[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 02 Dec 2008 19:34:15 -0500
From: Valdis.Kletnieks@...edu
To: Bastian Blank <bastian@...di.eu.org>
Cc: Sukadev Bhattiprolu <sukadev@...ux.vnet.ibm.com>, oleg@...hat.com,
ebiederm@...ssion.com, roland@...hat.com,
containers@...ts.osdl.org, linux-kernel@...r.kernel.org,
xemul@...nvz.org
Subject: Re: [RFC][PATCH 1/5] pid: Implement ns_of_pid
On Tue, 02 Dec 2008 12:58:41 +0100, Bastian Blank said:
> pid is a function argument and the function does not call detach_pid. So
> please try again.
What he means is that if some *other* function calls detach_pid() and then
something calls the ns_of_pid() function, it may be getting passed a null.
Note that the detach_pid() and ns_of_pid() may even be racing on two totally
separate CPUs, doing different things...
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists