From owner-cvs-src@FreeBSD.ORG Thu Sep 15 19:22:06 2005 Return-Path: X-Original-To: cvs-src@FreeBSD.org Delivered-To: cvs-src@FreeBSD.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CFBF816A420; Thu, 15 Sep 2005 19:22:06 +0000 (GMT) (envelope-from jhb@FreeBSD.org) Received: from mv.twc.weather.com (mv.twc.weather.com [65.212.71.225]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2F62443D46; Thu, 15 Sep 2005 19:22:06 +0000 (GMT) (envelope-from jhb@FreeBSD.org) Received: from [10.50.41.233] (Not Verified[10.50.41.233]) by mv.twc.weather.com with NetIQ MailMarshal (v6, 0, 3, 8) id ; Thu, 15 Sep 2005 15:37:46 -0400 From: John Baldwin To: Scott Long Date: Thu, 15 Sep 2005 15:22:37 -0400 User-Agent: KMail/1.8 References: <200509151709.j8FH9hkm099887@repoman.freebsd.org> <200509151341.42862.jhb@FreeBSD.org> <4329C835.6020104@samsco.org> In-Reply-To: <4329C835.6020104@samsco.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200509151522.39043.jhb@FreeBSD.org> Cc: cvs-src@FreeBSD.org, src-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/alpha/alpha busdma_machdep.c X-BeenThere: cvs-src@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Sep 2005 19:22:07 -0000 On Thursday 15 September 2005 03:15 pm, Scott Long wrote: > John Baldwin wrote: > > On Thursday 15 September 2005 01:09 pm, John Baldwin wrote: > >>jhb 2005-09-15 17:09:42 UTC > >> > >> FreeBSD src repository > >> > >> Modified files: > >> sys/alpha/alpha busdma_machdep.c > >> Log: > >> Add a memory barrier for PREWRITE operations to ensure all writes by > >> the CPU have drained before further writes to kick off the operation. > > > > Other architectures with relaxed memory ordering probably also need thus > > such as ia64, powerpc, and perhaps amd64. sparc64 already has a barrier. > > BTW, it seems that the bus_dma code could be somewhat simplified if > > there was a sys/kern/subr_busdma.c that contained routines to implement > > bounce buffers that could be used by the MD code since the MD code that > > does do bounce pages tends to duplicate a lot of stuff between the > > different archs. > > Indeed it does, and if you want to work on it, please feel free to send > it to me for > review. I can tack it onto my todo list but it'll be buried. :) I was hoping someone else might be able to hack on it and pass it off to you for review. :) -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" = http://www.FreeBSD.org