NEWS

@@ -310,44 +310,6 @@ CHANGES WITH 258 in spe:
           SO_PASSPIDFD socket option for AF_UNIX socket. There's also a new
           setting AcceptFileDescriptors= that controls the new SO_PASSRIGHTS.
 
-        * A new job type "lenient" has been added, that is similar to the
-          existing "fail" job mode, and which will fail the submitted
-          transaction immediately if it would stop any currently running unit.
-
-        * .socket units gained a new pair of settings DeferTrigger= and
-          DeferTriggerMaxSec= which modify triggering behaviour of the
-          socket. When used this will cause the triggered unit to be enqueued
-          with the new "lenient" job mode, and if the submission of the
-          transaction fails it is later retried to be submitted (up to a
-          configurable timeout), whenever a unit is stopped.
-
-        * The "preset" logic has been extended so that there are now three
-          preset directories: one that declares the default enablement state
-          for per-system services run on the host, one for per-user services,
-          and – now new – one for per-system services that are run in the
-          initrd. This reflects the fact that in many cases services that shall
-          be enabled by default on the host should not be enabled by default in
-          the initrd, or vice versa. Note that while the regular per-system
-          preset policy defaults to enabled, the one for the initrd defaults to
-          disabled.
-
-        * There are now new per-service settings
-          StateDirectoryQuota=/StateDirectoryAccounting=,
-          CacheDirectoryQuota=/CacheDirectoryAccounting=,
-          LogsDirectoryQuota=/LogsDirectoryAccounting= which allow doing
-          per-unit quota of the indicated per-unit directories. This is
-          implemented via project quota, as supported by xfs and ext4. This
-          does not support btrfs, currently. If quota accounting is enabled
-          this information is shown in the usual "systemct status" output.
-
-        * The service manager gained a new KillUnitSubgroup() syscall which may
-          be used to send a signal to a sub-control group of the unit's control
-          group. systemctl kill gained a new --kill-subgroup= switch to make
-          this available from the shell.
-
-        * A new PrivateBPF= switch has been added for unit files, which may be
-          used to mount a private bpffs instance for the unit's processes.
-
         systemd-journald & journal-remote:
 
         * journalctl's --setup-keys command now supports JSON output.
@@ -643,10 +605,6 @@ CHANGES WITH 258 in spe:
           servers. Delegate zones can be configured via drop-ins below
           /etc/systemd/dns-delegate.d/*.dns-delegate.
 
-        * "resolvectl query -t sshfp" will now decode the returned RR
-          information, and show the cryptographic algorithms by name instead of
-          number.
-
         systemd-hostnamed:
 
         * The system hardware's serial number may now be read from DeviceTree
@@ -1203,15 +1161,6 @@ CHANGES WITH 258 in spe:
           Hardware IDs" (CHIDs) of the local system. This is useful for
           preparing CHID-to-DeviceTree mappings when building UKIs.
 
-        * systemd-analyze gained a new "transient-settings" verb, which shows
-          all unit settings one can configure dynamically via the "-p" switch
-          when invoking transient units.
-
-        * systemd-analyze gained a new "unit-shell" verb that invokes an
-          interactive shell inside the processes namespaces of the main process
-          of a specified unit. This is useful for debugging unit sandboxes, and
-          getting an idea how things look like from the "inside" of a service.
-
         * The "package note" specification ELF binaries has been extended to
           cover PE binaries (i.e. UEFI binaries), too.
 
@@ -1376,17 +1325,6 @@ CHANGES WITH 258 in spe:
           specified binary is immediately invoked, and not delayed until a
           connection comes in.
 
-        * systemd-ssh-generator will now generate the AF_VSOCK ssh listener
-          .socket unit, so that a tiny new helper "systemd-ssh-issue" is
-          invoked when the socket is bound, that generates a drop-in file
-          /run/issue.d/50-ssh-vsock.issue that is shown by "login" and other
-          subsystems at login time. The file reports the AF_VSOCK CID of the
-          system, along with very brief information how to connect to the
-          system via ssh-over-AF_VSOCK. Or in other words: if the system is
-          booted up in an AF_VSOCK capable VM the console login screen shown
-          once boot-up is complete will tell you how to connect to the system
-          via SSH, if that's available.
-
         — <place>, <date>
 
