From 98d2b43081972abeb5bb5a087bc3e3197531c46e Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@redhat.com>
Date: Wed, 25 Oct 2023 16:01:59 +0200
Subject: add unique mount ID

If a mount is released then its mnt_id can immediately be reused.  This is
bad news for user interfaces that want to uniquely identify a mount.

Implementing a unique mount ID is trivial (use a 64bit counter).
Unfortunately userspace assumes 32bit size and would overflow after the
counter reaches 2^32.

Introduce a new 64bit ID alongside the old one.  Initialize the counter to
2^32, this guarantees that the old and new IDs are never mixed up.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Link: https://lore.kernel.org/r/20231025140205.3586473-2-mszeredi@redhat.com
Reviewed-by: Ian Kent <raven@themaw.net>
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 include/uapi/linux/stat.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/uapi')

diff --git a/include/uapi/linux/stat.h b/include/uapi/linux/stat.h
index 7cab2c65d3d7..2f2ee82d5517 100644
--- a/include/uapi/linux/stat.h
+++ b/include/uapi/linux/stat.h
@@ -154,6 +154,7 @@ struct statx {
 #define STATX_BTIME		0x00000800U	/* Want/got stx_btime */
 #define STATX_MNT_ID		0x00001000U	/* Got stx_mnt_id */
 #define STATX_DIOALIGN		0x00002000U	/* Want/got direct I/O alignment info */
+#define STATX_MNT_ID_UNIQUE	0x00004000U	/* Want/got extended stx_mount_id */
 
 #define STATX__RESERVED		0x80000000U	/* Reserved for future struct statx expansion */
 
-- 
cgit v1.2.3


From 46eae99ef73302f9fb3dddcd67c374b3dffe8fd6 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@redhat.com>
Date: Wed, 25 Oct 2023 16:02:02 +0200
Subject: add statmount(2) syscall

Add a way to query attributes of a single mount instead of having to parse
the complete /proc/$PID/mountinfo, which might be huge.

Lookup the mount the new 64bit mount ID.  If a mount needs to be queried
based on path, then statx(2) can be used to first query the mount ID
belonging to the path.

Design is based on a suggestion by Linus:

  "So I'd suggest something that is very much like "statfsat()", which gets
   a buffer and a length, and returns an extended "struct statfs" *AND*
   just a string description at the end."

The interface closely mimics that of statx.

Handle ASCII attributes by appending after the end of the structure (as per
above suggestion).  Pointers to strings are stored in u64 members to make
the structure the same regardless of pointer size.  Strings are nul
terminated.

Link: https://lore.kernel.org/all/CAHk-=wh5YifP7hzKSbwJj94+DZ2czjrZsczy6GBimiogZws=rg@mail.gmail.com/
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Link: https://lore.kernel.org/r/20231025140205.3586473-5-mszeredi@redhat.com
Reviewed-by: Ian Kent <raven@themaw.net>
[Christian Brauner <brauner@kernel.org>: various minor changes]
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/namespace.c             | 281 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/syscalls.h   |   5 +
 include/uapi/linux/mount.h |  53 +++++++++
 3 files changed, 339 insertions(+)

(limited to 'include/uapi')

diff --git a/fs/namespace.c b/fs/namespace.c
index d3665d025acb..ae35d8b6aca8 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -4683,6 +4683,287 @@ int show_path(struct seq_file *m, struct dentry *root)
 	return 0;
 }
 
