1 files changed, 255 insertions, 20 deletions

diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c
index 86602316422d..2b042c42fbc4 100644
--- a/kernel/user_namespace.c
+++ b/kernel/user_namespace.c
@@ -9,6 +9,7 @@
 #include <linux/nsproxy.h>
 #include <linux/slab.h>
 #include <linux/user_namespace.h>
+#include <linux/proc_fs.h>
 #include <linux/highuid.h>
 #include <linux/cred.h>
 #include <linux/securebits.h>
@@ -19,12 +20,31 @@
 #include <linux/fs.h>
 #include <linux/uaccess.h>
 #include <linux/ctype.h>
+#include <linux/projid.h>
 
 static struct kmem_cache *user_ns_cachep __read_mostly;
 
 static bool new_idmap_permitted(struct user_namespace *ns, int cap_setid,
 				struct uid_gid_map *map);
 
+static void set_cred_user_ns(struct cred *cred, struct user_namespace *user_ns)
+{
+	/* Start with the same capabilities as init but useless for doing
+	 * anything as the capabilities are bound to the new user namespace.
+	 */
+	cred->securebits = SECUREBITS_DEFAULT;
+	cred->cap_inheritable = CAP_EMPTY_SET;
+	cred->cap_permitted = CAP_FULL_SET;
+	cred->cap_effective = CAP_FULL_SET;
+	cred->cap_bset = CAP_FULL_SET;
+#ifdef CONFIG_KEYS
+	key_put(cred->request_key_auth);
+	cred->request_key_auth = NULL;
+#endif
+	/* tgcred will be cleared in our caller bc CLONE_THREAD won't be set */
+	cred->user_ns = user_ns;
+}
+
 /*
  * Create a new user namespace, deriving the creator from the user in the
  * passed credentials, and replacing that user with the new root user for the
@@ -38,6 +58,7 @@ int create_user_ns(struct cred *new)
 	struct user_namespace *ns, *parent_ns = new->user_ns;
 	kuid_t owner = new->euid;
 	kgid_t group = new->egid;
+	int ret;
 
 	/* The creator needs a mapping in the parent user namespace
 	 * or else we won't be able to reasonably tell userspace who
@@ -51,38 +72,45 @@ int create_user_ns(struct cred *new)
 	if (!ns)
 		return -ENOMEM;
 
+	ret = proc_alloc_inum(&ns->proc_inum);
+	if (ret) {
+		kmem_cache_free(user_ns_cachep, ns);
+		return ret;
+	}
+
 	kref_init(&ns->kref);
+	/* Leave the new->user_ns reference with the new user namespace. */
 	ns->parent = parent_ns;
 	ns->owner = owner;
 	ns->group = group;
 
-	/* Start with the same capabilities as init but useless for doing
-	 * anything as the capabilities are bound to the new user namespace.
-	 */
-	new->securebits = SECUREBITS_DEFAULT;
-	new->cap_inheritable = CAP_EMPTY_SET;
-	new->cap_permitted = CAP_FULL_SET;
-	new->cap_effective = CAP_FULL_SET;
-	new->cap_bset = CAP_FULL_SET;
-#ifdef CONFIG_KEYS
-	key_put(new->request_key_auth);
-	new->request_key_auth = NULL;
-#endif
-	/* tgcred will be cleared in our caller bc CLONE_THREAD won't be set */
-
-	/* Leave the new->user_ns reference with the new user namespace. */
-	/* Leave the reference to our user_ns with the new cred. */
-	new->user_ns = ns;
+	set_cred_user_ns(new, ns);
 
 	return 0;
 }
 
+int unshare_userns(unsigned long unshare_flags, struct cred **new_cred)
+{
+	struct cred *cred;
+
+	if (!(unshare_flags & CLONE_NEWUSER))
+		return 0;
+
+	cred = prepare_creds();
+	if (!cred)
+		return -ENOMEM;
+
+	*new_cred = cred;
+	return create_user_ns(cred);
+}
+
 void free_user_ns(struct kref *kref)
 {
 	struct user_namespace *parent, *ns =
 		container_of(kref, struct user_namespace, kref);
 
 	parent = ns->parent;
+	proc_free_inum(ns->proc_inum);
 	kmem_cache_free(user_ns_cachep, ns);
 	put_user_ns(parent);
 }
@@ -295,6 +323,75 @@ gid_t from_kgid_munged(struct user_namespace *targ, kgid_t kgid)
 }
 EXPORT_SYMBOL(from_kgid_munged);
 
