netdev - Re: [RFC v3 2/3] vxlan: add support for underlay in non-default VRF

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: <4918e368-5938-3f67-0720-ab558d8bff06@cumulusnetworks.com>
Date:   Tue, 20 Nov 2018 07:57:50 -0700
From:   David Ahern <dsa@...ulusnetworks.com>
To:     Alexis Bauvin <abauvin@...leway.com>, roopa@...ulusnetworks.com
Cc:     netdev@...r.kernel.org, akherbouche@...leway.com
Subject: Re: [RFC v3 2/3] vxlan: add support for underlay in non-default VRF

On 11/20/18 7:23 AM, Alexis Bauvin wrote:
> Creating a VXLAN device with is underlay in the non-default VRF makes
> egress route lookup fail or incorrect since it will resolve in the
> default VRF, and ingress fail because the socket listens in the default
> VRF.
> 
> This patch binds the underlying UDP tunnel socket to the l3mdev of the
> lower device of the VXLAN device. This will listen in the proper VRF and
> output traffic from said l3mdev, matching l3mdev routing rules and
> looking up the correct routing table.
> 
> When the VXLAN device does not have a lower device, or the lower device
> is in the default VRF, the socket will not be bound to any interface,
> keeping the previous behaviour.
> 
> The underlay l3mdev is deduced from the VXLAN lower device
> (IFLA_VXLAN_LINK).
> 
> The l3mdev_master_upper_ifindex_by_index function has been added to
> l3mdev. Its goal is to fetch the effective l3mdev of an interface which
> is not a direct slave of said l3mdev. It handles the following example,
> properly resolving the l3mdev of eth0 to vrf-blue:
> 
> +----------+                         +---------+
> |          |                         |         |
> | vrf-blue |                         | vrf-red |
> |          |                         |         |
> +----+-----+                         +----+----+
>      |                                    |
>      |                                    |
> +----+-----+                         +----+----+
> |          |                         |         |
> | br-blue  |                         | br-red  |
> |          |                         |         |
> +----+-----+                         +---+-+---+
>      |                                   | |
>      |                             +-----+ +-----+
>      |                             |             |
> +----+-----+                +------+----+   +----+----+
> |          |  lower device  |           |   |         |
> |   eth0   | <- - - - - - - | vxlan-red |   | tap-red | (... more taps)
> |          |                |           |   |         |
> +----------+                +-----------+   +---------+

same here. Very helpful diagram.


> diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
> index 27bd586b94b0..a3de08122269 100644
> --- a/drivers/net/vxlan.c
> +++ b/drivers/net/vxlan.c

The vxlan changes look ok to me. It would be good for someone who know
that code better than I do to review it.

Move the following l3mdev changes to a separate patch - introduce infra
changes separate from their use:

> diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
> index 3832099289c5..78fa0ac4613c 100644
> --- a/include/net/l3mdev.h
> +++ b/include/net/l3mdev.h
> @@ -101,6 +101,17 @@ struct net_device *l3mdev_master_dev_rcu(const struct net_device *_dev)
>  	return master;
>  }
>  
> +int l3mdev_master_upper_ifindex_by_index_rcu(struct net *net, int ifindex);
> +static inline
> +int l3mdev_master_upper_ifindex_by_index(struct net *net, int ifindex)
> +{
> +	rcu_read_lock();
> +	ifindex = l3mdev_master_upper_ifindex_by_index_rcu(net, ifindex);
> +	rcu_read_unlock();
> +
> +	return ifindex;
> +}
> +
>  u32 l3mdev_fib_table_rcu(const struct net_device *dev);
>  u32 l3mdev_fib_table_by_index(struct net *net, int ifindex);
>  static inline u32 l3mdev_fib_table(const struct net_device *dev)
> @@ -207,6 +218,17 @@ static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex)
>  	return 0;
>  }
>  
> +static inline
> +int l3mdev_master_upper_ifindex_by_index_rcu(struct net *net, int ifindex)
> +{
> +	return 0;
> +}
> +static inline
> +int l3mdev_master_upper_ifindex_by_index(struct net *net, int ifindex)
> +{
> +	return 0;
> +}
> +
>  static inline
>  struct net_device *l3mdev_master_dev_rcu(const struct net_device *dev)
>  {
> diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c
> index 8da86ceca33d..309dee76724e 100644
> --- a/net/l3mdev/l3mdev.c
> +++ b/net/l3mdev/l3mdev.c
> @@ -46,6 +46,24 @@ int l3mdev_master_ifindex_rcu(const struct net_device *dev)
>  }
>  EXPORT_SYMBOL_GPL(l3mdev_master_ifindex_rcu);
>  
> +/**
> + *	l3mdev_master_upper_ifindex_by_index - get index of upper l3 master
> + *					       device
> + *	@net: network namespace for device index lookup
> + *	@ifindex: targeted interface
> + */
> +int l3mdev_master_upper_ifindex_by_index_rcu(struct net *net, int ifindex)
> +{
> +	struct net_device *dev;
> +
> +	dev = dev_get_by_index_rcu(net, ifindex);
> +	while (dev && !netif_is_l3_master(dev))
> +		dev = netdev_master_upper_dev_get(dev);
> +
> +	return dev ? dev->ifindex : 0;
> +}
> +EXPORT_SYMBOL_GPL(l3mdev_master_upper_ifindex_by_index_rcu);
> +
>  /**
>   *	l3mdev_fib_table - get FIB table id associated with an L3
>   *                             master interface
>