From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [PATCH] mlx4_en: map entire pages to increase throughput Date: Mon, 16 Jul 2012 14:08:30 -0700 Message-ID: <500482CE.9000202@hp.com> References: <1342458113-10384-1-git-send-email-cascardo@linux.vnet.ibm.com> <50044F1D.6000703@hp.com> <20120716190611.GA1023@oc1711230544.ibm.com> <50046EB1.5040909@hp.com> <20120716204717.GA16137@oc1711230544.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "davem@davemloft.net" , "netdev@vger.kernel.org" , "yevgenyp@mellanox.co.il" , "ogerlitz@mellanox.com" , "amirv@mellanox.com" , "brking@linux.vnet.ibm.com" , "leitao@linux.vnet.ibm.com" , "klebers@linux.vnet.ibm.com" , "linuxppc-dev@lists.ozlabs.org" , "anton@samba.org" To: Thadeu Lima de Souza Cascardo Return-path: Received: from g6t0187.atlanta.hp.com ([15.193.32.64]:6621 "EHLO g6t0187.atlanta.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752291Ab2GPVIi (ORCPT ); Mon, 16 Jul 2012 17:08:38 -0400 In-Reply-To: <20120716204717.GA16137@oc1711230544.ibm.com> Sender: netdev-owner@vger.kernel.org List-ID: I was thinking more along the lines of an additional comparison, explicitly using netperf TCP_RR or something like it, not just the packets per second from a bulk transfer test. rick -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html > I used a uperf profile that is similar to TCP_RR. It writes, then reads > some bytes. I kept the TCP_NODELAY flag. > > Without the patch, I saw the following: > > packet size ops/s Gb/s > 1 337024 0.0027 > 90 276620 0.199 > 900 190455 1.37 > 4000 68863 2.20 > 9000 45638 3.29 > 60000 9409 4.52 > > With the patch: > > packet size ops/s Gb/s > 1 451738 0.0036 > 90 345682 0.248 > 900 272258 1.96 > 4000 127055 4.07 > 9000 106614 7.68 > 60000 30671 14.72 > So, on the surface it looks like it did good things for PPS, though it would be nice to know what the CPU utilizations/service demands were as a sanity check - does uperf not have that sort of functionality? I'm guessing there were several writes at a time - the 1 byte packet size (sic - that is payload, not packet, and without TCP_NODELAY not even payload necessarily) How many writes does it have outstanding before it does a read? And does it take care to build-up to that number of writes to avoid batching during slowstart, even with TCP_NODELAY set? rick jones From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from g6t0187.atlanta.hp.com (g6t0187.atlanta.hp.com [15.193.32.64]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client CN "smtp.hp.com", Issuer "VeriSign Class 3 Secure Server CA - G3" (not verified)) by ozlabs.org (Postfix) with ESMTPS id E70412C0110 for ; Tue, 17 Jul 2012 07:08:40 +1000 (EST) Message-ID: <500482CE.9000202@hp.com> Date: Mon, 16 Jul 2012 14:08:30 -0700 From: Rick Jones MIME-Version: 1.0 To: Thadeu Lima de Souza Cascardo Subject: Re: [PATCH] mlx4_en: map entire pages to increase throughput References: <1342458113-10384-1-git-send-email-cascardo@linux.vnet.ibm.com> <50044F1D.6000703@hp.com> <20120716190611.GA1023@oc1711230544.ibm.com> <50046EB1.5040909@hp.com> <20120716204717.GA16137@oc1711230544.ibm.com> In-Reply-To: <20120716204717.GA16137@oc1711230544.ibm.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: "netdev@vger.kernel.org" , "leitao@linux.vnet.ibm.com" , "amirv@mellanox.com" , "yevgenyp@mellanox.co.il" , "klebers@linux.vnet.ibm.com" , "anton@samba.org" , "brking@linux.vnet.ibm.com" , "ogerlitz@mellanox.com" , "linuxppc-dev@lists.ozlabs.org" , "davem@davemloft.net" List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , I was thinking more along the lines of an additional comparison, explicitly using netperf TCP_RR or something like it, not just the packets per second from a bulk transfer test. rick -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html > I used a uperf profile that is similar to TCP_RR. It writes, then reads > some bytes. I kept the TCP_NODELAY flag. > > Without the patch, I saw the following: > > packet size ops/s Gb/s > 1 337024 0.0027 > 90 276620 0.199 > 900 190455 1.37 > 4000 68863 2.20 > 9000 45638 3.29 > 60000 9409 4.52 > > With the patch: > > packet size ops/s Gb/s > 1 451738 0.0036 > 90 345682 0.248 > 900 272258 1.96 > 4000 127055 4.07 > 9000 106614 7.68 > 60000 30671 14.72 > So, on the surface it looks like it did good things for PPS, though it would be nice to know what the CPU utilizations/service demands were as a sanity check - does uperf not have that sort of functionality? I'm guessing there were several writes at a time - the 1 byte packet size (sic - that is payload, not packet, and without TCP_NODELAY not even payload necessarily) How many writes does it have outstanding before it does a read? And does it take care to build-up to that number of writes to avoid batching during slowstart, even with TCP_NODELAY set? rick jones