From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755756AbZKDM2i (ORCPT ); Wed, 4 Nov 2009 07:28:38 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755751AbZKDM2h (ORCPT ); Wed, 4 Nov 2009 07:28:37 -0500 Received: from vpn.id2.novell.com ([195.33.99.129]:32937 "EHLO vpn.id2.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755736AbZKDM2h convert rfc822-to-8bit (ORCPT ); Wed, 4 Nov 2009 07:28:37 -0500 Message-Id: <4AF18187020000780001D8AA@vpn.id2.novell.com> X-Mailer: Novell GroupWise Internet Agent 8.0.1 Date: Wed, 04 Nov 2009 12:28:39 +0000 From: "Jan Beulich" To: , , Cc: "Takashi Iwai" , , Subject: [PATCH] x86: adjust GFP mask handling for coherent allocations (take 2) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8BIT Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Rather than forcing GFP flags and DMA mask to be inconsistent, GFP flags should be determined even for the fallback device through dma_alloc_coherent_mask()/dma_alloc_coherent_gfp_flags(). This restores 64-bit behavior as it was prior to commits 8965eb19386fdf5ccd0ef8b02593eb8560aa3416 and 4a367f3a9dbf2e7ffcee4702203479809236ee6e (not sure why there are two of them), where GFP_DMA was forced on for 32-bit, but not for 64-bit, with the slight adjustment that afaict even 32-bit doesn't need this without CONFIG_ISA. Signed-off-by: Jan Beulich Cc: Takashi Iwai Cc: Jesse Barnes --- arch/x86/include/asm/dma-mapping.h | 10 +++++++--- arch/x86/kernel/pci-dma.c | 6 ++---- 2 files changed, 9 insertions(+), 7 deletions(-) --- linux-2.6.32-rc6/arch/x86/include/asm/dma-mapping.h 2009-11-04 13:23:36.000000000 +0100 +++ 2.6.32-rc6-x86-dma-mask-gfp/arch/x86/include/asm/dma-mapping.h 2009-10-27 11:10:28.000000000 +0100 @@ -14,6 +14,12 @@ #include #include +#ifdef CONFIG_ISA +#define ISA_DMA_BIT_MASK DMA_BIT_MASK(24) +#else +#define ISA_DMA_BIT_MASK DMA_BIT_MASK(32) +#endif + extern dma_addr_t bad_dma_address; extern int iommu_merge; extern struct device x86_dma_fallback_dev; @@ -124,10 +130,8 @@ dma_alloc_coherent(struct device *dev, s if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) return memory; - if (!dev) { + if (!dev) dev = &x86_dma_fallback_dev; - gfp |= GFP_DMA; - } if (!is_device_dma_capable(dev)) return NULL; --- linux-2.6.32-rc6/arch/x86/kernel/pci-dma.c 2009-11-04 13:23:36.000000000 +0100 +++ 2.6.32-rc6-x86-dma-mask-gfp/arch/x86/kernel/pci-dma.c 2009-10-27 11:14:11.000000000 +0100 @@ -45,12 +45,10 @@ int iommu_pass_through __read_mostly; dma_addr_t bad_dma_address __read_mostly = 0; EXPORT_SYMBOL(bad_dma_address); -/* Dummy device used for NULL arguments (normally ISA). Better would - be probably a smaller DMA mask, but this is bug-to-bug compatible - to older i386. */ +/* Dummy device used for NULL arguments (normally ISA). */ struct device x86_dma_fallback_dev = { .init_name = "fallback device", - .coherent_dma_mask = DMA_BIT_MASK(32), + .coherent_dma_mask = ISA_DMA_BIT_MASK, .dma_mask = &x86_dma_fallback_dev.coherent_dma_mask, }; EXPORT_SYMBOL(x86_dma_fallback_dev);