From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nir Tzachar Subject: Re: [RFCv4 PATCH 2/2] net: Allow protocols to provide an unlocked_recvmsg socket method Date: Thu, 1 Oct 2009 11:49:39 +0200 Message-ID: <9b2db90b0910010249h182bf5d4sc7fdaea9e1345720@mail.gmail.com> References: <20090916170745.GD7699@ghostprotocols.net> <9b2db90b0909170709n400859c6q13514b315970dde9@mail.gmail.com> <20090917212113.GC3691@ghostprotocols.net> <20090917215344.GE3691@ghostprotocols.net> <20090923000925.GA6011@ghostprotocols.net> <9b2db90b0909222123x7e547210p5755adf9f8ae875f@mail.gmail.com> <20090923043813.GA6464@ghostprotocols.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Cc: Ziv Ayalon To: Arnaldo Carvalho de Melo , Linux Networking Development Mailing List Return-path: Received: from mail-fx0-f218.google.com ([209.85.220.218]:52929 "EHLO mail-fx0-f218.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756050AbZJAJtg (ORCPT ); Thu, 1 Oct 2009 05:49:36 -0400 Received: by fxm18 with SMTP id 18so6035209fxm.17 for ; Thu, 01 Oct 2009 02:49:40 -0700 (PDT) In-Reply-To: <20090923043813.GA6464@ghostprotocols.net> Sender: netdev-owner@vger.kernel.org List-ID: Hi Arnaldo I have repeated the tests using net-next on top of linus' git tree (I hope I got it right..) and the patches you sent me. Things did not get better, and in most cases were even worse; the recvmmsg parts distinctly showed better throughput, but the latency has more than doubled. The simplest test of using a batch size of 1 results with recvmmsg's latency over 1000 micro, while regular recvmsg is around 450 micro. (note that to use 1 packet there is a small bug in the reg_recv which needs to be fixed. Namely, change ret = -1 to ret = 0). On the previous system config -- part 0001 of the patch, on top of 2.6.31 -- the latency of a single packet batch is 370 micro. So, there seems to be a regression with the kernel tree I am using, or with part 0002 of the path. I'll try running the net-next with only part 1 of the patch and report. Cheers.