Fix user namespace exiting OOPs

It turned out, that the user namespace is released during the do_exit() in exit_task_namespaces(), but the struct user_struct is released only during the put_task_struct(), i.e. MUCH later.

On debug kernels with poisoned slabs this will cause the oops in uid_hash_remove() because the head of the chain, which resides inside the struct user_namespace, will be already freed and poisoned.

Since the uid hash itself is required only when someone can search it, i.e. when the namespace is alive, we can safely unhash all the user_struct-s from it during the namespace exiting. The subsequent free_uid() will complete the user_struct destruction.

For example simple program

   #include <sched.h>
   
   char stack[2 * 1024 * 1024];
   
   int f(void *foo)
   {
   	return 0;
   }
   
   int main(void)
   {
   	clone(f, stack + 1 * 1024 * 1024, 0x10000000, 0);
   	return 0;
   }

run on kernel with CONFIG_USER_NS turned on will oops the kernel immediately.

This was spotted during OpenVZ kernel testing.

Signed-off-by Pavel Emelyanov <[email protected]>

Signed-off-by Alexey Dobriyan <[email protected]>

Signed-off-by Serge Hallyn <[email protected]>

---

 include/linux/sched.h   |    1 +
 kernel/user.c           |   26 +++++++++++++++++++++++++-
 kernel/user_namespace.c |    2 +-
 3 files changed, 27 insertions(+), 2 deletions(-)

include/linux/sched.h +1 -0

index a2afa88..b4a4211 100644
@@ -1530,6 +1530,7 @@ static inline struct user_struct *get_ui
 }
 extern void free_uid(struct user_struct *);
 extern void switch_uid(struct user_struct *);
+extern void release_uids(struct user_namespace *ns);
 
 #include <asm/current.h>
 

kernel/user.c +25 -1

index add57c7..e1f2d32 100644
@@ -62,7 +62,7 @@ static inline void uid_hash_insert(struc
 
 static inline void uid_hash_remove(struct user_struct *up)
 {
-	hlist_del(&up->uidhash_node);
+	hlist_del_init(&up->uidhash_node);
 }
 
 static inline struct user_struct *uid_hash_find(uid_t uid, struct hlist_head *hashent)

@@ -199,6 +199,30 @@ void switch_uid(struct user_struct *new_
 	suid_keys(current);
 }
 
+void release_uids(struct user_namespace *ns)
+{
+	int i;
+	unsigned long flags;
+	struct hlist_head *head;
+	struct hlist_node *nd;
+
+	spin_lock_irqsave(&uidhash_lock, flags);
+	/*
+	 * collapse the chains so that the user_struct-s will
+	 * be still alive, but not in hashes. subsequent free_uid()
+	 * will free them.
+	 */
+	for (i = 0; i < UIDHASH_SZ; i++) {
+		head = ns->uidhash_table + i;
+		while (!hlist_empty(head)) {
+			nd = head->first;
+			hlist_del_init(nd);
+		}
+	}
+	spin_unlock_irqrestore(&uidhash_lock, flags);
+
+	free_uid(ns->root_user);
+}
 
 static int __init uid_cache_init(void)
 {

kernel/user_namespace.c +1 -1

index 85af942..df1d2cf 100644
@@ -81,7 +81,7 @@ void free_user_ns(struct kref *kref)
 	struct user_namespace *ns;
 
 	ns = container_of(kref, struct user_namespace, kref);
-	free_uid(ns->root_user);
+	release_uids(ns);
 	kfree(ns);
 }

Quoted 7 lines

run on kernel with CONFIG_USER_NS turned on will oops the kernel immediately.

This was spotted during OpenVZ kernel testing.

Signed-off-by Pavel Emelyanov <[email protected]>

Signed-off-by Alexey Dobriyan <[email protected]>

Good spot. Interesting solution :)

Quoted 3 lines · On Fri, 14 Sep 2007 13:23:55 -0500 "Serge E. Hallyn" <[email protected]> wrote:

Quoted 3 lines

run on kernel with CONFIG_USER_NS turned on will oops the kernel immediately.

Quoted 6 lines

Quoted 4 lines

This was spotted during OpenVZ kernel testing.

Signed-off-by Pavel Emelyanov <[email protected]>

Signed-off-by Alexey Dobriyan <[email protected]>

Good spot. Interesting solution :)

Do we want to fix this in 2.6.23?

If so then at present I'll need to merge

kernel-userc-use-list_for_each_entry-instead-of-list_for_each.patch convert-uid-hash-to-hlist.patch fix-user-namespace-exiting-oops.patch

which is rather a lot of merging at this stage - surely more than is really needed?