From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751215AbWGCRb7 (ORCPT ); Mon, 3 Jul 2006 13:31:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751219AbWGCRb7 (ORCPT ); Mon, 3 Jul 2006 13:31:59 -0400 Received: from ug-out-1314.google.com ([66.249.92.174]:58153 "EHLO ug-out-1314.google.com") by vger.kernel.org with ESMTP id S1751215AbWGCRb6 (ORCPT ); Mon, 3 Jul 2006 13:31:58 -0400 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=c8eWn+8uE+RO6I4Gr9Ga7+GIOBYxMC/sun2D17VsiJAZPqAacEPFzFTPxmEp0P2Vp28VrRD5tbUeWnLP6Zf6gZzMUC/ohLffGPZGrGNJw/tFk/eXJeJSY2hJfKYTARPqESGeigcI2zs94CZUDgrFLf1naAjNxxljs/Fj1YAkqfE= Message-ID: <69304d110607031031j2b981ff5m571321deaf91f57d@mail.gmail.com> Date: Mon, 3 Jul 2006 19:31:57 +0200 From: "Antonio Vargas" To: "Nikita Danilov" , "Ananiev, Leonid I" , "Linux Kernel Mailing List" Subject: Re: [PATCH] mm: moving dirty pages balancing to pdfludh entirely In-Reply-To: <17570.26897.378682.724475@gargle.gargle.HOWL> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <6694B22B6436BC43B429958787E4549802394A55@mssmsx402nb> <17570.26897.378682.724475@gargle.gargle.HOWL> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On 6/28/06, Nikita Danilov wrote: > Ananiev, Leonid I writes: > > >From Leonid Ananiev > > Hello, > > > > > Moving dirty pages balancing from user to kernel thread pdfludh entirely > > reduces extra long write(2) latencies, increases performance. > > > > [...] > > > The benchmarks IOzone and Sysbench for file size 50% and 120% of > > RAM size on Pentium4, Xeon, Itanium have reported write and mix > > throughput increasing about 25%. The described Iozone > 0.1 sec write(2) > > Results are impressive. > > Wouldn't this interfere with current->backing_dev_info logic? This field > is set by __generic_file_aio_write_nolock() and checked by > may_write_to_queue() to force heavy writes to do more pageout. Maybe > pdflush threads should set this field too? > > > latencies are deleted. The condition writeback_in_progress() is tested > > earlier now. As a result extra pdflush works are not created and number > > of context switches increasing is inside variation limites. > > Nikita. Maybe we should keep the sync-write logic if there is only one online cpu, and thus avoiding extra context switches when they are not profitable? -- Greetz, Antonio Vargas aka winden of network