uglify kill_pid_info() to fix kill() vs exec() race

kill_pid_info()->pid_task() could be the old leader of the execing process.
In that case it is possible that the leader will be released before we take
siglock. This means that kill_pid_info() (and thus sys_kill()) can return a
false -ESRCH.

Change the code to retry when lock_task_sighand() fails. The endless loop is
not possible, __exit_signal() both clears ->sighand and does detach_pid().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kernel/signal.c

@@ -1050,17 +1050,26 @@ int kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp)
 
 int kill_pid_info(int sig, struct siginfo *info, struct pid *pid)
 {
-	int error;
+	int error = -ESRCH;
 	struct task_struct *p;
 
 	rcu_read_lock();
 	if (unlikely(sig_needs_tasklist(sig)))
 		read_lock(&tasklist_lock);
 
+retry:
 	p = pid_task(pid, PIDTYPE_PID);
-	error = -ESRCH;
-	if (p)
+	if (p) {
 		error = group_send_sig_info(sig, info, p);
+		if (unlikely(error == -ESRCH))
+			/*
+			 * The task was unhashed in between, try again.
+			 * If it is dead, pid_task() will return NULL,
+			 * if we race with de_thread() it will find the
+			 * new leader.
+			 */
+			goto retry;
+	}
 
 	if (unlikely(sig_needs_tasklist(sig)))
 		read_unlock(&tasklist_lock);