+static struct vfsmount *lookup_mnt_in_ns(u64 id, struct mnt_namespace *ns)
+{
+	struct mount *mnt = mnt_find_id_at(ns, id);
+
+	if (!mnt || mnt->mnt_id_unique != id)
+		return NULL;
+
+	return &mnt->mnt;
+}
+
+struct kstatmount {
+	struct statmount __user *const buf;
+	size_t const bufsize;
+	struct vfsmount *const mnt;
+	u64 const mask;
+	struct seq_file seq;
+	struct path root;
+	struct statmount sm;
+	size_t pos;
+	int err;
+};
+
+typedef int (*statmount_func_t)(struct kstatmount *);
+
+static int statmount_string_seq(struct kstatmount *s, statmount_func_t func)
+{
+	size_t rem = s->bufsize - s->pos - sizeof(s->sm);
+	struct seq_file *seq = &s->seq;
+	int ret;
+
+	seq->count = 0;
+	seq->size = min(seq->size, rem);
+	seq->buf = kvmalloc(seq->size, GFP_KERNEL_ACCOUNT);
+	if (!seq->buf)
+		return -ENOMEM;
+
+	ret = func(s);
+	if (ret)
+		return ret;
+
+	if (seq_has_overflowed(seq)) {
+		if (seq->size == rem)
+			return -EOVERFLOW;
+		seq->size *= 2;
+		if (seq->size > MAX_RW_COUNT)
+			return -ENOMEM;
+		kvfree(seq->buf);
+		return 0;
+	}
+
+	/* Done */
+	return 1;
+}
+
+static void statmount_string(struct kstatmount *s, u64 mask, statmount_func_t func,
+		       u32 *str)
+{
+	int ret = s->pos + sizeof(s->sm) >= s->bufsize ? -EOVERFLOW : 0;
+	struct statmount *sm = &s->sm;
+	struct seq_file *seq = &s->seq;
+
+	if (s->err || !(s->mask & mask))
+		return;
+
+	seq->size = PAGE_SIZE;
+	while (!ret)
+		ret = statmount_string_seq(s, func);
+
+	if (ret < 0) {
+		s->err = ret;
+	} else {
+		seq->buf[seq->count++] = '\0';
+		if (copy_to_user(s->buf->str + s->pos, seq->buf, seq->count)) {
+			s->err = -EFAULT;
+		} else {
+			*str = s->pos;
+			s->pos += seq->count;
+		}
+	}
+	kvfree(seq->buf);
+	sm->mask |= mask;
+}
+
+static void statmount_numeric(struct kstatmount *s, u64 mask, statmount_func_t func)
+{
+	if (s->err || !(s->mask & mask))
+		return;
+
+	s->err = func(s);
+	s->sm.mask |= mask;
+}
+
+static u64 mnt_to_attr_flags(struct vfsmount *mnt)
+{
+	unsigned int mnt_flags = READ_ONCE(mnt->mnt_flags);
+	u64 attr_flags = 0;
+
+	if (mnt_flags & MNT_READONLY)
+		attr_flags |= MOUNT_ATTR_RDONLY;
+	if (mnt_flags & MNT_NOSUID)
+		attr_flags |= MOUNT_ATTR_NOSUID;
+	if (mnt_flags & MNT_NODEV)
+		attr_flags |= MOUNT_ATTR_NODEV;
+	if (mnt_flags & MNT_NOEXEC)
+		attr_flags |= MOUNT_ATTR_NOEXEC;
+	if (mnt_flags & MNT_NODIRATIME)
+		attr_flags |= MOUNT_ATTR_NODIRATIME;
+	if (mnt_flags & MNT_NOSYMFOLLOW)
+		attr_flags |= MOUNT_ATTR_NOSYMFOLLOW;
+
+	if (mnt_flags & MNT_NOATIME)
+		attr_flags |= MOUNT_ATTR_NOATIME;
+	else if (mnt_flags & MNT_RELATIME)
+		attr_flags |= MOUNT_ATTR_RELATIME;
+	else
+		attr_flags |= MOUNT_ATTR_STRICTATIME;
+
+	if (is_idmapped_mnt(mnt))
+		attr_flags |= MOUNT_ATTR_IDMAP;
+
+	return attr_flags;
+}
+
+static u64 mnt_to_propagation_flags(struct mount *m)
+{
+	u64 propagation = 0;
+
+	if (IS_MNT_SHARED(m))
+		propagation |= MS_SHARED;
+	if (IS_MNT_SLAVE(m))
+		propagation |= MS_SLAVE;
+	if (IS_MNT_UNBINDABLE(m))
+		propagation |= MS_UNBINDABLE;
+	if (!propagation)
+		propagation |= MS_PRIVATE;
+
+	return propagation;
+}
+
+static int statmount_sb_basic(struct kstatmount *s)
+{
+	struct super_block *sb = s->mnt->mnt_sb;
+
+	s->sm.sb_dev_major = MAJOR(sb->s_dev);
+	s->sm.sb_dev_minor = MINOR(sb->s_dev);
+	s->sm.sb_magic = sb->s_magic;
+	s->sm.sb_flags = sb->s_flags & (SB_RDONLY|SB_SYNCHRONOUS|SB_DIRSYNC|SB_LAZYTIME);
+
+	return 0;
+}
+
+static int statmount_mnt_basic(struct kstatmount *s)
+{
+	struct mount *m = real_mount(s->mnt);
+
+	s->sm.mnt_id = m->mnt_id_unique;
