From owner-freebsd-current@FreeBSD.ORG Tue Jun 30 20:36:09 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 45B571065674 for ; Tue, 30 Jun 2009 20:36:09 +0000 (UTC) (envelope-from scottl@samsco.org) Received: from pooker.samsco.org (pooker.samsco.org [168.103.85.57]) by mx1.freebsd.org (Postfix) with ESMTP id EE0AB8FC16 for ; Tue, 30 Jun 2009 20:36:08 +0000 (UTC) (envelope-from scottl@samsco.org) Received: from phobos.local (pooker.samsco.org [168.103.85.57]) by pooker.samsco.org (8.14.2/8.14.2) with ESMTP id n5UKa5Tx054834; Tue, 30 Jun 2009 14:36:05 -0600 (MDT) (envelope-from scottl@samsco.org) Message-ID: <4A4A7735.9060301@samsco.org> Date: Tue, 30 Jun 2009 14:36:05 -0600 From: Scott Long User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en-US; rv:1.8.1.13) Gecko/20080313 SeaMonkey/1.1.9 MIME-Version: 1.0 To: =?ISO-8859-2?Q?Piotr_Zi=EAcik?= References: <200906251329.35200.kosmo@semihalf.com> <200906260929.40709.kosmo@semihalf.com> <4A44D4AD.8010307@samsco.org> <200906301529.42195.kosmo@semihalf.com> In-Reply-To: <200906301529.42195.kosmo@semihalf.com> X-Enigmail-Version: 0.95.6 Content-Type: text/plain; charset=ISO-8859-2; format=flowed Content-Transfer-Encoding: 8bit X-Spam-Status: No, score=-4.4 required=3.8 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on pooker.samsco.org Cc: freebsd-current@freebsd.org Subject: Re: [PATCH RFC]: Bus_dma eats all available memory X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 30 Jun 2009 20:36:09 -0000 Piotr Ziêcik wrote: > Friday 26 June 2009 16:01:17 Scott Long napisa³(a): >> Tags and maps should be allocated at driver initialization time, not >> every time a request comes in. The problem here isn't the MAX() test, >> it's that the MIN_ALLOC_COMP test is getting fooled because the tag >> keeps on getting recycled. The correct fix is likely to move the flag >> into the bounce zone object. But in general, you should not be >> allocating and freeing tags and maps so often, they are meant to have a >> long lifespan. >> > > I have fixed my driver and updated patch for bus_dma. Could I ask you for > review ? Patch was successfully tested on ARM. > This looks reasonable. I'll give it a closer review and commit it to the FreeBSD repo. Thanks a lot for working on it. Scott