+/**
+ *	make_kprojid - Map a user-namespace projid pair into a kprojid.
+ *	@ns:  User namespace that the projid is in
+ *	@projid: Project identifier
+ *
+ *	Maps a user-namespace uid pair into a kernel internal kuid,
+ *	and returns that kuid.
+ *
+ *	When there is no mapping defined for the user-namespace projid
+ *	pair INVALID_PROJID is returned.  Callers are expected to test
+ *	for and handle handle INVALID_PROJID being returned.  INVALID_PROJID
+ *	may be tested for using projid_valid().
+ */
+kprojid_t make_kprojid(struct user_namespace *ns, projid_t projid)
+{
+	/* Map the uid to a global kernel uid */
+	return KPROJIDT_INIT(map_id_down(&ns->projid_map, projid));
+}
+EXPORT_SYMBOL(make_kprojid);
+
+/**
+ *	from_kprojid - Create a projid from a kprojid user-namespace pair.
+ *	@targ: The user namespace we want a projid in.
+ *	@kprojid: The kernel internal project identifier to start with.
+ *
+ *	Map @kprojid into the user-namespace specified by @targ and
+ *	return the resulting projid.
+ *
+ *	There is always a mapping into the initial user_namespace.
+ *
+ *	If @kprojid has no mapping in @targ (projid_t)-1 is returned.
+ */
+projid_t from_kprojid(struct user_namespace *targ, kprojid_t kprojid)
+{
+	/* Map the uid from a global kernel uid */
+	return map_id_up(&targ->projid_map, __kprojid_val(kprojid));
+}
+EXPORT_SYMBOL(from_kprojid);
+
+/**
+ *	from_kprojid_munged - Create a projiid from a kprojid user-namespace pair.
+ *	@targ: The user namespace we want a projid in.
+ *	@kprojid: The kernel internal projid to start with.
+ *
+ *	Map @kprojid into the user-namespace specified by @targ and
+ *	return the resulting projid.
+ *
+ *	There is always a mapping into the initial user_namespace.
+ *
+ *	Unlike from_kprojid from_kprojid_munged never fails and always
+ *	returns a valid projid.  This makes from_kprojid_munged
+ *	appropriate for use in syscalls like stat and where
+ *	failing the system call and failing to provide a valid projid are
+ *	not an options.
+ *
+ *	If @kprojid has no mapping in @targ OVERFLOW_PROJID is returned.
+ */
+projid_t from_kprojid_munged(struct user_namespace *targ, kprojid_t kprojid)
+{
+	projid_t projid;
+	projid = from_kprojid(targ, kprojid);
+
+	if (projid == (projid_t) -1)
+		projid = OVERFLOW_PROJID;
+	return projid;
+}
+EXPORT_SYMBOL(from_kprojid_munged);
+
+
 static int uid_m_show(struct seq_file *seq, void *v)
 {
 	struct user_namespace *ns = seq->private;
@@ -302,7 +399,7 @@ static int uid_m_show(struct seq_file *seq, void *v)
 	struct user_namespace *lower_ns;
 	uid_t lower;
 
-	lower_ns = current_user_ns();
+	lower_ns = seq_user_ns(seq);
 	if ((lower_ns == ns) && lower_ns->parent)
 		lower_ns = lower_ns->parent;
 
@@ -323,7 +420,7 @@ static int gid_m_show(struct seq_file *seq, void *v)
 	struct user_namespace *lower_ns;
 	gid_t lower;
 
-	lower_ns = current_user_ns();
+	lower_ns = seq_user_ns(seq);
 	if ((lower_ns == ns) && lower_ns->parent)
 		lower_ns = lower_ns->parent;
 
@@ -337,6 +434,27 @@ static int gid_m_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
+static int projid_m_show(struct seq_file *seq, void *v)
+{
+	struct user_namespace *ns = seq->private;
+	struct uid_gid_extent *extent = v;
+	struct user_namespace *lower_ns;
+	projid_t lower;
+
+	lower_ns = seq_user_ns(seq);
+	if ((lower_ns == ns) && lower_ns->parent)
+		lower_ns = lower_ns->parent;
+
+	lower = from_kprojid(lower_ns, KPROJIDT_INIT(extent->lower_first));
+
+	seq_printf(seq, "%10u %10u %10u\n",
+		extent->first,
+		lower,
+		extent->count);
+
+	return 0;
+}
+
 static void *m_start(struct seq_file *seq, loff_t *ppos, struct uid_gid_map *map)
 {
 	struct uid_gid_extent *extent = NULL;
@@ -362,6 +480,13 @@ static void *gid_m_start(struct seq_file *seq, loff_t *ppos)
 	return m_start(seq, ppos, &ns->gid_map);
 }
 
+static void *projid_m_start(struct seq_file *seq, loff_t *ppos)
+{
+	struct user_namespace *ns = seq->private;
+
+	return m_start(seq, ppos, &ns->projid_map);
+}
+
 static void *m_next(struct seq_file *seq, void *v, loff_t *pos)
 {
 	(*pos)++;
@@ -387,6 +512,13 @@ struct seq_operations proc_gid_seq_operations = {
 	.show = gid_m_show,
 };
 
+struct seq_operations proc_projid_seq_operations = {
+	.start = projid_m_start,
+	.stop = m_stop,
+	.next = m_next,
+	.show = projid_m_show,
+};
+
 static DEFINE_MUTEX(id_map_mutex);
 
 static ssize_t map_write(struct file *file, const char __user *buf,
@@ -434,7 +566,7 @@ static ssize_t map_write(struct file *file, const char __user *buf,
 	/* Require the appropriate privilege CAP_SETUID or CAP_SETGID
 	 * over the user namespace in order to set the id mapping.
 	 */
-	if (!ns_capable(ns, cap_setid))
+	if (cap_valid(cap_setid) && !ns_capable(ns, cap_setid))
 		goto out;
 
