[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161028214228.7200fef2@jkicinski-Precision-T1700>
Date: Fri, 28 Oct 2016 21:42:28 +0100
From: Jakub Kicinski <kubakici@...pl>
To: Alexander Duyck <alexander.duyck@...il.com>
Cc: Alexei Starovoitov <alexei.starovoitov@...il.com>,
John Fastabend <john.fastabend@...il.com>,
David Miller <davem@...emloft.net>,
"Michael S. Tsirkin" <mst@...hat.com>,
Jesper Dangaard Brouer <brouer@...hat.com>,
shrijeet@...il.com, Tom Herbert <tom@...bertland.com>,
Netdev <netdev@...r.kernel.org>,
Shrijeet Mukherjee <shm@...ulusnetworks.com>,
roopa <roopa@...ulusnetworks.com>, nikolay@...ulusnetworks.com
Subject: Re: [PATCH net-next RFC WIP] Patch for XDP support for virtio_net
On Fri, 28 Oct 2016 13:35:02 -0700, Alexander Duyck wrote:
> > At the same time XDP doesn't require to use 4k buffer in something like Netronome.
> > If xdp bpf program can be offloaded into HW with 1800 byte buffers, great!
>
> So are you saying this is only really meant to be used with a full bpf
> hardware offload then?
I think Alexei just meant to say I don't have to worry about providing
4k buffers when program is offloaded but on the host a page is the
memory granularity for XDP...
Powered by blists - more mailing lists