[PATCH] uml: fix scheduler race

This fixes a use-after-free bug in the context switching.  A process going
out of context after exiting wakes up the next process and then kills
itself.  The problem is that when it gets around to killing itself is up to
the host and can happen a long time later, including after the incoming
process has freed its stack, and that memory is possibly being used for
something else.

The fix is to have the incoming process kill the exiting process just to
make sure it can't be running at the point that its stack is freed.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

arch/um/kernel/tt/process_kern.c

@@ -28,7 +28,7 @@
 
 void *switch_to_tt(void *prev, void *next, void *last)
 {
-	struct task_struct *from, *to;
+	struct task_struct *from, *to, *prev_sched;
 	unsigned long flags;
 	int err, vtalrm, alrm, prof, cpu;
 	char c;
@@ -72,6 +72,18 @@ void *switch_to_tt(void *prev, void *next, void *last)
 	if(err != sizeof(c))
 		panic("read of switch_pipe failed, errno = %d", -err);
 
+	/* If the process that we have just scheduled away from has exited,
+	 * then it needs to be killed here.  The reason is that, even though
+	 * it will kill itself when it next runs, that may be too late.  Its
+	 * stack will be freed, possibly before then, and if that happens,
+	 * we have a use-after-free situation.  So, it gets killed here
+	 * in case it has not already killed itself.
+	 */
+	prev_sched = current->thread.prev_sched;
+	if((prev_sched->state == TASK_ZOMBIE) ||
+	   (prev_sched->state == TASK_DEAD))
+		os_kill_process(prev_sched->thread.mode.tt.extern_pid, 1);
+
 	/* This works around a nasty race with 'jail'.  If we are switching
 	 * between two threads of a threaded app and the incoming process 
 	 * runs before the outgoing process reaches the read, and it makes