 CHANGES WITH 257:

src/basic/cgroup-util.c

@@ -142,9 +142,6 @@ int cg_read_pid(FILE *f, pid_t *ret, CGroupFlags flags) {
         assert(f);
         assert(ret);
 
-        /* NB: The kernel returns ENODEV if we tried to read from cgroup.procs of a cgroup that has been
-         * removed already. Callers should handle that! */
-
         for (;;) {
                 errno = 0;
                 if (fscanf(f, "%lu", &ul) != 1) {
@@ -303,13 +300,6 @@ int cg_kill(
                         _cleanup_(pidref_done) PidRef pidref = PIDREF_NULL;
 
                         r = cg_read_pidref(f, &pidref, flags);
-                        if (r == -ENODEV) {
-                                /* reading from cgroup.pids will result in ENODEV if the cgroup is
-                                 * concurrently removed. Just leave in that case, because a removed cgroup
-                                 * contains no processes anymore. */
-                                done = true;
-                                break;
-                        }
                         if (r < 0)
                                 return RET_GATHER(ret, log_debug_errno(r, "Failed to read pidref from cgroup '%s': %m", path));
                         if (r == 0)

src/basic/recurse-dir.c

@@ -175,7 +175,7 @@ int recurse_dir(
 
                 r = func(RECURSE_DIR_ENTER,
                          path,
-                         -EBADF, /* we have no parent fd */
+                         -1, /* we have no parent fd */
                          dir_fd,
                          NULL, /* we have no dirent */
                          statx_mask != 0 ? &root_sx : NULL,
@@ -191,8 +191,7 @@ int recurse_dir(
         if (r < 0)
                 return r;
 
-        FOREACH_ARRAY(entry, de->entries, de->n_entries) {
+        for (size_t i = 0; i < de->n_entries; i++) {
-                struct dirent *i = *entry;
                 _cleanup_close_ int inode_fd = -EBADF, subdir_fd = -EBADF;
                 _cleanup_free_ char *joined = NULL;
                 struct statx sx;
@@ -207,16 +206,16 @@ int recurse_dir(
                  */
 
                 if (path) {
-                        joined = path_join(path, i->d_name);
+                        joined = path_join(path, de->entries[i]->d_name);
                         if (!joined)
                                 return -ENOMEM;
 
                         p = joined;
                 } else
-                        p = i->d_name;
+                        p = de->entries[i]->d_name;
 
-                if (IN_SET(i->d_type, DT_UNKNOWN, DT_DIR)) {
+                if (IN_SET(de->entries[i]->d_type, DT_UNKNOWN, DT_DIR)) {
-                        subdir_fd = openat(dir_fd, i->d_name, O_DIRECTORY|O_NOFOLLOW|O_CLOEXEC);
+                        subdir_fd = openat(dir_fd, de->entries[i]->d_name, O_DIRECTORY|O_NOFOLLOW|O_CLOEXEC);
                         if (subdir_fd < 0) {
                                 if (errno == ENOENT) /* Vanished by now, go for next file immediately */
                                         continue;
@@ -230,9 +229,9 @@ int recurse_dir(
                                         r = func(RECURSE_DIR_SKIP_OPEN_DIR_ERROR_BASE + errno,
                                                  p,
                                                  dir_fd,
-                                                 /* inode_fd = */ -EBADF,
+                                                 -1,
-                                                 i,
+                                                 de->entries[i],
-                                                 /* sx = */ NULL,
+                                                 NULL,
                                                  userdata);
                                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
                                                 break;
@@ -246,7 +245,7 @@ int recurse_dir(
 
                         } else {
                                 /* If we managed to get a DIR* off the inode, it's definitely a directory. */
-                                i->d_type = DT_DIR;
+                                de->entries[i]->d_type = DT_DIR;
 
                                 if (statx_mask != 0 || (flags & RECURSE_DIR_SAME_MOUNT)) {
                                         if (statx(subdir_fd, "", AT_EMPTY_PATH, statx_mask, &sx) < 0)
@@ -262,7 +261,7 @@ int recurse_dir(
 
