releases/3.0.9/ptrace-don-t-clear-group_stop_sigmask-on-double-stop.patch

From oleg@redhat.com  Wed Nov  2 12:41:29 2011
From: Oleg Nesterov <oleg@redhat.com>
Date: Mon, 26 Sep 2011 19:06:32 +0200
Subject: ptrace: don't clear GROUP_STOP_SIGMASK on double-stop
To: Greg KH <gregkh@suse.de>
Cc: Tejun Heo <htejun@gmail.com>, Luke Macken <lmacken@redhat.com>, stable@kernel.org, linux-kernel@vger.kernel.org
Message-ID: <20110926170632.GA16936@redhat.com>
Content-Disposition: inline

From: Oleg Nesterov <oleg@redhat.com>

[This does not correspond to any specific patch in the upstream tree as it was
fixed accidentally by rewriting the code in the 3.1 release]

https://bugzilla.redhat.com/show_bug.cgi?id=740121

1. Luke Macken triggered WARN_ON(!(group_stop & GROUP_STOP_SIGMASK))
   in do_signal_stop().

   This is because do_signal_stop() clears GROUP_STOP_SIGMASK part
   unconditionally but doesn't update it if task_is_stopped().

2. Looking at this problem I noticed that WARN_ON_ONCE(!ptrace) is
   not right, a stopped-but-resumed tracee can clone the untraced
   thread in the SIGNAL_STOP_STOPPED group, the new thread can start
   another group-stop.

   Remove this warning, we need more fixes to make it true.

Reported-by: Luke Macken <lmacken@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 kernel/signal.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1894,21 +1894,19 @@ static int do_signal_stop(int signr)
 		 */
 		if (!(sig->flags & SIGNAL_STOP_STOPPED))
 			sig->group_exit_code = signr;
-		else
-			WARN_ON_ONCE(!task_ptrace(current));
 
 		current->group_stop &= ~GROUP_STOP_SIGMASK;
 		current->group_stop |= signr | gstop;
 		sig->group_stop_count = 1;
 		for (t = next_thread(current); t != current;
 		     t = next_thread(t)) {
-			t->group_stop &= ~GROUP_STOP_SIGMASK;
 			/*
 			 * Setting state to TASK_STOPPED for a group
 			 * stop is always done with the siglock held,
 			 * so this check has no races.
 			 */
 			if (!(t->flags & PF_EXITING) && !task_is_stopped(t)) {
+				t->group_stop &= ~GROUP_STOP_SIGMASK;
 				t->group_stop |= signr | gstop;
 				sig->group_stop_count++;
 				signal_wake_up(t, 0);