netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hans Schultz <schultz.hans@gmail.com>
To: Nikolay Aleksandrov <razor@blackwall.org>,
	Hans Schultz <schultz.hans@gmail.com>,
	davem@davemloft.net, kuba@kernel.org
Cc: netdev@vger.kernel.org, Andrew Lunn <andrew@lunn.ch>,
	Vivien Didelot <vivien.didelot@gmail.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Vladimir Oltean <olteanv@gmail.com>,
	Jiri Pirko <jiri@resnulli.us>, Ivan Vecera <ivecera@redhat.com>,
	Roopa Prabhu <roopa@nvidia.com>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Ido Schimmel <idosch@nvidia.com>,
	linux-kernel@vger.kernel.org, bridge@lists.linux-foundation.org
Subject: Re: [PATCH net-next 1/3] net: bridge: add fdb flag to extent locked port feature
Date: Thu, 10 Mar 2022 17:33:01 +0100	[thread overview]
Message-ID: <86tuc5939e.fsf@gmail.com> (raw)
In-Reply-To: <dc47275b-27f8-5de0-ae6e-e82013a03d1f@blackwall.org>

On tor, mar 10, 2022 at 18:14, Nikolay Aleksandrov <razor@blackwall.org> wrote:
> On 10/03/2022 18:11, Hans Schultz wrote:
>> On tor, mar 10, 2022 at 17:57, Nikolay Aleksandrov <razor@blackwall.org> wrote:
>>> On 10/03/2022 17:38, Hans Schultz wrote:
>>>> On tor, mar 10, 2022 at 16:42, Nikolay Aleksandrov <razor@blackwall.org> wrote:
>>>>> On 10/03/2022 16:23, Hans Schultz wrote:
>>>>>> Add an intermediate state for clients behind a locked port to allow for
>>>>>> possible opening of the port for said clients. This feature corresponds
>>>>>> to the Mac-Auth and MAC Authentication Bypass (MAB) named features. The
>>>>>> latter defined by Cisco.
>>>>>>
>>>>>> Signed-off-by: Hans Schultz <schultz.hans+netdev@gmail.com>
>>>>>> ---
>>>>>>     include/uapi/linux/neighbour.h |  1 +
>>>>>>     net/bridge/br_fdb.c            |  6 ++++++
>>>>>>     net/bridge/br_input.c          | 11 ++++++++++-
>>>>>>     net/bridge/br_private.h        |  3 ++-
>>>>>>     4 files changed, 19 insertions(+), 2 deletions(-)
>>>>>>
>>>>>> diff --git a/include/uapi/linux/neighbour.h b/include/uapi/linux/neighbour.h
>>>>>> index db05fb55055e..83115a592d58 100644
>>>>>> --- a/include/uapi/linux/neighbour.h
>>>>>> +++ b/include/uapi/linux/neighbour.h
>>>>>> @@ -208,6 +208,7 @@ enum {
>>>>>>     	NFEA_UNSPEC,
>>>>>>     	NFEA_ACTIVITY_NOTIFY,
>>>>>>     	NFEA_DONT_REFRESH,
>>>>>> +	NFEA_LOCKED,
>>>>>>     	__NFEA_MAX
>>>>>>     };
>>>>>
>>>>> Hmm, can you use NDA_FLAGS_EXT instead ?
>>>>> That should simplify things and reduce the nl size.
>>>>>
>>>>
>>>> I am using NDA_FDB_EXT_ATTRS. NFEA_LOCKED is just the
>>>> flag as the other flags section is full wrt the normal flags, but maybe it
>>>> doesn't fit in that section?
>>>>
>>>
>>> Actually wait a second, this is completely wrong use of NDA_FDB_EXT_ATTRS.
>>> That is a nested attribute, so the code below is wrong. More below..
>>>
>>>> I will just note that iproute2 support for parsing nested attributes
>>>> does not work, thus the BR_FDB_NOTIFY section (lines 150-165) are
>>>> obsolete with respect to iproute2 as it is now. I cannot rule out that
>>>> someone has some other tool that can handle this BR_FDB_NOTIFY, but I
>>>> could not make iproute2 as it stands handle nested attributes. And of
>>>> course there is no handling of NDA_FDB_EXT_ATTRS in iproute2 now.
>>>>>>>     #define NFEA_MAX (__NFEA_MAX - 1)
>>>>>> diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
>>>>>> index 6ccda68bd473..396dcf3084cf 100644
>>>>>> --- a/net/bridge/br_fdb.c
>>>>>> +++ b/net/bridge/br_fdb.c
>>>>>> @@ -105,6 +105,7 @@ static int fdb_fill_info(struct sk_buff *skb, const struct net_bridge *br,
>>>>>>     	struct nda_cacheinfo ci;
>>>>>>     	struct nlmsghdr *nlh;
>>>>>>     	struct ndmsg *ndm;
>>>>>> +	u8 ext_flags = 0;
>>>>>>     
>>>>>>     	nlh = nlmsg_put(skb, portid, seq, type, sizeof(*ndm), flags);
>>>>>>     	if (nlh == NULL)
>>>>>> @@ -125,11 +126,16 @@ static int fdb_fill_info(struct sk_buff *skb, const struct net_bridge *br,
>>>>>>     		ndm->ndm_flags |= NTF_EXT_LEARNED;
>>>>>>     	if (test_bit(BR_FDB_STICKY, &fdb->flags))
>>>>>>     		ndm->ndm_flags |= NTF_STICKY;
>>>>>> +	if (test_bit(BR_FDB_ENTRY_LOCKED, &fdb->flags))
>>>>>> +		ext_flags |= 1 << NFEA_LOCKED;
>>>>>>     
>>>>>>     	if (nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->key.addr))
>>>>>>     		goto nla_put_failure;
>>>>>>     	if (nla_put_u32(skb, NDA_MASTER, br->dev->ifindex))
>>>>>>     		goto nla_put_failure;
>>>>>> +	if (nla_put_u8(skb, NDA_FDB_EXT_ATTRS, ext_flags))
>>>>>> +		goto nla_put_failure;
>>>>>> +
>>>
>>> This is wrong. NDA_FDB_EXT_ATTRS is a nested attribute, you can't use it as a u8.
>>> You need to have this structure:
>>>    [ NDA_FDB_EXT_ATTRS ]
>>>     ` [ NFEA_LOCKED ]
>>>
>>> But that's why I asked if you could use the NDA_FLAGS_EXT attribute. You can see
>>> the logic from the neigh code.
>> 
>> Ahh yes, NDA_FLAGS_EXT was not there in the 5.15.x kernel I have
>> originally being making the patches in.
>> 
>> I hope that the handling of nested attributes has been fixed in
>> iproute2. ;-)
>> 
>
> It hasn't been broken, I'm guessing you're having issues with the nested bit being set.
> Check NLA_F_NESTED and NLA_TYPE_MASK.
>

