From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.8 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id EE2E7C43A1D for ; Thu, 12 Jul 2018 06:19:32 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id B0EFE20C0C for ; Thu, 12 Jul 2018 06:19:32 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org B0EFE20C0C Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=huawei.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1732425AbeGLG1e (ORCPT ); Thu, 12 Jul 2018 02:27:34 -0400 Received: from szxga05-in.huawei.com ([45.249.212.191]:9223 "EHLO huawei.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1732191AbeGLG1d (ORCPT ); Thu, 12 Jul 2018 02:27:33 -0400 Received: from DGGEMS410-HUB.china.huawei.com (unknown [172.30.72.58]) by Forcepoint Email with ESMTP id 7AB07ACFC0DCF; Thu, 12 Jul 2018 14:19:15 +0800 (CST) Received: from localhost (10.177.23.164) by DGGEMS410-HUB.china.huawei.com (10.3.19.210) with Microsoft SMTP Server id 14.3.382.0; Thu, 12 Jul 2018 14:19:07 +0800 From: Zhen Lei To: Jean-Philippe Brucker , Robin Murphy , Will Deacon , Joerg Roedel , linux-arm-kernel , iommu , linux-kernel CC: Zhen Lei Subject: [PATCH v3 6/6] iommu/arm-smmu-v3: add bootup option "iommu_strict_mode" Date: Thu, 12 Jul 2018 14:18:32 +0800 Message-ID: <1531376312-2192-7-git-send-email-thunder.leizhen@huawei.com> X-Mailer: git-send-email 1.9.5.msysgit.0 In-Reply-To: <1531376312-2192-1-git-send-email-thunder.leizhen@huawei.com> References: <1531376312-2192-1-git-send-email-thunder.leizhen@huawei.com> MIME-Version: 1.0 Content-Type: text/plain X-Originating-IP: [10.177.23.164] X-CFilter-Loop: Reflected Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Because the non-strict mode introduces a vulnerability window, so add a bootup option to make the manager can choose which mode to be used. The default mode is IOMMU_STRICT. Signed-off-by: Zhen Lei --- Documentation/admin-guide/kernel-parameters.txt | 12 ++++++++++ drivers/iommu/arm-smmu-v3.c | 32 ++++++++++++++++++++++--- 2 files changed, 41 insertions(+), 3 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index efc7aa7..0cc80bc 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1720,6 +1720,18 @@ nobypass [PPC/POWERNV] Disable IOMMU bypass, using IOMMU for PCI devices. + iommu_strict_mode= [arm-smmu-v3] + 0 - strict mode + Make sure all related TLBs to be invalidated before the + memory released. + 1 - non-strict mode + Put off TLBs invalidation and release memory first. This mode + introduces a vlunerability window, an untrusted device can + access the reused memory because the TLBs may still valid. + Please take full consideration before choosing this mode. + Note that, VFIO is always use strict mode. + others - strict mode + iommu.passthrough= [ARM64] Configure DMA to bypass the IOMMU by default. Format: { "0" | "1" } diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 4a198a0..9b72fc4 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -631,6 +631,24 @@ struct arm_smmu_option_prop { { 0, NULL}, }; +static u32 iommu_strict_mode __read_mostly = IOMMU_STRICT; + +static int __init setup_iommu_strict_mode(char *str) +{ + u32 strict_mode = IOMMU_STRICT; + + get_option(&str, &strict_mode); + if (strict_mode == IOMMU_NON_STRICT) { + iommu_strict_mode = strict_mode; + pr_warn("WARNING: iommu non-strict mode is chose.\n" + "It's good for scatter-gather performance but lacks full isolation\n"); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); + } + + return 0; +} +early_param("iommu_strict_mode", setup_iommu_strict_mode); + static inline void __iomem *arm_smmu_page1_fixup(unsigned long offset, struct arm_smmu_device *smmu) { @@ -1441,7 +1459,7 @@ static bool arm_smmu_capable(enum iommu_cap cap) case IOMMU_CAP_NOEXEC: return true; case IOMMU_CAP_NON_STRICT: - return true; + return (iommu_strict_mode == IOMMU_NON_STRICT) ? true : false; default: return false; } @@ -1750,6 +1768,14 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev) return ret; } +static u32 arm_smmu_strict_mode(struct iommu_domain *domain) +{ + if (iommu_strict_mode == IOMMU_NON_STRICT) + return IOMMU_DOMAIN_STRICT_MODE(domain); + + return IOMMU_STRICT; +} + static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova, phys_addr_t paddr, size_t size, int prot) { @@ -1769,7 +1795,7 @@ static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova, if (!ops) return 0; - return ops->unmap(ops, iova | IOMMU_DOMAIN_STRICT_MODE(domain), size); + return ops->unmap(ops, iova | arm_smmu_strict_mode(domain), size); } static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain) @@ -1784,7 +1810,7 @@ static void arm_smmu_iotlb_sync(struct iommu_domain *domain) { struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu; - if (smmu && (IOMMU_DOMAIN_STRICT_MODE(domain) == IOMMU_STRICT)) + if (smmu && (arm_smmu_strict_mode(domain) == IOMMU_STRICT)) __arm_smmu_tlb_sync(smmu); } -- 1.8.3