                         if (flags & RECURSE_DIR_INODE_FD) {
 
-                                inode_fd = openat(dir_fd, i->d_name, O_PATH|O_NOFOLLOW|O_CLOEXEC);
+                                inode_fd = openat(dir_fd, de->entries[i]->d_name, O_PATH|O_NOFOLLOW|O_CLOEXEC);
                                 if (inode_fd < 0) {
                                         if (errno == ENOENT) /* Vanished by now, go for next file immediately */
                                                 continue;
@@ -274,9 +273,9 @@ int recurse_dir(
                                         r = func(RECURSE_DIR_SKIP_OPEN_INODE_ERROR_BASE + errno,
                                                  p,
                                                  dir_fd,
-                                                 /* inode_fd = */ -EBADF,
+                                                 -1,
-                                                 i,
+                                                 de->entries[i],
-                                                 /* sx = */ NULL,
+                                                 NULL,
                                                  userdata);
                                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
                                                 break;
@@ -309,9 +308,9 @@ int recurse_dir(
                                         inode_fd = safe_close(inode_fd);
                                 }
 
-                        } else if (statx_mask != 0 || (i->d_type == DT_UNKNOWN && (flags & RECURSE_DIR_ENSURE_TYPE))) {
+                        } else if (statx_mask != 0 || (de->entries[i]->d_type == DT_UNKNOWN && (flags & RECURSE_DIR_ENSURE_TYPE))) {
 
-                                if (statx(dir_fd, i->d_name, AT_SYMLINK_NOFOLLOW, statx_mask | STATX_TYPE, &sx) < 0) {
+                                if (statx(dir_fd, de->entries[i]->d_name, AT_SYMLINK_NOFOLLOW, statx_mask | STATX_TYPE, &sx) < 0) {
                                         if (errno == ENOENT) /* Vanished by now? Go for next file immediately */
                                                 continue;
 
@@ -322,9 +321,9 @@ int recurse_dir(
                                         r = func(RECURSE_DIR_SKIP_STAT_INODE_ERROR_BASE + errno,
                                                  p,
                                                  dir_fd,
-                                                 /* inode_fd = */ -EBADF,
+                                                 -1,
-                                                 i,
+                                                 de->entries[i],
-                                                 /* sx = */ NULL,
+                                                 NULL,
                                                  userdata);
                                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
                                                 break;
@@ -352,9 +351,9 @@ int recurse_dir(
                                         r = func(RECURSE_DIR_SKIP_STAT_INODE_ERROR_BASE + EISDIR,
                                                  p,
                                                  dir_fd,
-                                                 /* inode_fd = */ -EBADF,
+                                                 -1,
-                                                 i,
+                                                 de->entries[i],
-                                                 /* sx = */ NULL,
+                                                 NULL,
                                                  userdata);
                                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
                                                 break;
@@ -370,11 +369,11 @@ int recurse_dir(
                         /* Copy over the data we acquired through statx() if we acquired any */
                         if (sx.stx_mask & STATX_TYPE) {
                                 assert((subdir_fd < 0) == !S_ISDIR(sx.stx_mode));
-                                i->d_type = IFTODT(sx.stx_mode);
+                                de->entries[i]->d_type = IFTODT(sx.stx_mode);
                         }
 
                         if (sx.stx_mask & STATX_INO)
-                                i->d_ino = sx.stx_ino;
+                                de->entries[i]->d_ino = sx.stx_ino;
                 }
 
                 if (subdir_fd >= 0) {
@@ -384,7 +383,7 @@ int recurse_dir(
                                 if (sx_valid && FLAGS_SET(sx.stx_attributes_mask, STATX_ATTR_MOUNT_ROOT))
                                         is_mount = FLAGS_SET(sx.stx_attributes, STATX_ATTR_MOUNT_ROOT);
                                 else {
-                                        r = is_mount_point_at(dir_fd, i->d_name, /* flags = */ 0);
+                                        r = is_mount_point_at(dir_fd, de->entries[i]->d_name, 0);
                                         if (r < 0)
                                                 log_debug_errno(r, "Failed to determine whether %s is a submount, assuming not: %m", p);
 
