From: Pavel Emelyanov <xemul@parallels.com>
To: Tejun Heo <tj@kernel.org>, Pedro Alves <pedro@codesourcery.com>,
Oleg Nesterov <oleg@redhat.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Cyrill Gorcunov <gorcunov@openvz.org>,
James Bottomley <jbottomley@parallels.com>
Subject: Re: [RFC][PATCH 0/3] fork: Add the ability to create tasks with given pids
Date: Thu, 24 Nov 2011 00:14:36 +0400 [thread overview]
Message-ID: <4ECD542C.7010705@parallels.com> (raw)
In-Reply-To: <4ECD3946.1030503@parallels.com>
On 11/23/2011 10:19 PM, Pavel Emelyanov wrote:
> On 11/23/2011 08:24 PM, Tejun Heo wrote:
>> Hello,
>>
>> On Wed, Nov 23, 2011 at 04:20:44PM +0000, Pedro Alves wrote:
>>>> Would CAP_CHECKPOINT be a shame too?
>>>
>>> I think CAP_CHECKPOINT (or something through some LSM) would be
>>> definitely better.
>>>
>>>> I'm reluctant about priviledge
>>>> through fd inheritance mostly because of its unusualness. I don't
>>>> think priv management is a good problem space for small creative
>>>> solutions. We're much better off with mundane mechanisms which people
>>>> are already familiar with and is easy to account for.
>>>
>>> fd inheritance wouldn't work for gdb; a user spawned gdb
>>> wouldn't inherit an open fd to kernel.ns_last_pid from anywhere.
>>
>> I see. So, let's do it for root for now and later add finer grained
>> CAP as necessary/viable. Pavel, what do you think?
>
> OK, I'll send the respective patches soon.
Hm... Started testing this stuff and thought about Pedro's wish to use this
in gdb one more time :(
The thing is, that we (in checkpoint/restore) are going to use this sysctl
when creating a pid namespace from scratch, thus having full control over
all the forks happening in this namespace.
But when it comes to the ability for gdb to create a task with a given pid in
a _living_ namespace this solution simply won't work! It doesn't guarantee,
that after setting the last_pid via sysctl this last_pid stays the same at
the time we do call fork()/clone(). Because there are other tasks that can call
fork themselves ignoring any lask_pid locking we can play with.
That said I see only two real-life scenarios of how to use _this_ API:
1. creating tasks in a new pid namespace, making sure all the fork-ers care
about the proper locking;
2. forking tasks in a loop checking that getpid() returns desired value and
hoping that other tasks do not fork() at speed high enough for spoiling
every single last_pid value set via sysctl.
Is any of these scenarios suitable for Pedro? Other thoughts on this?
>> Thanks.
>>
>
next prev parent reply other threads:[~2011-11-23 20:15 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-17 11:41 [RFC][PATCH 0/3] fork: Add the ability to create tasks with given pids Pavel Emelyanov
2011-11-17 11:42 ` [PATCH 1/3] pids: Make alloc_pid return error Pavel Emelyanov
2011-11-17 11:42 ` [PATCH 2/3] pids: Split alloc_pidmap into parts Pavel Emelyanov
2011-11-17 11:43 ` [PATCH 3/3] pids: Make it possible to clone tasks with given pids Pavel Emelyanov
2011-11-17 15:32 ` Oleg Nesterov
2011-11-17 15:49 ` Pavel Emelyanov
2011-11-17 16:00 ` Oleg Nesterov
2011-11-17 17:28 ` Linus Torvalds
2011-11-17 19:04 ` Oleg Nesterov
2011-11-17 18:36 ` Oleg Nesterov
2011-11-18 10:05 ` Pavel Emelyanov
2011-11-17 15:49 ` [RFC][PATCH 0/3] fork: Add the ability to create " Oleg Nesterov
2011-11-17 16:01 ` Pavel Emelyanov
2011-11-17 16:02 ` Oleg Nesterov
2011-11-18 23:30 ` Tejun Heo
2011-11-21 9:15 ` Pavel Emelyanov
2011-11-21 22:50 ` Tejun Heo
2011-11-22 11:11 ` Pavel Emelyanov
2011-11-22 12:04 ` Pedro Alves
2011-11-22 15:33 ` Tejun Heo
2011-11-23 16:20 ` Pedro Alves
2011-11-23 16:24 ` Tejun Heo
2011-11-23 17:26 ` Oleg Nesterov
2011-11-23 17:37 ` Tejun Heo
2011-11-23 18:19 ` Pavel Emelyanov
2011-11-23 20:14 ` Pavel Emelyanov [this message]
2011-11-24 17:31 ` Oleg Nesterov
2011-11-25 10:14 ` Pavel Emelyanov
2011-11-25 16:22 ` Oleg Nesterov
2011-11-25 16:44 ` Pavel Emelyanov
2011-11-25 16:54 ` Oleg Nesterov
2011-11-25 17:03 ` Pavel Emelyanov
2011-11-25 22:36 ` Pedro Alves
2011-11-27 16:24 ` [RFC][PATCH 0/3] fork: Add the ability to create tasks with?given pids Oleg Nesterov
2011-11-27 9:41 ` [RFC][PATCH 0/3] fork: Add the ability to create tasks with given pids Konstantin Khlebnikov
2011-11-27 17:34 ` Oleg Nesterov
2011-11-27 18:47 ` Tejun Heo
2011-11-28 10:38 ` Pavel Emelyanov
2011-11-28 16:25 ` Tejun Heo
2011-11-22 15:23 ` Tejun Heo
2011-11-22 15:29 ` Tejun Heo
2011-11-22 16:30 ` Pavel Emelyanov
2011-11-22 16:44 ` Linus Torvalds
2011-11-22 19:29 ` Pavel Emelyanov
2012-01-26 23:28 ` Kay Sievers
2011-11-22 21:16 ` Oleg Nesterov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4ECD542C.7010705@parallels.com \
--to=xemul@parallels.com \
--cc=gorcunov@openvz.org \
--cc=jbottomley@parallels.com \
--cc=linux-kernel@vger.kernel.org \
--cc=oleg@redhat.com \
--cc=pedro@codesourcery.com \
--cc=tj@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).