Merge branch 'for-joerg/arm-smmu/updates' of git://git.kernel.org/pub/scm/linux/kernel/git/will/linux into arm/smmu

6 files changed, 445 insertions, 228 deletions

diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 6ee3a25ae731..c88cfa7522b2 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -23,7 +23,7 @@ config IOMMU_IO_PGTABLE
 config IOMMU_IO_PGTABLE_LPAE
 	bool "ARMv7/v8 Long Descriptor Format"
 	select IOMMU_IO_PGTABLE
-	depends on HAS_DMA && (ARM || ARM64 || COMPILE_TEST)
+	depends on HAS_DMA && (ARM || ARM64 || (COMPILE_TEST && !GENERIC_ATOMIC64))
 	help
 	  Enable support for the ARM long descriptor pagetable format.
 	  This allocator supports 4K/2M/1G, 16K/32M and 64K/512M page
diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
index 380969aa60d5..568c400eeaed 100644
--- a/drivers/iommu/arm-smmu-v3.c
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -408,10 +408,20 @@
 
 /* High-level queue structures */
 #define ARM_SMMU_POLL_TIMEOUT_US	100
+#define ARM_SMMU_CMDQ_DRAIN_TIMEOUT_US	1000000 /* 1s! */
 
 #define MSI_IOVA_BASE			0x8000000
 #define MSI_IOVA_LENGTH			0x100000
 
+/* Until ACPICA headers cover IORT rev. C */
+#ifndef ACPI_IORT_SMMU_HISILICON_HI161X
+#define ACPI_IORT_SMMU_HISILICON_HI161X		0x1
+#endif
+
+#ifndef ACPI_IORT_SMMU_V3_CAVIUM_CN99XX
+#define ACPI_IORT_SMMU_V3_CAVIUM_CN99XX		0x2
+#endif
+
 static bool disable_bypass;
 module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
 MODULE_PARM_DESC(disable_bypass,
@@ -597,6 +607,7 @@ struct arm_smmu_device {
 	u32				features;
 
 #define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
+#define ARM_SMMU_OPT_PAGE0_REGS_ONLY	(1 << 1)
 	u32				options;
 
 	struct arm_smmu_cmdq		cmdq;
@@ -604,6 +615,7 @@ struct arm_smmu_device {
 	struct arm_smmu_priq		priq;
 
 	int				gerr_irq;
+	int				combined_irq;
 
 	unsigned long			ias; /* IPA */
 	unsigned long			oas; /* PA */
@@ -645,7 +657,6 @@ struct arm_smmu_domain {
 	struct mutex			init_mutex; /* Protects smmu pointer */
 
 	struct io_pgtable_ops		*pgtbl_ops;
-	spinlock_t			pgtbl_lock;
 
 	enum arm_smmu_domain_stage	stage;
 	union {
@@ -663,9 +674,20 @@ struct arm_smmu_option_prop {
 
 static struct arm_smmu_option_prop arm_smmu_options[] = {
 	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
+	{ ARM_SMMU_OPT_PAGE0_REGS_ONLY, "cavium,cn9900-broken-page1-regspace"},
 	{ 0, NULL},
 };
 
+static inline void __iomem *arm_smmu_page1_fixup(unsigned long offset,
+						 struct arm_smmu_device *smmu)
+{
+	if ((offset > SZ_64K) &&
+	    (smmu->options & ARM_SMMU_OPT_PAGE0_REGS_ONLY))
+		offset -= SZ_64K;
+
+	return smmu->base + offset;
+}
+
 static struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom)
 {
 	return container_of(dom, struct arm_smmu_domain, domain);
@@ -737,7 +759,13 @@ static void queue_inc_prod(struct arm_smmu_queue *q)
  */
 static int queue_poll_cons(struct arm_smmu_queue *q, bool drain, bool wfe)
 {
-	ktime_t timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
+	ktime_t timeout;
+	unsigned int delay = 1;
+
+	/* Wait longer if it's queue drain */
+	timeout = ktime_add_us(ktime_get(), drain ?
+					    ARM_SMMU_CMDQ_DRAIN_TIMEOUT_US :
+					    ARM_SMMU_POLL_TIMEOUT_US);
 
 	while (queue_sync_cons(q), (drain ? !queue_empty(q) : queue_full(q))) {
 		if (ktime_compare(ktime_get(), timeout) > 0)
@@ -747,7 +775,8 @@ static int queue_poll_cons(struct arm_smmu_queue *q, bool drain, bool wfe)
 			wfe();
 		} else {
 			cpu_relax();
-			udelay(1);
+			udelay(delay);
+			delay *= 2;
 		}
 	}
 
@@ -1302,6 +1331,24 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
 	return IRQ_HANDLED;
 }
 
+static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
+{
+	struct arm_smmu_device *smmu = dev;
+
+	arm_smmu_evtq_thread(irq, dev);
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		arm_smmu_priq_thread(irq, dev);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
+{
+	arm_smmu_gerror_handler(irq, dev);
+	arm_smmu_cmdq_sync_handler(irq, dev);
+	return IRQ_WAKE_THREAD;
+}
+
 /* IO_PGTABLE API */
 static void __arm_smmu_tlb_sync(struct arm_smmu_device *smmu)
 {
@@ -1406,7 +1453,6 @@ static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
 	}
 
 	mutex_init(&smmu_domain->init_mutex);
-	spin_lock_init(&smmu_domain->pgtbl_lock);
 	return &smmu_domain->domain;
 }
 
@@ -1555,6 +1601,9 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain)
 		.iommu_dev	= smmu->dev,
 	};
 
+	if (smmu->features & ARM_SMMU_FEAT_COHERENCY)
+		pgtbl_cfg.quirks = IO_PGTABLE_QUIRK_NO_DMA;
+
 	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
 	if (!pgtbl_ops)
 		return -ENOMEM;
@@ -1675,44 +1724,29 @@ out_unlock:
 static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
 			phys_addr_t paddr, size_t size, int prot)
 {
-	int ret;
-	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
 	if (!ops)
 		return -ENODEV;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
-	ret = ops->map(ops, iova, paddr, size, prot);
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
-	return ret;
+	return ops->map(ops, iova, paddr, size, prot);
 }
 
 static size_t
 arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
 {
-	size_t ret;
-	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
 	if (!ops)
 		return 0;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
-	ret = ops->unmap(ops, iova, size);
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
-	return ret;
+	return ops->unmap(ops, iova, size);
 }
 
 static phys_addr_t
 arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
 {
-	phys_addr_t ret;
-	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
 	if (domain->type == IOMMU_DOMAIN_IDENTITY)
 		return iova;
@@ -1720,11 +1754,7 @@ arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
 	if (!ops)
 		return 0;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
-	ret = ops->iova_to_phys(ops, iova);
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
-
-	return ret;
+	return ops->iova_to_phys(ops, iova);
 }
 
 static struct platform_driver arm_smmu_driver;
@@ -1961,8 +1991,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
 		return -ENOMEM;
 	}
 
