From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932621AbXGLQOi (ORCPT ); Thu, 12 Jul 2007 12:14:38 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758071AbXGLQOb (ORCPT ); Thu, 12 Jul 2007 12:14:31 -0400 Received: from extu-mxob-1.symantec.com ([216.10.194.28]:45733 "EHLO extu-mxob-1.symantec.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756868AbXGLQO3 (ORCPT ); Thu, 12 Jul 2007 12:14:29 -0400 Date: Thu, 12 Jul 2007 17:13:46 +0100 (BST) From: Hugh Dickins X-X-Sender: hugh@blonde.wat.veritas.com To: Bill Irwin cc: Ivan Kokshaysky , Bob Tracy , Michal Piotrowski , linux-kernel@vger.kernel.org, linux-alpha@vger.kernel.org, Andrew Morton Subject: Re: 2.6.22-rc6 bad page error In-Reply-To: <20070712134606.A10012@jurassic.park.msu.ru> Message-ID: References: <20070709185324.387ACDBA1@gherkin.frus.com> <20070711170245.A1148@jurassic.park.msu.ru> <20070712134606.A10012@jurassic.park.msu.ru> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Brightmail-Verdict: VlJEQwAAAAIAAAABAAAAAAAAAAEAAAAAAAAAB2luYm94AGxpbnV4LWtlcm5lbEB2Z2VyLmtlcm5lbC5vcmcAbGludXgtYWxwaGFAdmdlci5rZXJuZWwub3JnAGFrcG1AbGludXgtZm91bmRhdGlvbi5vcmcAaW5rQGp1cmFzc2ljLnBhcmsubXN1LnJ1AHdsaUBob2xvbW9ycGh5LmNvbQBtaWNoYWwuay5rLnBpb3Ryb3dza2lAZ21haWwuY29tAHJjdEBnaGVya2luLmZydXMuY29tAA== X-Brightmail-Tracker: AAAAAA== Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 12 Jul 2007, Ivan Kokshaysky wrote: > On Wed, Jul 11, 2007 at 06:07:59PM +0100, Hugh Dickins wrote: > > > > + return __pci_alloc_consistent(dev, size, dma, GFP_ATOMIC); > > > > I was going to ask you why that needs to be GFP_ATOMIC on Alpha. > > But find you're following the example of asm-generic and others. > > So really should be asking Sparc how it gets away with GFP_KERNEL. > > According to DMA-mapping.txt GFP_ATOMIC is mandatory for > pci_alloc_consistent() since it can be called from interrupt context. > So sparc code is probably incorrect... Hi Bill, we've veered away from the original topic, and are wondering how the sparc32 pci_alloc_consistent() gets away with using GFP_KERNEL, when GFP_ATOMIC seems to be required? Hugh