From: "Paul E. McKenney" <paulmck@us.ibm.com>

Bug in attach_pid() can result in RCU readers in find_pid() getting
confused if they race with process creation.

Signed-off-by: <paulmck@us.ibm.com>
Acked-by: William Irwin <wli@holomorphy.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---

 kernel/pid.c |    2 +-
 1 files changed, 1 insertion(+), 1 deletion(-)

diff -puN kernel/pid.c~rcu-signal-handling-fix-in-attach_pid kernel/pid.c
--- devel/kernel/pid.c~rcu-signal-handling-fix-in-attach_pid	2005-11-09 20:17:32.000000000 -0800
+++ devel-akpm/kernel/pid.c	2005-11-09 20:17:32.000000000 -0800
@@ -150,6 +150,7 @@ int fastcall attach_pid(task_t *task, en
 
 	task_pid = &task->pids[type];
 	pid = find_pid(type, nr);
+	task_pid->nr = nr;
 	if (pid == NULL) {
 		INIT_LIST_HEAD(&task_pid->pid_list);
 		hlist_add_head_rcu(&task_pid->pid_chain,
@@ -158,7 +159,6 @@ int fastcall attach_pid(task_t *task, en
 		INIT_HLIST_NODE(&task_pid->pid_chain);
 		list_add_tail_rcu(&task_pid->pid_list, &pid->pid_list);
 	}
-	task_pid->nr = nr;
 
 	return 0;
 }
_