@@ -396,7 +395,7 @@ int recurse_dir(
                                                  p,
                                                  dir_fd,
                                                  subdir_fd,
-                                                 i,
+                                                 de->entries[i],
                                                  statx_mask != 0 ? &sx : NULL, /* only pass sx if user asked for it */
                                                  userdata);
                                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
@@ -415,7 +414,7 @@ int recurse_dir(
                                          p,
                                          dir_fd,
                                          subdir_fd,
-                                         i,
+                                         de->entries[i],
                                          statx_mask != 0 ? &sx : NULL, /* only pass sx if user asked for it */
                                          userdata);
                                 if (r == RECURSE_DIR_LEAVE_DIRECTORY)
@@ -430,7 +429,7 @@ int recurse_dir(
                                  p,
                                  dir_fd,
                                  subdir_fd,
-                                 i,
+                                 de->entries[i],
                                  statx_mask != 0 ? &sx : NULL, /* only pass sx if user asked for it */
                                  userdata);
                         if (r == RECURSE_DIR_LEAVE_DIRECTORY)
@@ -444,7 +443,7 @@ int recurse_dir(
                                         p,
                                         statx_mask,
                                         n_depth_max - 1,
-                                        flags & ~RECURSE_DIR_TOPLEVEL, /* we already called the callback for this entry */
+                                        flags &~ RECURSE_DIR_TOPLEVEL, /* we already called the callback for this entry */
                                         func,
                                         userdata);
                         if (r != 0)
@@ -454,7 +453,7 @@ int recurse_dir(
                                  p,
                                  dir_fd,
                                  subdir_fd,
-                                 i,
+                                 de->entries[i],
                                  statx_mask != 0 ? &sx : NULL, /* only pass sx if user asked for it */
                                  userdata);
                 } else
@@ -463,7 +462,7 @@ int recurse_dir(
                                  p,
                                  dir_fd,
                                  inode_fd,
-                                 i,
+                                 de->entries[i],
                                  statx_mask != 0 ? &sx : NULL, /* only pass sx if user asked for it */
                                  userdata);
 
@@ -477,9 +476,9 @@ int recurse_dir(
 
                 r = func(RECURSE_DIR_LEAVE,
                          path,
-                         -EBADF, /* we have no parent fd */
+                         -1,
                          dir_fd,
-                         NULL, /* we have no dirent */
+                         NULL,
                          statx_mask != 0 ? &root_sx : NULL,
                          userdata);
                 if (!IN_SET(r, RECURSE_DIR_LEAVE_DIRECTORY, RECURSE_DIR_SKIP_ENTRY, RECURSE_DIR_CONTINUE))

src/core/dbus-unit.c

@@ -1351,13 +1351,10 @@ static int append_cgroup(sd_bus_message *reply, const char *p, Set *pids) {
                 /* libvirt / qemu uses threaded mode and cgroup.procs cannot be read at the lower levels.
                  * From https://docs.kernel.org/admin-guide/cgroup-v2.html#threads, “cgroup.procs” in a
                  * threaded domain cgroup contains the PIDs of all processes in the subtree and is not
-                 * readable in the subtree proper.
+                 * readable in the subtree proper. */
-                 *
-                 * We'll see ENODEV when trying to enumerate processes and the cgroup is removed at the same
-                 * time. Handle this gracefully. */
 
                 r = cg_read_pidref(f, &pidref, /* flags = */ 0);
-                if (IN_SET(r, 0, -EOPNOTSUPP, -ENODEV))
+                if (IN_SET(r, 0, -EOPNOTSUPP))
                         break;
                 if (r < 0)
                         return r;

src/nspawn/nspawn-mount.c

@@ -824,7 +824,7 @@ static int mount_bind(const char *dest, CustomMount *m, uid_t uid_shift, uid_t u
          * caller's userns *without* any mount idmapping in place. To get that uid, we clone the
          * mount source tree and clear any existing idmapping and temporarily mount that tree over
          * the mount source before we stat the mount source to figure out the source uid. */
-        _cleanup_close_ int fd_clone = open_tree_attr_with_fallback(
+        _cleanup_close_ int fd_clone = open_tree_attr_fallback(
                         AT_FDCWD,
                         m->source,
                         OPEN_TREE_CLONE|OPEN_TREE_CLOEXEC,

src/shared/cgroup-setup.c

@@ -367,8 +367,6 @@ int cg_migrate(
                         if (r < 0)
                                 return RET_GATHER(ret, r);
                 }
-                if (r == -ENODEV)
-                        continue;
                 if (r < 0)
                         return RET_GATHER(ret, r);
         } while (!done);

