From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Nasonov Subject: Re: TCP/IP stack for DPDK Date: Tue, 9 Sep 2014 13:16:59 +0100 Message-ID: <20140909121659.GB11499@neva> References: <4a71bb41.1307.14857e341d5.Coremail.zimeiw@163.com> <20140909062016.GA7050@mhcomputing.net> <20140909065852.GB7240@mhcomputing.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "dev-VfR2kkLFssw@public.gmane.org" To: Matthew Hall Return-path: Content-Disposition: inline In-Reply-To: <20140909065852.GB7240-Hv3ogNYU3JfZZajBQzqCxQ@public.gmane.org> List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org Sender: "dev" Matthew Hall wrote: > However despite this issue, there are some cases where the Linux stack is > greatly superior to the BSD one although normally the opposite is the case... > AF_NETLINK for configuring 10,000+ IP addresses, especially for L4-L7 > performance testing, would be one possible example of this. Another potential > example would be the BPF JIT compiler if you want to combine BPF filters with > DPDK (something I'm doing right now in my own code actually). BPF JIT is available in NetBSD too. It should be quite staightforward to enable it in the rump-dpdk kernel. Alex