linux-kernel - Re: + cgroups-add-functionality-to-read-write-lock-clone_thread-forking-pe r-threadgroup.patch added to -mm tree

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20100103190613.GA13423@andrew.cmu.edu>
Date:	Sun, 3 Jan 2010 14:06:14 -0500
From:	Ben Blum <bblum@...rew.cmu.edu>
To:	Oleg Nesterov <oleg@...hat.com>
Cc:	Paul Menage <menage@...gle.com>, akpm@...ux-foundation.org,
	linux-kernel@...r.kernel.org, bblum@...gle.com,
	ebiederm@...ssion.com, lizf@...fujitsu.com, matthltc@...ibm.com,
	containers@...ts.linux-foundation.org, bblum@...rew.cmu.edu
Subject: Re: + cgroups-add-functionality-to-read-write-lock-clone_thread-forking-pe r-threadgroup.patch added to -mm tree

On Sat, Aug 22, 2009 at 03:09:52PM +0200, Oleg Nesterov wrote:
> 
> OK, cgroups-add-ability-to-move-all-threads-in-a-process-to-a-new-cgroup-atomically.patch
> does
> 
> 	threadgroup_sighand = threadgroup_fork_lock(leader);
> 
> 	rcu_read_lock();
> 	list_for_each_entry_rcu(tsk, &leader->thread_group, thread_group)
> 
> and this is equally wrong afais. Hmm, and other similar
> list_for_each_entry_rcu's doesn't look right. This code can do something
> like
> 
> 	rcu_read_lock();
> 	if (!tsk->sighand)	// tsk is leader or not, doesn't matter
> 		fail;
> 	list_for_each_rcu(tsk->thread_group) {}
> 	rcu_read_unlock();
> 
> though.
> 
> 
> 
> And why do we need sighand->threadgroup_fork_lock ? I gueess, to protect
> against clone(CLONE_THREAD).
> 
> But. threadgroup_fork_lock() has another problem. Even if the process
> P is singlethreaded, I can't see how ->threadgroup_fork_lock work.
> 
> threadgroup_fork_lock() bumps P->sighand->count. If P exec, it will
> notice sighand->count != 1 and switch to another ->sighand.
> 
> Oleg.

So how about this: Each time we take tasklist_lock to iterate over
thread_group (once in getting the sighand, once to move all the tasks),
check if we raced with exec. The two problems are 1) group_leader
changes - we'll need to find the new leader's task_struct anyway - means
we can't iterate over thread_group, and 2) sighand changes after we take
the old one, means we've taken a useless lock.

I put together draft revisions of the old patches that check for racing
with exec in both cases, and if so, returning EAGAIN. I have the wrapper
function cgroup_procs_write loop around the return value, but it could
also possibly give EAGAIN back to userspace. Hopefully the code is safe
and sane this time :)

-- bblum

---
 Documentation/cgroups/cgroups.txt |    7
 include/linux/cgroup.h            |   14 -
 include/linux/init_task.h         |    9
 include/linux/sched.h             |   15 +
 kernel/cgroup.c                   |  519 ++++++++++++++++++++++++++++++++++----
 kernel/fork.c                     |    9
 6 files changed, 524 insertions(+), 49 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/