+	s->sm.mnt_parent_id = m->mnt_parent->mnt_id_unique;
+	s->sm.mnt_id_old = m->mnt_id;
+	s->sm.mnt_parent_id_old = m->mnt_parent->mnt_id;
+	s->sm.mnt_attr = mnt_to_attr_flags(&m->mnt);
+	s->sm.mnt_propagation = mnt_to_propagation_flags(m);
+	s->sm.mnt_peer_group = IS_MNT_SHARED(m) ? m->mnt_group_id : 0;
+	s->sm.mnt_master = IS_MNT_SLAVE(m) ? m->mnt_master->mnt_group_id : 0;
+
+	return 0;
+}
+
+static int statmount_propagate_from(struct kstatmount *s)
+{
+	struct mount *m = real_mount(s->mnt);
+
+	if (!IS_MNT_SLAVE(m))
+		return 0;
+
+	s->sm.propagate_from = get_dominating_id(m, &current->fs->root);
+
+	return 0;
+}
+
+static int statmount_mnt_root(struct kstatmount *s)
+{
+	struct seq_file *seq = &s->seq;
+	int err = show_path(seq, s->mnt->mnt_root);
+
+	if (!err && !seq_has_overflowed(seq)) {
+		seq->buf[seq->count] = '\0';
+		seq->count = string_unescape_inplace(seq->buf, UNESCAPE_OCTAL);
+	}
+	return err;
+}
+
+static int statmount_mnt_point(struct kstatmount *s)
+{
+	struct vfsmount *mnt = s->mnt;
+	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
+	int err = seq_path_root(&s->seq, &mnt_path, &s->root, "");
+
+	return err == SEQ_SKIP ? 0 : err;
+}
+
+static int statmount_fs_type(struct kstatmount *s)
+{
+	struct seq_file *seq = &s->seq;
+	struct super_block *sb = s->mnt->mnt_sb;
+
+	seq_puts(seq, sb->s_type->name);
+	return 0;
+}
+
+static int do_statmount(struct kstatmount *s)
+{
+	struct statmount *sm = &s->sm;
+	struct mount *m = real_mount(s->mnt);
+	size_t copysize = min_t(size_t, s->bufsize, sizeof(*sm));
+	int err;
+
+	/*
+	 * Don't trigger audit denials. We just want to determine what
+	 * mounts to show users.
+	 */
+	if (!is_path_reachable(m, m->mnt.mnt_root, &s->root) &&
+	    !ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN))
+		return -EPERM;
+
+	err = security_sb_statfs(s->mnt->mnt_root);
+	if (err)
+		return err;
+
+	statmount_numeric(s, STATMOUNT_SB_BASIC, statmount_sb_basic);
+	statmount_numeric(s, STATMOUNT_MNT_BASIC, statmount_mnt_basic);
+	statmount_numeric(s, STATMOUNT_PROPAGATE_FROM, statmount_propagate_from);
+	statmount_string(s, STATMOUNT_FS_TYPE, statmount_fs_type, &sm->fs_type);
+	statmount_string(s, STATMOUNT_MNT_ROOT, statmount_mnt_root, &sm->mnt_root);
+	statmount_string(s, STATMOUNT_MNT_POINT, statmount_mnt_point, &sm->mnt_point);
+
+	if (s->err)
+		return s->err;
+
+	/* Return the number of bytes copied to the buffer */
+	sm->size = copysize + s->pos;
+
+	if (copy_to_user(s->buf, sm, copysize))
+		return -EFAULT;
+
+	return 0;
+}
+
+SYSCALL_DEFINE4(statmount, const struct mnt_id_req __user *, req,
+		struct statmount __user *, buf, size_t, bufsize,
+		unsigned int, flags)
+{
+	struct vfsmount *mnt;
+	struct mnt_id_req kreq;
+	int ret;
+
+	if (flags)
+		return -EINVAL;
+
+	if (copy_from_user(&kreq, req, sizeof(kreq)))
+		return -EFAULT;
+
+	down_read(&namespace_sem);
+	mnt = lookup_mnt_in_ns(kreq.mnt_id, current->nsproxy->mnt_ns);
+	ret = -ENOENT;
+	if (mnt) {
+		struct kstatmount s = {
+			.mask = kreq.request_mask,
+			.buf = buf,
+			.bufsize = bufsize,
+			.mnt = mnt,
+		};
+
+		get_fs_root(current->fs, &s.root);
+		ret = do_statmount(&s);
+		path_put(&s.root);
+	}
+	up_read(&namespace_sem);
+
+	return ret;
+}
+
 static void __init init_mount_tree(void)
 {
 	struct vfsmount *mnt;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index fd9d12de7e92..530ca9adf5f1 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -74,6 +74,8 @@ struct landlock_ruleset_attr;
 enum landlock_rule_type;
 struct cachestat_range;
 struct cachestat;
+struct statmount;
+struct mnt_id_req;
 