-	q->prod_reg	= smmu->base + prod_off;
-	q->cons_reg	= smmu->base + cons_off;
+	q->prod_reg	= arm_smmu_page1_fixup(prod_off, smmu);
+	q->cons_reg	= arm_smmu_page1_fixup(cons_off, smmu);
 	q->ent_dwords	= dwords;
 
 	q->q_base  = Q_BASE_RWA;
@@ -2218,18 +2248,9 @@ static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
 	devm_add_action(dev, arm_smmu_free_msis, dev);
 }
 
-static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
+static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 {
-	int ret, irq;
-	u32 irqen_flags = IRQ_CTRL_EVTQ_IRQEN | IRQ_CTRL_GERROR_IRQEN;
-
-	/* Disable IRQs first */
-	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_IRQ_CTRL,
-				      ARM_SMMU_IRQ_CTRLACK);
-	if (ret) {
-		dev_err(smmu->dev, "failed to disable irqs\n");
-		return ret;
-	}
+	int irq, ret;
 
 	arm_smmu_setup_msis(smmu);
 
@@ -2272,10 +2293,41 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
 			if (ret < 0)
 				dev_warn(smmu->dev,
 					 "failed to enable priq irq\n");
-			else
-				irqen_flags |= IRQ_CTRL_PRIQ_IRQEN;
 		}
 	}
+}
+
+static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
+{
+	int ret, irq;
+	u32 irqen_flags = IRQ_CTRL_EVTQ_IRQEN | IRQ_CTRL_GERROR_IRQEN;
+
+	/* Disable IRQs first */
+	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_IRQ_CTRL,
+				      ARM_SMMU_IRQ_CTRLACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to disable irqs\n");
+		return ret;
+	}
+
+	irq = smmu->combined_irq;
+	if (irq) {
+		/*
+		 * Cavium ThunderX2 implementation doesn't not support unique
+		 * irq lines. Use single irq line for all the SMMUv3 interrupts.
+		 */
+		ret = devm_request_threaded_irq(smmu->dev, irq,
+					arm_smmu_combined_irq_handler,
+					arm_smmu_combined_irq_thread,
+					IRQF_ONESHOT,
+					"arm-smmu-v3-combined-irq", smmu);
+		if (ret < 0)
+			dev_warn(smmu->dev, "failed to enable combined irq\n");
+	} else
+		arm_smmu_setup_unique_irqs(smmu);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		irqen_flags |= IRQ_CTRL_PRIQ_IRQEN;
 
 	/* Enable interrupt generation on the SMMU */
 	ret = arm_smmu_write_reg_sync(smmu, irqen_flags,
@@ -2363,8 +2415,10 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
 
 	/* Event queue */
 	writeq_relaxed(smmu->evtq.q.q_base, smmu->base + ARM_SMMU_EVTQ_BASE);
-	writel_relaxed(smmu->evtq.q.prod, smmu->base + ARM_SMMU_EVTQ_PROD);
-	writel_relaxed(smmu->evtq.q.cons, smmu->base + ARM_SMMU_EVTQ_CONS);
+	writel_relaxed(smmu->evtq.q.prod,
+		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_PROD, smmu));
+	writel_relaxed(smmu->evtq.q.cons,
+		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_CONS, smmu));
 
 	enables |= CR0_EVTQEN;
 	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
@@ -2379,9 +2433,9 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
 		writeq_relaxed(smmu->priq.q.q_base,
 			       smmu->base + ARM_SMMU_PRIQ_BASE);
 		writel_relaxed(smmu->priq.q.prod,
-			       smmu->base + ARM_SMMU_PRIQ_PROD);
+			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_PROD, smmu));
 		writel_relaxed(smmu->priq.q.cons,
-			       smmu->base + ARM_SMMU_PRIQ_CONS);
+			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_CONS, smmu));
 
 		enables |= CR0_PRIQEN;
 		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
@@ -2605,6 +2659,20 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 }
 
 #ifdef CONFIG_ACPI
+static void acpi_smmu_get_options(u32 model, struct arm_smmu_device *smmu)
+{
+	switch (model) {
+	case ACPI_IORT_SMMU_V3_CAVIUM_CN99XX:
+		smmu->options |= ARM_SMMU_OPT_PAGE0_REGS_ONLY;
+		break;
+	case ACPI_IORT_SMMU_HISILICON_HI161X:
+		smmu->options |= ARM_SMMU_OPT_SKIP_PREFETCH;
+		break;
+	}
+
+	dev_notice(smmu->dev, "option mask 0x%x\n", smmu->options);
+}
+
 static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
 				      struct arm_smmu_device *smmu)
 {
@@ -2617,6 +2685,8 @@ static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
 	/* Retrieve SMMUv3 specific data */
 	iort_smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
 
+	acpi_smmu_get_options(iort_smmu->model, smmu);
+
 	if (iort_smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE)
 		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
 
@@ -2652,6 +2722,14 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
 	return ret;
 }
 
+static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
+{
+	if (smmu->options & ARM_SMMU_OPT_PAGE0_REGS_ONLY)
+		return SZ_64K;
+	else
+		return SZ_128K;
+}
+
 static int arm_smmu_device_probe(struct platform_device *pdev)
 {
 	int irq, ret;
@@ -2668,9 +2746,20 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 	}
 	smmu->dev = dev;
 
+	if (dev->of_node) {
+		ret = arm_smmu_device_dt_probe(pdev, smmu);
+	} else {
+		ret = arm_smmu_device_acpi_probe(pdev, smmu);
+		if (ret == -ENODEV)
+			return ret;
+	}
+
+	/* Set bypass mode according to firmware probing result */
+	bypass = !!ret;
+
 	/* Base address */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (resource_size(res) + 1 < SZ_128K) {
+	if (resource_size(res) + 1 < arm_smmu_resource_size(smmu)) {
 		dev_err(dev, "MMIO region too small (%pr)\n", res);
 		return -EINVAL;
 	}
@@ -2681,33 +2770,27 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 		return PTR_ERR(smmu->base);
 
 	/* Interrupt lines */
-	irq = platform_get_irq_byname(pdev, "eventq");
-	if (irq > 0)
-		smmu->evtq.q.irq = irq;
 
-	irq = platform_get_irq_byname(pdev, "priq");
+	irq = platform_get_irq_byname(pdev, "combined");
 	if (irq > 0)
-		smmu->priq.q.irq = irq;
+		smmu->combined_irq = irq;
+	else {
+		irq = platform_get_irq_byname(pdev, "eventq");
+		if (irq > 0)
+			smmu->evtq.q.irq = irq;
 
-	irq = platform_get_irq_byname(pdev, "cmdq-sync");
-	if (irq > 0)
-		smmu->cmdq.q.irq = irq;
+		irq = platform_get_irq_byname(pdev, "priq");
+		if (irq > 0)
+			smmu->priq.q.irq = irq;
 
-	irq = platform_get_irq_byname(pdev, "gerror");
-	if (irq > 0)
-		smmu->gerr_irq = irq;
+		irq = platform_get_irq_byname(pdev, "cmdq-sync");
+		if (irq > 0)
+			smmu->cmdq.q.irq = irq;
 
