[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d8af0493-064a-2ef7-722c-79d1ac8cb6c7@gmail.com>
Date: Wed, 3 Apr 2019 08:07:24 -0700
From: Florian Fainelli <f.fainelli@...il.com>
To: Vladimir Oltean <olteanv@...il.com>, Andrew Lunn <andrew@...n.ch>
Cc: vivien.didelot@...il.com, davem@...emloft.net,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
linus.walleij@...aro.org, georg.waibel@...sor-technik.de
Subject: Re: [PATCH net-next 16/17] Documentation: networking: dsa: Add
details about NXP SJA1105 driver
On 4/3/2019 3:09 AM, Vladimir Oltean wrote:
> On 4/3/19 12:24 AM, Andrew Lunn wrote:
>>> +The switches do not support switch tagging in hardware. But they do
>>> support
>>> +customizing the TPID by which VLAN traffic is identified as such.
>>> The switch
>>> +driver is leveraging CONFIG_NET_DSA_TAG_8021Q by requesting that
>>> special VLANs
>>> +(with a custom TPID of ETH_P_EDSA instead of ETH_P_8021Q)
>>
>> Hi Vladimir
>>
>> Reusing ETH_P_EDSA is possibility going to case problems in tcpdump.
>> It has code which looks for this Ethertype and uses it to call the
>> Marvell EDSA header decoder code. This is not going to work for your
>> switch, where you have a normal VLAN header.
>>
>> I wounder if we should allocate a new EtherType for this application
>> of VLANs?
>>
>> Andrew
>>
>
> Hi Andrew,
>
> This makes me wonder whether I would be able to use an EtherType range
> to support cascading switch tags on SJA1105.
> In theory if I were to consistently lie to all switches within a switch
> tree, I could make all of them see each other's fake VLAN-tagged traffic
> as untagged.
> By the way what is the simple ETH_P_DSA used for?
ETH_P_DSA is not used since commit
3e8a72d1dae374cf6fc1dba97cec663585845ff9 ("net: dsa: reduce number of
protocol hooks") but since it is in an UAPI exported file we probably
cannot remove it.
--
Florian
Powered by blists - more mailing lists