From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755779AbYC1Lby (ORCPT ); Fri, 28 Mar 2008 07:31:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754835AbYC1Lbo (ORCPT ); Fri, 28 Mar 2008 07:31:44 -0400 Received: from one.firstfloor.org ([213.235.205.2]:57019 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753278AbYC1Lbn (ORCPT ); Fri, 28 Mar 2008 07:31:43 -0400 Date: Fri, 28 Mar 2008 12:35:08 +0100 From: Andi Kleen To: Pavel Machek Cc: Andi Kleen , Ingo Molnar , kernel list Subject: Re: agpgart: when telling user you'll corrupt his data, at least do it at KERN_CRIT Message-ID: <20080328113508.GP29105@one.firstfloor.org> References: <20080328112507.GA10774@elf.ucw.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20080328112507.GA10774@elf.ucw.cz> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Mar 28, 2008 at 12:25:07PM +0100, Pavel Machek wrote: > > IOMMU off means very bad stuff may happen, like data corruption on Actually swiotlb takes over then in this case. The printk really predates the introduction of swiotlb and could be actually removed. That is why swiotlb is always SELECTed for gart too. > your hard drives. At least tells users this is serious... No there shouldn't be any data corruption with the default options. Also especially hard disk drivers are expected (and generally do) check the dma_map_sg() etc. return values and bail out if the IOMMU fails e.g. due to overflow. -Andi