Hmmm, then I wonder why I could not make the same code as in the said
lines (150-165) in br_fdb.c give any parsed attributes in iproute2 under
tb[NDA_FDB_EXT_ATTR].

Did I miss something, or are those lines incorrect?

>>>
>>> Also note that you need to account for the new attribute's size in fdb_nlmsg_size().
>>>
>>>
>>>>>>     	ci.ndm_used	 = jiffies_to_clock_t(now - fdb->used);
>>>>>>     	ci.ndm_confirmed = 0;
>>>>>>     	ci.ndm_updated	 = jiffies_to_clock_t(now - fdb->updated);
>>>>>> diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
>>>>>> index e0c13fcc50ed..897908484b18 100644
>>>>>> --- a/net/bridge/br_input.c
>>>>>> +++ b/net/bridge/br_input.c
>>>>>> @@ -75,6 +75,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb
>>>>>>     	struct net_bridge_mcast *brmctx;
>>>>>>     	struct net_bridge_vlan *vlan;
>>>>>>     	struct net_bridge *br;
>>>>>> +	unsigned long flags = 0;
>>>>>
>>>>> Please move this below...
>>>>>
>>>>>>     	u16 vid = 0;
>>>>>>     	u8 state;
>>>>>>     
>>>>>> @@ -94,8 +95,16 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb
>>>>>>     			br_fdb_find_rcu(br, eth_hdr(skb)->h_source, vid);
>>>>>>     
>>>>>>     		if (!fdb_src || READ_ONCE(fdb_src->dst) != p ||
>>>>>> -		    test_bit(BR_FDB_LOCAL, &fdb_src->flags))
>>>>>> +		    test_bit(BR_FDB_LOCAL, &fdb_src->flags)) {
>>>>>> +			if (!fdb_src) {
>>>>>
>>>>> ... here where it's only used.
>>>>>
>>>>
>>>> Forgot that one. Shall do!
>>>>
>>>>>> +				set_bit(BR_FDB_ENTRY_LOCKED, &flags);
>>>>>> +				br_fdb_update(br, p, eth_hdr(skb)->h_source, vid, flags);
>>>>>> +			}
>>>>>>     			goto drop;
>>>>>> +		} else {
>>>>>> +			if (test_bit(BR_FDB_ENTRY_LOCKED, &fdb_src->flags))
>>>>>> +				goto drop;
>>>>>> +		}
>>>>>>     	}
>>>>>>     
>>>>>>     	nbp_switchdev_frame_mark(p, skb);
>>>>>> diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
>>>>>> index 48bc61ebc211..f5a0b68c4857 100644
>>>>>> --- a/net/bridge/br_private.h
>>>>>> +++ b/net/bridge/br_private.h
>>>>>> @@ -248,7 +248,8 @@ enum {
>>>>>>     	BR_FDB_ADDED_BY_EXT_LEARN,
>>>>>>     	BR_FDB_OFFLOADED,
>>>>>>     	BR_FDB_NOTIFY,
>>>>>> -	BR_FDB_NOTIFY_INACTIVE
>>>>>> +	BR_FDB_NOTIFY_INACTIVE,
>>>>>> +	BR_FDB_ENTRY_LOCKED,
>>>>>>     };
>>>>>>     
>>>>>>     struct net_bridge_fdb_key {

