From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eliezer Tamir Subject: Re: [PATCH v3 net-next 0/4] net: low latency Ethernet device polling Date: Thu, 23 May 2013 14:06:51 +0300 Message-ID: <519DF84B.10401@linux.intel.com> References: <20130520101552.14133.45953.stgit@ladj378.jer.intel.com> <519B632F.7040202@mellanox.com> <519B68A4.3010406@linux.intel.com> <519B7355.9090303@mellanox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Dave Miller , linux-kernel@vger.kernel.org, netdev@vger.kernel.org, Jesse Brandeburg , Don Skidmore , e1000-devel@lists.sourceforge.net, Willem de Bruijn , Andi Kleen , HPA , Eliezer Tamir To: Alex Rosenbaum Return-path: In-Reply-To: <519B7355.9090303@mellanox.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 21/05/2013 16:15, Alex Rosenbaum wrote: > On 5/21/2013 3:29 PM, Eliezer Tamir wrote: >> What benchmarks are you using to test poll/select/epoll? > for epoll/select latency tests we are using sockperf as performance > latency tool: https://code.google.com/p/sockperf/ > It is a client-server based tool and it supported ping-pong, throughput, > and under-load test type. > For epoll, you will need to define a 'feedfile' ("-f filepathname") > which has a list of TCP and/or UDP socket and defined your IO mux type > ("-F epoll"). Thank you! This is very helpful. With sockperf i can directly observe how poll/select/epoll are behaving. I can see some improvement in all of them but clearly more work is needed here. -Eliezer