From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753517AbZK0N4J (ORCPT ); Fri, 27 Nov 2009 08:56:09 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753486AbZK0N4G (ORCPT ); Fri, 27 Nov 2009 08:56:06 -0500 Received: from tx2ehsobe003.messaging.microsoft.com ([65.55.88.13]:35942 "EHLO TX2EHSOBE005.bigfish.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753156AbZK0Nz7 (ORCPT ); Fri, 27 Nov 2009 08:55:59 -0500 X-SpamScore: 1 X-BigFish: VPS1(zzzz1202hzzz32i6bh61h) X-Spam-TCS-SCL: 0:0 X-WSS-ID: 0KTRTD9-01-6S2-02 X-M-MSG: From: Joerg Roedel To: Ingo Molnar CC: iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org, Joerg Roedel Subject: [PATCH 16/34] x86/amd-iommu: Make alloc_new_range aware of multiple IOMMUs Date: Fri, 27 Nov 2009 14:55:27 +0100 Message-ID: <1259330145-14865-17-git-send-email-joerg.roedel@amd.com> X-Mailer: git-send-email 1.6.5.3 In-Reply-To: <1259330145-14865-1-git-send-email-joerg.roedel@amd.com> References: <1259330145-14865-1-git-send-email-joerg.roedel@amd.com> X-OriginalArrivalTime: 27 Nov 2009 13:55:48.0295 (UTC) FILETIME=[52AF4D70:01CA6F69] MIME-Version: 1.0 Content-Type: text/plain X-Reverse-DNS: ausb3extmailp02.amd.com Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Since the assumption that an dma_ops domain is only bound to one IOMMU was given up we need to make alloc_new_range aware of it. Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu.c | 27 +++++++++++++++------------ 1 files changed, 15 insertions(+), 12 deletions(-) diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index da3f9d8..687f617 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -788,11 +788,11 @@ static u64 *fetch_pte(struct protection_domain *domain, * aperture in case of dma_ops domain allocation or address allocation * failure. */ -static int alloc_new_range(struct amd_iommu *iommu, - struct dma_ops_domain *dma_dom, +static int alloc_new_range(struct dma_ops_domain *dma_dom, bool populate, gfp_t gfp) { int index = dma_dom->aperture_size >> APERTURE_RANGE_SHIFT; + struct amd_iommu *iommu; int i; #ifdef CONFIG_IOMMU_STRESS @@ -832,14 +832,17 @@ static int alloc_new_range(struct amd_iommu *iommu, dma_dom->aperture_size += APERTURE_RANGE_SIZE; /* Intialize the exclusion range if necessary */ - if (iommu->exclusion_start && - iommu->exclusion_start >= dma_dom->aperture[index]->offset && - iommu->exclusion_start < dma_dom->aperture_size) { - unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT; - int pages = iommu_num_pages(iommu->exclusion_start, - iommu->exclusion_length, - PAGE_SIZE); - dma_ops_reserve_addresses(dma_dom, startpage, pages); + for_each_iommu(iommu) { + if (iommu->exclusion_start && + iommu->exclusion_start >= dma_dom->aperture[index]->offset + && iommu->exclusion_start < dma_dom->aperture_size) { + unsigned long startpage; + int pages = iommu_num_pages(iommu->exclusion_start, + iommu->exclusion_length, + PAGE_SIZE); + startpage = iommu->exclusion_start >> PAGE_SHIFT; + dma_ops_reserve_addresses(dma_dom, startpage, pages); + } } /* @@ -1143,7 +1146,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu) add_domain_to_list(&dma_dom->domain); - if (alloc_new_range(iommu, dma_dom, true, GFP_KERNEL)) + if (alloc_new_range(dma_dom, true, GFP_KERNEL)) goto free_dma_dom; /* @@ -1686,7 +1689,7 @@ retry: */ dma_dom->next_address = dma_dom->aperture_size; - if (alloc_new_range(iommu, dma_dom, false, GFP_ATOMIC)) + if (alloc_new_range(dma_dom, false, GFP_ATOMIC)) goto out; /* -- 1.6.5.3