 #include <linux/types.h>
 #include <linux/aio_abi.h>
@@ -407,6 +409,9 @@ asmlinkage long sys_statfs64(const char __user *path, size_t sz,
 asmlinkage long sys_fstatfs(unsigned int fd, struct statfs __user *buf);
 asmlinkage long sys_fstatfs64(unsigned int fd, size_t sz,
 				struct statfs64 __user *buf);
+asmlinkage long sys_statmount(const struct mnt_id_req __user *req,
+			      struct statmount __user *buf, size_t bufsize,
+			      unsigned int flags);
 asmlinkage long sys_truncate(const char __user *path, long length);
 asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length);
 #if BITS_PER_LONG == 32
diff --git a/include/uapi/linux/mount.h b/include/uapi/linux/mount.h
index bb242fdcfe6b..afdf4f2f6672 100644
--- a/include/uapi/linux/mount.h
+++ b/include/uapi/linux/mount.h
@@ -138,4 +138,57 @@ struct mount_attr {
 /* List of all mount_attr versions. */
 #define MOUNT_ATTR_SIZE_VER0	32 /* sizeof first published struct */
 
+
+/*
+ * Structure for getting mount/superblock/filesystem info with statmount(2).
+ *
+ * The interface is similar to statx(2): individual fields or groups can be
+ * selected with the @mask argument of statmount().  Kernel will set the @mask
+ * field according to the supported fields.
+ *
+ * If string fields are selected, then the caller needs to pass a buffer that
+ * has space after the fixed part of the structure.  Nul terminated strings are
+ * copied there and offsets relative to @str are stored in the relevant fields.
+ * If the buffer is too small, then EOVERFLOW is returned.  The actually used
+ * size is returned in @size.
+ */
+struct statmount {
+	__u32 size;		/* Total size, including strings */
+	__u32 __spare1;
+	__u64 mask;		/* What results were written */
+	__u32 sb_dev_major;	/* Device ID */
+	__u32 sb_dev_minor;
+	__u64 sb_magic;		/* ..._SUPER_MAGIC */
+	__u32 sb_flags;		/* SB_{RDONLY,SYNCHRONOUS,DIRSYNC,LAZYTIME} */
+	__u32 fs_type;		/* [str] Filesystem type */
+	__u64 mnt_id;		/* Unique ID of mount */
+	__u64 mnt_parent_id;	/* Unique ID of parent (for root == mnt_id) */
+	__u32 mnt_id_old;	/* Reused IDs used in proc/.../mountinfo */
+	__u32 mnt_parent_id_old;
+	__u64 mnt_attr;		/* MOUNT_ATTR_... */
+	__u64 mnt_propagation;	/* MS_{SHARED,SLAVE,PRIVATE,UNBINDABLE} */
+	__u64 mnt_peer_group;	/* ID of shared peer group */
+	__u64 mnt_master;	/* Mount receives propagation from this ID */
+	__u64 propagate_from;	/* Propagation from in current namespace */
+	__u32 mnt_root;		/* [str] Root of mount relative to root of fs */
+	__u32 mnt_point;	/* [str] Mountpoint relative to current root */
+	__u64 __spare2[50];
+	char str[];		/* Variable size part containing strings */
+};
+
+struct mnt_id_req {
+	__u64 mnt_id;
+	__u64 request_mask;
+};
+
+/*
+ * @mask bits for statmount(2)
+ */
+#define STATMOUNT_SB_BASIC		0x00000001U     /* Want/got sb_... */
+#define STATMOUNT_MNT_BASIC		0x00000002U	/* Want/got mnt_... */
+#define STATMOUNT_PROPAGATE_FROM	0x00000004U	/* Want/got propagate_from */
+#define STATMOUNT_MNT_ROOT		0x00000008U	/* Want/got mnt_root  */
+#define STATMOUNT_MNT_POINT		0x00000010U	/* Want/got mnt_point */
+#define STATMOUNT_FS_TYPE		0x00000020U	/* Want/got fs_type */
+
 #endif /* _UAPI_LINUX_MOUNT_H */
-- 
cgit v1.2.3


From b4c2bea8ceaa50cd42a8f73667389d801a3ecf2d Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@redhat.com>
Date: Wed, 25 Oct 2023 16:02:03 +0200
Subject: add listmount(2) syscall

Add way to query the children of a particular mount.  This is a more
flexible way to iterate the mount tree than having to parse
/proc/self/mountinfo.

Lookup the mount by the new 64bit mount ID. If a mount needs to be
queried based on path, then statx(2) can be used to first query the
mount ID belonging to the path.

Return an array of new (64bit) mount ID's. Without privileges only
mounts are listed which are reachable from the task's root.

Folded into this patch are several later improvements. Keeping them
separate would make the history pointlessly confusing:

* Recursive listing of mounts is the default now (cf. [1]).
* Remove explicit LISTMOUNT_UNREACHABLE flag (cf. [1]) and fail if mount
  is unreachable from current root. This also makes permission checking
  consistent with statmount() (cf. [3]).
* Start listing mounts in unique mount ID order (cf. [2]) to allow
  continuing listmount() from a midpoint.
* Allow to continue listmount(). The @request_mask parameter is renamed
  and to @param to be usable by both statmount() and listmount().
  If @param is set to a mount id then listmount() will continue listing
  mounts from that id on. This allows listing mounts in multiple
  listmount invocations without having to resize the buffer. If @param
  is zero then the listing starts from the beginning (cf. [4]).
* Don't return EOVERFLOW, instead return the buffer size which allows to
  detect a full buffer as well (cf. [4]).

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Link: https://lore.kernel.org/r/20231025140205.3586473-6-mszeredi@redhat.com
Reviewed-by: Ian Kent <raven@themaw.net>
Link: https://lore.kernel.org/r/20231128160337.29094-2-mszeredi@redhat.com [1] (folded)
Link: https://lore.kernel.org/r/20231128160337.29094-3-mszeredi@redhat.com [2] (folded)
Link: https://lore.kernel.org/r/20231128160337.29094-4-mszeredi@redhat.com [3] (folded)
Link: https://lore.kernel.org/r/20231128160337.29094-5-mszeredi@redhat.com [4] (folded)
[Christian Brauner <brauner@kernel.org>: various smaller fixes]
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/namespace.c             | 86 ++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/syscalls.h   |  3 ++
 include/uapi/linux/mount.h | 14 +++++++-
 3 files changed, 100 insertions(+), 3 deletions(-)

(limited to 'include/uapi')

diff --git a/fs/namespace.c b/fs/namespace.c
index 7f1618ed2aba..873185b8a84b 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -32,6 +32,7 @@
 #include <linux/fs_context.h>
 #include <linux/shmem_fs.h>
 #include <linux/mnt_idmapping.h>
+#include <linux/nospec.h>
 
