From patchwork Thu Apr 30 14:34:17 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jean-Philippe Brucker X-Patchwork-Id: 1280489 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=vger.kernel.org (client-ip=23.128.96.18; helo=vger.kernel.org; envelope-from=linux-pci-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=none dis=none) header.from=linaro.org Authentication-Results: ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=linaro.org header.i=@linaro.org header.a=rsa-sha256 header.s=google header.b=hdTx9q9/; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by ozlabs.org (Postfix) with ESMTP id 49CdNW5ltVz9sV9 for ; Fri, 1 May 2020 00:40:59 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728302AbgD3Okp (ORCPT ); Thu, 30 Apr 2020 10:40:45 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34328 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1728284AbgD3Okp (ORCPT ); Thu, 30 Apr 2020 10:40:45 -0400 Received: from mail-wm1-x344.google.com (mail-wm1-x344.google.com [IPv6:2a00:1450:4864:20::344]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 76A0DC035494 for ; Thu, 30 Apr 2020 07:40:41 -0700 (PDT) Received: by mail-wm1-x344.google.com with SMTP id g12so2207559wmh.3 for ; Thu, 30 Apr 2020 07:40:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=5zUk9NV0nU5TmSz+uWZW+0J3z6ZO8c45bE/F41UldBY=; b=hdTx9q9/ZBbgUC8/ufZMC68Zeis9W/RYxXDui8aFftSuMllKim0WGr4wx8hjminehP w1g/QaD2kaDN7e4b/Oot4dMemmy6yU9/xYXZzDgPDqvydC0M+3O369m72jEiXRiwoh4Q i9sPe28+qOx1cVjOwtZfZAPHopc0wdyPkjIQIMQ2c7FH+MZaoaRXEYOgYTEQXDFsaKUR fnu9Ofq0Ad+jqiOB9RX3dVIfO3pE3uTEbw4YMKfVwEHw08Ev0DJweclrvLf+op6gOheV b6oa+L4yDKBH7DkfS+f+u0eVik2GiUJxKykIvfc4MOaWRpfZjXdvgAuODWilHbJw3q8r Gh5w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=5zUk9NV0nU5TmSz+uWZW+0J3z6ZO8c45bE/F41UldBY=; b=VuzMj3PTMJZuPqMHi5cS7Ugwpa0uuB8a1otpXu08rf5TMhpQIfd31U7/2jLsXZL0E2 dbf+fQJr4/xQHlo9weIbr8MRLNsSUI7kXgeo62E6fx9wme0XsZ34MvfHcL7NGw4r1ihL /nwgq/CD3DLddn7HIPtQit+HWieaK7R9oWPIybLSxiPyw06gN0OyDAqIY4Z4ssbXA9g3 92KgNsgyGzvyvUYDGTef0kd8Jf+XC/Dvndfybqc5Vde7qNGFdIUd/JYC2Ux/yuykK1lU 5gQWaPRGOnPSP+PlCj3yOl9MDw5szy/dOJDRYK459rE//mO5rwgFR+S2m5PxRcDuPKmL J1Ag== X-Gm-Message-State: AGi0Pua+KF8h9lsA285O865ZWTJUnfb3q84zhTfPjQCbuEvVXiGmiblX jwp8DB1uyCWJiKOOfMiYXAyEsg== X-Google-Smtp-Source: APiQypLYnxQfwV9vUitaM0N+BAtQ0WTyZbQIbMQN53sHeaH1WhNxiOl1IRhWF/dZdB3Dd06HM1dwfw== X-Received: by 2002:a1c:32c7:: with SMTP id y190mr3503540wmy.13.1588257640161; Thu, 30 Apr 2020 07:40:40 -0700 (PDT) Received: from localhost.localdomain ([2001:171b:226e:c200:c43b:ef78:d083:b355]) by smtp.gmail.com with ESMTPSA id n2sm4153286wrt.33.2020.04.30.07.40.38 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 30 Apr 2020 07:40:39 -0700 (PDT) From: Jean-Philippe Brucker To: iommu@lists.linux-foundation.org, devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-pci@vger.kernel.org, linux-mm@kvack.org Cc: joro@8bytes.org, catalin.marinas@arm.com, will@kernel.org, robin.murphy@arm.com, kevin.tian@intel.com, baolu.lu@linux.intel.com, Jonathan.Cameron@huawei.com, jacob.jun.pan@linux.intel.com, christian.koenig@amd.com, felix.kuehling@amd.com, zhangfei.gao@linaro.org, jgg@ziepe.ca, xuzaibo@huawei.com, fenghua.yu@intel.com, hch@infradead.org, Jean-Philippe Brucker Subject: [PATCH v6 18/25] iommu/arm-smmu-v3: Hook up ATC invalidation to mm ops Date: Thu, 30 Apr 2020 16:34:17 +0200 Message-Id: <20200430143424.2787566-19-jean-philippe@linaro.org> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200430143424.2787566-1-jean-philippe@linaro.org> References: <20200430143424.2787566-1-jean-philippe@linaro.org> MIME-Version: 1.0 Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org The invalidate_range() notifier is called for any change to the address space. Perform the required ATC invalidations. Signed-off-by: Jean-Philippe Brucker --- drivers/iommu/arm-smmu-v3.c | 56 ++++++++++++++++++++++++++++++------- 1 file changed, 46 insertions(+), 10 deletions(-) diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c index 00e5b69bb81a5..c65937d953b5f 100644 --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -742,7 +742,7 @@ struct