-	if (dev->of_node) {
-		ret = arm_smmu_device_dt_probe(pdev, smmu);
-	} else {
-		ret = arm_smmu_device_acpi_probe(pdev, smmu);
-		if (ret == -ENODEV)
-			return ret;
+		irq = platform_get_irq_byname(pdev, "gerror");
+		if (irq > 0)
+			smmu->gerr_irq = irq;
 	}
-
-	/* Set bypass mode according to firmware probing result */
-	bypass = !!ret;
-
 	/* Probe the h/w */
 	ret = arm_smmu_device_hw_probe(smmu);
 	if (ret)
@@ -2736,6 +2819,10 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
 
 	ret = iommu_device_register(&smmu->iommu);
+	if (ret) {
+		dev_err(dev, "Failed to register iommu\n");
+		return ret;
+	}
 
 #ifdef CONFIG_PCI
 	if (pci_bus_type.iommu_ops != &arm_smmu_ops) {
@@ -2768,7 +2855,7 @@ static int arm_smmu_device_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static struct of_device_id arm_smmu_of_match[] = {
+static const struct of_device_id arm_smmu_of_match[] = {
 	{ .compatible = "arm,smmu-v3", },
 	{ },
 };
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 7ec30b08b3bd..bc89b4d6c043 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -312,6 +312,14 @@ enum arm_smmu_implementation {
 	CAVIUM_SMMUV2,
 };
 
+/* Until ACPICA headers cover IORT rev. C */
+#ifndef ACPI_IORT_SMMU_CORELINK_MMU401
+#define ACPI_IORT_SMMU_CORELINK_MMU401	0x4
+#endif
+#ifndef ACPI_IORT_SMMU_CAVIUM_THUNDERX
+#define ACPI_IORT_SMMU_CAVIUM_THUNDERX	0x5
+#endif
+
 struct arm_smmu_s2cr {
 	struct iommu_group		*group;
 	int				count;
@@ -425,10 +433,10 @@ enum arm_smmu_domain_stage {
 struct arm_smmu_domain {
 	struct arm_smmu_device		*smmu;
 	struct io_pgtable_ops		*pgtbl_ops;
-	spinlock_t			pgtbl_lock;
 	struct arm_smmu_cfg		cfg;
 	enum arm_smmu_domain_stage	stage;
 	struct mutex			init_mutex; /* Protects smmu pointer */
+	spinlock_t			cb_lock; /* Serialises ATS1* ops */
 	struct iommu_domain		domain;
 };
 
@@ -1010,6 +1018,9 @@ static int arm_smmu_init_domain_context(struct iommu_domain *domain,
 		.iommu_dev	= smmu->dev,
 	};
 
+	if (smmu->features & ARM_SMMU_FEAT_COHERENT_WALK)
+		pgtbl_cfg.quirks = IO_PGTABLE_QUIRK_NO_DMA;
+
 	smmu_domain->smmu = smmu;
 	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
 	if (!pgtbl_ops) {
@@ -1102,7 +1113,7 @@ static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
 	}
 
 	mutex_init(&smmu_domain->init_mutex);
-	spin_lock_init(&smmu_domain->pgtbl_lock);
+	spin_lock_init(&smmu_domain->cb_lock);
 
 	return &smmu_domain->domain;
 }
@@ -1380,35 +1391,23 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
 			phys_addr_t paddr, size_t size, int prot)
 {
-	int ret;
-	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops= smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
 	if (!ops)
 		return -ENODEV;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
-	ret = ops->map(ops, iova, paddr, size, prot);
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
-	return ret;
+	return ops->map(ops, iova, paddr, size, prot);
 }
 
 static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
 			     size_t size)
 {
-	size_t ret;
-	unsigned long flags;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops= smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
 
 	if (!ops)
 		return 0;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
-	ret = ops->unmap(ops, iova, size);
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
-	return ret;
+	return ops->unmap(ops, iova, size);
 }
 
 static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
@@ -1422,10 +1421,11 @@ static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
 	void __iomem *cb_base;
 	u32 tmp;
 	u64 phys;
-	unsigned long va;
+	unsigned long va, flags;
 
 	cb_base = ARM_SMMU_CB(smmu, cfg->cbndx);
 
+	spin_lock_irqsave(&smmu_domain->cb_lock, flags);
 	/* ATS1 registers can only be written atomically */
 	va = iova & ~0xfffUL;
 	if (smmu->version == ARM_SMMU_V2)
@@ -1435,6 +1435,7 @@ static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
 
 	if (readl_poll_timeout_atomic(cb_base + ARM_SMMU_CB_ATSR, tmp,
 				      !(tmp & ATSR_ACTIVE), 5, 50)) {
+		spin_unlock_irqrestore(&smmu_domain->cb_lock, flags);
 		dev_err(dev,
 			"iova to phys timed out on %pad. Falling back to software table walk.\n",
 			&iova);
@@ -1442,6 +1443,7 @@ static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
 	}
 
 	phys = readq_relaxed(cb_base + ARM_SMMU_CB_PAR);
+	spin_unlock_irqrestore(&smmu_domain->cb_lock, flags);
 	if (phys & CB_PAR_F) {
 		dev_err(dev, "translation fault!\n");
 		dev_err(dev, "PAR = 0x%llx\n", phys);
@@ -1454,10 +1456,8 @@ static phys_addr_t arm_smmu_iova_to_phys_hard(struct iommu_domain *domain,
 static phys_addr_t arm_smmu_iova_to_phys(struct iommu_domain *domain,
 					dma_addr_t iova)
 {
-	phys_addr_t ret;
-	unsigned long flags;
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops= smmu_domain->pgtbl_ops;
+	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
 
 	if (domain->type == IOMMU_DOMAIN_IDENTITY)
 		return iova;
@@ -1465,17 +1465,11 @@ static phys_addr_t arm_smmu_iova_to_phys(struct iommu_domain *domain,
 	if (!ops)
 		return 0;
 
-	spin_lock_irqsave(&smmu_domain->pgtbl_lock, flags);
 	if (smmu_domain->smmu->features & ARM_SMMU_FEAT_TRANS_OPS &&
-			smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		ret = arm_smmu_iova_to_phys_hard(domain, iova);
-	} else {
-		ret = ops->iova_to_phys(ops, iova);
-	}
-
-	spin_unlock_irqrestore(&smmu_domain->pgtbl_lock, flags);
+			smmu_domain->stage == ARM_SMMU_DOMAIN_S1)
+		return arm_smmu_iova_to_phys_hard(domain, iova);
 
-	return ret;
+	return ops->iova_to_phys(ops, iova);
 }
 
 static bool arm_smmu_capable(enum iommu_cap cap)
@@ -2073,6 +2067,10 @@ static int acpi_smmu_get_data(u32 model, struct arm_smmu_device *smmu)
 		smmu->version = ARM_SMMU_V1;
 		smmu->model = GENERIC_SMMU;
 		break;
+	case ACPI_IORT_SMMU_CORELINK_MMU401:
+		smmu->version = ARM_SMMU_V1_64K;
+		smmu->model = GENERIC_SMMU;
+		break;
 	case ACPI_IORT_SMMU_V2:
 		smmu->version = ARM_SMMU_V2;
 		smmu->model = GENERIC_SMMU;
@@ -2081,6 +2079,10 @@ static int acpi_smmu_get_data(u32 model, struct arm_smmu_device *smmu)
 		smmu->version = ARM_SMMU_V2;
 		smmu->model = ARM_MMU500;
 		break;
+	case ACPI_IORT_SMMU_CAVIUM_THUNDERX:
+		smmu->version = ARM_SMMU_V2;
+		smmu->model = CAVIUM_SMMUV2;
+		break;
 	default:
 		ret = -ENODEV;
 	}
diff --git a/drivers/iommu/io-pgtable-arm-v7s.c b/drivers/iommu/io-pgtable-arm-v7s.c
index 8d6ca28c3e1f..af330f513653 100644
--- a/drivers/iommu/io-pgtable-arm-v7s.c
+++ b/drivers/iommu/io-pgtable-arm-v7s.c
@@ -32,6 +32,7 @@
 
