Linux-libre 4.2-gnu

19 files changed, 5002 insertions, 1772 deletions

diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 1ae4e547b..f1fb1d3cc 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -178,6 +178,9 @@ config OMAP_IOMMU
 	depends on ARM && MMU
 	depends on ARCH_OMAP2PLUS || COMPILE_TEST
 	select IOMMU_API
+	---help---
+	  The OMAP3 media platform drivers depend on iommu support,
+	  if you need them say Y here.
 
 config OMAP_IOMMU_DEBUG
 	bool "Export OMAP IOMMU internals in DebugFS"
@@ -219,7 +222,7 @@ config TEGRA_IOMMU_SMMU
 	select IOMMU_API
 	help
 	  This driver supports the IOMMU hardware (SMMU) found on NVIDIA Tegra
-	  SoCs (Tegra30 up to Tegra124).
+	  SoCs (Tegra30 up to Tegra132).
 
 config EXYNOS_IOMMU
 	bool "Exynos IOMMU Support"
@@ -339,6 +342,7 @@ config SPAPR_TCE_IOMMU
 	  Enables bits of IOMMU API required by VFIO. The iommu_ops
 	  is not implemented as it is not necessary for VFIO.
 
+# ARM IOMMU support
 config ARM_SMMU
 	bool "ARM Ltd. System MMU (SMMU) Support"
 	depends on (ARM64 || ARM) && MMU
@@ -352,4 +356,16 @@ config ARM_SMMU
 	  Say Y here if your SoC includes an IOMMU device implementing
 	  the ARM SMMU architecture.
 
+config ARM_SMMU_V3
+	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support"
+	depends on ARM64 && PCI
+	select IOMMU_API
+	select IOMMU_IO_PGTABLE_LPAE
+	help
+	  Support for implementations of the ARM System MMU architecture
+	  version 3 providing translation support to a PCIe root complex.
+
+	  Say Y here if your system includes an IOMMU device implementing
+	  the ARM SMMUv3 architecture.
+
 endif # IOMMU_SUPPORT
diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile
index 080ffab4e..c6dcc513d 100644
--- a/drivers/iommu/Makefile
+++ b/drivers/iommu/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_MSM_IOMMU) += msm_iommu.o msm_iommu_dev.o
 obj-$(CONFIG_AMD_IOMMU) += amd_iommu.o amd_iommu_init.o
 obj-$(CONFIG_AMD_IOMMU_V2) += amd_iommu_v2.o
 obj-$(CONFIG_ARM_SMMU) += arm-smmu.o
+obj-$(CONFIG_ARM_SMMU_V3) += arm-smmu-v3.o
 obj-$(CONFIG_DMAR_TABLE) += dmar.o
 obj-$(CONFIG_INTEL_IOMMU) += intel-iommu.o
 obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index ca9f4edbb..658ee39e6 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -34,6 +34,7 @@
 #include <linux/irq.h>
 #include <linux/msi.h>
 #include <linux/dma-contiguous.h>
+#include <linux/irqdomain.h>
 #include <asm/irq_remapping.h>
 #include <asm/io_apic.h>
 #include <asm/apic.h>
@@ -64,10 +65,6 @@
 
 static DEFINE_RWLOCK(amd_iommu_devtable_lock);
 
-/* A list of preallocated protection domains */
-static LIST_HEAD(iommu_pd_list);
-static DEFINE_SPINLOCK(iommu_pd_list_lock);
-
 /* List of all available dev_data structures */
 static LIST_HEAD(dev_data_list);
 static DEFINE_SPINLOCK(dev_data_list_lock);
@@ -79,8 +76,6 @@ LIST_HEAD(hpet_map);
  * Domain for untranslated devices - only allocated
  * if iommu=pt passed on kernel cmd line.
  */
-static struct protection_domain *pt_domain;
-
 static const struct iommu_ops amd_iommu_ops;
 
 static ATOMIC_NOTIFIER_HEAD(ppr_notifier);
@@ -99,7 +94,7 @@ struct iommu_dev_data {
 	struct protection_domain *domain; /* Domain the device is bound to */
 	u16 devid;			  /* PCI Device ID */
 	bool iommu_v2;			  /* Device can make use of IOMMUv2 */
-	bool passthrough;		  /* Default for device is pt_domain */
+	bool passthrough;		  /* Device is identity mapped */
 	struct {
 		bool enabled;
 		int qdep;
@@ -119,7 +114,7 @@ struct iommu_cmd {
 struct kmem_cache *amd_iommu_irq_cache;
 
 static void update_domain(struct protection_domain *domain);
-static int __init alloc_passthrough_domain(void);
+static int protection_domain_init(struct protection_domain *domain);
 
 /****************************************************************************
  *
@@ -234,31 +229,38 @@ static bool pdev_pri_erratum(struct pci_dev *pdev, u32 erratum)
 }
 
 /*
- * In this function the list of preallocated protection domains is traversed to
- * find the domain for a specific device
+ * This function actually applies the mapping to the page table of the
+ * dma_ops domain.
  */
-static struct dma_ops_domain *find_protection_domain(u16 devid)
+static void alloc_unity_mapping(struct dma_ops_domain *dma_dom,
+				struct unity_map_entry *e)
 {
-	struct dma_ops_domain *entry, *ret = NULL;
-	unsigned long flags;
-	u16 alias = amd_iommu_alias_table[devid];
-
-	if (list_empty(&iommu_pd_list))
-		return NULL;
-
-	spin_lock_irqsave(&iommu_pd_list_lock, flags);
+	u64 addr;
 
-	list_for_each_entry(entry, &iommu_pd_list, list) {
-		if (entry->target_dev == devid ||
-		    entry->target_dev == alias) {
-			ret = entry;
-			break;
-		}
+	for (addr = e->address_start; addr < e->address_end;
+	     addr += PAGE_SIZE) {
+		if (addr < dma_dom->aperture_size)
+			__set_bit(addr >> PAGE_SHIFT,
+				  dma_dom->aperture[0]->bitmap);
 	}
+}
 
-	spin_unlock_irqrestore(&iommu_pd_list_lock, flags);
+/*
+ * Inits the unity mappings required for a specific device
+ */
+static void init_unity_mappings_for_device(struct device *dev,
+					   struct dma_ops_domain *dma_dom)
+{
+	struct unity_map_entry *e;
+	u16 devid;
 
-	return ret;
+	devid = get_device_id(dev);
+
+	list_for_each_entry(e, &amd_iommu_unity_map, list) {
+		if (!(devid >= e->devid_start && devid <= e->devid_end))
+			continue;
+		alloc_unity_mapping(dma_dom, e);
+	}
 }
 
 /*
@@ -290,11 +292,23 @@ static bool check_device(struct device *dev)
 
 static void init_iommu_group(struct device *dev)
 {
+	struct dma_ops_domain *dma_domain;
+	struct iommu_domain *domain;
 	struct iommu_group *group;
 
 	group = iommu_group_get_for_dev(dev);
-	if (!IS_ERR(group))
-		iommu_group_put(group);
+	if (IS_ERR(group))
+		return;
+
+	domain = iommu_group_default_domain(group);
+	if (!domain)
+		goto out;
+
+	dma_domain = to_pdomain(domain)->priv;
+
+	init_unity_mappings_for_device(dev, dma_domain);
+out:
+	iommu_group_put(group);
 }
 
 static int __last_alias(struct pci_dev *pdev, u16 alias, void *data)
@@ -434,64 +448,15 @@ static void iommu_uninit_device(struct device *dev)
 	/* Unlink from alias, it may change if another device is re-plugged */
 	dev_data->alias_data = NULL;
 
+	/* Remove dma-ops */
+	dev->archdata.dma_ops = NULL;
+
 	/*
 	 * We keep dev_data around for unplugged devices and reuse it when the
 	 * device is re-plugged - not doing so would introduce a ton of races.
 	 */
 }
 
-void __init amd_iommu_uninit_devices(void)
-{
-	struct iommu_dev_data *dev_data, *n;
-	struct pci_dev *pdev = NULL;
-
-	for_each_pci_dev(pdev) {
-
-		if (!check_device(&pdev->dev))
-			continue;
-
-		iommu_uninit_device(&pdev->dev);
-	}
-
-	/* Free all of our dev_data structures */
-	list_for_each_entry_safe(dev_data, n, &dev_data_list, dev_data_list)
-		free_dev_data(dev_data);
-}
-
-int __init amd_iommu_init_devices(void)
-{
-	struct pci_dev *pdev = NULL;
-	int ret = 0;
-
-	for_each_pci_dev(pdev) {
-
-		if (!check_device(&pdev->dev))
-			continue;
-
-		ret = iommu_init_device(&pdev->dev);
-		if (ret == -ENOTSUPP)
-			iommu_ignore_device(&pdev->dev);
-		else if (ret)
-			goto out_free;
-	}
-
-	/*
-	 * Initialize IOMMU groups only after iommu_init_device() has
-	 * had a chance to populate any IVRS defined aliases.
-	 */
-	for_each_pci_dev(pdev) {
-		if (check_device(&pdev->dev))
-			init_iommu_group(&pdev->dev);
-	}
-
-	return 0;
-
-out_free:
-
-	amd_iommu_uninit_devices();
-
-	return ret;
-}
 #ifdef CONFIG_AMD_IOMMU_STATS
 
 /*
@@ -1463,94 +1428,6 @@ static unsigned long iommu_unmap_page(struct protection_domain *dom,
 	return unmapped;
 }
 