  reply	other threads:[~2022-03-10 16:33 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-10 14:23 [PATCH net-next 0/3] Extend locked port feature with FDB locked flag (MAC-Auth/MAB) Hans Schultz
2022-03-10 14:23 ` [PATCH net-next 1/3] net: bridge: add fdb flag to extent locked port feature Hans Schultz
2022-03-10 14:42   ` Nikolay Aleksandrov
2022-03-10 15:38     ` Hans Schultz
2022-03-10 15:57       ` Nikolay Aleksandrov
2022-03-10 16:11         ` Hans Schultz
2022-03-10 16:14           ` Nikolay Aleksandrov
2022-03-10 16:33             ` Hans Schultz [this message]
2022-03-14 15:30   ` Ido Schimmel
2022-03-15  8:48     ` Hans Schultz
2022-03-15 11:00       ` Ido Schimmel
2022-03-10 14:23 ` [PATCH net-next 2/3] net: switchdev: add support for offloading of fdb locked flag Hans Schultz
2022-03-10 14:23 ` [PATCH net-next 3/3] net: dsa: mv88e6xxx: mac-auth/MAB implementation Hans Schultz
2022-03-10 14:28   ` Vladimir Oltean
2022-03-10 15:00     ` Hans Schultz
2022-03-10 15:07       ` Vladimir Oltean
2022-03-10 15:51         ` Hans Schultz
2022-03-10 16:05           ` Vladimir Oltean
2022-03-10 16:40             ` Hans Schultz
2022-03-10 15:57     ` Hans Schultz
2022-03-14 10:46     ` Hans Schultz
2022-03-16 23:34       ` Vladimir Oltean
2022-03-17  8:52         ` Hans Schultz
2022-03-17 14:19           ` Andrew Lunn
2022-03-17 15:36             ` Vladimir Oltean
2022-03-17 16:07               ` Hans Schultz
2022-03-17 16:18                 ` Vladimir Oltean
2022-03-17 16:58                   ` Hans Schultz
2022-03-17 17:20                     ` Vladimir Oltean
2022-03-18 10:04                       ` Hans Schultz
2022-03-18 12:14                         ` Vladimir Oltean
2022-03-18 13:10                           ` Hans Schultz
2022-03-18 13:19                             ` Vladimir Oltean
2022-03-22 11:01                               ` Hans Schultz
2022-03-22 11:08                                 ` Vladimir Oltean
2022-03-22 13:21                                   ` Hans Schultz
2022-03-22 14:47                                     ` Hans Schultz
2022-03-23 10:13                                   ` Hans Schultz
2022-03-23 10:16                                     ` Vladimir Oltean
2022-03-23 10:46                                       ` Hans Schultz
2022-03-23 10:57                                       ` Hans Schultz
2022-03-23 11:21                                         ` Vladimir Oltean
2022-03-23 11:43                                           ` Hans Schultz
2022-03-23 11:54                                             ` Vladimir Oltean
2022-03-21 14:51             ` Hans Schultz
2022-03-10 14:54   ` Andrew Lunn
2022-03-11  7:59     ` Hans Schultz
2022-03-14 15:50 ` [PATCH net-next 0/3] Extend locked port feature with FDB locked flag (MAC-Auth/MAB) Ido Schimmel
2022-03-15  8:59   ` Hans Schultz
2022-03-15 11:11     ` Ido Schimmel
2022-03-17  0:18 ` Florian Fainelli
2022-03-17  8:29   ` Hans Schultz
2022-03-17 18:42     ` Vladimir Oltean

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=86tuc5939e.fsf@gmail.com \
    --to=schultz.hans@gmail.com \
    --cc=andrew@lunn.ch \
    --cc=bridge@lists.linux-foundation.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=idosch@nvidia.com \
    --cc=ivecera@redhat.com \
    --cc=jiri@resnulli.us \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=olteanv@gmail.com \
    --cc=razor@blackwall.org \
    --cc=roopa@nvidia.com \
    --cc=vivien.didelot@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).