[PATCH] net namespace: wrap for_each_net with rtnl_lock

[PATCH] net namespace: wrap for_each_net with rtnl_lock

[Fan Du]

The read access to net_namespace_list with for_each_net should always
be protected with rtnl_lock agiast any adding/removing operation from
the list.

Signed-off-by: Fan Du <fan.du@windriver.com>
---
 net/core/net_namespace.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index f976520..f3808ff 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -445,12 +445,14 @@ static int __register_pernet_operations(struct list_head *list,
 
 	list_add_tail(&ops->list, list);
 	if (ops->init || (ops->id && ops->size)) {
+		rtnl_lock();
 		for_each_net(net) {
 			error = ops_init(ops, net);
 			if (error)
 				goto out_undo;
 			list_add_tail(&net->exit_list, &net_exit_list);
 		}
+		rtnl_unlock();
 	}
 	return 0;
 
@@ -468,8 +470,10 @@ static void __unregister_pernet_operations(struct pernet_operations *ops)
 	LIST_HEAD(net_exit_list);
 
 	list_del(&ops->list);
+	rtnl_lock();
 	for_each_net(net)
 		list_add_tail(&net->exit_list, &net_exit_list);
+	rtnl_unlock();
 	ops_exit_list(ops, &net_exit_list);
 	ops_free_list(ops, &net_exit_list);
 }
-- 
1.7.1

Re: [PATCH] net namespace: wrap for_each_net with rtnl_lock

[Eric W. Biederman]

Fan Du <fan.du@windriver.com> writes:

> The read access to net_namespace_list with for_each_net should always
> be protected with rtnl_lock agiast any adding/removing operation from
> the list.

That is not correct.  The rtnl_lock does not protect the
net_namespace_list.  The net_mutex provides that protection.

Modifications to the net_namespace_list are under both the rtnl_lock
and the net_mutex which removes the need of grabbing the net_mutex when
you just need to traverse the list of network namespaces.  This avoids a
lock ordering problem as most places it is desirable to traverse the
net namespace list the rtnl_lock is already held.

In general the init methods will deadlock if you call them with the 
rtnl_lock held, as they grab the rtnl_lock when creating network devices
etc.

The methods you change are protected by the net_mutex so I don't see any
problems here.

Was this patch inspired by code review or was there an actual problem
that inspired it?

> Signed-off-by: Fan Du <fan.du@windriver.com>
> ---
>  net/core/net_namespace.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
>
> diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
> index f976520..f3808ff 100644
> --- a/net/core/net_namespace.c
> +++ b/net/core/net_namespace.c
> @@ -445,12 +445,14 @@ static int __register_pernet_operations(struct list_head *list,
>  
>  	list_add_tail(&ops->list, list);
>  	if (ops->init || (ops->id && ops->size)) {
> +		rtnl_lock();
>  		for_each_net(net) {
>  			error = ops_init(ops, net);
>  			if (error)
>  				goto out_undo;
>  			list_add_tail(&net->exit_list, &net_exit_list);
>  		}
> +		rtnl_unlock();
>  	}
>  	return 0;
>  
> @@ -468,8 +470,10 @@ static void __unregister_pernet_operations(struct pernet_operations *ops)
>  	LIST_HEAD(net_exit_list);
>  
>  	list_del(&ops->list);
> +	rtnl_lock();
>  	for_each_net(net)
>  		list_add_tail(&net->exit_list, &net_exit_list);
> +	rtnl_unlock();
>  	ops_exit_list(ops, &net_exit_list);
>  	ops_free_list(ops, &net_exit_list);
>  }

Re: [PATCH] net namespace: wrap for_each_net with rtnl_lock

[Fan Du]

Hi, Eric

Thanks for your reply!

On 2013年07月04日 15:04, Eric W. Biederman wrote:
> Fan Du<fan.du@windriver.com>  writes:
>
>> The read access to net_namespace_list with for_each_net should always
>> be protected with rtnl_lock agiast any adding/removing operation from
>> the list.
>
> That is not correct.  The rtnl_lock does not protect the
> net_namespace_list.  The net_mutex provides that protection.
>
> Modifications to the net_namespace_list are under both the rtnl_lock
> and the net_mutex which removes the need of grabbing the net_mutex when
> you just need to traverse the list of network namespaces.  This avoids a
> lock ordering problem as most places it is desirable to traverse the
> net namespace list the rtnl_lock is already held.


By my understanding, net_mutex protects operations on pernet_list, and rtln_lock
protects net_namespace_list. net_mutex has side effects on net_namespace_list,
because we try to hold rtnl_lock to modify net_namespace_list after already holding
net_mutex(copy_net_ns). Sorry, I cann't understand the necessity by doing so.

(1)
         mutex_lock(&net_mutex);
         rv = setup_net(net, user_ns);
         if (rv == 0) {
                 rtnl_lock();
                 list_add_tail_rcu(&net->list, &net_namespace_list);
                 rtnl_unlock();
         }
         mutex_unlock(&net_mutex);

why could we do it separately as below?

(2)
         mutex_lock(&net_mutex);
         rv = setup_net(net, user_ns);
         mutex_unlock(&net_mutex);
         if (rv == 0) {
                 rtnl_lock();
                 list_add_tail_rcu(&net->list, &net_namespace_list);
                 rtnl_unlock();
         }


>
> In general the init methods will deadlock if you call them with the
> rtnl_lock held, as they grab the rtnl_lock when creating network devices
> etc.

Yes, I understand. This is reason why we do it in (1) style.


> The methods you change are protected by the net_mutex so I don't see any
> problems here.
> Was this patch inspired by code review or was there an actual problem
> that inspired it?

It's code review, no real alarm :)

>> Signed-off-by: Fan Du<fan.du@windriver.com>
>> ---
>>   net/core/net_namespace.c |    4 ++++
>>   1 files changed, 4 insertions(+), 0 deletions(-)
>>
>> diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
>> index f976520..f3808ff 100644
>> --- a/net/core/net_namespace.c
>> +++ b/net/core/net_namespace.c
>> @@ -445,12 +445,14 @@ static int __register_pernet_operations(struct list_head *list,
>>
>>   	list_add_tail(&ops->list, list);
>>   	if (ops->init || (ops->id&&  ops->size)) {
>> +		rtnl_lock();
>>   		for_each_net(net) {
>>   			error = ops_init(ops, net);
>>   			if (error)
>>   				goto out_undo;
>>   			list_add_tail(&net->exit_list,&net_exit_list);
>>   		}
>> +		rtnl_unlock();
>>   	}
>>   	return 0;
>>
>> @@ -468,8 +470,10 @@ static void __unregister_pernet_operations(struct pernet_operations *ops)
>>   	LIST_HEAD(net_exit_list);
>>
>>   	list_del(&ops->list);
>> +	rtnl_lock();
>>   	for_each_net(net)
>>   		list_add_tail(&net->exit_list,&net_exit_list);
>> +	rtnl_unlock();
>>   	ops_exit_list(ops,&net_exit_list);
>>   	ops_free_list(ops,&net_exit_list);
>>   }
>

-- 
浮沉随浪只记今朝笑

--fan