-/*
- * This function checks if a specific unity mapping entry is needed for
- * this specific IOMMU.
- */
-static int iommu_for_unity_map(struct amd_iommu *iommu,
-			       struct unity_map_entry *entry)
-{
-	u16 bdf, i;
-
-	for (i = entry->devid_start; i <= entry->devid_end; ++i) {
-		bdf = amd_iommu_alias_table[i];
-		if (amd_iommu_rlookup_table[bdf] == iommu)
-			return 1;
-	}
-
-	return 0;
-}
-
-/*
- * This function actually applies the mapping to the page table of the
- * dma_ops domain.
- */
-static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
-			     struct unity_map_entry *e)
-{
-	u64 addr;
-	int ret;
-
-	for (addr = e->address_start; addr < e->address_end;
-	     addr += PAGE_SIZE) {
-		ret = iommu_map_page(&dma_dom->domain, addr, addr, e->prot,
-				     PAGE_SIZE);
-		if (ret)
-			return ret;
-		/*
-		 * if unity mapping is in aperture range mark the page
-		 * as allocated in the aperture
-		 */
-		if (addr < dma_dom->aperture_size)
-			__set_bit(addr >> PAGE_SHIFT,
-				  dma_dom->aperture[0]->bitmap);
-	}
-
-	return 0;
-}
-
-/*
- * Init the unity mappings for a specific IOMMU in the system
- *
- * Basically iterates over all unity mapping entries and applies them to
- * the default domain DMA of that IOMMU if necessary.
- */
-static int iommu_init_unity_mappings(struct amd_iommu *iommu)
-{
-	struct unity_map_entry *entry;
-	int ret;
-
-	list_for_each_entry(entry, &amd_iommu_unity_map, list) {
-		if (!iommu_for_unity_map(iommu, entry))
-			continue;
-		ret = dma_ops_unity_map(iommu->default_dom, entry);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-/*
- * Inits the unity mappings required for a specific device
- */
-static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom,
-					  u16 devid)
-{
-	struct unity_map_entry *e;
-	int ret;
-
-	list_for_each_entry(e, &amd_iommu_unity_map, list) {
-		if (!(devid >= e->devid_start && devid <= e->devid_end))
-			continue;
-		ret = dma_ops_unity_map(dma_dom, e);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
 /****************************************************************************
  *
  * The next functions belong to the address allocator for the dma_ops
@@ -1704,14 +1581,16 @@ static unsigned long dma_ops_area_alloc(struct device *dev,
 	unsigned long next_bit = dom->next_address % APERTURE_RANGE_SIZE;
 	int max_index = dom->aperture_size >> APERTURE_RANGE_SHIFT;
 	int i = start >> APERTURE_RANGE_SHIFT;
-	unsigned long boundary_size;
+	unsigned long boundary_size, mask;
 	unsigned long address = -1;
 	unsigned long limit;
 
 	next_bit >>= PAGE_SHIFT;
 
-	boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1,
-			PAGE_SIZE) >> PAGE_SHIFT;
+	mask = dma_get_seg_boundary(dev);
+
+	boundary_size = mask + 1 ? ALIGN(mask + 1, PAGE_SIZE) >> PAGE_SHIFT :
+				   1UL << (BITS_PER_LONG - PAGE_SHIFT);
 
 	for (;i < max_index; ++i) {
 		unsigned long offset = dom->aperture[i]->offset >> PAGE_SHIFT;
@@ -2000,12 +1879,9 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void)
 	if (!dma_dom)
 		return NULL;
 
-	spin_lock_init(&dma_dom->domain.lock);
-
-	dma_dom->domain.id = domain_id_alloc();
-	if (dma_dom->domain.id == 0)
+	if (protection_domain_init(&dma_dom->domain))
 		goto free_dma_dom;
-	INIT_LIST_HEAD(&dma_dom->domain.dev_list);
+
 	dma_dom->domain.mode = PAGE_MODE_2_LEVEL;
 	dma_dom->domain.pt_root = (void *)get_zeroed_page(GFP_KERNEL);
 	dma_dom->domain.flags = PD_DMA_OPS_MASK;
@@ -2014,7 +1890,6 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void)
 		goto free_dma_dom;
 
 	dma_dom->need_flush = false;
-	dma_dom->target_dev = 0xffff;
 
 	add_domain_to_list(&dma_dom->domain);
 
@@ -2289,15 +2164,17 @@ static int attach_device(struct device *dev,
 	dev_data = get_dev_data(dev);
 
 	if (domain->flags & PD_IOMMUV2_MASK) {
-		if (!dev_data->iommu_v2 || !dev_data->passthrough)
+		if (!dev_data->passthrough)
 			return -EINVAL;
 
-		if (pdev_iommuv2_enable(pdev) != 0)
-			return -EINVAL;
+		if (dev_data->iommu_v2) {
+			if (pdev_iommuv2_enable(pdev) != 0)
+				return -EINVAL;
 
-		dev_data->ats.enabled = true;
-		dev_data->ats.qdep    = pci_ats_queue_depth(pdev);
-		dev_data->pri_tlp     = pci_pri_tlp_required(pdev);
+			dev_data->ats.enabled = true;
+			dev_data->ats.qdep    = pci_ats_queue_depth(pdev);
+			dev_data->pri_tlp     = pci_pri_tlp_required(pdev);
+		}
 	} else if (amd_iommu_iotlb_sup &&
 		   pci_enable_ats(pdev, PAGE_SHIFT) == 0) {
 		dev_data->ats.enabled = true;
@@ -2343,15 +2220,6 @@ static void __detach_device(struct iommu_dev_data *dev_data)
 	do_detach(head);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
-
-	/*
-	 * If we run in passthrough mode the device must be assigned to the
-	 * passthrough domain if it is detached from any other domain.
-	 * Make sure we can deassign from the pt_domain itself.
-	 */
-	if (dev_data->passthrough &&
-	    (dev_data->domain == NULL && domain != pt_domain))
-		__attach_device(dev_data, pt_domain);
 }
 
 /*
@@ -2371,7 +2239,7 @@ static void detach_device(struct device *dev)
 	__detach_device(dev_data);
 	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
 
-	if (domain->flags & PD_IOMMUV2_MASK)
+	if (domain->flags & PD_IOMMUV2_MASK && dev_data->iommu_v2)
 		pdev_iommuv2_disable(to_pci_dev(dev));
 	else if (dev_data->ats.enabled)
 		pci_disable_ats(to_pci_dev(dev));
@@ -2379,110 +2247,65 @@ static void detach_device(struct device *dev)
 	dev_data->ats.enabled = false;
 }
 
-/*
- * Find out the protection domain structure for a given PCI device. This
- * will give us the pointer to the page table root for example.
- */
-static struct protection_domain *domain_for_device(struct device *dev)
-{
-	struct iommu_dev_data *dev_data;
-	struct protection_domain *dom = NULL;
-	unsigned long flags;
-
-	dev_data   = get_dev_data(dev);
-
-	if (dev_data->domain)
-		return dev_data->domain;
-
-	if (dev_data->alias_data != NULL) {
-		struct iommu_dev_data *alias_data = dev_data->alias_data;
-
-		read_lock_irqsave(&amd_iommu_devtable_lock, flags);
-		if (alias_data->domain != NULL) {
-			__attach_device(dev_data, alias_data->domain);
-			dom = alias_data->domain;
-		}
-		read_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
-	}
-
-	return dom;
-}
-
-static int device_change_notifier(struct notifier_block *nb,
-				  unsigned long action, void *data)
+static int amd_iommu_add_device(struct device *dev)
 {
-	struct dma_ops_domain *dma_domain;
-	struct protection_domain *domain;
 	struct iommu_dev_data *dev_data;
-	struct device *dev = data;
+	struct iommu_domain *domain;
 	struct amd_iommu *iommu;
-	unsigned long flags;
 	u16 devid;
+	int ret;
 
-	if (!check_device(dev))
+	if (!check_device(dev) || get_dev_data(dev))
 		return 0;
 
-	devid    = get_device_id(dev);
-	iommu    = amd_iommu_rlookup_table[devid];
-	dev_data = get_dev_data(dev);
-
-	switch (action) {
-	case BUS_NOTIFY_ADD_DEVICE:
+	devid = get_device_id(dev);
+	iommu = amd_iommu_rlookup_table[devid];
 
-		iommu_init_device(dev);
-		init_iommu_group(dev);
+	ret = iommu_init_device(dev);
+	if (ret) {
+		if (ret != -ENOTSUPP)
+			pr_err("Failed to initialize device %s - trying to proceed anyway\n",
+				dev_name(dev));
 
-		/*
-		 * dev_data is still NULL and
-		 * got initialized in iommu_init_device
-		 */
-		dev_data = get_dev_data(dev);
+		iommu_ignore_device(dev);
+		dev->archdata.dma_ops = &nommu_dma_ops;
+		goto out;
+	}
+	init_iommu_group(dev);
 
-		if (iommu_pass_through || dev_data->iommu_v2) {
-			dev_data->passthrough = true;
-			attach_device(dev, pt_domain);
-			break;
-		}
+	dev_data = get_dev_data(dev);
 
-		domain = domain_for_device(dev);
+	BUG_ON(!dev_data);
 
-		/* allocate a protection domain if a device is added */
-		dma_domain = find_protection_domain(devid);
-		if (!dma_domain) {
-			dma_domain = dma_ops_domain_alloc();
-			if (!dma_domain)
-				goto out;
-			dma_domain->target_dev = devid;
-
-			spin_lock_irqsave(&iommu_pd_list_lock, flags);
-			list_add_tail(&dma_domain->list, &iommu_pd_list);
-			spin_unlock_irqrestore(&iommu_pd_list_lock, flags);
-		}
+	if (iommu_pass_through || dev_data->iommu_v2)
+		iommu_request_dm_for_dev(dev);
 
+	/* Domains are initialized for this device - have a look what we ended up with */
+	domain = iommu_get_domain_for_dev(dev);
+	if (domain->type == IOMMU_DOMAIN_IDENTITY)
+		dev_data->passthrough = true;
+	else
 		dev->archdata.dma_ops = &amd_iommu_dma_ops;
 
-		break;
-	case BUS_NOTIFY_REMOVED_DEVICE:
-
-		iommu_uninit_device(dev);
-
-	default:
-		goto out;
-	}
-
+out:
 	iommu_completion_wait(iommu);
 
-out:
 	return 0;
 }
 
-static struct notifier_block device_nb = {
-	.notifier_call = device_change_notifier,
-};
-
-void amd_iommu_init_notifier(void)
+static void amd_iommu_remove_device(struct device *dev)
 {
-	bus_register_notifier(&pci_bus_type, &device_nb);
+	struct amd_iommu *iommu;
+	u16 devid;
+
+	if (!check_device(dev))
+		return;
+
+	devid = get_device_id(dev);
+	iommu = amd_iommu_rlookup_table[devid];
+
+	iommu_uninit_device(dev);
+	iommu_completion_wait(iommu);
 }
 
