fs: support mapped mounts of mapped filesystems

In previous patches we added new and modified existing helpers to handle
idmapped mounts of filesystems mounted with an idmapping. In this final
patch we convert all relevant places in the vfs to actually pass the
filesystem's idmapping into these helpers.

With this the vfs is in shape to handle idmapped mounts of filesystems
mounted with an idmapping. Note that this is just the generic
infrastructure. Actually adding support for idmapped mounts to a
filesystem mountable with an idmapping is follow-up work.

In this patch we extend the definition of an idmapped mount from a mount
that that has the initial idmapping attached to it to a mount that has
an idmapping attached to it which is not the same as the idmapping the
filesystem was mounted with.

As before we do not allow the initial idmapping to be attached to a
mount. In addition this patch prevents that the idmapping the filesystem
was mounted with can be attached to a mount created based on this
filesystem.

This has multiple reasons and advantages. First, attaching the initial
idmapping or the filesystem's idmapping doesn't make much sense as in
both cases the values of the i_{g,u}id and other places where k{g,u}ids
are used do not change. Second, a user that really wants to do this for
whatever reason can just create a separate dedicated identical idmapping
to attach to the mount. Third, we can continue to use the initial
idmapping as an indicator that a mount is not idmapped allowing us to
continue to keep passing the initial idmapping into the mapping helpers
to tell them that something isn't an idmapped mount even if the
filesystem is mounted with an idmapping.

Link: https://lore.kernel.org/r/20211123114227.3124056-11-brauner@kernel.org (v1)
Link: https://lore.kernel.org/r/20211130121032.3753852-11-brauner@kernel.org (v2)
Link: https://lore.kernel.org/r/20211203111707.3901969-11-brauner@kernel.org
Cc: Seth Forshee <sforshee@digitalocean.com>
Cc: Amir Goldstein <amir73il@gmail.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
CC: linux-fsdevel@vger.kernel.org
Reviewed-by: Seth Forshee <sforshee@digitalocean.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>

3 files changed, 46 insertions, 20 deletions

diff --git a/fs/namespace.c b/fs/namespace.c
index 4994b816a74c..08266a35c0c1 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -31,6 +31,7 @@
 #include <uapi/linux/mount.h>
 #include <linux/fs_context.h>
 #include <linux/shmem_fs.h>
+#include <linux/mnt_idmapping.h>
 
 #include "pnode.h"
 #include "internal.h"
@@ -561,7 +562,7 @@ static void free_vfsmnt(struct mount *mnt)
 	struct user_namespace *mnt_userns;
 
 	mnt_userns = mnt_user_ns(&mnt->mnt);
-	if (mnt_userns != &init_user_ns)
+	if (!initial_idmapping(mnt_userns))
 		put_user_ns(mnt_userns);
 	kfree_const(mnt->mnt_devname);
 #ifdef CONFIG_SMP
@@ -965,6 +966,7 @@ static struct mount *skip_mnt_tree(struct mount *p)
 struct vfsmount *vfs_create_mount(struct fs_context *fc)
 {
 	struct mount *mnt;
+	struct user_namespace *fs_userns;
 
 	if (!fc->root)
 		return ERR_PTR(-EINVAL);
@@ -982,6 +984,10 @@ struct vfsmount *vfs_create_mount(struct fs_context *fc)
 	mnt->mnt_mountpoint	= mnt->mnt.mnt_root;
 	mnt->mnt_parent		= mnt;
 
+	fs_userns = mnt->mnt.mnt_sb->s_user_ns;
+	if (!initial_idmapping(fs_userns))
+		mnt->mnt.mnt_userns = get_user_ns(fs_userns);
+
 	lock_mount_hash();
 	list_add_tail(&mnt->mnt_instance, &mnt->mnt.mnt_sb->s_mounts);
 	unlock_mount_hash();
@@ -1072,7 +1078,7 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 
 	atomic_inc(&sb->s_active);
 	mnt->mnt.mnt_userns = mnt_user_ns(&old->mnt);
-	if (mnt->mnt.mnt_userns != &init_user_ns)
+	if (!initial_idmapping(mnt->mnt.mnt_userns))
 		mnt->mnt.mnt_userns = get_user_ns(mnt->mnt.mnt_userns);
 	mnt->mnt.mnt_sb = sb;
 	mnt->mnt.mnt_root = dget(root);
@@ -3927,11 +3933,19 @@ static unsigned int recalc_flags(struct mount_kattr *kattr, struct mount *mnt)
 static int can_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt)
 {
 	struct vfsmount *m = &mnt->mnt;
+	struct user_namespace *fs_userns = m->mnt_sb->s_user_ns;
 
 	if (!kattr->mnt_userns)
 		return 0;
 
 	/*
+	 * Creating an idmapped mount with the filesystem wide idmapping
+	 * doesn't make sense so block that. We don't allow mushy semantics.
+	 */
+	if (kattr->mnt_userns == fs_userns)
+		return -EINVAL;
+
+	/*
 	 * Once a mount has been idmapped we don't allow it to change its
 	 * mapping. It makes things simpler and callers can just create
 	 * another bind-mount they can idmap if they want to.
@@ -3943,12 +3957,8 @@ static int can_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt)
 	if (!(m->mnt_sb->s_type->fs_flags & FS_ALLOW_IDMAP))
 		return -EINVAL;
 