 #define pr_fmt(fmt)	"arm-v7s io-pgtable: " fmt
 
+#include <linux/atomic.h>
 #include <linux/dma-mapping.h>
 #include <linux/gfp.h>
 #include <linux/iommu.h>
@@ -39,6 +40,7 @@
 #include <linux/kmemleak.h>
 #include <linux/sizes.h>
 #include <linux/slab.h>
+#include <linux/spinlock.h>
 #include <linux/types.h>
 
 #include <asm/barrier.h>
@@ -92,7 +94,8 @@
 #define ARM_V7S_PTE_TYPE_CONT_PAGE	0x1
 
 #define ARM_V7S_PTE_IS_VALID(pte)	(((pte) & 0x3) != 0)
-#define ARM_V7S_PTE_IS_TABLE(pte, lvl)	(lvl == 1 && ((pte) & ARM_V7S_PTE_TYPE_TABLE))
+#define ARM_V7S_PTE_IS_TABLE(pte, lvl) \
+	((lvl) == 1 && (((pte) & 0x3) == ARM_V7S_PTE_TYPE_TABLE))
 
 /* Page table bits */
 #define ARM_V7S_ATTR_XN(lvl)		BIT(4 * (2 - (lvl)))
@@ -167,6 +170,7 @@ struct arm_v7s_io_pgtable {
 
 	arm_v7s_iopte		*pgd;
 	struct kmem_cache	*l2_tables;
+	spinlock_t		split_lock;
 };
 
 static dma_addr_t __arm_v7s_dma_addr(void *pages)
@@ -186,7 +190,8 @@ static arm_v7s_iopte *iopte_deref(arm_v7s_iopte pte, int lvl)
 static void *__arm_v7s_alloc_table(int lvl, gfp_t gfp,
 				   struct arm_v7s_io_pgtable *data)
 {
-	struct device *dev = data->iop.cfg.iommu_dev;
+	struct io_pgtable_cfg *cfg = &data->iop.cfg;
+	struct device *dev = cfg->iommu_dev;
 	dma_addr_t dma;
 	size_t size = ARM_V7S_TABLE_SIZE(lvl);
 	void *table = NULL;
@@ -195,7 +200,7 @@ static void *__arm_v7s_alloc_table(int lvl, gfp_t gfp,
 		table = (void *)__get_dma_pages(__GFP_ZERO, get_order(size));
 	else if (lvl == 2)
 		table = kmem_cache_zalloc(data->l2_tables, gfp | GFP_DMA);
-	if (table && !selftest_running) {
+	if (table && !(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA)) {
 		dma = dma_map_single(dev, table, size, DMA_TO_DEVICE);
 		if (dma_mapping_error(dev, dma))
 			goto out_free;
@@ -224,10 +229,11 @@ out_free:
 static void __arm_v7s_free_table(void *table, int lvl,
 				 struct arm_v7s_io_pgtable *data)
 {
-	struct device *dev = data->iop.cfg.iommu_dev;
+	struct io_pgtable_cfg *cfg = &data->iop.cfg;
+	struct device *dev = cfg->iommu_dev;
 	size_t size = ARM_V7S_TABLE_SIZE(lvl);
 
-	if (!selftest_running)
+	if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA))
 		dma_unmap_single(dev, __arm_v7s_dma_addr(table), size,
 				 DMA_TO_DEVICE);
 	if (lvl == 1)
@@ -239,7 +245,7 @@ static void __arm_v7s_free_table(void *table, int lvl,
 static void __arm_v7s_pte_sync(arm_v7s_iopte *ptep, int num_entries,
 			       struct io_pgtable_cfg *cfg)
 {
-	if (selftest_running)
+	if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA))
 		return;
 
 	dma_sync_single_for_device(cfg->iommu_dev, __arm_v7s_dma_addr(ptep),
@@ -280,6 +286,13 @@ static arm_v7s_iopte arm_v7s_prot_to_pte(int prot, int lvl,
 	else if (prot & IOMMU_CACHE)
 		pte |= ARM_V7S_ATTR_B | ARM_V7S_ATTR_C;
 
+	pte |= ARM_V7S_PTE_TYPE_PAGE;
+	if (lvl == 1 && (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS))
+		pte |= ARM_V7S_ATTR_NS_SECTION;
+
+	if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_MTK_4GB)
+		pte |= ARM_V7S_ATTR_MTK_4GB;
+
 	return pte;
 }
 
@@ -352,7 +365,7 @@ static int arm_v7s_init_pte(struct arm_v7s_io_pgtable *data,
 			    int lvl, int num_entries, arm_v7s_iopte *ptep)
 {
 	struct io_pgtable_cfg *cfg = &data->iop.cfg;
-	arm_v7s_iopte pte = arm_v7s_prot_to_pte(prot, lvl, cfg);
+	arm_v7s_iopte pte;
 	int i;
 
 	for (i = 0; i < num_entries; i++)
@@ -374,13 +387,7 @@ static int arm_v7s_init_pte(struct arm_v7s_io_pgtable *data,
 			return -EEXIST;
 		}
 
-	pte |= ARM_V7S_PTE_TYPE_PAGE;
-	if (lvl == 1 && (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS))
-		pte |= ARM_V7S_ATTR_NS_SECTION;
-
-	if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_MTK_4GB)
-		pte |= ARM_V7S_ATTR_MTK_4GB;
-
+	pte = arm_v7s_prot_to_pte(prot, lvl, cfg);
 	if (num_entries > 1)
 		pte = arm_v7s_pte_to_cont(pte, lvl);
 
@@ -390,6 +397,30 @@ static int arm_v7s_init_pte(struct arm_v7s_io_pgtable *data,
 	return 0;
 }
 
+static arm_v7s_iopte arm_v7s_install_table(arm_v7s_iopte *table,
+					   arm_v7s_iopte *ptep,
+					   arm_v7s_iopte curr,
+					   struct io_pgtable_cfg *cfg)
+{
+	arm_v7s_iopte old, new;
+
+	new = virt_to_phys(table) | ARM_V7S_PTE_TYPE_TABLE;
+	if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS)
+		new |= ARM_V7S_ATTR_NS_TABLE;
+
+	/*
+	 * Ensure the table itself is visible before its PTE can be.
+	 * Whilst we could get away with cmpxchg64_release below, this
+	 * doesn't have any ordering semantics when !CONFIG_SMP.
+	 */
+	dma_wmb();
+
+	old = cmpxchg_relaxed(ptep, curr, new);
+	__arm_v7s_pte_sync(ptep, 1, cfg);
+
+	return old;
+}
+
 static int __arm_v7s_map(struct arm_v7s_io_pgtable *data, unsigned long iova,
 			 phys_addr_t paddr, size_t size, int prot,
 			 int lvl, arm_v7s_iopte *ptep)
@@ -411,20 +442,23 @@ static int __arm_v7s_map(struct arm_v7s_io_pgtable *data, unsigned long iova,
 		return -EINVAL;
 
