Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262294AbVBBT7A (ORCPT ); Wed, 2 Feb 2005 14:59:00 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262297AbVBBTq6 (ORCPT ); Wed, 2 Feb 2005 14:46:58 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:22477 "EHLO parcelfarce.linux.theplanet.co.uk") by vger.kernel.org with ESMTP id S262679AbVBBTkG (ORCPT ); Wed, 2 Feb 2005 14:40:06 -0500 Date: Wed, 2 Feb 2005 14:31:10 -0200 From: Marcelo Tosatti To: Christoph Lameter Cc: David Woodhouse , linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org Subject: Re: A scrub daemon (prezeroing) Message-ID: <20050202163110.GB23132@logos.cnet> References: <1106828124.19262.45.camel@hades.cambridge.redhat.com> <20050202153256.GA19615@logos.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.5.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1469 Lines: 36 On Wed, Feb 02, 2005 at 11:05:14AM -0800, Christoph Lameter wrote: > On Wed, 2 Feb 2005, Marcelo Tosatti wrote: > > > Sounds very interesting idea to me. Guess it depends on whether the cost of > > DMA write for memory zeroing, which is memory architecture/DMA engine dependant, > > offsets the cost of CPU zeroing. > > > > Do you have any thoughts on that? > > > > I wonder if such thing (using unrelated devices DMA engine's for zeroing) ever been > > done on other OS'es? > > > > AFAIK SGI's BTE is special purpose hardware for memory zeroing. > > Nope the BTE is a block transfer engine. Its an inter numa node DMA thing > that is being abused to zero blocks. Ah, OK. Is there a driver for normal BTE operation or is not kernel-controlled ? > The same can be done with most DMA chips (I have done so on some other > platforms not on i386) Nice! What kind of DMA chip was that and through which kind of bus was it connected to CPU ? I wonder what has to be done to have active DMA engines be abused for zeroing when idle and what are the implications of that. Some kind of notification mechanism is necessary to inform idleness ? Someone should try implementing the zeroing driver for a fast x86 PCI device. :) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/