netdev - Re: [PATCH v4 net-next 10/15] net: dsa: Validate hardware support for MST

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]

Date:   Wed, 16 Mar 2022 10:15:18 +0100
From:   Tobias Waldekranz <tobias@...dekranz.com>
To:     Vladimir Oltean <olteanv@...il.com>
Cc:     davem@...emloft.net, kuba@...nel.org, Andrew Lunn <andrew@...n.ch>,
        Vivien Didelot <vivien.didelot@...il.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        Jiri Pirko <jiri@...nulli.us>,
        Ivan Vecera <ivecera@...hat.com>,
        Roopa Prabhu <roopa@...dia.com>,
        Nikolay Aleksandrov <razor@...ckwall.org>,
        Russell King <linux@...linux.org.uk>,
        Petr Machata <petrm@...dia.com>,
        Ido Schimmel <idosch@...dia.com>,
        Matt Johnston <matt@...econstruct.com.au>,
        Cooper Lees <me@...perlees.com>, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org, bridge@...ts.linux-foundation.org
Subject: Re: [PATCH v4 net-next 10/15] net: dsa: Validate hardware support
 for MST

On Tue, Mar 15, 2022 at 19:11, Vladimir Oltean <olteanv@...il.com> wrote:
> On Tue, Mar 15, 2022 at 01:25:38AM +0100, Tobias Waldekranz wrote:
>> When joining a bridge where MST is enabled, we validate that the
>> proper offloading support is in place, otherwise we fallback to
>> software bridging.
>> 
>> When then mode is changed on a bridge in which we are members, we
>> refuse the change if offloading is not supported.
>> 
>> At the moment we only check for configurable learning, but this will
>> be further restricted as we support more MST related switchdev events.
>> 
>> Signed-off-by: Tobias Waldekranz <tobias@...dekranz.com>
>> ---
>>  net/dsa/dsa_priv.h |  2 ++
>>  net/dsa/port.c     | 22 ++++++++++++++++++++++
>>  net/dsa/slave.c    |  6 ++++++
>>  3 files changed, 30 insertions(+)
>> 
>> diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
>> index f20bdd8ea0a8..2aba420696ef 100644
>> --- a/net/dsa/dsa_priv.h
>> +++ b/net/dsa/dsa_priv.h
>> @@ -234,6 +234,8 @@ int dsa_port_vlan_filtering(struct dsa_port *dp, bool vlan_filtering,
>>  			    struct netlink_ext_ack *extack);
>>  bool dsa_port_skip_vlan_configuration(struct dsa_port *dp);
>>  int dsa_port_ageing_time(struct dsa_port *dp, clock_t ageing_clock);
>> +int dsa_port_mst_enable(struct dsa_port *dp, bool on,
>> +			struct netlink_ext_ack *extack);
>>  int dsa_port_mtu_change(struct dsa_port *dp, int new_mtu,
>>  			bool targeted_match);
>>  int dsa_port_fdb_add(struct dsa_port *dp, const unsigned char *addr,
>> diff --git a/net/dsa/port.c b/net/dsa/port.c
>> index 58291df14cdb..02214033cec0 100644
>> --- a/net/dsa/port.c
>> +++ b/net/dsa/port.c
>> @@ -321,6 +321,11 @@ static void dsa_port_bridge_destroy(struct dsa_port *dp,
>>  	kfree(bridge);
>>  }
>>  
>> +static bool dsa_port_supports_mst(struct dsa_port *dp)
>> +{
>> +	return dsa_port_can_configure_learning(dp);
>> +}
>> +
>>  int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
>>  			 struct netlink_ext_ack *extack)
>>  {
>> @@ -334,6 +339,9 @@ int dsa_port_bridge_join(struct dsa_port *dp, struct net_device *br,
>>  	struct net_device *brport_dev;
>>  	int err;
>>  
>> +	if (br_mst_enabled(br) && !dsa_port_supports_mst(dp))
>> +		return -EOPNOTSUPP;
>> +
>>  	/* Here the interface is already bridged. Reflect the current
>>  	 * configuration so that drivers can program their chips accordingly.
>>  	 */
>> @@ -735,6 +743,20 @@ int dsa_port_ageing_time(struct dsa_port *dp, clock_t ageing_clock)
>>  	return 0;
>>  }
>>  
>> +int dsa_port_mst_enable(struct dsa_port *dp, bool on,
>> +			struct netlink_ext_ack *extack)
>> +{
>> +	if (!on)
>> +		return 0;
>> +
>> +	if (!dsa_port_supports_mst(dp)) {
>> +		NL_SET_ERR_MSG_MOD(extack, "Hardware does not support MST");
>> +		return -EINVAL;
>> +	}
>> +
>> +	return 0;
>> +}
>
> Experimenting a bit... maybe this looks tidier? We make the "if" condition
> have the same basic structure as the previous "if (br_mst_enabled(br) &&
> !dsa_port_supports_mst(dp))", albeit transformed using De Morgan's rules.
>
> {
> 	if (!on || dsa_port_supports_mst(dp))
> 		return 0;
>
> 	NL_SET_ERR_MSG_MOD(extack, "Hardware does not support MST");
> 	return -EINVAL;
> }

I initially had it like this. It looks tidier, yes - but to me the
intent is less obvious when reading it. How about:

{
	if (on && !dsa_port_supports_mst(dp)) {
		NL_SET_ERR_MSG_MOD(extack, "Hardware does not support MST");
		return -EINVAL;
	}

	return 0;
}

>> +
>>  int dsa_port_pre_bridge_flags(const struct dsa_port *dp,
>>  			      struct switchdev_brport_flags flags,
>>  			      struct netlink_ext_ack *extack)
>> diff --git a/net/dsa/slave.c b/net/dsa/slave.c
>> index 647adee97f7f..879d18cc99cb 100644
>> --- a/net/dsa/slave.c
>> +++ b/net/dsa/slave.c
>> @@ -463,6 +463,12 @@ static int dsa_slave_port_attr_set(struct net_device *dev, const void *ctx,
>>  
>>  		ret = dsa_port_ageing_time(dp, attr->u.ageing_time);
>>  		break;
>> +	case SWITCHDEV_ATTR_ID_BRIDGE_MST:
>> +		if (!dsa_port_offloads_bridge_dev(dp, attr->orig_dev))
>> +			return -EOPNOTSUPP;
>> +
>> +		ret = dsa_port_mst_enable(dp, attr->u.mst, extack);
>> +		break;
>>  	case SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS:
>>  		if (!dsa_port_offloads_bridge_port(dp, attr->orig_dev))
>>  			return -EOPNOTSUPP;
>> -- 
>> 2.25.1
>>