 	/* Grab a pointer to the next level */
-	pte = *ptep;
+	pte = READ_ONCE(*ptep);
 	if (!pte) {
 		cptep = __arm_v7s_alloc_table(lvl + 1, GFP_ATOMIC, data);
 		if (!cptep)
 			return -ENOMEM;
 
-		pte = virt_to_phys(cptep) | ARM_V7S_PTE_TYPE_TABLE;
-		if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS)
-			pte |= ARM_V7S_ATTR_NS_TABLE;
+		pte = arm_v7s_install_table(cptep, ptep, 0, cfg);
+		if (pte)
+			__arm_v7s_free_table(cptep, lvl + 1, data);
+	} else {
+		/* We've no easy way of knowing if it's synced yet, so... */
+		__arm_v7s_pte_sync(ptep, 1, cfg);
+	}
 
-		__arm_v7s_set_pte(ptep, pte, 1, cfg);
-	} else if (ARM_V7S_PTE_IS_TABLE(pte, lvl)) {
+	if (ARM_V7S_PTE_IS_TABLE(pte, lvl)) {
 		cptep = iopte_deref(pte, lvl);
-	} else {
+	} else if (pte) {
 		/* We require an unmap first */
 		WARN_ON(!selftest_running);
 		return -EEXIST;
@@ -477,66 +511,73 @@ static void arm_v7s_free_pgtable(struct io_pgtable *iop)
 	kfree(data);
 }
 
-static void arm_v7s_split_cont(struct arm_v7s_io_pgtable *data,
-			       unsigned long iova, int idx, int lvl,
-			       arm_v7s_iopte *ptep)
+static arm_v7s_iopte arm_v7s_split_cont(struct arm_v7s_io_pgtable *data,
+					unsigned long iova, int idx, int lvl,
+					arm_v7s_iopte *ptep)
 {
 	struct io_pgtable *iop = &data->iop;
 	arm_v7s_iopte pte;
 	size_t size = ARM_V7S_BLOCK_SIZE(lvl);
 	int i;
 
+	/* Check that we didn't lose a race to get the lock */
+	pte = *ptep;
+	if (!arm_v7s_pte_is_cont(pte, lvl))
+		return pte;
+
 	ptep -= idx & (ARM_V7S_CONT_PAGES - 1);
-	pte = arm_v7s_cont_to_pte(*ptep, lvl);
-	for (i = 0; i < ARM_V7S_CONT_PAGES; i++) {
-		ptep[i] = pte;
-		pte += size;
-	}
+	pte = arm_v7s_cont_to_pte(pte, lvl);
+	for (i = 0; i < ARM_V7S_CONT_PAGES; i++)
+		ptep[i] = pte + i * size;
 
 	__arm_v7s_pte_sync(ptep, ARM_V7S_CONT_PAGES, &iop->cfg);
 
 	size *= ARM_V7S_CONT_PAGES;
 	io_pgtable_tlb_add_flush(iop, iova, size, size, true);
 	io_pgtable_tlb_sync(iop);
+	return pte;
 }
 
 static int arm_v7s_split_blk_unmap(struct arm_v7s_io_pgtable *data,
 				   unsigned long iova, size_t size,
-				   arm_v7s_iopte *ptep)
+				   arm_v7s_iopte blk_pte, arm_v7s_iopte *ptep)
 {
-	unsigned long blk_start, blk_end, blk_size;
-	phys_addr_t blk_paddr;
-	arm_v7s_iopte table = 0;
-	int prot = arm_v7s_pte_to_prot(*ptep, 1);
+	struct io_pgtable_cfg *cfg = &data->iop.cfg;
+	arm_v7s_iopte pte, *tablep;
+	int i, unmap_idx, num_entries, num_ptes;
 
-	blk_size = ARM_V7S_BLOCK_SIZE(1);
-	blk_start = iova & ARM_V7S_LVL_MASK(1);
-	blk_end = blk_start + ARM_V7S_BLOCK_SIZE(1);
-	blk_paddr = *ptep & ARM_V7S_LVL_MASK(1);
+	tablep = __arm_v7s_alloc_table(2, GFP_ATOMIC, data);
+	if (!tablep)
+		return 0; /* Bytes unmapped */
 
-	for (; blk_start < blk_end; blk_start += size, blk_paddr += size) {
-		arm_v7s_iopte *tablep;
+	num_ptes = ARM_V7S_PTES_PER_LVL(2);
+	num_entries = size >> ARM_V7S_LVL_SHIFT(2);
+	unmap_idx = ARM_V7S_LVL_IDX(iova, 2);
 
+	pte = arm_v7s_prot_to_pte(arm_v7s_pte_to_prot(blk_pte, 1), 2, cfg);
+	if (num_entries > 1)
+		pte = arm_v7s_pte_to_cont(pte, 2);
+
+	for (i = 0; i < num_ptes; i += num_entries, pte += size) {
 		/* Unmap! */
-		if (blk_start == iova)
+		if (i == unmap_idx)
 			continue;
 
-		/* __arm_v7s_map expects a pointer to the start of the table */
-		tablep = &table - ARM_V7S_LVL_IDX(blk_start, 1);
-		if (__arm_v7s_map(data, blk_start, blk_paddr, size, prot, 1,
-				  tablep) < 0) {
-			if (table) {
-				/* Free the table we allocated */
-				tablep = iopte_deref(table, 1);
-				__arm_v7s_free_table(tablep, 2, data);
-			}
-			return 0; /* Bytes unmapped */
-		}
+		__arm_v7s_set_pte(&tablep[i], pte, num_entries, cfg);
 	}
 
-	__arm_v7s_set_pte(ptep, table, 1, &data->iop.cfg);
-	iova &= ~(blk_size - 1);
-	io_pgtable_tlb_add_flush(&data->iop, iova, blk_size, blk_size, true);
+	pte = arm_v7s_install_table(tablep, ptep, blk_pte, cfg);
+	if (pte != blk_pte) {
+		__arm_v7s_free_table(tablep, 2, data);
+
+		if (!ARM_V7S_PTE_IS_TABLE(pte, 1))
+			return 0;
+
+		tablep = iopte_deref(pte, 1);
+		return __arm_v7s_unmap(data, iova, size, 2, tablep);
+	}
+
+	io_pgtable_tlb_add_flush(&data->iop, iova, size, size, true);
 	return size;
 }
 
@@ -555,17 +596,28 @@ static int __arm_v7s_unmap(struct arm_v7s_io_pgtable *data,
 	idx = ARM_V7S_LVL_IDX(iova, lvl);
 	ptep += idx;
 	do {
-		if (WARN_ON(!ARM_V7S_PTE_IS_VALID(ptep[i])))
+		pte[i] = READ_ONCE(ptep[i]);
+		if (WARN_ON(!ARM_V7S_PTE_IS_VALID(pte[i])))
 			return 0;
-		pte[i] = ptep[i];
 	} while (++i < num_entries);
 