 #include "pnode.h"
 #include "internal.h"
@@ -1009,7 +1010,7 @@ void mnt_change_mountpoint(struct mount *parent, struct mountpoint *mp, struct m
 
 static inline struct mount *node_to_mount(struct rb_node *node)
 {
-	return rb_entry(node, struct mount, mnt_node);
+	return node ? rb_entry(node, struct mount, mnt_node) : NULL;
 }
 
 static void mnt_add_to_ns(struct mnt_namespace *ns, struct mount *mnt)
@@ -4945,7 +4946,7 @@ static int prepare_kstatmount(struct kstatmount *ks, struct mnt_id_req *kreq,
 		return -EFAULT;
 
 	memset(ks, 0, sizeof(*ks));
-	ks->mask = kreq->request_mask;
+	ks->mask = kreq->param;
 	ks->buf = buf;
 	ks->bufsize = bufsize;
 	ks->seq.size = seq_size;
@@ -4999,6 +5000,87 @@ retry:
 	return ret;
 }
 
+static struct mount *listmnt_next(struct mount *curr)
+{
+	return node_to_mount(rb_next(&curr->mnt_node));
+}
+
+static ssize_t do_listmount(struct mount *first, struct path *orig, u64 mnt_id,
+			    u64 __user *buf, size_t bufsize,
+			    const struct path *root)
+{
+	struct mount *r;
+	ssize_t ctr;
+	int err;
+
+	/*
+	 * Don't trigger audit denials. We just want to determine what
+	 * mounts to show users.
+	 */
+	if (!is_path_reachable(real_mount(orig->mnt), orig->dentry, root) &&
+	    !ns_capable_noaudit(&init_user_ns, CAP_SYS_ADMIN))
+		return -EPERM;
+
+	err = security_sb_statfs(orig->dentry);
+	if (err)
+		return err;
+
+	for (ctr = 0, r = first; r && ctr < bufsize; r = listmnt_next(r)) {
+		if (r->mnt_id_unique == mnt_id)
+			continue;
+		if (!is_path_reachable(r, r->mnt.mnt_root, orig))
+			continue;
+		ctr = array_index_nospec(ctr, bufsize);
+		if (put_user(r->mnt_id_unique, buf + ctr))
+			return -EFAULT;
+		if (check_add_overflow(ctr, 1, &ctr))
+			return -ERANGE;
+	}
+	return ctr;
+}
+
+SYSCALL_DEFINE4(listmount, const struct mnt_id_req __user *, req,
+		u64 __user *, buf, size_t, bufsize, unsigned int, flags)
+{
+	struct mnt_namespace *ns = current->nsproxy->mnt_ns;
+	struct mnt_id_req kreq;
+	struct mount *first;
+	struct path root, orig;
+	u64 mnt_id, last_mnt_id;
+	ssize_t ret;
+
+	if (flags)
+		return -EINVAL;
+
+	if (copy_from_user(&kreq, req, sizeof(kreq)))
+		return -EFAULT;
+	mnt_id = kreq.mnt_id;
+	last_mnt_id = kreq.param;
+
+	down_read(&namespace_sem);
+	get_fs_root(current->fs, &root);
+	if (mnt_id == LSMT_ROOT) {
+		orig = root;
+	} else {
+		ret = -ENOENT;
+		orig.mnt  = lookup_mnt_in_ns(mnt_id, ns);
+		if (!orig.mnt)
+			goto err;
+		orig.dentry = orig.mnt->mnt_root;
+	}
+	if (!last_mnt_id)
+		first = node_to_mount(rb_first(&ns->mounts));
+	else
+		first = mnt_find_id_at(ns, last_mnt_id + 1);
+
+	ret = do_listmount(first, &orig, mnt_id, buf, bufsize, &root);
+err:
+	path_put(&root);
+	up_read(&namespace_sem);
+	return ret;
+}
+
+
 static void __init init_mount_tree(void)
 {
 	struct vfsmount *mnt;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 530ca9adf5f1..2d6d3e76e3f7 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -412,6 +412,9 @@ asmlinkage long sys_fstatfs64(unsigned int fd, size_t sz,
 asmlinkage long sys_statmount(const struct mnt_id_req __user *req,
 			      struct statmount __user *buf, size_t bufsize,
 			      unsigned int flags);
+asmlinkage long sys_listmount(const struct mnt_id_req __user *req,
+			      u64 __user *buf, size_t bufsize,
+			      unsigned int flags);
 asmlinkage long sys_truncate(const char __user *path, long length);
 asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length);
 #if BITS_PER_LONG == 32
diff --git a/include/uapi/linux/mount.h b/include/uapi/linux/mount.h
index afdf4f2f6672..dc9a0112d819 100644
--- a/include/uapi/linux/mount.h
+++ b/include/uapi/linux/mount.h
@@ -176,9 +176,16 @@ struct statmount {
 	char str[];		/* Variable size part containing strings */
 };
 