 /*****************************************************************************
@@ -2501,28 +2324,20 @@ void amd_iommu_init_notifier(void)
 static struct protection_domain *get_domain(struct device *dev)
 {
 	struct protection_domain *domain;
-	struct dma_ops_domain *dma_dom;
-	u16 devid = get_device_id(dev);
+	struct iommu_domain *io_domain;
 
 	if (!check_device(dev))
 		return ERR_PTR(-EINVAL);
 
-	domain = domain_for_device(dev);
-	if (domain != NULL && !dma_ops_domain(domain))
-		return ERR_PTR(-EBUSY);
-
-	if (domain != NULL)
-		return domain;
+	io_domain = iommu_get_domain_for_dev(dev);
+	if (!io_domain)
+		return NULL;
 
-	/* Device not bound yet - bind it */
-	dma_dom = find_protection_domain(devid);
-	if (!dma_dom)
-		dma_dom = amd_iommu_rlookup_table[devid]->default_dom;
-	attach_device(dev, &dma_dom->domain);
-	DUMP_printk("Using protection domain %d for device %s\n",
-		    dma_dom->domain.id, dev_name(dev));
+	domain = to_pdomain(io_domain);
+	if (!dma_ops_domain(domain))
+		return ERR_PTR(-EBUSY);
 
-	return &dma_dom->domain;
+	return domain;
 }
 
 static void update_device_table(struct protection_domain *domain)
@@ -3018,54 +2833,6 @@ static int amd_iommu_dma_supported(struct device *dev, u64 mask)
 	return check_device(dev);
 }
 
-/*
- * The function for pre-allocating protection domains.
- *
- * If the driver core informs the DMA layer if a driver grabs a device
- * we don't need to preallocate the protection domains anymore.
- * For now we have to.
- */
-static void __init prealloc_protection_domains(void)
-{
-	struct iommu_dev_data *dev_data;
-	struct dma_ops_domain *dma_dom;
-	struct pci_dev *dev = NULL;
-	u16 devid;
-
-	for_each_pci_dev(dev) {
-
-		/* Do we handle this device? */
-		if (!check_device(&dev->dev))
-			continue;
-
-		dev_data = get_dev_data(&dev->dev);
-		if (!amd_iommu_force_isolation && dev_data->iommu_v2) {
-			/* Make sure passthrough domain is allocated */
-			alloc_passthrough_domain();
-			dev_data->passthrough = true;
-			attach_device(&dev->dev, pt_domain);
-			pr_info("AMD-Vi: Using passthrough domain for device %s\n",
-				dev_name(&dev->dev));
-		}
-
-		/* Is there already any domain for it? */
-		if (domain_for_device(&dev->dev))
-			continue;
-
-		devid = get_device_id(&dev->dev);
-
-		dma_dom = dma_ops_domain_alloc();
-		if (!dma_dom)
-			continue;
-		init_unity_mappings_for_device(dma_dom, devid);
-		dma_dom->target_dev = devid;
-
-		attach_device(&dev->dev, &dma_dom->domain);
-
-		list_add_tail(&dma_dom->list, &iommu_pd_list);
-	}
-}
-
 static struct dma_map_ops amd_iommu_dma_ops = {
 	.alloc = alloc_coherent,
 	.free = free_coherent,
@@ -3076,75 +2843,24 @@ static struct dma_map_ops amd_iommu_dma_ops = {
 	.dma_supported = amd_iommu_dma_supported,
 };
 
-static unsigned device_dma_ops_init(void)
-{
-	struct iommu_dev_data *dev_data;
-	struct pci_dev *pdev = NULL;
-	unsigned unhandled = 0;
-
-	for_each_pci_dev(pdev) {
-		if (!check_device(&pdev->dev)) {
-
-			iommu_ignore_device(&pdev->dev);
-
-			unhandled += 1;
-			continue;
-		}
-
-		dev_data = get_dev_data(&pdev->dev);
-
-		if (!dev_data->passthrough)
-			pdev->dev.archdata.dma_ops = &amd_iommu_dma_ops;
-		else
-			pdev->dev.archdata.dma_ops = &nommu_dma_ops;
-	}
-
-	return unhandled;
-}
-
-/*
- * The function which clues the AMD IOMMU driver into dma_ops.
- */
-
-void __init amd_iommu_init_api(void)
+int __init amd_iommu_init_api(void)
 {
-	bus_set_iommu(&pci_bus_type, &amd_iommu_ops);
+	return bus_set_iommu(&pci_bus_type, &amd_iommu_ops);
 }
 
 int __init amd_iommu_init_dma_ops(void)
 {
-	struct amd_iommu *iommu;
-	int ret, unhandled;
-
-	/*
-	 * first allocate a default protection domain for every IOMMU we
-	 * found in the system. Devices not assigned to any other
-	 * protection domain will be assigned to the default one.
-	 */
-	for_each_iommu(iommu) {
-		iommu->default_dom = dma_ops_domain_alloc();
-		if (iommu->default_dom == NULL)
-			return -ENOMEM;
-		iommu->default_dom->domain.flags |= PD_DEFAULT_MASK;
-		ret = iommu_init_unity_mappings(iommu);
-		if (ret)
-			goto free_domains;
-	}
+	swiotlb        = iommu_pass_through ? 1 : 0;
+	iommu_detected = 1;
 
 	/*
-	 * Pre-allocate the protection domains for each device.
+	 * In case we don't initialize SWIOTLB (actually the common case
+	 * when AMD IOMMU is enabled), make sure there are global
+	 * dma_ops set as a fall-back for devices not handled by this
+	 * driver (for example non-PCI devices).
 	 */
-	prealloc_protection_domains();
-
-	iommu_detected = 1;
-	swiotlb = 0;
-
-	/* Make the driver finally visible to the drivers */
-	unhandled = device_dma_ops_init();
-	if (unhandled && max_pfn > MAX_DMA32_PFN) {
-		/* There are unhandled devices - initialize swiotlb for them */
-		swiotlb = 1;
-	}
+	if (!swiotlb)
+		dma_ops = &nommu_dma_ops;
 
 	amd_iommu_stats_init();
 
@@ -3154,14 +2870,6 @@ int __init amd_iommu_init_dma_ops(void)
 		pr_info("AMD-Vi: Lazy IO/TLB flushing enabled\n");
 
 	return 0;
-
-free_domains:
-
-	for_each_iommu(iommu) {
-		dma_ops_domain_free(iommu->default_dom);
-	}
-
-	return ret;
 }
 
 /*****************************************************************************
@@ -3203,6 +2911,18 @@ static void protection_domain_free(struct protection_domain *domain)
 	kfree(domain);
 }
 
+static int protection_domain_init(struct protection_domain *domain)
+{
+	spin_lock_init(&domain->lock);
+	mutex_init(&domain->api_lock);
+	domain->id = domain_id_alloc();
+	if (!domain->id)
+		return -ENOMEM;
+	INIT_LIST_HEAD(&domain->dev_list);
+
+	return 0;
+}
+
 static struct protection_domain *protection_domain_alloc(void)
 {
 	struct protection_domain *domain;
@@ -3211,12 +2931,8 @@ static struct protection_domain *protection_domain_alloc(void)
 	if (!domain)
 		return NULL;
 
-	spin_lock_init(&domain->lock);
-	mutex_init(&domain->api_lock);
-	domain->id = domain_id_alloc();
-	if (!domain->id)
+	if (protection_domain_init(domain))
 		goto out_err;
-	INIT_LIST_HEAD(&domain->dev_list);
 
 	add_domain_to_list(domain);
 
@@ -3228,48 +2944,49 @@ out_err:
 	return NULL;
 }
 
-static int __init alloc_passthrough_domain(void)
-{
-	if (pt_domain != NULL)
-		return 0;
-
-	/* allocate passthrough domain */
-	pt_domain = protection_domain_alloc();
-	if (!pt_domain)
-		return -ENOMEM;
-
-	pt_domain->mode = PAGE_MODE_NONE;
-
-	return 0;
-}
-
 static struct iommu_domain *amd_iommu_domain_alloc(unsigned type)
 {
 	struct protection_domain *pdomain;
+	struct dma_ops_domain *dma_domain;
 
-	/* We only support unmanaged domains for now */
-	if (type != IOMMU_DOMAIN_UNMANAGED)
-		return NULL;
-
-	pdomain = protection_domain_alloc();
-	if (!pdomain)
-		goto out_free;
+	switch (type) {
+	case IOMMU_DOMAIN_UNMANAGED:
+		pdomain = protection_domain_alloc();
+		if (!pdomain)
+			return NULL;
 
-	pdomain->mode    = PAGE_MODE_3_LEVEL;
-	pdomain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
-	if (!pdomain->pt_root)
-		goto out_free;
+		pdomain->mode    = PAGE_MODE_3_LEVEL;
+		pdomain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
+		if (!pdomain->pt_root) {
+			protection_domain_free(pdomain);
+			return NULL;
+		}
 
-	pdomain->domain.geometry.aperture_start = 0;
-	pdomain->domain.geometry.aperture_end   = ~0ULL;
-	pdomain->domain.geometry.force_aperture = true;
+		pdomain->domain.geometry.aperture_start = 0;
+		pdomain->domain.geometry.aperture_end   = ~0ULL;
+		pdomain->domain.geometry.force_aperture = true;
 
-	return &pdomain->domain;
+		break;
+	case IOMMU_DOMAIN_DMA:
+		dma_domain = dma_ops_domain_alloc();
+		if (!dma_domain) {
+			pr_err("AMD-Vi: Failed to allocate\n");
+			return NULL;
+		}
+		pdomain = &dma_domain->domain;
+		break;
+	case IOMMU_DOMAIN_IDENTITY:
+		pdomain = protection_domain_alloc();
+		if (!pdomain)
+			return NULL;
 
-out_free:
-	protection_domain_free(pdomain);
+		pdomain->mode = PAGE_MODE_NONE;
+		break;
+	default:
+		return NULL;
+	}
 
-	return NULL;
+	return &pdomain->domain;
 }
 
 static void amd_iommu_domain_free(struct iommu_domain *dom)
@@ -3419,6 +3136,47 @@ static bool amd_iommu_capable(enum iommu_cap cap)
 	return false;
 }
 
