Re: [PATCH net-next] vxlan: Provide means for obtaining port information

[Stephen Hemminger <stephen@networkplumber.org>]

On Wed, 27 Mar 2013 14:33:11 -0700
Alexander Duyck <alexander.h.duyck@intel.com> wrote:

> This patch provides a very simple means for device drivers to obtain access to
> the VXLAN destination port number configuration.  This information is needed
> in order to provide devices with a means of determining if a given UDP flow is
> a VXLAN flow or if it is a standard flow for offload purposes.
> 
> The main disadvantage with this approach is that loading a driver that calls
> this interface will mean that it will also load the vxlan module and any other
> modules it depends on.  To that end I really hope at some point the value
> becomes static as we could then avoid this extra overhead.  The alternative to
> this though is significantly more complicated as it would likely require
> registering notifiers to capture when the vxlan module is loaded and then
> updating the value.
> 
> This patch adds a new header, if_vxlan.h which I suspect will slowly be
> populated with more details such as the header layout as more devices end up
> needing to figure out how to parse VXLAN for various offloads.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> ---
> 
>  drivers/net/vxlan.c      |    7 +++++++
>  include/linux/if_vxlan.h |   28 ++++++++++++++++++++++++++++
>  2 files changed, 35 insertions(+), 0 deletions(-)
>  create mode 100644 include/linux/if_vxlan.h
> 
> diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
> index 62a4438..5b3bb9a 100644
> --- a/drivers/net/vxlan.c
> +++ b/drivers/net/vxlan.c
> @@ -898,6 +898,13 @@ static u16 vxlan_src_port(const struct vxlan_dev *vxlan, struct sk_buff *skb)
>  	return (((u64) hash * range) >> 32) + vxlan->port_min;
>  }
>  
> +/* simple accessor function for getting VXLAN destination port number */
> +int vxlan_dst_port(void)
> +{
> +	return vxlan_port;
> +}
> +EXPORT_SYMBOL_GPL(vxlan_dst_port);
> +
>  static int handle_offloads(struct sk_buff *skb)
>  {
>  	if (skb_is_gso(skb)) {
> diff --git a/include/linux/if_vxlan.h b/include/linux/if_vxlan.h
> new file mode 100644
> index 0000000..b997eac
> --- /dev/null
> +++ b/include/linux/if_vxlan.h
> @@ -0,0 +1,28 @@
> +/*
> + * VXLAN: Virtual eXtensible Local Area Network
> + *
> + * Copyright (c) 2013, Intel Corporation.
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
> + * Place - Suite 330, Boston, MA 02111-1307 USA.
> + *
> + */
> +
> +#if IS_ENABLED(CONFIG_VXLAN)
> +extern int vxlan_dst_port(void);
> +#else
> +static inline int vxlan_dst_port(void)
> +{
> +	return 0;
> +}
> +#endif

Won't this all break if we what Dave wants and allow listening on multiple ports.
The port should be a property of the vxlan device not global.