tmpfs: Overhaul option handling on remount

On remount, we must accept all the same options as on initial mount.
For parameters which we're unable to modify on the fly, fail only if
the new value is different from the existing one.

PR:		295096
MFC after:	1 week
Reviewed by:	kib
Differential Revision:	https://reviews.freebsd.org/D57044

(cherry picked from commit 1cfc208)

Author: dag-erling

sys/fs/tmpfs/tmpfs_vfsops.c

@@ -93,10 +93,6 @@ static const char *tmpfs_opts[] = {
 	"export", "union", "nonc", "nomtime", "nosymfollow", "pgread", NULL
 };
 
-static const char *tmpfs_updateopts[] = {
-	"from", "easize", "export", "nomtime", "size", "nosymfollow", NULL
-};
-
 static int
 tmpfs_update_mtime_lazy_filter(struct vnode *vp, void *arg)
 {
@@ -325,6 +321,7 @@ tmpfs_mount(struct mount *mp)
 	    sizeof(struct tmpfs_dirent) + sizeof(struct tmpfs_node));
 	struct tmpfs_mount *tmp;
 	struct tmpfs_node *root;
+	struct vfsoptlist *opts;
 	int error;
 	bool nomtime, nonc, pgread;
 	/* Size counters. */
@@ -338,44 +335,71 @@ tmpfs_mount(struct mount *mp)
 
 	struct vattr va;
 
-	if (vfs_filteropt(mp->mnt_optnew, tmpfs_opts))
+	opts = mp->mnt_optnew;
+	if (vfs_filteropt(opts, tmpfs_opts))
 		return (EINVAL);
 
 	if (mp->mnt_flag & MNT_UPDATE) {
-		/* Only support update mounts for certain options. */
-		if (vfs_filteropt(mp->mnt_optnew, tmpfs_updateopts) != 0)
-			return (EOPNOTSUPP);
 		tmp = VFS_TO_TMPFS(mp);
-		if (vfs_getopt_size(mp->mnt_optnew, "size", &size_max) == 0) {
-			/*
-			 * On-the-fly resizing is not supported (yet). We still
-			 * need to have "size" listed as "supported", otherwise
-			 * trying to update fs that is listed in fstab with size
-			 * parameter, say trying to change rw to ro or vice
-			 * versa, would cause vfs_filteropt() to bail.
-			 */
-			if (size_max != tmp->tm_size_max)
-				return (EOPNOTSUPP);
-		}
-		if (vfs_getopt_size(mp->mnt_optnew, "easize", &ea_max_size) == 0) {
+
+		/*
+		 * These options cannot (yet) be modified on the fly, but
+		 * mount(8) will still pass them when remounting, so we
+		 * will silently ignore them as long as the value is
+		 * unchanged.
+		 */
+		if (vfs_scanopt(opts, "gid", "%d", &root_gid) == 1 &&
+		    root_gid != tmp->tm_root->tn_gid)
+			return (EOPNOTSUPP);
+		if (vfs_scanopt(opts, "uid", "%d", &root_uid) == 1 &&
+		    root_uid != tmp->tm_root->tn_uid)
+			return (EOPNOTSUPP);
+		if (vfs_scanopt(opts, "mode", "%ho", &root_mode) == 1 &&
+		    (root_mode & S_IFMT) != (tmp->tm_root->tn_mode & S_IFMT))
+			return (EOPNOTSUPP);
+		if (vfs_getopt_size(opts, "inodes", &nodes_max) == 0 &&
+		    nodes_max != 0 && nodes_max != tmp->tm_nodes_max)
+			return (EOPNOTSUPP);
+		if (vfs_getopt_size(opts, "size", &size_max) == 0 &&
+		    size_max != 0 && size_max != tmp->tm_size_max)
+			return (EOPNOTSUPP);
+		if (vfs_getopt_size(opts, "maxfilesize", &maxfilesize) == 0 &&
+		    maxfilesize != 0 && maxfilesize != tmp->tm_maxfilesize)
+			return (EOPNOTSUPP);
+		if (tmp->tm_nonc !=
+		    (vfs_getopt(opts, "nonc", NULL, NULL) == 0))
+			return (EOPNOTSUPP);
+		if (tmp->tm_pgread !=
+		    (vfs_getopt(opts, "pgread", NULL, NULL) == 0))
+			return (EOPNOTSUPP);
+
+		/*
+		 * These options can be modified.
+		 */
+		if (vfs_getopt_size(opts, "easize", &ea_max_size) != 0)
 			tmp->tm_ea_memory_max = ea_max_size;
-		}
-		if (vfs_flagopt(mp->mnt_optnew, "ro", NULL, 0) &&
-		    !tmp->tm_ronly) {
+		tmp->tm_nomtime = (vfs_getopt(opts, "nomtime", NULL, 0) == 0);
+
+		/*
+		 * Handle read-write to read-only or vice versa.
+		 */
+		if (vfs_flagopt(opts, "ro", NULL, 0) && !tmp->tm_ronly) {
 			/* RW -> RO */
 			return (tmpfs_rw_to_ro(mp));
-		} else if (!vfs_flagopt(mp->mnt_optnew, "ro", NULL, 0) &&
-		    tmp->tm_ronly) {
+		}
+		if (!vfs_flagopt(opts, "ro", NULL, 0) && tmp->tm_ronly) {
 			/* RO -> RW */
 			tmp->tm_ronly = 0;
 			MNT_ILOCK(mp);
 			mp->mnt_flag &= ~MNT_RDONLY;
 			MNT_IUNLOCK(mp);
 		}
-		tmp->tm_nomtime = vfs_getopt(mp->mnt_optnew, "nomtime", NULL,
-		    0) == 0;
+
+		/*
+		 * Check if fast path lookup is still supported.
+		 */
 		MNT_ILOCK(mp);
-		if ((mp->mnt_flag & MNT_UNION) == 0) {
+		if (!tmp->tm_nonc && (mp->mnt_flag & MNT_UNION) == 0) {
 			mp->mnt_kern_flag |= MNTK_FPLOOKUP;
 		} else {
 			mp->mnt_kern_flag &= ~MNTK_FPLOOKUP;
@@ -391,25 +415,25 @@ tmpfs_mount(struct mount *mp)
 		return (error);
 
