Oleg Nesterov abd96ecb29 exec: kill unsafe BUG_ON(sig->count) checks
de_thread:

	if (atomic_read(&oldsighand->count) <= 1)
		BUG_ON(atomic_read(&sig->count) != 1);

This is not safe without the rmb() in between.  The results of two
correctly ordered __exit_signal()->atomic_dec_and_test()'s could be seen
out of order on our CPU.

The same is true for the "thread_group_empty()" case, __unhash_process()'s
changes could be seen before atomic_dec_and_test(&sig->count).

On some platforms (including i386) atomic_read() doesn't provide even the
compiler barrier, in that case these checks are simply racy.

Remove these BUG_ON()'s. Alternatively, we can do something like

	BUG_ON( ({ smp_rmb(); atomic_read(&sig->count) != 1; }) );

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-08-22 19:52:47 -07:00
..
2007-07-16 16:03:25 -05:00
2007-07-31 15:39:40 -07:00
2007-08-22 19:52:46 -07:00
2007-05-08 11:14:59 -07:00
2007-08-20 22:44:27 -07:00
2007-07-16 09:05:52 -07:00
2007-07-31 15:39:36 -07:00
2007-07-26 11:11:57 -07:00
2007-07-21 17:49:14 -07:00
2006-10-04 06:51:26 -06:00
2007-07-20 08:44:19 -07:00
2007-07-19 10:04:43 -07:00
2007-07-19 10:04:45 -07:00
2007-07-29 17:09:29 -07:00
2007-05-18 13:09:34 -07:00
2007-05-21 09:18:19 -07:00
2007-05-08 11:15:01 -07:00
2007-05-08 11:15:09 -07:00
2007-07-16 09:05:48 -07:00
2007-05-09 06:44:57 +02:00
2007-07-31 15:39:43 -07:00
2007-05-11 08:29:36 -07:00
2007-07-19 10:04:45 -07:00
2007-07-19 10:04:45 -07:00
2007-07-27 08:08:51 +02:00
2006-10-01 00:39:19 -07:00
2007-07-27 15:40:13 -07:00
2007-07-10 08:04:15 +02:00
2007-05-09 07:10:02 +02:00
2007-07-27 08:08:51 +02:00
2007-07-16 09:05:46 -07:00