 	/* Get a buffer */
@@ -564,10 +696,14 @@ ssize_t proc_uid_map_write(struct file *file, const char __user *buf, size_t siz
 {
 	struct seq_file *seq = file->private_data;
 	struct user_namespace *ns = seq->private;
+	struct user_namespace *seq_ns = seq_user_ns(seq);
 
 	if (!ns->parent)
 		return -EPERM;
 
+	if ((seq_ns != ns) && (seq_ns != ns->parent))
+		return -EPERM;
+
 	return map_write(file, buf, size, ppos, CAP_SETUID,
 			 &ns->uid_map, &ns->parent->uid_map);
 }
@@ -576,17 +712,57 @@ ssize_t proc_gid_map_write(struct file *file, const char __user *buf, size_t siz
 {
 	struct seq_file *seq = file->private_data;
 	struct user_namespace *ns = seq->private;
+	struct user_namespace *seq_ns = seq_user_ns(seq);
 
 	if (!ns->parent)
 		return -EPERM;
 
+	if ((seq_ns != ns) && (seq_ns != ns->parent))
+		return -EPERM;
+
 	return map_write(file, buf, size, ppos, CAP_SETGID,
 			 &ns->gid_map, &ns->parent->gid_map);
 }
 
+ssize_t proc_projid_map_write(struct file *file, const char __user *buf, size_t size, loff_t *ppos)
+{
+	struct seq_file *seq = file->private_data;
+	struct user_namespace *ns = seq->private;
+	struct user_namespace *seq_ns = seq_user_ns(seq);
+
+	if (!ns->parent)
+		return -EPERM;
+
+	if ((seq_ns != ns) && (seq_ns != ns->parent))
+		return -EPERM;
+
+	/* Anyone can set any valid project id no capability needed */
+	return map_write(file, buf, size, ppos, -1,
+			 &ns->projid_map, &ns->parent->projid_map);
+}
+
 static bool new_idmap_permitted(struct user_namespace *ns, int cap_setid,
 				struct uid_gid_map *new_map)
 {
+	/* Allow mapping to your own filesystem ids */
+	if ((new_map->nr_extents == 1) && (new_map->extent[0].count == 1)) {
+		u32 id = new_map->extent[0].lower_first;
+		if (cap_setid == CAP_SETUID) {
+			kuid_t uid = make_kuid(ns->parent, id);
+			if (uid_eq(uid, current_fsuid()))
+				return true;
+		}
+		else if (cap_setid == CAP_SETGID) {
+			kgid_t gid = make_kgid(ns->parent, id);
+			if (gid_eq(gid, current_fsgid()))
+				return true;
+		}
+	}
+
+	/* Allow anyone to set a mapping that doesn't require privilege */
+	if (!cap_valid(cap_setid))
+		return true;
+
 	/* Allow the specified ids if we have the appropriate capability
 	 * (CAP_SETUID or CAP_SETGID) over the parent user namespace.
 	 */
@@ -596,6 +772,65 @@ static bool new_idmap_permitted(struct user_namespace *ns, int cap_setid,
 	return false;
 }
 
+static void *userns_get(struct task_struct *task)
+{
+	struct user_namespace *user_ns;
+
+	rcu_read_lock();
+	user_ns = get_user_ns(__task_cred(task)->user_ns);
+	rcu_read_unlock();
+
+	return user_ns;
+}
+
+static void userns_put(void *ns)
+{
+	put_user_ns(ns);
+}
+
+static int userns_install(struct nsproxy *nsproxy, void *ns)
+{
+	struct user_namespace *user_ns = ns;
+	struct cred *cred;
+
+	/* Don't allow gaining capabilities by reentering
+	 * the same user namespace.
+	 */
+	if (user_ns == current_user_ns())
+		return -EINVAL;
+
+	/* Threaded processes may not enter a different user namespace */
+	if (atomic_read(&current->mm->mm_users) > 1)
+		return -EINVAL;
+
+	if (!ns_capable(user_ns, CAP_SYS_ADMIN))
+		return -EPERM;
+
+	cred = prepare_creds();
+	if (!cred)
+		return -ENOMEM;
+
+	put_user_ns(cred->user_ns);
+	set_cred_user_ns(cred, get_user_ns(user_ns));
+
+	return commit_creds(cred);
+}
+
+static unsigned int userns_inum(void *ns)
+{
+	struct user_namespace *user_ns = ns;
+	return user_ns->proc_inum;
+}
+
+const struct proc_ns_operations userns_operations = {
+	.name		= "user",
+	.type		= CLONE_NEWUSER,
+	.get		= userns_get,
+	.put		= userns_put,
+	.install	= userns_install,
+	.inum		= userns_inum,
+};
+
 static __init int user_namespaces_init(void)
 {
 	user_ns_cachep = KMEM_CACHE(user_namespace, SLAB_PANIC);