arm_smmu_mmu_notifier { struct arm_smmu_ctx_desc *cd; bool cleared; refcount_t refs; - struct arm_smmu_domain *domain; + struct arm_smmu_domain __rcu *domain; }; #define mn_to_smmu(mn) container_of(mn, struct arm_smmu_mmu_notifier, mn) @@ -2396,6 +2396,20 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size, size_t inval_grain_shift = 12; unsigned long page_start, page_end; + /* + * ATS and PASID: + * + * If substream_valid is clear, the PCIe TLP is sent without a PASID + * prefix. In that case all ATC entries within the address range are + * invalidated, including those that were requested with a PASID! There + * is no way to invalidate only entries without PASID. + * + * When using STRTAB_STE_1_S1DSS_SSID0 (reserving CD 0 for non-PASID + * traffic), translation requests without PASID create ATC entries + * without PASID, which must be invalidated with substream_valid clear. + * This has the unpleasant side-effect of invalidating all PASID-tagged + * ATC entries within the address range. + */ *cmd = (struct arm_smmu_cmdq_ent) { .opcode = CMDQ_OP_ATC_INV, .substream_valid = !!ssid, @@ -2439,12 +2453,12 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size, cmd->atc.size = log2_span; } -static int arm_smmu_atc_inv_master(struct arm_smmu_master *master) +static int arm_smmu_atc_inv_master(struct arm_smmu_master *master, int ssid) { int i; struct arm_smmu_cmdq_ent cmd; - arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd); + arm_smmu_atc_inv_to_cmd(ssid, 0, 0, &cmd); for (i = 0; i < master->num_sids; i++) { cmd.atc.sid = master->sids[i]; @@ -2958,7 +2972,7 @@ static void arm_smmu_disable_ats(struct arm_smmu_master *master) * ATC invalidation via the SMMU. */ wmb(); - arm_smmu_atc_inv_master(master); + arm_smmu_atc_inv_master(master, 0); atomic_dec(&smmu_domain->nr_ats_masters); } @@ -3187,7 +3201,22 @@ static void arm_smmu_mm_invalidate_range(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long start, unsigned long end) { - /* TODO: invalidate ATS */ + struct arm_smmu_mmu_notifier *smmu_mn = mn_to_smmu(mn); + struct arm_smmu_domain *smmu_domain; + + rcu_read_lock(); + smmu_domain = rcu_dereference(smmu_mn->domain); + if (smmu_domain) { + /* + * Ensure that mm->pasid is valid. Pairs with the + * smp_store_release() from rcu_assign_pointer() in + * __arm_smmu_sva_bind() + */ + smp_rmb(); + arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, start, + end - start + 1); + } + rcu_read_unlock(); } static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm) @@ -3201,7 +3230,8 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm) return; } - smmu_domain = smmu_mn->domain; + smmu_domain = rcu_dereference_protected(smmu_mn->domain, + lockdep_is_held(&arm_smmu_sva_lock)); /* * DMA may still be running. Keep the cd valid but disable @@ -3210,7 +3240,7 @@ static void arm_smmu_mm_release(struct mmu_notifier *mn, struct mm_struct *mm) arm_smmu_write_ctx_desc(smmu_domain, mm->pasid, &invalid_cd); arm_smmu_tlb_inv_asid(smmu_domain->smmu, smmu_mn->cd->asid); - /* TODO: invalidate ATS */ + arm_smmu_atc_inv_domain(smmu_domain, mm->pasid, 0, 0); smmu_mn->cleared = true; mutex_unlock(&arm_smmu_sva_lock); @@ -3251,7 +3281,8 @@ __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm) return ERR_CAST(mn); smmu_mn = mn_to_smmu(mn); - if (smmu_mn->domain) + if (rcu_dereference_protected(smmu_mn->domain, + lockdep_is_held(&arm_smmu_sva_lock))) refcount_inc(&smmu_mn->refs); bond = kzalloc(sizeof(*bond), GFP_KERNEL); @@ -3277,7 +3308,11 @@ __arm_smmu_sva_bind(struct device *dev, struct mm_struct *mm) bond->sva.dev = dev; list_add(&bond->list, &master->bonds); - smmu_mn->domain = smmu_domain; + /* + * Initialize domain last, since the invalidate() notifier assumes a + * valid mm->pasid after fetching a valid domain. + */ + rcu_assign_pointer(smmu_mn->domain, smmu_domain); return &bond->sva; err_free_pasid: @@ -3318,7 +3353,8 @@ static void __arm_smmu_sva_unbind(struct iommu_sva *handle) if (!smmu_mn->cleared) { arm_smmu_tlb_inv_asid(smmu_domain->smmu, smmu_mn->cd->asid); - /* TODO: invalidate ATS */ + arm_smmu_atc_inv_domain(smmu_domain, bond->mm->pasid, + 0, 0); } }