+static void amd_iommu_get_dm_regions(struct device *dev,
+				     struct list_head *head)
+{
+	struct unity_map_entry *entry;
+	u16 devid;
+
+	devid = get_device_id(dev);
+
+	list_for_each_entry(entry, &amd_iommu_unity_map, list) {
+		struct iommu_dm_region *region;
+
+		if (devid < entry->devid_start || devid > entry->devid_end)
+			continue;
+
+		region = kzalloc(sizeof(*region), GFP_KERNEL);
+		if (!region) {
+			pr_err("Out of memory allocating dm-regions for %s\n",
+				dev_name(dev));
+			return;
+		}
+
+		region->start = entry->address_start;
+		region->length = entry->address_end - entry->address_start;
+		if (entry->prot & IOMMU_PROT_IR)
+			region->prot |= IOMMU_READ;
+		if (entry->prot & IOMMU_PROT_IW)
+			region->prot |= IOMMU_WRITE;
+
+		list_add_tail(&region->list, head);
+	}
+}
+
+static void amd_iommu_put_dm_regions(struct device *dev,
+				     struct list_head *head)
+{
+	struct iommu_dm_region *entry, *next;
+
+	list_for_each_entry_safe(entry, next, head, list)
+		kfree(entry);
+}
+
 static const struct iommu_ops amd_iommu_ops = {
 	.capable = amd_iommu_capable,
 	.domain_alloc = amd_iommu_domain_alloc,
@@ -3429,6 +3187,10 @@ static const struct iommu_ops amd_iommu_ops = {
 	.unmap = amd_iommu_unmap,
 	.map_sg = default_iommu_map_sg,
 	.iova_to_phys = amd_iommu_iova_to_phys,
+	.add_device = amd_iommu_add_device,
+	.remove_device = amd_iommu_remove_device,
+	.get_dm_regions = amd_iommu_get_dm_regions,
+	.put_dm_regions = amd_iommu_put_dm_regions,
 	.pgsize_bitmap	= AMD_IOMMU_PGSIZES,
 };
 
@@ -3442,33 +3204,6 @@ static const struct iommu_ops amd_iommu_ops = {
  *
  *****************************************************************************/
 
-int __init amd_iommu_init_passthrough(void)
-{
-	struct iommu_dev_data *dev_data;
-	struct pci_dev *dev = NULL;
-	int ret;
-
-	ret = alloc_passthrough_domain();
-	if (ret)
-		return ret;
-
-	for_each_pci_dev(dev) {
-		if (!check_device(&dev->dev))
-			continue;
-
-		dev_data = get_dev_data(&dev->dev);
-		dev_data->passthrough = true;
-
-		attach_device(&dev->dev, pt_domain);
-	}
-
-	amd_iommu_stats_init();
-
-	pr_info("AMD-Vi: Initialized for Passthrough Mode\n");
-
-	return 0;
-}
-
 /* IOMMUv2 specific functions */
 int amd_iommu_register_ppr_notifier(struct notifier_block *nb)
 {
@@ -3583,7 +3318,12 @@ static int __flush_pasid(struct protection_domain *domain, int pasid,
 		struct amd_iommu *iommu;
 		int qdep;
 
-		BUG_ON(!dev_data->ats.enabled);
+		/*
+		   There might be non-IOMMUv2 capable devices in an IOMMUv2
+		 * domain.
+		 */
+		if (!dev_data->ats.enabled)
+			continue;
 
 		qdep  = dev_data->ats.qdep;
 		iommu = amd_iommu_rlookup_table[dev_data->devid];
@@ -3858,6 +3598,21 @@ union irte {
 	} fields;
 };
 
+struct irq_2_irte {
+	u16 devid; /* Device ID for IRTE table */
+	u16 index; /* Index into IRTE table*/
+};
+
+struct amd_ir_data {
+	struct irq_2_irte			irq_2_irte;
+	union irte				irte_entry;
+	union {
+		struct msi_msg			msi_entry;
+	};
+};
+
+static struct irq_chip amd_ir_chip;
+
 #define DTE_IRQ_PHYS_ADDR_MASK	(((1ULL << 45)-1) << 6)
 #define DTE_IRQ_REMAP_INTCTL    (2ULL << 60)
 #define DTE_IRQ_TABLE_LEN       (8ULL << 1)
@@ -3951,7 +3706,7 @@ out_unlock:
 	return table;
 }
 
-static int alloc_irq_index(struct irq_cfg *cfg, u16 devid, int count)
+static int alloc_irq_index(u16 devid, int count)
 {
 	struct irq_remap_table *table;
 	unsigned long flags;
@@ -3973,18 +3728,10 @@ static int alloc_irq_index(struct irq_cfg *cfg, u16 devid, int count)
 			c = 0;
 
 		if (c == count)	{
-			struct irq_2_irte *irte_info;
-
 			for (; c != 0; --c)
 				table->table[index - c + 1] = IRTE_ALLOCATED;
 
 			index -= count - 1;
-
-			cfg->remapped	      = 1;
-			irte_info             = &cfg->irq_2_irte;
-			irte_info->devid      = devid;
-			irte_info->index      = index;
-
 			goto out;
 		}
 	}
@@ -3997,22 +3744,6 @@ out:
 	return index;
 }
 
-static int get_irte(u16 devid, int index, union irte *irte)
-{
-	struct irq_remap_table *table;
-	unsigned long flags;
-
-	table = get_irq_table(devid, false);
-	if (!table)
-		return -ENOMEM;
-
-	spin_lock_irqsave(&table->lock, flags);
-	irte->val = table->table[index];
-	spin_unlock_irqrestore(&table->lock, flags);
-
-	return 0;
-}
-
 static int modify_irte(u16 devid, int index, union irte irte)
 {
 	struct irq_remap_table *table;
@@ -4059,243 +3790,316 @@ static void free_irte(u16 devid, int index)
 	iommu_completion_wait(iommu);
 }
 
-static int setup_ioapic_entry(int irq, struct IO_APIC_route_entry *entry,
-			      unsigned int destination, int vector,
-			      struct io_apic_irq_attr *attr)
+static int get_devid(struct irq_alloc_info *info)
 {
-	struct irq_remap_table *table;
-	struct irq_2_irte *irte_info;
-	struct irq_cfg *cfg;
-	union irte irte;
-	int ioapic_id;
-	int index;
-	int devid;
-	int ret;
+	int devid = -1;
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
-		return -EINVAL;
-
-	irte_info = &cfg->irq_2_irte;
-	ioapic_id = mpc_ioapic_id(attr->ioapic);
-	devid     = get_ioapic_devid(ioapic_id);
-
-	if (devid < 0)
-		return devid;
-
-	table = get_irq_table(devid, true);
-	if (table == NULL)
-		return -ENOMEM;
-
-	index = attr->ioapic_pin;
-
-	/* Setup IRQ remapping info */
-	cfg->remapped	      = 1;
-	irte_info->devid      = devid;
-	irte_info->index      = index;
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_IOAPIC:
+		devid     = get_ioapic_devid(info->ioapic_id);
+		break;
+	case X86_IRQ_ALLOC_TYPE_HPET:
+		devid     = get_hpet_devid(info->hpet_id);
+		break;
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		devid = get_device_id(&info->msi_dev->dev);
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
 
-	/* Setup IRTE for IOMMU */
-	irte.val		= 0;
-	irte.fields.vector      = vector;
-	irte.fields.int_type    = apic->irq_delivery_mode;
-	irte.fields.destination = destination;
-	irte.fields.dm          = apic->irq_dest_mode;
-	irte.fields.valid       = 1;
-
-	ret = modify_irte(devid, index, irte);
-	if (ret)
-		return ret;
+	return devid;
+}
 
-	/* Setup IOAPIC entry */
-	memset(entry, 0, sizeof(*entry));
+static struct irq_domain *get_ir_irq_domain(struct irq_alloc_info *info)
+{
+	struct amd_iommu *iommu;
+	int devid;
 
-	entry->vector        = index;
-	entry->mask          = 0;
-	entry->trigger       = attr->trigger;
-	entry->polarity      = attr->polarity;
+	if (!info)
+		return NULL;
 
-	/*
-	 * Mask level triggered irqs.
-	 */
-	if (attr->trigger)
-		entry->mask = 1;
+	devid = get_devid(info);
+	if (devid >= 0) {
+		iommu = amd_iommu_rlookup_table[devid];
+		if (iommu)
+			return iommu->ir_domain;
+	}
 
-	return 0;
+	return NULL;
 }
 
-static int set_affinity(struct irq_data *data, const struct cpumask *mask,
-			bool force)
+static struct irq_domain *get_irq_domain(struct irq_alloc_info *info)
 {
-	struct irq_2_irte *irte_info;
-	unsigned int dest, irq;
-	struct irq_cfg *cfg;
-	union irte irte;
-	int err;
-
-	if (!config_enabled(CONFIG_SMP))
-		return -1;
-
-	cfg       = irqd_cfg(data);
-	irq       = data->irq;
-	irte_info = &cfg->irq_2_irte;
+	struct amd_iommu *iommu;
+	int devid;
 
-	if (!cpumask_intersects(mask, cpu_online_mask))
-		return -EINVAL;
+	if (!info)
+		return NULL;
 
-	if (get_irte(irte_info->devid, irte_info->index, &irte))
-		return -EBUSY;
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		devid = get_device_id(&info->msi_dev->dev);
+		if (devid >= 0) {
+			iommu = amd_iommu_rlookup_table[devid];
+			if (iommu)
+				return iommu->msi_domain;
+		}
+		break;
+	default:
+		break;
+	}
 
-	if (assign_irq_vector(irq, cfg, mask))
-		return -EBUSY;
+	return NULL;
+}
 
-	err = apic->cpu_mask_to_apicid_and(cfg->domain, mask, &dest);
-	if (err) {
-		if (assign_irq_vector(irq, cfg, data->affinity))
-			pr_err("AMD-Vi: Failed to recover vector for irq %d\n", irq);
-		return err;
-	}
+struct irq_remap_ops amd_iommu_irq_ops = {
+	.prepare		= amd_iommu_prepare,
+	.enable			= amd_iommu_enable,
+	.disable		= amd_iommu_disable,
+	.reenable		= amd_iommu_reenable,
+	.enable_faulting	= amd_iommu_enable_faulting,
+	.get_ir_irq_domain	= get_ir_irq_domain,
+	.get_irq_domain		= get_irq_domain,
+};
 
