[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1477084478.7065.89.camel@edumazet-glaptop3.roam.corp.google.com>
Date: Fri, 21 Oct 2016 14:14:38 -0700
From: Eric Dumazet <eric.dumazet@...il.com>
To: Paolo Abeni <pabeni@...hat.com>
Cc: netdev@...r.kernel.org, "David S. Miller" <davem@...emloft.net>,
James Morris <jmorris@...ei.org>,
Trond Myklebust <trond.myklebust@...marydata.com>,
Alexander Duyck <aduyck@...antis.com>,
Daniel Borkmann <daniel@...earbox.net>,
Eric Dumazet <edumazet@...gle.com>,
Tom Herbert <tom@...bertland.com>,
Hannes Frederic Sowa <hannes@...essinduktion.org>,
Edward Cree <ecree@...arflare.com>, linux-nfs@...r.kernel.org
Subject: Re: [PATCH net-next v6 3/3] udp: use it's own memory accounting
schema
On Fri, 2016-10-21 at 13:55 +0200, Paolo Abeni wrote:
> Completely avoid default sock memory accounting and replace it
> with udp-specific accounting.
>
> Since the new memory accounting model encapsulates completely
> the required locking, remove the socket lock on both enqueue and
> dequeue, and avoid using the backlog on enqueue.
>
> Be sure to clean-up rx queue memory on socket destruction, using
> udp its own sk_destruct.
>
> Tested using pktgen with random src port, 64 bytes packet,
> wire-speed on a 10G link as sender and udp_sink as the receiver,
> using an l4 tuple rxhash to stress the contention, and one or more
> udp_sink instances with reuseport.
>
> nr readers Kpps (vanilla) Kpps (patched)
> 1 170 440
Acked-by: Eric Dumazet <edumazet@...gle.com>
Thanks a lot guys.
Powered by blists - more mailing lists