-	/* Don't yet support filesystem mountable in user namespaces. */
-	if (m->mnt_sb->s_user_ns != &init_user_ns)
-		return -EINVAL;
-
 	/* We're not controlling the superblock. */
-	if (!capable(CAP_SYS_ADMIN))
+	if (!ns_capable(fs_userns, CAP_SYS_ADMIN))
 		return -EPERM;
 
 	/* Mount has already been visible in the filesystem hierarchy. */
@@ -4002,14 +4012,27 @@ out:
 
 static void do_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt)
 {
-	struct user_namespace *mnt_userns;
+	struct user_namespace *mnt_userns, *old_mnt_userns;
 
 	if (!kattr->mnt_userns)
 		return;
 
+	/*
+	 * We're the only ones able to change the mount's idmapping. So
+	 * mnt->mnt.mnt_userns is stable and we can retrieve it directly.
+	 */
+	old_mnt_userns = mnt->mnt.mnt_userns;
+
 	mnt_userns = get_user_ns(kattr->mnt_userns);
 	/* Pairs with smp_load_acquire() in mnt_user_ns(). */
 	smp_store_release(&mnt->mnt.mnt_userns, mnt_userns);
+
+	/*
+	 * If this is an idmapped filesystem drop the reference we've taken
+	 * in vfs_create_mount() before.
+	 */
+	if (!initial_idmapping(old_mnt_userns))
+		put_user_ns(old_mnt_userns);
 }
 
 static void mount_setattr_commit(struct mount_kattr *kattr,
@@ -4133,13 +4156,15 @@ static int build_mount_idmapped(const struct mount_attr *attr, size_t usize,
 	}
 
 	/*
-	 * The init_user_ns is used to indicate that a vfsmount is not idmapped.
-	 * This is simpler than just having to treat NULL as unmapped. Users
-	 * wanting to idmap a mount to init_user_ns can just use a namespace
-	 * with an identity mapping.
+	 * The initial idmapping cannot be used to create an idmapped
+	 * mount. We use the initial idmapping as an indicator of a mount
+	 * that is not idmapped. It can simply be passed into helpers that
+	 * are aware of idmapped mounts as a convenient shortcut. A user
+	 * can just create a dedicated identity mapping to achieve the same
+	 * result.
 	 */
 	mnt_userns = container_of(ns, struct user_namespace, ns);
-	if (mnt_userns == &init_user_ns) {
+	if (initial_idmapping(mnt_userns)) {
 		err = -EPERM;
 		goto out_fput;
 	}
diff --git a/fs/open.c b/fs/open.c
index 40a00e71865b..9ff2f621b760 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -641,7 +641,7 @@ SYSCALL_DEFINE2(chmod, const char __user *, filename, umode_t, mode)
 
 int chown_common(const struct path *path, uid_t user, gid_t group)
 {
-	struct user_namespace *mnt_userns;
+	struct user_namespace *mnt_userns, *fs_userns;
 	struct inode *inode = path->dentry->d_inode;
 	struct inode *delegated_inode = NULL;
 	int error;
@@ -653,8 +653,9 @@ int chown_common(const struct path *path, uid_t user, gid_t group)
 	gid = make_kgid(current_user_ns(), group);
 
 	mnt_userns = mnt_user_ns(path->mnt);
-	uid = mapped_kuid_user(mnt_userns, &init_user_ns, uid);
-	gid = mapped_kgid_user(mnt_userns, &init_user_ns, gid);
+	fs_userns = i_user_ns(inode);
+	uid = mapped_kuid_user(mnt_userns, fs_userns, uid);
+	gid = mapped_kgid_user(mnt_userns, fs_userns, gid);
 
 retry_deleg:
 	newattrs.ia_valid =  ATTR_CTIME;
diff --git a/fs/posix_acl.c b/fs/posix_acl.c
index 4b5fb9a9b90f..80acb6885cf9 100644
--- a/fs/posix_acl.c
+++ b/fs/posix_acl.c
@@ -376,8 +376,8 @@ posix_acl_permission(struct user_namespace *mnt_userns, struct inode *inode,
                                 break;
                         case ACL_USER:
 				uid = mapped_kuid_fs(mnt_userns,
-						      &init_user_ns,
-						      pa->e_uid);
+						     i_user_ns(inode),
+						     pa->e_uid);
 				if (uid_eq(uid, current_fsuid()))
                                         goto mask;
 				break;
@@ -391,8 +391,8 @@ posix_acl_permission(struct user_namespace *mnt_userns, struct inode *inode,
 				break;
                         case ACL_GROUP:
 				gid = mapped_kgid_fs(mnt_userns,
-						      &init_user_ns,
-						      pa->e_gid);
+						     i_user_ns(inode),
+						     pa->e_gid);
 				if (in_group_p(gid)) {
 					found = 1;
 					if ((pa->e_perm & want) == want)