From owner-soc-status@FreeBSD.ORG Fri Aug 19 16:15:52 2011 Return-Path: Delivered-To: soc-status@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D7B54106566C for ; Fri, 19 Aug 2011 16:15:52 +0000 (UTC) (envelope-from gockzy@gmail.com) Received: from mail-ey0-f172.google.com (mail-ey0-f172.google.com [209.85.215.172]) by mx1.freebsd.org (Postfix) with ESMTP id 6F1A08FC1C for ; Fri, 19 Aug 2011 16:15:52 +0000 (UTC) Received: by eye4 with SMTP id 4so1739030eye.31 for ; Fri, 19 Aug 2011 09:15:51 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:from:date:message-id:subject:to:content-type; bh=iZwPyqq4V19RFxlK3zmO+BXotVOCIseC/10DA+FpldY=; b=XVxzGb162zYbQr2PxGkiAjzx2bIdlXElu/ZwfNeTwoDMrlAyHobavee6AyCVU0Pntc wlPkuzHAGmiKzmK+r0AfPes+t2o4Qph7OIK98S7eoypsp4QrEY1k6GsTlylo5IELqCWR r5qg4EH9g4x+yyRWDlOV7dTlBwgVV9yDx+2ZA= Received: by 10.14.147.141 with SMTP id t13mr874816eej.36.1313770551176; Fri, 19 Aug 2011 09:15:51 -0700 (PDT) MIME-Version: 1.0 Received: by 10.14.186.6 with HTTP; Fri, 19 Aug 2011 09:15:31 -0700 (PDT) From: Kazuya Goda Date: Sat, 20 Aug 2011 01:15:31 +0900 Message-ID: To: soc-status@freebsd.org Content-Type: text/plain; charset=ISO-8859-1 Subject: [status report] RPS/RFS #week12 X-BeenThere: soc-status@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Summer of Code Status Reports and Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Aug 2011 16:15:52 -0000 Hi, The project goal is to implement RPS/RFS on FreeBSD. RPS solves the problem of mono-queue NIC which can't distribute packets across multiple processors. RFS is the extension of RPS, which deliver packets to the CPU where application running. This week status: Performance does not improve, so I investigate RPS bottlenecks. The results of survey showed this is caused by "lock competition". Multi queue NIC has the same tendency as this. I may have to improve protocol stack with lock-free for performance improvement. --Kazuya Goda