + if (!mntns) + goto path_put_and_out; + user_ns = mntns->user_ns; + obj = mntns; + } + + ret = -EPERM; + if (user_ns && !ns_capable(user_ns, CAP_SYS_ADMIN)) + goto path_put_and_out; + + ret = -EINVAL; + if (!obj) + goto path_put_and_out; + + /* + * If some other task has this inode open for write we should not add + * an ignore mask, unless that ignore mask is supposed to survive + * modification changes anyway. + */ + if (mark_cmd == FAN_MARK_ADD && (flags & FANOTIFY_MARK_IGNORE_BITS) && + !(flags & FAN_MARK_IGNORED_SURV_MODIFY)) { + ret = !inode ? -EINVAL : -EISDIR; + /* FAN_MARK_IGNORE requires SURV_MODIFY for sb/mount/dir marks */ + if (ignore == FAN_MARK_IGNORE && + (!inode || S_ISDIR(inode->i_mode))) + goto path_put_and_out; + + ret = 0; + if (inode && inode_is_open_for_write(inode)) + goto path_put_and_out; + } /* Mask out FAN_EVENT_ON_CHILD flag for sb/mount/non-dir marks */ - if (mnt || !S_ISDIR(inode->i_mode)) { + if (!inode || !S_ISDIR(inode->i_mode)) { mask &= ~FAN_EVENT_ON_CHILD; umask = FAN_EVENT_ON_CHILD; /* @@ -1650,33 +2128,19 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, * events with parent/name info for non-directory. */ if ((fid_mode & FAN_REPORT_DIR_FID) && - (flags & FAN_MARK_ADD) && !ignored) + (flags & FAN_MARK_ADD) && !ignore) mask |= FAN_EVENT_ON_CHILD; } /* create/update an inode mark */ - switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE)) { + switch (mark_cmd) { case FAN_MARK_ADD: - if (mark_type == FAN_MARK_MOUNT) - ret = fanotify_add_vfsmount_mark(group, mnt, mask, - flags, fsid); - else if (mark_type == FAN_MARK_FILESYSTEM) - ret = fanotify_add_sb_mark(group, mnt->mnt_sb, mask, - flags, fsid); - else - ret = fanotify_add_inode_mark(group, inode, mask, - flags, fsid); + ret = fanotify_add_mark(group, obj, obj_type, mask, flags, + fsid); break; case FAN_MARK_REMOVE: - if (mark_type == FAN_MARK_MOUNT) - ret = fanotify_remove_vfsmount_mark(group, mnt, mask, - flags, umask); - else if (mark_type == FAN_MARK_FILESYSTEM) - ret = fanotify_remove_sb_mark(group, mnt->mnt_sb, mask, - flags, umask); - else - ret = fanotify_remove_inode_mark(group, inode, mask, - flags, umask); + ret = fanotify_remove_mark(group, obj, obj_type, mask, flags, + umask); break; default: ret = -EINVAL; @@ -1684,8 +2148,6 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask, path_put_and_out: path_put(&path); -fput_and_out: - fdput(f); return ret; } @@ -1732,10 +2194,10 @@ static int __init fanotify_user_setup(void) FANOTIFY_DEFAULT_MAX_USER_MARKS); BUILD_BUG_ON(FANOTIFY_INIT_FLAGS & FANOTIFY_INTERNAL_GROUP_FLAGS); - BUILD_BUG_ON(HWEIGHT32(FANOTIFY_INIT_FLAGS) != 12); - BUILD_BUG_ON(HWEIGHT32(FANOTIFY_MARK_FLAGS) != 9); + BUILD_BUG_ON(HWEIGHT32(FANOTIFY_INIT_FLAGS) != 14); + BUILD_BUG_ON(HWEIGHT32(FANOTIFY_MARK_FLAGS) != 11); - fanotify_mark_cache = KMEM_CACHE(fsnotify_mark, + fanotify_mark_cache = KMEM_CACHE(fanotify_mark, SLAB_PANIC|SLAB_ACCOUNT); fanotify_fid_event_cachep = KMEM_CACHE(fanotify_fid_event, SLAB_PANIC); @@ -1745,6 +2207,7 @@ static int __init fanotify_user_setup(void) fanotify_perm_event_cachep = KMEM_CACHE(fanotify_perm_event, SLAB_PANIC); } + fanotify_mnt_event_cachep = KMEM_CACHE(fanotify_mnt_event, SLAB_PANIC); fanotify_max_queued_events = FANOTIFY_DEFAULT_MAX_EVENTS; init_user_ns.ucount_max[UCOUNT_FANOTIFY_GROUPS] = |