-	irte.fields.vector      = cfg->vector;
-	irte.fields.destination = dest;
+static void irq_remapping_prepare_irte(struct amd_ir_data *data,
+				       struct irq_cfg *irq_cfg,
+				       struct irq_alloc_info *info,
+				       int devid, int index, int sub_handle)
+{
+	struct irq_2_irte *irte_info = &data->irq_2_irte;
+	struct msi_msg *msg = &data->msi_entry;
+	union irte *irte = &data->irte_entry;
+	struct IO_APIC_route_entry *entry;
 
-	modify_irte(irte_info->devid, irte_info->index, irte);
+	data->irq_2_irte.devid = devid;
+	data->irq_2_irte.index = index + sub_handle;
 
-	if (cfg->move_in_progress)
-		send_cleanup_vector(cfg);
+	/* Setup IRTE for IOMMU */
+	irte->val = 0;
+	irte->fields.vector      = irq_cfg->vector;
+	irte->fields.int_type    = apic->irq_delivery_mode;
+	irte->fields.destination = irq_cfg->dest_apicid;
+	irte->fields.dm          = apic->irq_dest_mode;
+	irte->fields.valid       = 1;
+
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_IOAPIC:
+		/* Setup IOAPIC entry */
+		entry = info->ioapic_entry;
+		info->ioapic_entry = NULL;
+		memset(entry, 0, sizeof(*entry));
+		entry->vector        = index;
+		entry->mask          = 0;
+		entry->trigger       = info->ioapic_trigger;
+		entry->polarity      = info->ioapic_polarity;
+		/* Mask level triggered irqs. */
+		if (info->ioapic_trigger)
+			entry->mask = 1;
+		break;
 
-	cpumask_copy(data->affinity, mask);
+	case X86_IRQ_ALLOC_TYPE_HPET:
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		msg->address_hi = MSI_ADDR_BASE_HI;
+		msg->address_lo = MSI_ADDR_BASE_LO;
+		msg->data = irte_info->index;
+		break;
 
-	return 0;
+	default:
+		BUG_ON(1);
+		break;
+	}
 }
 
-static int free_irq(int irq)
+static int irq_remapping_alloc(struct irq_domain *domain, unsigned int virq,
+			       unsigned int nr_irqs, void *arg)
 {
-	struct irq_2_irte *irte_info;
+	struct irq_alloc_info *info = arg;
+	struct irq_data *irq_data;
+	struct amd_ir_data *data;
 	struct irq_cfg *cfg;
+	int i, ret, devid;
+	int index = -1;
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
+	if (!info)
+		return -EINVAL;
+	if (nr_irqs > 1 && info->type != X86_IRQ_ALLOC_TYPE_MSI &&
+	    info->type != X86_IRQ_ALLOC_TYPE_MSIX)
 		return -EINVAL;
 
-	irte_info = &cfg->irq_2_irte;
-
-	free_irte(irte_info->devid, irte_info->index);
+	/*
+	 * With IRQ remapping enabled, don't need contiguous CPU vectors
+	 * to support multiple MSI interrupts.
+	 */
+	if (info->type == X86_IRQ_ALLOC_TYPE_MSI)
+		info->flags &= ~X86_IRQ_ALLOC_CONTIGUOUS_VECTORS;
 
-	return 0;
-}
+	devid = get_devid(info);
+	if (devid < 0)
+		return -EINVAL;
 
-static void compose_msi_msg(struct pci_dev *pdev,
-			    unsigned int irq, unsigned int dest,
-			    struct msi_msg *msg, u8 hpet_id)
-{
-	struct irq_2_irte *irte_info;
-	struct irq_cfg *cfg;
-	union irte irte;
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
+	if (ret < 0)
+		return ret;
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
-		return;
+	ret = -ENOMEM;
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		goto out_free_parent;
 
-	irte_info = &cfg->irq_2_irte;
+	if (info->type == X86_IRQ_ALLOC_TYPE_IOAPIC) {
+		if (get_irq_table(devid, true))
+			index = info->ioapic_pin;
+		else
+			ret = -ENOMEM;
+	} else {
+		index = alloc_irq_index(devid, nr_irqs);
+	}
+	if (index < 0) {
+		pr_warn("Failed to allocate IRTE\n");
+		kfree(data);
+		goto out_free_parent;
+	}
 
-	irte.val		= 0;
-	irte.fields.vector	= cfg->vector;
-	irte.fields.int_type    = apic->irq_delivery_mode;
-	irte.fields.destination	= dest;
-	irte.fields.dm		= apic->irq_dest_mode;
-	irte.fields.valid	= 1;
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(domain, virq + i);
+		cfg = irqd_cfg(irq_data);
+		if (!irq_data || !cfg) {
+			ret = -EINVAL;
+			goto out_free_data;
+		}
 
-	modify_irte(irte_info->devid, irte_info->index, irte);
+		if (i > 0) {
+			data = kzalloc(sizeof(*data), GFP_KERNEL);
+			if (!data)
+				goto out_free_data;
+		}
+		irq_data->hwirq = (devid << 16) + i;
+		irq_data->chip_data = data;
+		irq_data->chip = &amd_ir_chip;
+		irq_remapping_prepare_irte(data, cfg, info, devid, index, i);
+		irq_set_status_flags(virq + i, IRQ_MOVE_PCNTXT);
+	}
+	return 0;
 
-	msg->address_hi = MSI_ADDR_BASE_HI;
-	msg->address_lo = MSI_ADDR_BASE_LO;
-	msg->data       = irte_info->index;
+out_free_data:
+	for (i--; i >= 0; i--) {
+		irq_data = irq_domain_get_irq_data(domain, virq + i);
+		if (irq_data)
+			kfree(irq_data->chip_data);
+	}
+	for (i = 0; i < nr_irqs; i++)
+		free_irte(devid, index + i);
+out_free_parent:
+	irq_domain_free_irqs_common(domain, virq, nr_irqs);
+	return ret;
 }
 
-static int msi_alloc_irq(struct pci_dev *pdev, int irq, int nvec)
+static void irq_remapping_free(struct irq_domain *domain, unsigned int virq,
+			       unsigned int nr_irqs)
 {
-	struct irq_cfg *cfg;
-	int index;
-	u16 devid;
-
-	if (!pdev)
-		return -EINVAL;
+	struct irq_2_irte *irte_info;
+	struct irq_data *irq_data;
+	struct amd_ir_data *data;
+	int i;
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
-		return -EINVAL;
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(domain, virq  + i);
+		if (irq_data && irq_data->chip_data) {
+			data = irq_data->chip_data;
+			irte_info = &data->irq_2_irte;
+			free_irte(irte_info->devid, irte_info->index);
+			kfree(data);
+		}
+	}
+	irq_domain_free_irqs_common(domain, virq, nr_irqs);
+}
 
-	devid = get_device_id(&pdev->dev);
-	index = alloc_irq_index(cfg, devid, nvec);
+static void irq_remapping_activate(struct irq_domain *domain,
+				   struct irq_data *irq_data)
+{
+	struct amd_ir_data *data = irq_data->chip_data;
+	struct irq_2_irte *irte_info = &data->irq_2_irte;
 
-	return index < 0 ? MAX_IRQS_PER_TABLE : index;
+	modify_irte(irte_info->devid, irte_info->index, data->irte_entry);
 }
 
-static int msi_setup_irq(struct pci_dev *pdev, unsigned int irq,
-			 int index, int offset)
+static void irq_remapping_deactivate(struct irq_domain *domain,
+				     struct irq_data *irq_data)
 {
-	struct irq_2_irte *irte_info;
-	struct irq_cfg *cfg;
-	u16 devid;
+	struct amd_ir_data *data = irq_data->chip_data;
+	struct irq_2_irte *irte_info = &data->irq_2_irte;
+	union irte entry;
 
-	if (!pdev)
-		return -EINVAL;
+	entry.val = 0;
+	modify_irte(irte_info->devid, irte_info->index, data->irte_entry);
+}
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
-		return -EINVAL;
+static struct irq_domain_ops amd_ir_domain_ops = {
+	.alloc = irq_remapping_alloc,
+	.free = irq_remapping_free,
+	.activate = irq_remapping_activate,
+	.deactivate = irq_remapping_deactivate,
+};
 
-	if (index >= MAX_IRQS_PER_TABLE)
-		return 0;
+static int amd_ir_set_affinity(struct irq_data *data,
+			       const struct cpumask *mask, bool force)
+{
+	struct amd_ir_data *ir_data = data->chip_data;
+	struct irq_2_irte *irte_info = &ir_data->irq_2_irte;
+	struct irq_cfg *cfg = irqd_cfg(data);
+	struct irq_data *parent = data->parent_data;
+	int ret;
+
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret < 0 || ret == IRQ_SET_MASK_OK_DONE)
+		return ret;
 
-	devid		= get_device_id(&pdev->dev);
-	irte_info	= &cfg->irq_2_irte;
+	/*
+	 * Atomically updates the IRTE with the new destination, vector
+	 * and flushes the interrupt entry cache.
+	 */
+	ir_data->irte_entry.fields.vector = cfg->vector;
+	ir_data->irte_entry.fields.destination = cfg->dest_apicid;
+	modify_irte(irte_info->devid, irte_info->index, ir_data->irte_entry);
 
-	cfg->remapped	      = 1;
-	irte_info->devid      = devid;
-	irte_info->index      = index + offset;
+	/*
+	 * After this point, all the interrupts will start arriving
+	 * at the new destination. So, time to cleanup the previous
+	 * vector allocation.
+	 */
+	send_cleanup_vector(cfg);
 
-	return 0;
+	return IRQ_SET_MASK_OK_DONE;
 }
 
-static int alloc_hpet_msi(unsigned int irq, unsigned int id)
+static void ir_compose_msi_msg(struct irq_data *irq_data, struct msi_msg *msg)
 {
-	struct irq_2_irte *irte_info;
-	struct irq_cfg *cfg;
-	int index, devid;
+	struct amd_ir_data *ir_data = irq_data->chip_data;
 
-	cfg = irq_cfg(irq);
-	if (!cfg)
-		return -EINVAL;
+	*msg = ir_data->msi_entry;
+}
 
-	irte_info = &cfg->irq_2_irte;
-	devid     = get_hpet_devid(id);
-	if (devid < 0)
-		return devid;
+static struct irq_chip amd_ir_chip = {
+	.irq_ack = ir_ack_apic_edge,
+	.irq_set_affinity = amd_ir_set_affinity,
+	.irq_compose_msi_msg = ir_compose_msi_msg,
+};
 