src/shared/cgroup-show.c

@@ -107,12 +107,9 @@ static int show_cgroup_one_by_path(
                 /* libvirt / qemu uses threaded mode and cgroup.procs cannot be read at the lower levels.
                  * From https://docs.kernel.org/admin-guide/cgroup-v2.html#threads,
                  * “cgroup.procs” in a threaded domain cgroup contains the PIDs of all processes in
-                 * the subtree and is not readable in the subtree proper.
+                 * the subtree and is not readable in the subtree proper. */
-                 *
-                 * ENODEV is generated when we enumerate processes from a cgroup and the cgroup is removed
-                 * concurrently. */
                 r = cg_read_pid(f, &pid, /* flags = */ 0);
-                if (IN_SET(r, 0, -EOPNOTSUPP, -ENODEV))
+                if (IN_SET(r, 0, -EOPNOTSUPP))
                         break;
                 if (r < 0)
                         return r;

src/shared/mount-util.c

@@ -1441,18 +1441,10 @@ int make_userns(uid_t uid_shift,
         return TAKE_FD(userns_fd);
 }
 
-int open_tree_attr_with_fallback(int dir_fd, const char *path, unsigned int flags, struct mount_attr *attr) {
+int open_tree_attr_fallback(int dir_fd, const char *path, unsigned int flags, struct mount_attr *attr) {
-        _cleanup_close_ int fd = -EBADF;
-
-        assert(dir_fd >= 0 || dir_fd == AT_FDCWD);
         assert(attr);
 
-        if (isempty(path)) {
+        _cleanup_close_ int fd = open_tree_attr(dir_fd, path, flags, attr, sizeof(struct mount_attr));
-                path = "";
-                flags |= AT_EMPTY_PATH;
-        }
-
-        fd = open_tree_attr(dir_fd, path, flags, attr, sizeof(struct mount_attr));
         if (fd >= 0)
                 return TAKE_FD(fd);
         if (!ERRNO_IS_NOT_SUPPORTED(errno))
@@ -1500,8 +1492,8 @@ int remount_idmap_fd(
 
         for (size_t i = 0; i < n; i++) {
                 /* Clone the mount point and et the user namespace mapping attribute on the cloned mount point. */
-                mount_fds[n_mounts_fds] = open_tree_attr_with_fallback(
+                mount_fds[n_mounts_fds] = open_tree_attr_fallback(
-                                AT_FDCWD,
+                                /* dir_fd= */ -EBADF,
                                 paths[i],
                                 OPEN_TREE_CLONE | OPEN_TREE_CLOEXEC,
                                 &(struct mount_attr) {

src/shared/mount-util.h

@@ -13,13 +13,11 @@ typedef struct SubMount {
 void sub_mount_array_free(SubMount *s, size_t n);
 
 int get_sub_mounts(const char *prefix, SubMount **ret_mounts, size_t *ret_n_mounts);
-int bind_mount_submounts(
-                const char *source,
-                const char *target);
 
 int repeat_unmount(const char *path, int flags);
 
 int umount_recursive_full(const char *target, int flags, char **keep);
+
 static inline int umount_recursive(const char *target, int flags) {
         return umount_recursive_full(target, flags, NULL);
 }
@@ -150,12 +148,16 @@ typedef enum RemountIdmapping {
         _REMOUNT_IDMAPPING_INVALID = -EINVAL,
 } RemountIdmapping;
 
-int open_tree_attr_with_fallback(int dir_fd, const char *path, unsigned int flags, struct mount_attr *attr);
+int open_tree_attr_fallback(int dir_fd, const char *path, unsigned int flags, struct mount_attr *attr);
 
 int make_userns(uid_t uid_shift, uid_t uid_range, uid_t host_owner, uid_t dest_owner, RemountIdmapping idmapping);
 int remount_idmap_fd(char **p, int userns_fd, uint64_t extra_mount_attr_set);
 int remount_idmap(char **p, uid_t uid_shift, uid_t uid_range, uid_t host_owner, uid_t dest_owner, RemountIdmapping idmapping);
 
+int bind_mount_submounts(
+                const char *source,
+                const char *target);
+
 /* Creates a mount point (without any parents) based on the source path or mode - i.e., a file or a directory */
 int make_mount_point_inode_from_mode(int dir_fd, const char *dest, mode_t source_mode, mode_t target_mode);
 int make_mount_point_inode_from_path(const char *source, const char *dest, mode_t mode);