+/*
+ * Structure for passing mount ID and miscellaneous parameters to statmount(2)
+ * and listmount(2).
+ *
+ * For statmount(2) @param represents the request mask.
+ * For listmount(2) @param represents the last listed mount id (or zero).
+ */
 struct mnt_id_req {
 	__u64 mnt_id;
-	__u64 request_mask;
+	__u64 param;
 };
 
 /*
@@ -191,4 +198,9 @@ struct mnt_id_req {
 #define STATMOUNT_MNT_POINT		0x00000010U	/* Want/got mnt_point */
 #define STATMOUNT_FS_TYPE		0x00000020U	/* Want/got fs_type */
 
+/*
+ * Special @mnt_id values that can be passed to listmount
+ */
+#define LSMT_ROOT		0xffffffffffffffff	/* root mount */
+
 #endif /* _UAPI_LINUX_MOUNT_H */
-- 
cgit v1.2.3


From d8b0f5465012538cc4bb10ddc4affadbab73465b Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@redhat.com>
Date: Wed, 25 Oct 2023 16:02:04 +0200
Subject: wire up syscalls for statmount/listmount

Wire up all archs.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Link: https://lore.kernel.org/r/20231025140205.3586473-7-mszeredi@redhat.com
Reviewed-by: Ian Kent <raven@themaw.net>
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 arch/alpha/kernel/syscalls/syscall.tbl      | 2 ++
 arch/arm/tools/syscall.tbl                  | 2 ++
 arch/arm64/include/asm/unistd32.h           | 4 ++++
 arch/m68k/kernel/syscalls/syscall.tbl       | 2 ++
 arch/microblaze/kernel/syscalls/syscall.tbl | 2 ++
 arch/mips/kernel/syscalls/syscall_n32.tbl   | 2 ++
 arch/mips/kernel/syscalls/syscall_n64.tbl   | 2 ++
 arch/mips/kernel/syscalls/syscall_o32.tbl   | 2 ++
 arch/parisc/kernel/syscalls/syscall.tbl     | 2 ++
 arch/powerpc/kernel/syscalls/syscall.tbl    | 2 ++
 arch/s390/kernel/syscalls/syscall.tbl       | 2 ++
 arch/sh/kernel/syscalls/syscall.tbl         | 2 ++
 arch/sparc/kernel/syscalls/syscall.tbl      | 2 ++
 arch/x86/entry/syscalls/syscall_32.tbl      | 2 ++
 arch/x86/entry/syscalls/syscall_64.tbl      | 2 ++
 arch/xtensa/kernel/syscalls/syscall.tbl     | 2 ++
 include/uapi/asm-generic/unistd.h           | 8 +++++++-
 17 files changed, 41 insertions(+), 1 deletion(-)

(limited to 'include/uapi')

diff --git a/arch/alpha/kernel/syscalls/syscall.tbl b/arch/alpha/kernel/syscalls/syscall.tbl
index 18c842ca6c32..186e785f5b56 100644
--- a/arch/alpha/kernel/syscalls/syscall.tbl
+++ b/arch/alpha/kernel/syscalls/syscall.tbl
@@ -496,3 +496,5 @@
 564	common	futex_wake			sys_futex_wake
 565	common	futex_wait			sys_futex_wait
 566	common	futex_requeue			sys_futex_requeue
+567	common	statmount			sys_statmount
+568	common	listmount			sys_listmount
diff --git a/arch/arm/tools/syscall.tbl b/arch/arm/tools/syscall.tbl
index 584f9528c996..d6a324dbff2e 100644
--- a/arch/arm/tools/syscall.tbl
+++ b/arch/arm/tools/syscall.tbl
@@ -470,3 +470,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/arm64/include/asm/unistd32.h b/arch/arm64/include/asm/unistd32.h
index 9f7c1bf99526..8a191423c316 100644
--- a/arch/arm64/include/asm/unistd32.h
+++ b/arch/arm64/include/asm/unistd32.h
@@ -919,6 +919,10 @@ __SYSCALL(__NR_futex_wake, sys_futex_wake)
 __SYSCALL(__NR_futex_wait, sys_futex_wait)
 #define __NR_futex_requeue 456
 __SYSCALL(__NR_futex_requeue, sys_futex_requeue)
+#define __NR_statmount 457
+__SYSCALL(__NR_statmount, sys_statmount)
+#define __NR_listmount 458
+__SYSCALL(__NR_listmount, sys_listmount)
 