-	index = alloc_irq_index(cfg, devid, 1);
-	if (index < 0)
-		return index;
+int amd_iommu_create_irq_domain(struct amd_iommu *iommu)
+{
+	iommu->ir_domain = irq_domain_add_tree(NULL, &amd_ir_domain_ops, iommu);
+	if (!iommu->ir_domain)
+		return -ENOMEM;
 
-	cfg->remapped	      = 1;
-	irte_info->devid      = devid;
-	irte_info->index      = index;
+	iommu->ir_domain->parent = arch_get_ir_parent_domain();
+	iommu->msi_domain = arch_create_msi_irq_domain(iommu->ir_domain);
 
 	return 0;
 }
-
-struct irq_remap_ops amd_iommu_irq_ops = {
-	.prepare		= amd_iommu_prepare,
-	.enable			= amd_iommu_enable,
-	.disable		= amd_iommu_disable,
-	.reenable		= amd_iommu_reenable,
-	.enable_faulting	= amd_iommu_enable_faulting,
-	.setup_ioapic_entry	= setup_ioapic_entry,
-	.set_affinity		= set_affinity,
-	.free_irq		= free_irq,
-	.compose_msi_msg	= compose_msi_msg,
-	.msi_alloc_irq		= msi_alloc_irq,
-	.msi_setup_irq		= msi_setup_irq,
-	.alloc_hpet_msi		= alloc_hpet_msi,
-};
 #endif
diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c
index 450ef5001..a24495eb4 100644
--- a/drivers/iommu/amd_iommu_init.c
+++ b/drivers/iommu/amd_iommu_init.c
@@ -226,6 +226,7 @@ static enum iommu_init_state init_state = IOMMU_START_STATE;
 
 static int amd_iommu_enable_interrupts(void);
 static int __init iommu_go_to_state(enum iommu_init_state state);
+static void init_device_table_dma(void);
 
 static inline void update_last_devid(u16 devid)
 {
@@ -1124,6 +1125,10 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h)
 	if (ret)
 		return ret;
 
+	ret = amd_iommu_create_irq_domain(iommu);
+	if (ret)
+		return ret;
+
 	/*
 	 * Make sure IOMMU is not considered to translate itself. The IVRS
 	 * table tells us so, but this is a lie!
@@ -1385,9 +1390,15 @@ static int __init amd_iommu_init_pci(void)
 			break;
 	}
 
-	ret = amd_iommu_init_devices();
+	init_device_table_dma();
+
+	for_each_iommu(iommu)
+		iommu_flush_all_caches(iommu);
+
+	ret = amd_iommu_init_api();
 
-	print_iommu_info();
+	if (!ret)
+		print_iommu_info();
 
 	return ret;
 }
@@ -1825,8 +1836,6 @@ static bool __init check_ioapic_information(void)
 
 static void __init free_dma_resources(void)
 {
-	amd_iommu_uninit_devices();
-
 	free_pages((unsigned long)amd_iommu_pd_alloc_bitmap,
 		   get_order(MAX_DOMAIN_ID/8));
 
@@ -2017,31 +2026,6 @@ static bool detect_ivrs(void)
 	return true;
 }
 
-static int amd_iommu_init_dma(void)
-{
-	struct amd_iommu *iommu;
-	int ret;
-
-	if (iommu_pass_through)
-		ret = amd_iommu_init_passthrough();
-	else
-		ret = amd_iommu_init_dma_ops();
-
-	if (ret)
-		return ret;
-
-	init_device_table_dma();
-
-	for_each_iommu(iommu)
-		iommu_flush_all_caches(iommu);
-
-	amd_iommu_init_api();
-
-	amd_iommu_init_notifier();
-
-	return 0;
-}
-
 /****************************************************************************
  *
  * AMD IOMMU Initialization State Machine
@@ -2081,7 +2065,7 @@ static int __init state_next(void)
 		init_state = ret ? IOMMU_INIT_ERROR : IOMMU_INTERRUPTS_EN;
 		break;
 	case IOMMU_INTERRUPTS_EN:
-		ret = amd_iommu_init_dma();
+		ret = amd_iommu_init_dma_ops();
 		init_state = ret ? IOMMU_INIT_ERROR : IOMMU_DMA_OPS;
 		break;
 	case IOMMU_DMA_OPS:
diff --git a/drivers/iommu/amd_iommu_proto.h b/drivers/iommu/amd_iommu_proto.h
index 72b0fd455..0bd9eb374 100644
--- a/drivers/iommu/amd_iommu_proto.h
+++ b/drivers/iommu/amd_iommu_proto.h
@@ -30,7 +30,7 @@ extern void amd_iommu_reset_cmd_buffer(struct amd_iommu *iommu);
 extern int amd_iommu_init_devices(void);
 extern void amd_iommu_uninit_devices(void);
 extern void amd_iommu_init_notifier(void);
-extern void amd_iommu_init_api(void);
+extern int amd_iommu_init_api(void);
 
 /* Needed for interrupt remapping */
 extern int amd_iommu_prepare(void);
@@ -62,6 +62,15 @@ extern u8 amd_iommu_pc_get_max_counters(u16 devid);
 extern int amd_iommu_pc_get_set_reg_val(u16 devid, u8 bank, u8 cntr, u8 fxn,
 				    u64 *value, bool is_write);
 
+#ifdef CONFIG_IRQ_REMAP
+extern int amd_iommu_create_irq_domain(struct amd_iommu *iommu);
+#else
+static inline int amd_iommu_create_irq_domain(struct amd_iommu *iommu)
+{
+	return 0;
+}
+#endif
+
 #define PPR_SUCCESS			0x0
 #define PPR_INVALID			0x1
 #define PPR_FAILURE			0xf
diff --git a/drivers/iommu/amd_iommu_types.h b/drivers/iommu/amd_iommu_types.h
index 05030e523..f65908841 100644
--- a/drivers/iommu/amd_iommu_types.h
+++ b/drivers/iommu/amd_iommu_types.h
@@ -398,6 +398,7 @@ struct amd_iommu_fault {
 
 
 struct iommu_domain;
+struct irq_domain;
 
 /*
  * This structure contains generic data for  IOMMU protection domains
@@ -446,8 +447,6 @@ struct aperture_range {
  * Data container for a dma_ops specific protection domain
  */
 struct dma_ops_domain {
-	struct list_head list;
-
 	/* generic protection domain information */
 	struct protection_domain domain;
 
@@ -462,12 +461,6 @@ struct dma_ops_domain {
 
 	/* This will be set to true when TLB needs to be flushed */
 	bool need_flush;
-
-	/*
-	 * if this is a preallocated domain, keep the device for which it was
-	 * preallocated in this variable
-	 */
-	u16 target_dev;
 };
 
 /*
@@ -552,9 +545,6 @@ struct amd_iommu {
 	/* if one, we need to send a completion wait command */
 	bool need_sync;
 
-	/* default dma_ops domain for that IOMMU */
-	struct dma_ops_domain *default_dom;
-
 	/* IOMMU sysfs device */
 	struct device *iommu_dev;
 
@@ -579,6 +569,10 @@ struct amd_iommu {
 	/* The maximum PC banks and counters/bank (PCSup=1) */
 	u8 max_banks;
 	u8 max_counters;
+#ifdef CONFIG_IRQ_REMAP
+	struct irq_domain *ir_domain;
+	struct irq_domain *msi_domain;
+#endif
 };
 
 struct devid_map {
diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c
index 3465faf18..f7b875bb7 100644
--- a/drivers/iommu/amd_iommu_v2.c
+++ b/drivers/iommu/amd_iommu_v2.c
@@ -132,11 +132,19 @@ static struct device_state *get_device_state(u16 devid)
 
 static void free_device_state(struct device_state *dev_state)
 {
+	struct iommu_group *group;
+
 	/*
 	 * First detach device from domain - No more PRI requests will arrive
 	 * from that device after it is unbound from the IOMMUv2 domain.
 	 */
-	iommu_detach_device(dev_state->domain, &dev_state->pdev->dev);
+	group = iommu_group_get(&dev_state->pdev->dev);
+	if (WARN_ON(!group))
+		return;
+
+	iommu_detach_group(dev_state->domain, group);
+
+	iommu_group_put(group);
 
 	/* Everything is down now, free the IOMMUv2 domain */
 	iommu_domain_free(dev_state->domain);
@@ -731,6 +739,7 @@ EXPORT_SYMBOL(amd_iommu_unbind_pasid);
 int amd_iommu_init_device(struct pci_dev *pdev, int pasids)
 {
 	struct device_state *dev_state;
+	struct iommu_group *group;
 	unsigned long flags;
 	int ret, tmp;
 	u16 devid;
@@ -776,10 +785,16 @@ int amd_iommu_init_device(struct pci_dev *pdev, int pasids)
 	if (ret)
 		goto out_free_domain;
 
-	ret = iommu_attach_device(dev_state->domain, &pdev->dev);
-	if (ret != 0)
+	group = iommu_group_get(&pdev->dev);
+	if (!group)
 		goto out_free_domain;
 
+	ret = iommu_attach_group(dev_state->domain, group);
+	if (ret != 0)
+		goto out_drop_group;
+
+	iommu_group_put(group);
+
 	spin_lock_irqsave(&state_lock, flags);
 
 	if (__get_device_state(devid) != NULL) {
@@ -794,6 +809,9 @@ int amd_iommu_init_device(struct pci_dev *pdev, int pasids)
 
 	return 0;
 
