Re: [RFC PATCH 2/4] net: dsa: link aggregation support

[Vladimir Oltean <olteanv@gmail.com>]

On Tue, Oct 27, 2020 at 11:51:15AM +0100, Tobias Waldekranz wrote:
> Monitor the following events and notify the driver when:
> 
> - A DSA port joins/leaves a LAG.
> - A LAG, made up of DSA ports, joins/leaves a bridge.
> - A DSA port in a LAG is enabled/disabled (enabled meaning
>   "distributing" in 802.3ad LACP terms).
> 
> Each LAG interface to which a DSA port is attached is represented by a
> `struct dsa_lag` which is globally reachable from the switch tree and

When you use dsa_broadcast, it is reachable from _all_ switch trees, not
from "the" switch tree. This was added to support "islands" of
inter-compatible DSA switches separated by other DSA switches with
incompatible taggers. Not sure if it was a voluntary decision to use
that as opposed to plain dsa_port_notify. Not a problem either way.

> from each associated port.
> 
> When a LAG joins a bridge, the DSA subsystem will treat that as each
> individual port joining the bridge. The driver may look at the port's
> LAG pointer to see if it is associated with any LAG, if that is
> required. This is analogue to how switchdev events are replicated out
> to all lower devices when reaching e.g. a LAG.
> 
> Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
> ---
>  include/net/dsa.h  |  68 +++++++++++++++++++++
>  net/dsa/dsa2.c     |   3 +
>  net/dsa/dsa_priv.h |  16 +++++
>  net/dsa/port.c     | 146 +++++++++++++++++++++++++++++++++++++++++++++
>  net/dsa/slave.c    |  53 ++++++++++++++--
>  net/dsa/switch.c   |  64 ++++++++++++++++++++
>  6 files changed, 346 insertions(+), 4 deletions(-)
> 
> diff --git a/include/net/dsa.h b/include/net/dsa.h
> index 35429a140dfa..58d73eafe891 100644
> --- a/include/net/dsa.h
> +++ b/include/net/dsa.h
> @@ -145,6 +145,9 @@ struct dsa_switch_tree {
>  	/* List of switch ports */
>  	struct list_head ports;
>  
> +	/* List of configured LAGs */
> +	struct list_head lags;
> +
>  	/* List of DSA links composing the routing table */
>  	struct list_head rtable;
>  };
> @@ -178,6 +181,48 @@ struct dsa_mall_tc_entry {
>  	};
>  };
>  
> +struct dsa_lag {
> +	struct net_device *dev;
> +	int id;
> +
> +	struct list_head ports;
> +
> +	/* For multichip systems, we must ensure that each hash bucket
> +	 * is only enabled on a single egress port throughout the
> +	 * whole tree.

Or else?
I don't really understand this statement.

> --- a/net/dsa/port.c
> +++ b/net/dsa/port.c
> @@ -193,6 +193,152 @@ void dsa_port_bridge_leave(struct dsa_port *dp, struct net_device *br)
>  	dsa_port_set_state_now(dp, BR_STATE_FORWARDING);
>  }
>  
> +static struct dsa_lag *dsa_lag_get(struct dsa_switch_tree *dst,
> +				   struct net_device *dev)
> +{
> +	struct dsa_lag *lag;
> +	unsigned long busy = 0;

Reverse Christmas notation please?

> +	int id;
> +
> +	list_for_each_entry(lag, &dst->lags, list) {
> +		set_bit(lag->id, &busy);
> +
> +		if (lag->dev == dev) {
> +			kref_get(&lag->refcount);
> +			return lag;
> +		}
> +	}
> +
> +	id = find_first_zero_bit(&busy, BITS_PER_LONG);
> +	if (id >= BITS_PER_LONG)
> +		return ERR_PTR(-ENOSPC);
> +
> +	lag = kzalloc(sizeof(*lag), GFP_KERNEL);
> +	if (!lag)
> +		return ERR_PTR(-ENOMEM);
> +
> +	kref_init(&lag->refcount);
> +	lag->id = id;
> +	lag->dev = dev;
> +	INIT_LIST_HEAD(&lag->ports);
> +	INIT_LIST_HEAD(&lag->tx_ports);
> +
> +	INIT_LIST_HEAD(&lag->list);
> +	list_add_tail_rcu(&lag->list, &dst->lags);
> +	return lag;
> +}
> +
> diff --git a/net/dsa/slave.c b/net/dsa/slave.c
> index 3bc5ca40c9fb..e5e4f3d096c0 100644
> --- a/net/dsa/slave.c
> +++ b/net/dsa/slave.c
> @@ -334,7 +334,8 @@ static int dsa_slave_vlan_add(struct net_device *dev,
>  	struct switchdev_obj_port_vlan vlan;
>  	int vid, err;
>  
> -	if (obj->orig_dev != dev)
> +	if (!(obj->orig_dev == dev ||
> +	      (dp->lag && obj->orig_dev == dp->lag->dev)))

A small comment here maybe?

>  		return -EOPNOTSUPP;
>  
>  	if (dsa_port_skip_vlan_configuration(dp))
> @@ -421,7 +422,8 @@ static int dsa_slave_vlan_del(struct net_device *dev,
>  	struct switchdev_obj_port_vlan *vlan;
>  	int vid, err;
>  
> -	if (obj->orig_dev != dev)
> +	if (!(obj->orig_dev == dev ||
> +	      (dp->lag && obj->orig_dev == dp->lag->dev)))
>  		return -EOPNOTSUPP;
>  
>  	if (dsa_port_skip_vlan_configuration(dp))