 /*
  * Please add new compat syscalls above this comment and update
diff --git a/arch/m68k/kernel/syscalls/syscall.tbl b/arch/m68k/kernel/syscalls/syscall.tbl
index 7a4b780e82cb..37db1a810b67 100644
--- a/arch/m68k/kernel/syscalls/syscall.tbl
+++ b/arch/m68k/kernel/syscalls/syscall.tbl
@@ -456,3 +456,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/microblaze/kernel/syscalls/syscall.tbl b/arch/microblaze/kernel/syscalls/syscall.tbl
index 5b6a0b02b7de..07fff5ad1c9c 100644
--- a/arch/microblaze/kernel/syscalls/syscall.tbl
+++ b/arch/microblaze/kernel/syscalls/syscall.tbl
@@ -462,3 +462,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/mips/kernel/syscalls/syscall_n32.tbl b/arch/mips/kernel/syscalls/syscall_n32.tbl
index a842b41c8e06..134ea054b1c7 100644
--- a/arch/mips/kernel/syscalls/syscall_n32.tbl
+++ b/arch/mips/kernel/syscalls/syscall_n32.tbl
@@ -395,3 +395,5 @@
 454	n32	futex_wake			sys_futex_wake
 455	n32	futex_wait			sys_futex_wait
 456	n32	futex_requeue			sys_futex_requeue
+457	n32	statmount			sys_statmount
+458	n32	listmount			sys_listmount
diff --git a/arch/mips/kernel/syscalls/syscall_n64.tbl b/arch/mips/kernel/syscalls/syscall_n64.tbl
index 116ff501bf92..959a21664703 100644
--- a/arch/mips/kernel/syscalls/syscall_n64.tbl
+++ b/arch/mips/kernel/syscalls/syscall_n64.tbl
@@ -371,3 +371,5 @@
 454	n64	futex_wake			sys_futex_wake
 455	n64	futex_wait			sys_futex_wait
 456	n64	futex_requeue			sys_futex_requeue
+457	n64	statmount			sys_statmount
+458	n64	listmount			sys_listmount
diff --git a/arch/mips/kernel/syscalls/syscall_o32.tbl b/arch/mips/kernel/syscalls/syscall_o32.tbl
index 525cc54bc63b..e55bc1d4bf0f 100644
--- a/arch/mips/kernel/syscalls/syscall_o32.tbl
+++ b/arch/mips/kernel/syscalls/syscall_o32.tbl
@@ -444,3 +444,5 @@
 454	o32	futex_wake			sys_futex_wake
 455	o32	futex_wait			sys_futex_wait
 456	o32	futex_requeue			sys_futex_requeue
+457	o32	statmount			sys_statmount
+458	o32	listmount			sys_listmount
diff --git a/arch/parisc/kernel/syscalls/syscall.tbl b/arch/parisc/kernel/syscalls/syscall.tbl
index a47798fed54e..9c84470c31c7 100644
--- a/arch/parisc/kernel/syscalls/syscall.tbl
+++ b/arch/parisc/kernel/syscalls/syscall.tbl
@@ -455,3 +455,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/powerpc/kernel/syscalls/syscall.tbl b/arch/powerpc/kernel/syscalls/syscall.tbl
index 7fab411378f2..6988ecbc316e 100644
--- a/arch/powerpc/kernel/syscalls/syscall.tbl
+++ b/arch/powerpc/kernel/syscalls/syscall.tbl
@@ -543,3 +543,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/s390/kernel/syscalls/syscall.tbl b/arch/s390/kernel/syscalls/syscall.tbl
index 86fec9b080f6..5f5cd20ebb34 100644
--- a/arch/s390/kernel/syscalls/syscall.tbl
+++ b/arch/s390/kernel/syscalls/syscall.tbl
@@ -459,3 +459,5 @@
 454  common	futex_wake		sys_futex_wake			sys_futex_wake
 455  common	futex_wait		sys_futex_wait			sys_futex_wait
 456  common	futex_requeue		sys_futex_requeue		sys_futex_requeue
+457  common	statmount		sys_statmount			sys_statmount
+458  common	listmount		sys_listmount			sys_listmount
diff --git a/arch/sh/kernel/syscalls/syscall.tbl b/arch/sh/kernel/syscalls/syscall.tbl
index 363fae0fe9bf..3103ebd2e4cb 100644
--- a/arch/sh/kernel/syscalls/syscall.tbl
+++ b/arch/sh/kernel/syscalls/syscall.tbl
@@ -459,3 +459,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/sparc/kernel/syscalls/syscall.tbl b/arch/sparc/kernel/syscalls/syscall.tbl
index 7bcaa3d5ea44..ba147d7ad19a 100644
--- a/arch/sparc/kernel/syscalls/syscall.tbl
+++ b/arch/sparc/kernel/syscalls/syscall.tbl
@@ -502,3 +502,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/arch/x86/entry/syscalls/syscall_32.tbl b/arch/x86/entry/syscalls/syscall_32.tbl
index c8fac5205803..56e6c2f3ee9c 100644
--- a/arch/x86/entry/syscalls/syscall_32.tbl
+++ b/arch/x86/entry/syscalls/syscall_32.tbl
@@ -461,3 +461,5 @@
 454	i386	futex_wake		sys_futex_wake
 455	i386	futex_wait		sys_futex_wait
 456	i386	futex_requeue		sys_futex_requeue
+457	i386	statmount		sys_statmount
+458	i386	listmount		sys_listmount
diff --git a/arch/x86/entry/syscalls/syscall_64.tbl b/arch/x86/entry/syscalls/syscall_64.tbl
index 8cb8bf68721c..3a22eef585c2 100644
--- a/arch/x86/entry/syscalls/syscall_64.tbl
+++ b/arch/x86/entry/syscalls/syscall_64.tbl
@@ -378,6 +378,8 @@
 454	common	futex_wake		sys_futex_wake
 455	common	futex_wait		sys_futex_wait
 456	common	futex_requeue		sys_futex_requeue
+457	common	statmount		sys_statmount
+458	common	listmount		sys_listmount
 