+out_drop_group:
+	iommu_group_put(group);
+
 out_free_domain:
 	iommu_domain_free(dev_state->domain);
 
diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
new file mode 100644
index 000000000..da902baaa
--- /dev/null
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -0,0 +1,2701 @@
+/*
+ * IOMMU API for ARM architected SMMUv3 implementations.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (C) 2015 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ *
+ * This driver is powered by bad coffee and bombay mix.
+ */
+
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/iommu.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+
+#include "io-pgtable.h"
+
+/* MMIO registers */
+#define ARM_SMMU_IDR0			0x0
+#define IDR0_ST_LVL_SHIFT		27
+#define IDR0_ST_LVL_MASK		0x3
+#define IDR0_ST_LVL_2LVL		(1 << IDR0_ST_LVL_SHIFT)
+#define IDR0_STALL_MODEL		(3 << 24)
+#define IDR0_TTENDIAN_SHIFT		21
+#define IDR0_TTENDIAN_MASK		0x3
+#define IDR0_TTENDIAN_LE		(2 << IDR0_TTENDIAN_SHIFT)
+#define IDR0_TTENDIAN_BE		(3 << IDR0_TTENDIAN_SHIFT)
+#define IDR0_TTENDIAN_MIXED		(0 << IDR0_TTENDIAN_SHIFT)
+#define IDR0_CD2L			(1 << 19)
+#define IDR0_VMID16			(1 << 18)
+#define IDR0_PRI			(1 << 16)
+#define IDR0_SEV			(1 << 14)
+#define IDR0_MSI			(1 << 13)
+#define IDR0_ASID16			(1 << 12)
+#define IDR0_ATS			(1 << 10)
+#define IDR0_HYP			(1 << 9)
+#define IDR0_COHACC			(1 << 4)
+#define IDR0_TTF_SHIFT			2
+#define IDR0_TTF_MASK			0x3
+#define IDR0_TTF_AARCH64		(2 << IDR0_TTF_SHIFT)
+#define IDR0_S1P			(1 << 1)
+#define IDR0_S2P			(1 << 0)
+
+#define ARM_SMMU_IDR1			0x4
+#define IDR1_TABLES_PRESET		(1 << 30)
+#define IDR1_QUEUES_PRESET		(1 << 29)
+#define IDR1_REL			(1 << 28)
+#define IDR1_CMDQ_SHIFT			21
+#define IDR1_CMDQ_MASK			0x1f
+#define IDR1_EVTQ_SHIFT			16
+#define IDR1_EVTQ_MASK			0x1f
+#define IDR1_PRIQ_SHIFT			11
+#define IDR1_PRIQ_MASK			0x1f
+#define IDR1_SSID_SHIFT			6
+#define IDR1_SSID_MASK			0x1f
+#define IDR1_SID_SHIFT			0
+#define IDR1_SID_MASK			0x3f
+
+#define ARM_SMMU_IDR5			0x14
+#define IDR5_STALL_MAX_SHIFT		16
+#define IDR5_STALL_MAX_MASK		0xffff
+#define IDR5_GRAN64K			(1 << 6)
+#define IDR5_GRAN16K			(1 << 5)
+#define IDR5_GRAN4K			(1 << 4)
+#define IDR5_OAS_SHIFT			0
+#define IDR5_OAS_MASK			0x7
+#define IDR5_OAS_32_BIT			(0 << IDR5_OAS_SHIFT)
+#define IDR5_OAS_36_BIT			(1 << IDR5_OAS_SHIFT)
+#define IDR5_OAS_40_BIT			(2 << IDR5_OAS_SHIFT)
+#define IDR5_OAS_42_BIT			(3 << IDR5_OAS_SHIFT)
+#define IDR5_OAS_44_BIT			(4 << IDR5_OAS_SHIFT)
+#define IDR5_OAS_48_BIT			(5 << IDR5_OAS_SHIFT)
+
+#define ARM_SMMU_CR0			0x20
+#define CR0_CMDQEN			(1 << 3)
+#define CR0_EVTQEN			(1 << 2)
+#define CR0_PRIQEN			(1 << 1)
+#define CR0_SMMUEN			(1 << 0)
+
+#define ARM_SMMU_CR0ACK			0x24
+
+#define ARM_SMMU_CR1			0x28
+#define CR1_SH_NSH			0
+#define CR1_SH_OSH			2
+#define CR1_SH_ISH			3
+#define CR1_CACHE_NC			0
+#define CR1_CACHE_WB			1
+#define CR1_CACHE_WT			2
+#define CR1_TABLE_SH_SHIFT		10
+#define CR1_TABLE_OC_SHIFT		8
+#define CR1_TABLE_IC_SHIFT		6
+#define CR1_QUEUE_SH_SHIFT		4
+#define CR1_QUEUE_OC_SHIFT		2
+#define CR1_QUEUE_IC_SHIFT		0
+
+#define ARM_SMMU_CR2			0x2c
+#define CR2_PTM				(1 << 2)
+#define CR2_RECINVSID			(1 << 1)
+#define CR2_E2H				(1 << 0)
+
+#define ARM_SMMU_IRQ_CTRL		0x50
+#define IRQ_CTRL_EVTQ_IRQEN		(1 << 2)
+#define IRQ_CTRL_GERROR_IRQEN		(1 << 0)
+
+#define ARM_SMMU_IRQ_CTRLACK		0x54
+
+#define ARM_SMMU_GERROR			0x60
+#define GERROR_SFM_ERR			(1 << 8)
+#define GERROR_MSI_GERROR_ABT_ERR	(1 << 7)
+#define GERROR_MSI_PRIQ_ABT_ERR		(1 << 6)
+#define GERROR_MSI_EVTQ_ABT_ERR		(1 << 5)
+#define GERROR_MSI_CMDQ_ABT_ERR		(1 << 4)
+#define GERROR_PRIQ_ABT_ERR		(1 << 3)
+#define GERROR_EVTQ_ABT_ERR		(1 << 2)
+#define GERROR_CMDQ_ERR			(1 << 0)
+#define GERROR_ERR_MASK			0xfd
+
+#define ARM_SMMU_GERRORN		0x64
+
+#define ARM_SMMU_GERROR_IRQ_CFG0	0x68
+#define ARM_SMMU_GERROR_IRQ_CFG1	0x70
+#define ARM_SMMU_GERROR_IRQ_CFG2	0x74
+
+#define ARM_SMMU_STRTAB_BASE		0x80
+#define STRTAB_BASE_RA			(1UL << 62)
+#define STRTAB_BASE_ADDR_SHIFT		6
+#define STRTAB_BASE_ADDR_MASK		0x3ffffffffffUL
+
+#define ARM_SMMU_STRTAB_BASE_CFG	0x88
+#define STRTAB_BASE_CFG_LOG2SIZE_SHIFT	0
+#define STRTAB_BASE_CFG_LOG2SIZE_MASK	0x3f
+#define STRTAB_BASE_CFG_SPLIT_SHIFT	6
+#define STRTAB_BASE_CFG_SPLIT_MASK	0x1f
+#define STRTAB_BASE_CFG_FMT_SHIFT	16
+#define STRTAB_BASE_CFG_FMT_MASK	0x3
+#define STRTAB_BASE_CFG_FMT_LINEAR	(0 << STRTAB_BASE_CFG_FMT_SHIFT)
+#define STRTAB_BASE_CFG_FMT_2LVL	(1 << STRTAB_BASE_CFG_FMT_SHIFT)
+
+#define ARM_SMMU_CMDQ_BASE		0x90
+#define ARM_SMMU_CMDQ_PROD		0x98
+#define ARM_SMMU_CMDQ_CONS		0x9c
+
+#define ARM_SMMU_EVTQ_BASE		0xa0
+#define ARM_SMMU_EVTQ_PROD		0x100a8
+#define ARM_SMMU_EVTQ_CONS		0x100ac
+#define ARM_SMMU_EVTQ_IRQ_CFG0		0xb0
+#define ARM_SMMU_EVTQ_IRQ_CFG1		0xb8
+#define ARM_SMMU_EVTQ_IRQ_CFG2		0xbc
+
+#define ARM_SMMU_PRIQ_BASE		0xc0
+#define ARM_SMMU_PRIQ_PROD		0x100c8
+#define ARM_SMMU_PRIQ_CONS		0x100cc
+#define ARM_SMMU_PRIQ_IRQ_CFG0		0xd0
+#define ARM_SMMU_PRIQ_IRQ_CFG1		0xd8
+#define ARM_SMMU_PRIQ_IRQ_CFG2		0xdc
+
+/* Common MSI config fields */
+#define MSI_CFG0_SH_SHIFT		60
+#define MSI_CFG0_SH_NSH			(0UL << MSI_CFG0_SH_SHIFT)
+#define MSI_CFG0_SH_OSH			(2UL << MSI_CFG0_SH_SHIFT)
+#define MSI_CFG0_SH_ISH			(3UL << MSI_CFG0_SH_SHIFT)
+#define MSI_CFG0_MEMATTR_SHIFT		56
+#define MSI_CFG0_MEMATTR_DEVICE_nGnRE	(0x1 << MSI_CFG0_MEMATTR_SHIFT)
+#define MSI_CFG0_ADDR_SHIFT		2
+#define MSI_CFG0_ADDR_MASK		0x3fffffffffffUL
+
+#define Q_IDX(q, p)			((p) & ((1 << (q)->max_n_shift) - 1))
+#define Q_WRP(q, p)			((p) & (1 << (q)->max_n_shift))
+#define Q_OVERFLOW_FLAG			(1 << 31)
+#define Q_OVF(q, p)			((p) & Q_OVERFLOW_FLAG)
+#define Q_ENT(q, p)			((q)->base +			\
+					 Q_IDX(q, p) * (q)->ent_dwords)
+
+#define Q_BASE_RWA			(1UL << 62)
+#define Q_BASE_ADDR_SHIFT		5
+#define Q_BASE_ADDR_MASK		0xfffffffffffUL
+#define Q_BASE_LOG2SIZE_SHIFT		0
+#define Q_BASE_LOG2SIZE_MASK		0x1fUL
+
+/*
+ * Stream table.
+ *
+ * Linear: Enough to cover 1 << IDR1.SIDSIZE entries
+ * 2lvl: 128k L1 entries,
