From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 12 Feb 2001 14:06:46 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 12 Feb 2001 14:06:37 -0500 Received: from smtp1.cern.ch ([137.138.128.38]:16 "EHLO smtp1.cern.ch") by vger.kernel.org with ESMTP id ; Mon, 12 Feb 2001 14:06:04 -0500 To: Gérard Roudier Cc: Alan Cox , Ion Badulescu , Alan Cox , Donald Becker , Jeff Garzik , linux-kernel@vger.kernel.org Subject: Re: [PATCH] starfire reads irq before pci_enable_device. In-Reply-To: From: Jes Sorensen Date: 12 Feb 2001 20:01:48 +0100 In-Reply-To: Gérard Roudier's message of "Sat, 10 Feb 2001 09:48:41 +0100 (CET)" Message-ID: User-Agent: Gnus/5.070096 (Pterodactyl Gnus v0.96) Emacs/20.4 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >>>>> "Gérard" == Gérard Roudier writes: Gérard> On Fri, 9 Feb 2001, Alan Cox wrote: >> DMA to main memory normally invalidates those lines in the CPU >> cache rather than the cache snooping and updating its view of them. Gérard> In PCI, it is the Memory Write and Invalidate PCI transaction Gérard> that is intended to allow core-logics to optimize DMA this Gérard> way. For normal Memory Write PCI transactions or when the Gérard> core-logic is aliasing MWI to MW, the snooping may well Gérard> happen. All that stuff, very probably, varies a lot depending Gérard> on the core-logic. In fact one has to look out for this and disable the feature in some cases. On the acenic not disabling Memory Write and Invalidate costs ~20% on performance on some systems. Jes - 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://vger.kernel.org/lkml/