 #
 # Due to a historical design error, certain syscalls are numbered differently
diff --git a/arch/xtensa/kernel/syscalls/syscall.tbl b/arch/xtensa/kernel/syscalls/syscall.tbl
index 06eefa9c1458..497b5d32f457 100644
--- a/arch/xtensa/kernel/syscalls/syscall.tbl
+++ b/arch/xtensa/kernel/syscalls/syscall.tbl
@@ -427,3 +427,5 @@
 454	common	futex_wake			sys_futex_wake
 455	common	futex_wait			sys_futex_wait
 456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
diff --git a/include/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h
index 756b013fb832..b67b18e71fbd 100644
--- a/include/uapi/asm-generic/unistd.h
+++ b/include/uapi/asm-generic/unistd.h
@@ -829,8 +829,14 @@ __SYSCALL(__NR_futex_wait, sys_futex_wait)
 #define __NR_futex_requeue 456
 __SYSCALL(__NR_futex_requeue, sys_futex_requeue)
 
+#define __NR_statmount   457
+__SYSCALL(__NR_statmount, sys_statmount)
+
+#define __NR_listmount   458
+__SYSCALL(__NR_listmount, sys_listmount)
+
 #undef __NR_syscalls
-#define __NR_syscalls 457
+#define __NR_syscalls 459
 
 /*
  * 32 bit systems traditionally used different
-- 
cgit v1.2.3


From 35e27a5744131996061e6e323f1bcb4c827ae867 Mon Sep 17 00:00:00 2001
From: Christian Brauner <brauner@kernel.org>
Date: Wed, 29 Nov 2023 12:27:15 +0100
Subject: fs: keep struct mnt_id_req extensible

Make it extensible so that we have the liberty to reuse it in future
mount-id based apis. Treat zero size as the first published struct.

Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/namespace.c             | 34 ++++++++++++++++++++++++++++++----
 include/uapi/linux/mount.h |  5 +++++
 2 files changed, 35 insertions(+), 4 deletions(-)

(limited to 'include/uapi')

diff --git a/fs/namespace.c b/fs/namespace.c
index 873185b8a84b..918e8f89ce35 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -4956,6 +4956,30 @@ static int prepare_kstatmount(struct kstatmount *ks, struct mnt_id_req *kreq,
 	return 0;
 }
 
+static int copy_mnt_id_req(const struct mnt_id_req __user *req,
+			   struct mnt_id_req *kreq)
+{
+	int ret;
+	size_t usize;
+
+	BUILD_BUG_ON(sizeof(struct mnt_id_req) != MNT_ID_REQ_SIZE_VER0);
+
+	ret = get_user(usize, &req->size);
+	if (ret)
+		return -EFAULT;
+	if (unlikely(usize > PAGE_SIZE))
+		return -E2BIG;
+	if (unlikely(usize < MNT_ID_REQ_SIZE_VER0))
+		return -EINVAL;
+	memset(kreq, 0, sizeof(*kreq));
+	ret = copy_struct_from_user(kreq, sizeof(*kreq), req, usize);
+	if (ret)
+		return ret;
+	if (kreq->spare != 0)
+		return -EINVAL;
+	return 0;
+}
+
 SYSCALL_DEFINE4(statmount, const struct mnt_id_req __user *, req,
 		struct statmount __user *, buf, size_t, bufsize,
 		unsigned int, flags)
@@ -4970,8 +4994,9 @@ SYSCALL_DEFINE4(statmount, const struct mnt_id_req __user *, req,
 	if (flags)
 		return -EINVAL;
 
-	if (copy_from_user(&kreq, req, sizeof(kreq)))
-		return -EFAULT;
+	ret = copy_mnt_id_req(req, &kreq);
+	if (ret)
+		return ret;
 
 retry:
 	ret = prepare_kstatmount(&ks, &kreq, buf, bufsize, seq_size);
@@ -5052,8 +5077,9 @@ SYSCALL_DEFINE4(listmount, const struct mnt_id_req __user *, req,
 	if (flags)
 		return -EINVAL;
 
-	if (copy_from_user(&kreq, req, sizeof(kreq)))
-		return -EFAULT;
+	ret = copy_mnt_id_req(req, &kreq);
+	if (ret)
+		return ret;
 	mnt_id = kreq.mnt_id;
 	last_mnt_id = kreq.param;
 
diff --git a/include/uapi/linux/mount.h b/include/uapi/linux/mount.h
index dc9a0112d819..ad5478dbad00 100644
--- a/include/uapi/linux/mount.h
+++ b/include/uapi/linux/mount.h
@@ -184,10 +184,15 @@ struct statmount {
  * For listmount(2) @param represents the last listed mount id (or zero).
  */
 struct mnt_id_req {
+	__u32 size;
+	__u32 spare;
 	__u64 mnt_id;
 	__u64 param;
 };
 
+/* List of all mnt_id_req versions. */
+#define MNT_ID_REQ_SIZE_VER0	24 /* sizeof first published struct */
+
 /*
  * @mask bits for statmount(2)
  */
-- 
cgit v1.2.3