[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <61b24c47.1c69fb81.3e958.1990@mx.google.com>
Date: Thu, 9 Dec 2021 19:34:45 +0100
From: Ansuel Smith <ansuelsmth@...il.com>
To: Vladimir Oltean <vladimir.oltean@....com>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>, Andrew Lunn <andrew@...n.ch>,
Vivien Didelot <vivien.didelot@...il.com>,
Florian Fainelli <f.fainelli@...il.com>
Subject: Re: [RFC PATCH net-next 0/7] DSA master state tracking
On Thu, Dec 09, 2021 at 05:57:20PM +0000, Vladimir Oltean wrote:
> On Thu, Dec 09, 2021 at 06:44:38PM +0100, Ansuel Smith wrote:
> > Ok will send a pcap. Any preferred way to send it?
>
> Email attachment should be fine.
This is a pcap done on the cpu port eth0.
All the packet with lenght 60 are the Ethernet mdio packet.
The capture is done with no cable connected so don't know why there are
some broadcast for ipv6.
Special care about the fact that the qca tag is always present in the
EtherType.
Mdio request qca tag 8181
mdio ack qca tag b887
(I should really investigate the extra packet... In theory they are not
sent (and i remember checking this) by the tagger as they have bit 5:4
set to 0x2... From Documentation these bit are reserved and in our code
we always set them to zero. The switch by itself sends broadcast tagged
packet and respond itself?)
I attached the pcap.
--
Ansuel
Download attachment "pkts.pcap" of type "application/vnd.tcpdump.pcap" (29944 bytes)
Powered by blists - more mailing lists