[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160516090039.GA2038@nanopsycho.orion>
Date: Mon, 16 May 2016 11:00:39 +0200
From: Jiri Pirko <jiri@...nulli.us>
To: Andrew Lunn <andrew@...n.ch>
Cc: Roopa Prabhu <roopa@...ulusnetworks.com>, netdev@...r.kernel.org,
davem@...emloft.net, nogahf@...lanox.com, idosch@...lanox.com,
eladr@...lanox.com, yotamg@...lanox.com, ogerlitz@...lanox.com,
nikolay@...ulusnetworks.com, linville@...driver.com, tgraf@...g.ch,
gospo@...ulusnetworks.com, sfeldma@...il.com, sd@...asysnail.net,
eranbe@...lanox.com, ast@...mgrid.com, edumazet@...gle.com,
hannes@...essinduktion.org
Subject: Re: [patch net-next 1/4] netdevice: add SW statistics ndo
Sun, May 15, 2016 at 05:44:11PM CEST, andrew@...n.ch wrote:
>> I think we don/t understand each other. HW stats always include SW
>> stats. Because whatever goes in or out goes through HW.
>
>Hi Jiri
>
>Bit of a corner case, but what about multicast and broadcast? Can you
>do the replication in the switch, so that only a single copy is sent
>from the host to the switch?
If kernel is originator of such packets, it always sends them through
specific netdevs.
Powered by blists - more mailing lists