From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753172AbZBCBwU (ORCPT ); Mon, 2 Feb 2009 20:52:20 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750977AbZBCBwK (ORCPT ); Mon, 2 Feb 2009 20:52:10 -0500 Received: from fgwmail6.fujitsu.co.jp ([192.51.44.36]:50101 "EHLO fgwmail6.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750799AbZBCBwJ (ORCPT ); Mon, 2 Feb 2009 20:52:09 -0500 Date: Tue, 03 Feb 2009 10:52:03 +0900 From: Yasunori Goto To: FUJITA Tomonori , mingo@elte.hu Subject: Re: [PATCH -tip 1/3] IA64: fix swiotlb alloc_coherent for non DMA_64BIT_MASK devices Cc: tony.luck@intel.com, linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org In-Reply-To: <20090203082825J.fujita.tomonori@lab.ntt.co.jp> References: <20090130184210.4A8E.E1E9C6FF@jp.fujitsu.com> <20090203082825J.fujita.tomonori@lab.ntt.co.jp> X-Mailer-Plugin: BkASPil for Becky!2 Ver.2.068 Message-Id: <20090203104439.8ADF.E1E9C6FF@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: Becky! ver. 2.48.02 [ja] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > On Fri, 30 Jan 2009 20:24:13 +0900 > Yasunori Goto wrote: > > > > > swiotlb_dma_ops is set by pci_swiotlb_init(). > > > > However, it seems to be not called when CONFIG_IA64_DIG/Generic configuration. > > > > > > You are talking about setting dma_ops to swiotlb_dma_ops? > > > > > > I think that swiotlb_dma_init() also sets dma_ops to swiotlb_dma_ops. > > > > Certainly. > > I overlooked the 7th patch of your unifying dma_ops patch set. > > ia64_swiotlb_alloc_coherent was called when I tested again. > > I'm sorry. > > > > However, GFP_DMA is still always on in ia64_swiotlb_alloc_coherent() > > regardless of the value of dev->coherent_dma_mask. > > > > After following patch applied, GFP_DMA is set correctly by your patch. > > Am I still overlooking other your patch? > > No, you are not. We need the following change. Somehow I thought that > I already made this change. > > Can you resend this to Ingo with the proper description and > Signed-off-by (you can add my Acked-by if you like)? Sure! Ingo-san. Could you apply the following patch? Bye. ----- Because dma_alloc_coherent() always required DMA zone even if DMA is NOT necessary, FUJITA Tomonori posted a patch to fix it. http://marc.info/?l=linux-ia64&m=123314730923356&w=2 However, this fix needs one more patch to fix completely. I tested and confirmed dma_alloc_coherent() returns correct zone after applied following patch. Signed-off-by: Yasunori Goto Acked-by: FUJITA Tomonori --- arch/ia64/include/asm/dma-mapping.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Index: dma_fix_test/arch/ia64/include/asm/dma-mapping.h =================================================================== --- dma_fix_test.orig/arch/ia64/include/asm/dma-mapping.h 2009-01-30 19:36:33.000000000 +0900 +++ dma_fix_test/arch/ia64/include/asm/dma-mapping.h 2009-01-30 20:20:06.000000000 +0900 @@ -71,7 +71,7 @@ static inline void *dma_alloc_coherent(s dma_addr_t *daddr, gfp_t gfp) { struct dma_mapping_ops *ops = platform_dma_get_ops(dev); - return ops->alloc_coherent(dev, size, daddr, gfp | GFP_DMA); + return ops->alloc_coherent(dev, size, daddr, gfp); } static inline void dma_free_coherent(struct device *dev, size_t size, -- Yasunori Goto