 	/*
 	 * If we've hit a contiguous 'large page' entry at this level, it
 	 * needs splitting first, unless we're unmapping the whole lot.
+	 *
+	 * For splitting, we can't rewrite 16 PTEs atomically, and since we
+	 * can't necessarily assume TEX remap we don't have a software bit to
+	 * mark live entries being split. In practice (i.e. DMA API code), we
+	 * will never be splitting large pages anyway, so just wrap this edge
+	 * case in a lock for the sake of correctness and be done with it.
 	 */
-	if (num_entries <= 1 && arm_v7s_pte_is_cont(pte[0], lvl))
-		arm_v7s_split_cont(data, iova, idx, lvl, ptep);
+	if (num_entries <= 1 && arm_v7s_pte_is_cont(pte[0], lvl)) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&data->split_lock, flags);
+		pte[0] = arm_v7s_split_cont(data, iova, idx, lvl, ptep);
+		spin_unlock_irqrestore(&data->split_lock, flags);
+	}
 
 	/* If the size matches this level, we're in the right place */
 	if (num_entries) {
@@ -593,7 +645,7 @@ static int __arm_v7s_unmap(struct arm_v7s_io_pgtable *data,
 		 * Insert a table at the next level to map the old region,
 		 * minus the part we want to unmap
 		 */
-		return arm_v7s_split_blk_unmap(data, iova, size, ptep);
+		return arm_v7s_split_blk_unmap(data, iova, size, pte[0], ptep);
 	}
 
 	/* Keep on walkin' */
@@ -623,7 +675,8 @@ static phys_addr_t arm_v7s_iova_to_phys(struct io_pgtable_ops *ops,
 	u32 mask;
 
 	do {
-		pte = ptep[ARM_V7S_LVL_IDX(iova, ++lvl)];
+		ptep += ARM_V7S_LVL_IDX(iova, ++lvl);
+		pte = READ_ONCE(*ptep);
 		ptep = iopte_deref(pte, lvl);
 	} while (ARM_V7S_PTE_IS_TABLE(pte, lvl));
 
@@ -651,7 +704,8 @@ static struct io_pgtable *arm_v7s_alloc_pgtable(struct io_pgtable_cfg *cfg,
 	if (cfg->quirks & ~(IO_PGTABLE_QUIRK_ARM_NS |
 			    IO_PGTABLE_QUIRK_NO_PERMS |
 			    IO_PGTABLE_QUIRK_TLBI_ON_MAP |
-			    IO_PGTABLE_QUIRK_ARM_MTK_4GB))
+			    IO_PGTABLE_QUIRK_ARM_MTK_4GB |
+			    IO_PGTABLE_QUIRK_NO_DMA))
 		return NULL;
 
 	/* If ARM_MTK_4GB is enabled, the NO_PERMS is also expected. */
@@ -663,6 +717,7 @@ static struct io_pgtable *arm_v7s_alloc_pgtable(struct io_pgtable_cfg *cfg,
 	if (!data)
 		return NULL;
 
+	spin_lock_init(&data->split_lock);
 	data->l2_tables = kmem_cache_create("io-pgtable_armv7s_l2",
 					    ARM_V7S_TABLE_SIZE(2),
 					    ARM_V7S_TABLE_SIZE(2),
@@ -749,7 +804,7 @@ static void dummy_tlb_sync(void *cookie)
 	WARN_ON(cookie != cfg_cookie);
 }
 
-static struct iommu_gather_ops dummy_tlb_ops = {
+static const struct iommu_gather_ops dummy_tlb_ops = {
 	.tlb_flush_all	= dummy_tlb_flush_all,
 	.tlb_add_flush	= dummy_tlb_add_flush,
 	.tlb_sync	= dummy_tlb_sync,
@@ -768,7 +823,7 @@ static int __init arm_v7s_do_selftests(void)
 		.tlb = &dummy_tlb_ops,
 		.oas = 32,
 		.ias = 32,
-		.quirks = IO_PGTABLE_QUIRK_ARM_NS,
+		.quirks = IO_PGTABLE_QUIRK_ARM_NS | IO_PGTABLE_QUIRK_NO_DMA,
 		.pgsize_bitmap = SZ_4K | SZ_64K | SZ_1M | SZ_16M,
 	};
 	unsigned int iova, size, iova_start;
diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
index 6e5df5e0a3bd..b182039862c5 100644
--- a/drivers/iommu/io-pgtable-arm.c
+++ b/drivers/iommu/io-pgtable-arm.c
@@ -20,6 +20,7 @@
 
 #define pr_fmt(fmt)	"arm-lpae io-pgtable: " fmt
 
+#include <linux/atomic.h>
 #include <linux/iommu.h>
 #include <linux/kernel.h>
 #include <linux/sizes.h>
@@ -99,6 +100,8 @@
 #define ARM_LPAE_PTE_ATTR_HI_MASK	(((arm_lpae_iopte)6) << 52)
 #define ARM_LPAE_PTE_ATTR_MASK		(ARM_LPAE_PTE_ATTR_LO_MASK |	\
 					 ARM_LPAE_PTE_ATTR_HI_MASK)
+/* Software bit for solving coherency races */
+#define ARM_LPAE_PTE_SW_SYNC		(((arm_lpae_iopte)1) << 55)
 
 /* Stage-1 PTE */
 #define ARM_LPAE_PTE_AP_UNPRIV		(((arm_lpae_iopte)1) << 6)
@@ -217,7 +220,7 @@ static void *__arm_lpae_alloc_pages(size_t size, gfp_t gfp,
 	if (!pages)
 		return NULL;
 
-	if (!selftest_running) {
+	if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA)) {
 		dma = dma_map_single(dev, pages, size, DMA_TO_DEVICE);
 		if (dma_mapping_error(dev, dma))
 			goto out_free;
@@ -243,40 +246,64 @@ out_free:
 static void __arm_lpae_free_pages(void *pages, size_t size,
 				  struct io_pgtable_cfg *cfg)
 {
-	if (!selftest_running)
+	if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA))
 		dma_unmap_single(cfg->iommu_dev, __arm_lpae_dma_addr(pages),
 				 size, DMA_TO_DEVICE);
 	free_pages_exact(pages, size);
 }
 
+static void __arm_lpae_sync_pte(arm_lpae_iopte *ptep,
+				struct io_pgtable_cfg *cfg)
+{
+	dma_sync_single_for_device(cfg->iommu_dev, __arm_lpae_dma_addr(ptep),
+				   sizeof(*ptep), DMA_TO_DEVICE);
+}
+
 static void __arm_lpae_set_pte(arm_lpae_iopte *ptep, arm_lpae_iopte pte,
 			       struct io_pgtable_cfg *cfg)
 {
 	*ptep = pte;
 
-	if (!selftest_running)
-		dma_sync_single_for_device(cfg->iommu_dev,
-					   __arm_lpae_dma_addr(ptep),
-					   sizeof(pte), DMA_TO_DEVICE);
+	if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA))
+		__arm_lpae_sync_pte(ptep, cfg);
 }
 
 static int __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
 			    unsigned long iova, size_t size, int lvl,
 			    arm_lpae_iopte *ptep);
 