 	if (mp->mnt_cred->cr_ruid != 0 ||
-	    vfs_scanopt(mp->mnt_optnew, "gid", "%d", &root_gid) != 1)
+	    vfs_scanopt(opts, "gid", "%d", &root_gid) != 1)
 		root_gid = va.va_gid;
 	if (mp->mnt_cred->cr_ruid != 0 ||
-	    vfs_scanopt(mp->mnt_optnew, "uid", "%d", &root_uid) != 1)
+	    vfs_scanopt(opts, "uid", "%d", &root_uid) != 1)
 		root_uid = va.va_uid;
 	if (mp->mnt_cred->cr_ruid != 0 ||
-	    vfs_scanopt(mp->mnt_optnew, "mode", "%ho", &root_mode) != 1)
+	    vfs_scanopt(opts, "mode", "%ho", &root_mode) != 1)
 		root_mode = va.va_mode;
-	if (vfs_getopt_size(mp->mnt_optnew, "inodes", &nodes_max) != 0)
+	if (vfs_getopt_size(opts, "inodes", &nodes_max) != 0)
 		nodes_max = 0;
-	if (vfs_getopt_size(mp->mnt_optnew, "size", &size_max) != 0)
+	if (vfs_getopt_size(opts, "size", &size_max) != 0)
 		size_max = 0;
-	if (vfs_getopt_size(mp->mnt_optnew, "maxfilesize", &maxfilesize) != 0)
+	if (vfs_getopt_size(opts, "maxfilesize", &maxfilesize) != 0)
 		maxfilesize = 0;
-	if (vfs_getopt_size(mp->mnt_optnew, "easize", &ea_max_size) != 0)
+	if (vfs_getopt_size(opts, "easize", &ea_max_size) != 0)
 		ea_max_size = 0;
-	nonc = vfs_getopt(mp->mnt_optnew, "nonc", NULL, NULL) == 0;
-	nomtime = vfs_getopt(mp->mnt_optnew, "nomtime", NULL, NULL) == 0;
-	pgread = vfs_getopt(mp->mnt_optnew, "pgread", NULL, NULL) == 0;
+	nonc = vfs_getopt(opts, "nonc", NULL, NULL) == 0;
+	nomtime = vfs_getopt(opts, "nomtime", NULL, NULL) == 0;
+	pgread = vfs_getopt(opts, "pgread", NULL, NULL) == 0;
 
 	/* Do not allow mounts if we do not have enough memory to preserve
 	 * the minimum reserved pages. */