From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Emelyanov Subject: Re: [PATCH 10/15] Make each namespace has its own proc tree Date: Mon, 30 Jul 2007 10:45:46 +0400 Message-ID: <46AD891A.3070706@openvz.org> References: <46A8B37B.6050108@openvz.org> <46A8B59E.7050009@openvz.org> <20070729155841.GI120@tv-sign.ru> <20070729170436.GA941@tv-sign.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20070729170436.GA941-6lXkIZvqkOAvJsYlp49lxw@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Oleg Nesterov Cc: Linux Containers List-Id: containers.vger.kernel.org Oleg Nesterov wrote: > On 07/29, Oleg Nesterov wrote: >> Look, proc_flush_task() doesn't need pid, it can use active pid_ns > > Oops. Yes, proc_flush_task() doesn't use ->numbers[].nr, but it should? > > proc_flush_task_mnt() uses task->pid to flush /proc/$pid, this looks wrong, > we should use ->number[ns->level].nr, no? Indeed. Hmm, why then pids were released during my tests... Looks like we do still need struct pid for proc_flush_task(). > Oleg. Thanks, Pavel