+static void __arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
+				phys_addr_t paddr, arm_lpae_iopte prot,
+				int lvl, arm_lpae_iopte *ptep)
+{
+	arm_lpae_iopte pte = prot;
+
+	if (data->iop.cfg.quirks & IO_PGTABLE_QUIRK_ARM_NS)
+		pte |= ARM_LPAE_PTE_NS;
+
+	if (lvl == ARM_LPAE_MAX_LEVELS - 1)
+		pte |= ARM_LPAE_PTE_TYPE_PAGE;
+	else
+		pte |= ARM_LPAE_PTE_TYPE_BLOCK;
+
+	pte |= ARM_LPAE_PTE_AF | ARM_LPAE_PTE_SH_IS;
+	pte |= pfn_to_iopte(paddr >> data->pg_shift, data);
+
+	__arm_lpae_set_pte(ptep, pte, &data->iop.cfg);
+}
+
 static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
 			     unsigned long iova, phys_addr_t paddr,
 			     arm_lpae_iopte prot, int lvl,
 			     arm_lpae_iopte *ptep)
 {
-	arm_lpae_iopte pte = prot;
-	struct io_pgtable_cfg *cfg = &data->iop.cfg;
+	arm_lpae_iopte pte = *ptep;
 
-	if (iopte_leaf(*ptep, lvl)) {
+	if (iopte_leaf(pte, lvl)) {
 		/* We require an unmap first */
 		WARN_ON(!selftest_running);
 		return -EEXIST;
-	} else if (iopte_type(*ptep, lvl) == ARM_LPAE_PTE_TYPE_TABLE) {
+	} else if (iopte_type(pte, lvl) == ARM_LPAE_PTE_TYPE_TABLE) {
 		/*
 		 * We need to unmap and free the old table before
 		 * overwriting it with a block entry.
@@ -289,19 +316,40 @@ static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
 			return -EINVAL;
 	}
 
+	__arm_lpae_init_pte(data, paddr, prot, lvl, ptep);
+	return 0;
+}
+
+static arm_lpae_iopte arm_lpae_install_table(arm_lpae_iopte *table,
+					     arm_lpae_iopte *ptep,
+					     arm_lpae_iopte curr,
+					     struct io_pgtable_cfg *cfg)
+{
+	arm_lpae_iopte old, new;
+
+	new = __pa(table) | ARM_LPAE_PTE_TYPE_TABLE;
 	if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS)
-		pte |= ARM_LPAE_PTE_NS;
+		new |= ARM_LPAE_PTE_NSTABLE;
 
-	if (lvl == ARM_LPAE_MAX_LEVELS - 1)
-		pte |= ARM_LPAE_PTE_TYPE_PAGE;
-	else
-		pte |= ARM_LPAE_PTE_TYPE_BLOCK;
+	/*
+	 * Ensure the table itself is visible before its PTE can be.
+	 * Whilst we could get away with cmpxchg64_release below, this
+	 * doesn't have any ordering semantics when !CONFIG_SMP.
+	 */
+	dma_wmb();
 
-	pte |= ARM_LPAE_PTE_AF | ARM_LPAE_PTE_SH_IS;
-	pte |= pfn_to_iopte(paddr >> data->pg_shift, data);
+	old = cmpxchg64_relaxed(ptep, curr, new);
 
-	__arm_lpae_set_pte(ptep, pte, cfg);
-	return 0;
+	if ((cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA) ||
+	    (old & ARM_LPAE_PTE_SW_SYNC))
+		return old;
+
+	/* Even if it's not ours, there's no point waiting; just kick it */
+	__arm_lpae_sync_pte(ptep, cfg);
+	if (old == curr)
+		WRITE_ONCE(*ptep, new | ARM_LPAE_PTE_SW_SYNC);
+
+	return old;
 }
 
 static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova,
@@ -310,6 +358,7 @@ static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova,
 {
 	arm_lpae_iopte *cptep, pte;
 	size_t block_size = ARM_LPAE_BLOCK_SIZE(lvl, data);
+	size_t tblsz = ARM_LPAE_GRANULE(data);
 	struct io_pgtable_cfg *cfg = &data->iop.cfg;
 
 	/* Find our entry at the current level */
@@ -324,20 +373,23 @@ static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova,
 		return -EINVAL;
 
 	/* Grab a pointer to the next level */
-	pte = *ptep;
+	pte = READ_ONCE(*ptep);
 	if (!pte) {
-		cptep = __arm_lpae_alloc_pages(ARM_LPAE_GRANULE(data),
-					       GFP_ATOMIC, cfg);
+		cptep = __arm_lpae_alloc_pages(tblsz, GFP_ATOMIC, cfg);
 		if (!cptep)
 			return -ENOMEM;
 
-		pte = __pa(cptep) | ARM_LPAE_PTE_TYPE_TABLE;
-		if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS)
-			pte |= ARM_LPAE_PTE_NSTABLE;
-		__arm_lpae_set_pte(ptep, pte, cfg);
-	} else if (!iopte_leaf(pte, lvl)) {
+		pte = arm_lpae_install_table(cptep, ptep, 0, cfg);
+		if (pte)
+			__arm_lpae_free_pages(cptep, tblsz, cfg);
+	} else if (!(cfg->quirks & IO_PGTABLE_QUIRK_NO_DMA) &&
+		   !(pte & ARM_LPAE_PTE_SW_SYNC)) {
+		__arm_lpae_sync_pte(ptep, cfg);
+	}
+
+	if (pte && !iopte_leaf(pte, lvl)) {
 		cptep = iopte_deref(pte, data);
-	} else {
+	} else if (pte) {
 		/* We require an unmap first */
 		WARN_ON(!selftest_running);
 		return -EEXIST;
@@ -452,40 +504,55 @@ static void arm_lpae_free_pgtable(struct io_pgtable *iop)
 
 static int arm_lpae_split_blk_unmap(struct arm_lpae_io_pgtable *data,
 				    unsigned long iova, size_t size,
-				    arm_lpae_iopte prot, int lvl,
-				    arm_lpae_iopte *ptep, size_t blk_size)
+				    arm_lpae_iopte blk_pte, int lvl,
+				    arm_lpae_iopte *ptep)
 {
-	unsigned long blk_start, blk_end;
+	struct io_pgtable_cfg *cfg = &data->iop.cfg;
+	arm_lpae_iopte pte, *tablep;
 	phys_addr_t blk_paddr;
-	arm_lpae_iopte table = 0;
+	size_t tablesz = ARM_LPAE_GRANULE(data);
+	size_t split_sz = ARM_LPAE_BLOCK_SIZE(lvl, data);
+	int i, unmap_idx = -1;
+
+	if (WARN_ON(lvl == ARM_LPAE_MAX_LEVELS))
+		return 0;
 
-	blk_start = iova & ~(blk_size - 1);
-	blk_end = blk_start + blk_size;
-	blk_paddr = iopte_to_pfn(*ptep, data) << data->pg_shift;
+	tablep = __arm_lpae_alloc_pages(tablesz, GFP_ATOMIC, cfg);
+	if (!tablep)
+		return 0; /* Bytes unmapped */
 
