[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1453987543.7627.27.camel@edumazet-glaptop2.roam.corp.google.com>
Date: Thu, 28 Jan 2016 05:25:43 -0800
From: Eric Dumazet <eric.dumazet@...il.com>
To: Jesper Dangaard Brouer <brouer@...hat.com>
Cc: Alexei Starovoitov <alexei.starovoitov@...il.com>,
John Fastabend <john.fastabend@...il.com>,
Tom Herbert <tom@...bertland.com>,
"Michael S. Tsirkin" <mst@...hat.com>,
David Miller <davem@...emloft.net>,
Or Gerlitz <gerlitz.or@...il.com>,
Eric Dumazet <edumazet@...gle.com>,
Linux Kernel Network Developers <netdev@...r.kernel.org>,
Alexander Duyck <alexander.duyck@...il.com>,
Daniel Borkmann <borkmann@...earbox.net>,
Marek Majkowski <marek@...udflare.com>,
Hannes Frederic Sowa <hannes@...essinduktion.org>,
Florian Westphal <fw@...len.de>,
Paolo Abeni <pabeni@...hat.com>,
John Fastabend <john.r.fastabend@...el.com>,
Amir Vadai <amirva@...il.com>,
Daniel Borkmann <daniel@...earbox.net>,
Vladislav Yasevich <vyasevich@...il.com>
Subject: Re: Bypass at packet-page level (Was: Optimizing instruction-cache,
more packets at each stage)
On Thu, 2016-01-28 at 10:52 +0100, Jesper Dangaard Brouer wrote:
> I'm still in flux/undecided how long we should delay the first touching
> of pkt-data, which happens when calling eth_type_trans(). Should it
> stay in the driver or not(?).
Some cpus have limited prefetch capabilities.
Sometimes, prefetches need to be spaced, otherwise they are ignored.
A driver author might be tempted to 'optimize' its rx handler for few
cpus.
Also, removing eth_type_trans() from the drivers would require quite
some work, but would be generic and certainly helpful.
Powered by blists - more mailing lists