-	for (; blk_start < blk_end; blk_start += size, blk_paddr += size) {
-		arm_lpae_iopte *tablep;
+	if (size == split_sz)
+		unmap_idx = ARM_LPAE_LVL_IDX(iova, lvl, data);
 
+	blk_paddr = iopte_to_pfn(blk_pte, data) << data->pg_shift;
+	pte = iopte_prot(blk_pte);
+
+	for (i = 0; i < tablesz / sizeof(pte); i++, blk_paddr += split_sz) {
 		/* Unmap! */
-		if (blk_start == iova)
+		if (i == unmap_idx)
 			continue;
 
-		/* __arm_lpae_map expects a pointer to the start of the table */
-		tablep = &table - ARM_LPAE_LVL_IDX(blk_start, lvl, data);
-		if (__arm_lpae_map(data, blk_start, blk_paddr, size, prot, lvl,
-				   tablep) < 0) {
-			if (table) {
-				/* Free the table we allocated */
-				tablep = iopte_deref(table, data);
-				__arm_lpae_free_pgtable(data, lvl + 1, tablep);
-			}
-			return 0; /* Bytes unmapped */
-		}
+		__arm_lpae_init_pte(data, blk_paddr, pte, lvl, &tablep[i]);
 	}
 
-	__arm_lpae_set_pte(ptep, table, &data->iop.cfg);
-	iova &= ~(blk_size - 1);
-	io_pgtable_tlb_add_flush(&data->iop, iova, blk_size, blk_size, true);
+	pte = arm_lpae_install_table(tablep, ptep, blk_pte, cfg);
+	if (pte != blk_pte) {
+		__arm_lpae_free_pages(tablep, tablesz, cfg);
+		/*
+		 * We may race against someone unmapping another part of this
+		 * block, but anything else is invalid. We can't misinterpret
+		 * a page entry here since we're never at the last level.
+		 */
+		if (iopte_type(pte, lvl - 1) != ARM_LPAE_PTE_TYPE_TABLE)
+			return 0;
+
+		tablep = iopte_deref(pte, data);
+	}
+
+	if (unmap_idx < 0)
+		return __arm_lpae_unmap(data, iova, size, lvl, tablep);
+
+	io_pgtable_tlb_add_flush(&data->iop, iova, size, size, true);
 	return size;
 }
 
@@ -495,19 +562,18 @@ static int __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
 {
 	arm_lpae_iopte pte;
 	struct io_pgtable *iop = &data->iop;
-	size_t blk_size = ARM_LPAE_BLOCK_SIZE(lvl, data);
 
 	/* Something went horribly wrong and we ran out of page table */
 	if (WARN_ON(lvl == ARM_LPAE_MAX_LEVELS))
 		return 0;
 
 	ptep += ARM_LPAE_LVL_IDX(iova, lvl, data);
-	pte = *ptep;
+	pte = READ_ONCE(*ptep);
 	if (WARN_ON(!pte))
 		return 0;
 
 	/* If the size matches this level, we're in the right place */
-	if (size == blk_size) {
+	if (size == ARM_LPAE_BLOCK_SIZE(lvl, data)) {
 		__arm_lpae_set_pte(ptep, 0, &iop->cfg);
 
 		if (!iopte_leaf(pte, lvl)) {
@@ -527,9 +593,8 @@ static int __arm_lpae_unmap(struct arm_lpae_io_pgtable *data,
 		 * Insert a table at the next level to map the old region,
 		 * minus the part we want to unmap
 		 */
-		return arm_lpae_split_blk_unmap(data, iova, size,
-						iopte_prot(pte), lvl, ptep,
-						blk_size);
+		return arm_lpae_split_blk_unmap(data, iova, size, pte,
+						lvl + 1, ptep);
 	}
 
 	/* Keep on walkin' */
@@ -565,7 +630,8 @@ static phys_addr_t arm_lpae_iova_to_phys(struct io_pgtable_ops *ops,
 			return 0;
 
 		/* Grab the IOPTE we're interested in */
-		pte = *(ptep + ARM_LPAE_LVL_IDX(iova, lvl, data));
+		ptep += ARM_LPAE_LVL_IDX(iova, lvl, data);
+		pte = READ_ONCE(*ptep);
 
 		/* Valid entry? */
 		if (!pte)
@@ -673,7 +739,7 @@ arm_64_lpae_alloc_pgtable_s1(struct io_pgtable_cfg *cfg, void *cookie)
 	u64 reg;
 	struct arm_lpae_io_pgtable *data;
 
-	if (cfg->quirks & ~IO_PGTABLE_QUIRK_ARM_NS)
+	if (cfg->quirks & ~(IO_PGTABLE_QUIRK_ARM_NS | IO_PGTABLE_QUIRK_NO_DMA))
 		return NULL;
 
 	data = arm_lpae_alloc_pgtable(cfg);
@@ -762,7 +828,7 @@ arm_64_lpae_alloc_pgtable_s2(struct io_pgtable_cfg *cfg, void *cookie)
 	struct arm_lpae_io_pgtable *data;
 
 	/* The NS quirk doesn't apply at stage 2 */
-	if (cfg->quirks)
+	if (cfg->quirks & ~IO_PGTABLE_QUIRK_NO_DMA)
 		return NULL;
 
 	data = arm_lpae_alloc_pgtable(cfg);
@@ -1066,6 +1132,7 @@ static int __init arm_lpae_do_selftests(void)
 	struct io_pgtable_cfg cfg = {
 		.tlb = &dummy_tlb_ops,
 		.oas = 48,
+		.quirks = IO_PGTABLE_QUIRK_NO_DMA,
 	};
 
 	for (i = 0; i < ARRAY_SIZE(pgsize); ++i) {
diff --git a/drivers/iommu/io-pgtable.h b/drivers/iommu/io-pgtable.h
index 969d82cc92ca..524263a7ae6f 100644
--- a/drivers/iommu/io-pgtable.h
+++ b/drivers/iommu/io-pgtable.h
@@ -65,11 +65,17 @@ struct io_pgtable_cfg {
 	 *	PTEs, for Mediatek IOMMUs which treat it as a 33rd address bit
 	 *	when the SoC is in "4GB mode" and they can only access the high
 	 *	remap of DRAM (0x1_00000000 to 0x1_ffffffff).
+	 *
+	 * IO_PGTABLE_QUIRK_NO_DMA: Guarantees that the tables will only ever
+	 *	be accessed by a fully cache-coherent IOMMU or CPU (e.g. for a
+	 *	software-emulated IOMMU), such that pagetable updates need not
+	 *	be treated as explicit DMA data.
 	 */
 	#define IO_PGTABLE_QUIRK_ARM_NS		BIT(0)
 	#define IO_PGTABLE_QUIRK_NO_PERMS	BIT(1)
 	#define IO_PGTABLE_QUIRK_TLBI_ON_MAP	BIT(2)
 	#define IO_PGTABLE_QUIRK_ARM_MTK_4GB	BIT(3)
+	#define IO_PGTABLE_QUIRK_NO_DMA		BIT(4)
 	unsigned long			quirks;
 	unsigned long			pgsize_bitmap;
 	unsigned int			ias;