+ *       256 lazy entries per table (each table covers a PCI bus)
+ */
+#define STRTAB_L1_SZ_SHIFT		20
+#define STRTAB_SPLIT			8
+
+#define STRTAB_L1_DESC_DWORDS		1
+#define STRTAB_L1_DESC_SPAN_SHIFT	0
+#define STRTAB_L1_DESC_SPAN_MASK	0x1fUL
+#define STRTAB_L1_DESC_L2PTR_SHIFT	6
+#define STRTAB_L1_DESC_L2PTR_MASK	0x3ffffffffffUL
+
+#define STRTAB_STE_DWORDS		8
+#define STRTAB_STE_0_V			(1UL << 0)
+#define STRTAB_STE_0_CFG_SHIFT		1
+#define STRTAB_STE_0_CFG_MASK		0x7UL
+#define STRTAB_STE_0_CFG_ABORT		(0UL << STRTAB_STE_0_CFG_SHIFT)
+#define STRTAB_STE_0_CFG_BYPASS		(4UL << STRTAB_STE_0_CFG_SHIFT)
+#define STRTAB_STE_0_CFG_S1_TRANS	(5UL << STRTAB_STE_0_CFG_SHIFT)
+#define STRTAB_STE_0_CFG_S2_TRANS	(6UL << STRTAB_STE_0_CFG_SHIFT)
+
+#define STRTAB_STE_0_S1FMT_SHIFT	4
+#define STRTAB_STE_0_S1FMT_LINEAR	(0UL << STRTAB_STE_0_S1FMT_SHIFT)
+#define STRTAB_STE_0_S1CTXPTR_SHIFT	6
+#define STRTAB_STE_0_S1CTXPTR_MASK	0x3ffffffffffUL
+#define STRTAB_STE_0_S1CDMAX_SHIFT	59
+#define STRTAB_STE_0_S1CDMAX_MASK	0x1fUL
+
+#define STRTAB_STE_1_S1C_CACHE_NC	0UL
+#define STRTAB_STE_1_S1C_CACHE_WBRA	1UL
+#define STRTAB_STE_1_S1C_CACHE_WT	2UL
+#define STRTAB_STE_1_S1C_CACHE_WB	3UL
+#define STRTAB_STE_1_S1C_SH_NSH		0UL
+#define STRTAB_STE_1_S1C_SH_OSH		2UL
+#define STRTAB_STE_1_S1C_SH_ISH		3UL
+#define STRTAB_STE_1_S1CIR_SHIFT	2
+#define STRTAB_STE_1_S1COR_SHIFT	4
+#define STRTAB_STE_1_S1CSH_SHIFT	6
+
+#define STRTAB_STE_1_S1STALLD		(1UL << 27)
+
+#define STRTAB_STE_1_EATS_ABT		0UL
+#define STRTAB_STE_1_EATS_TRANS		1UL
+#define STRTAB_STE_1_EATS_S1CHK		2UL
+#define STRTAB_STE_1_EATS_SHIFT		28
+
+#define STRTAB_STE_1_STRW_NSEL1		0UL
+#define STRTAB_STE_1_STRW_EL2		2UL
+#define STRTAB_STE_1_STRW_SHIFT		30
+
+#define STRTAB_STE_2_S2VMID_SHIFT	0
+#define STRTAB_STE_2_S2VMID_MASK	0xffffUL
+#define STRTAB_STE_2_VTCR_SHIFT		32
+#define STRTAB_STE_2_VTCR_MASK		0x7ffffUL
+#define STRTAB_STE_2_S2AA64		(1UL << 51)
+#define STRTAB_STE_2_S2ENDI		(1UL << 52)
+#define STRTAB_STE_2_S2PTW		(1UL << 54)
+#define STRTAB_STE_2_S2R		(1UL << 58)
+
+#define STRTAB_STE_3_S2TTB_SHIFT	4
+#define STRTAB_STE_3_S2TTB_MASK		0xfffffffffffUL
+
+/* Context descriptor (stage-1 only) */
+#define CTXDESC_CD_DWORDS		8
+#define CTXDESC_CD_0_TCR_T0SZ_SHIFT	0
+#define ARM64_TCR_T0SZ_SHIFT		0
+#define ARM64_TCR_T0SZ_MASK		0x1fUL
+#define CTXDESC_CD_0_TCR_TG0_SHIFT	6
+#define ARM64_TCR_TG0_SHIFT		14
+#define ARM64_TCR_TG0_MASK		0x3UL
+#define CTXDESC_CD_0_TCR_IRGN0_SHIFT	8
+#define ARM64_TCR_IRGN0_SHIFT		8
+#define ARM64_TCR_IRGN0_MASK		0x3UL
+#define CTXDESC_CD_0_TCR_ORGN0_SHIFT	10
+#define ARM64_TCR_ORGN0_SHIFT		10
+#define ARM64_TCR_ORGN0_MASK		0x3UL
+#define CTXDESC_CD_0_TCR_SH0_SHIFT	12
+#define ARM64_TCR_SH0_SHIFT		12
+#define ARM64_TCR_SH0_MASK		0x3UL
+#define CTXDESC_CD_0_TCR_EPD0_SHIFT	14
+#define ARM64_TCR_EPD0_SHIFT		7
+#define ARM64_TCR_EPD0_MASK		0x1UL
+#define CTXDESC_CD_0_TCR_EPD1_SHIFT	30
+#define ARM64_TCR_EPD1_SHIFT		23
+#define ARM64_TCR_EPD1_MASK		0x1UL
+
+#define CTXDESC_CD_0_ENDI		(1UL << 15)
+#define CTXDESC_CD_0_V			(1UL << 31)
+
+#define CTXDESC_CD_0_TCR_IPS_SHIFT	32
+#define ARM64_TCR_IPS_SHIFT		32
+#define ARM64_TCR_IPS_MASK		0x7UL
+#define CTXDESC_CD_0_TCR_TBI0_SHIFT	38
+#define ARM64_TCR_TBI0_SHIFT		37
+#define ARM64_TCR_TBI0_MASK		0x1UL
+
+#define CTXDESC_CD_0_AA64		(1UL << 41)
+#define CTXDESC_CD_0_R			(1UL << 45)
+#define CTXDESC_CD_0_A			(1UL << 46)
+#define CTXDESC_CD_0_ASET_SHIFT		47
+#define CTXDESC_CD_0_ASET_SHARED	(0UL << CTXDESC_CD_0_ASET_SHIFT)
+#define CTXDESC_CD_0_ASET_PRIVATE	(1UL << CTXDESC_CD_0_ASET_SHIFT)
+#define CTXDESC_CD_0_ASID_SHIFT		48
+#define CTXDESC_CD_0_ASID_MASK		0xffffUL
+
+#define CTXDESC_CD_1_TTB0_SHIFT		4
+#define CTXDESC_CD_1_TTB0_MASK		0xfffffffffffUL
+
+#define CTXDESC_CD_3_MAIR_SHIFT		0
+
+/* Convert between AArch64 (CPU) TCR format and SMMU CD format */
+#define ARM_SMMU_TCR2CD(tcr, fld)					\
+	(((tcr) >> ARM64_TCR_##fld##_SHIFT & ARM64_TCR_##fld##_MASK)	\
+	 << CTXDESC_CD_0_TCR_##fld##_SHIFT)
+
+/* Command queue */
+#define CMDQ_ENT_DWORDS			2
+#define CMDQ_MAX_SZ_SHIFT		8
+
+#define CMDQ_ERR_SHIFT			24
+#define CMDQ_ERR_MASK			0x7f
+#define CMDQ_ERR_CERROR_NONE_IDX	0
+#define CMDQ_ERR_CERROR_ILL_IDX		1
+#define CMDQ_ERR_CERROR_ABT_IDX		2
+
+#define CMDQ_0_OP_SHIFT			0
+#define CMDQ_0_OP_MASK			0xffUL
+#define CMDQ_0_SSV			(1UL << 11)
+
+#define CMDQ_PREFETCH_0_SID_SHIFT	32
+#define CMDQ_PREFETCH_1_SIZE_SHIFT	0
+#define CMDQ_PREFETCH_1_ADDR_MASK	~0xfffUL
+
+#define CMDQ_CFGI_0_SID_SHIFT		32
+#define CMDQ_CFGI_0_SID_MASK		0xffffffffUL
+#define CMDQ_CFGI_1_LEAF		(1UL << 0)
+#define CMDQ_CFGI_1_RANGE_SHIFT		0
+#define CMDQ_CFGI_1_RANGE_MASK		0x1fUL
+
+#define CMDQ_TLBI_0_VMID_SHIFT		32
+#define CMDQ_TLBI_0_ASID_SHIFT		48
+#define CMDQ_TLBI_1_LEAF		(1UL << 0)
+#define CMDQ_TLBI_1_ADDR_MASK		~0xfffUL
+
+#define CMDQ_PRI_0_SSID_SHIFT		12
+#define CMDQ_PRI_0_SSID_MASK		0xfffffUL
+#define CMDQ_PRI_0_SID_SHIFT		32
+#define CMDQ_PRI_0_SID_MASK		0xffffffffUL
+#define CMDQ_PRI_1_GRPID_SHIFT		0
+#define CMDQ_PRI_1_GRPID_MASK		0x1ffUL
+#define CMDQ_PRI_1_RESP_SHIFT		12
+#define CMDQ_PRI_1_RESP_DENY		(0UL << CMDQ_PRI_1_RESP_SHIFT)
+#define CMDQ_PRI_1_RESP_FAIL		(1UL << CMDQ_PRI_1_RESP_SHIFT)
+#define CMDQ_PRI_1_RESP_SUCC		(2UL << CMDQ_PRI_1_RESP_SHIFT)
+
+#define CMDQ_SYNC_0_CS_SHIFT		12
+#define CMDQ_SYNC_0_CS_NONE		(0UL << CMDQ_SYNC_0_CS_SHIFT)
+#define CMDQ_SYNC_0_CS_SEV		(2UL << CMDQ_SYNC_0_CS_SHIFT)
+
+/* Event queue */
+#define EVTQ_ENT_DWORDS			4
+#define EVTQ_MAX_SZ_SHIFT		7
+
+#define EVTQ_0_ID_SHIFT			0
+#define EVTQ_0_ID_MASK			0xffUL
+
+/* PRI queue */
+#define PRIQ_ENT_DWORDS			2
+#define PRIQ_MAX_SZ_SHIFT		8
+
+#define PRIQ_0_SID_SHIFT		0
+#define PRIQ_0_SID_MASK			0xffffffffUL
+#define PRIQ_0_SSID_SHIFT		32
+#define PRIQ_0_SSID_MASK		0xfffffUL
+#define PRIQ_0_OF			(1UL << 57)
+#define PRIQ_0_PERM_PRIV		(1UL << 58)
+#define PRIQ_0_PERM_EXEC		(1UL << 59)
+#define PRIQ_0_PERM_READ		(1UL << 60)
+#define PRIQ_0_PERM_WRITE		(1UL << 61)
+#define PRIQ_0_PRG_LAST			(1UL << 62)
+#define PRIQ_0_SSID_V			(1UL << 63)
+
+#define PRIQ_1_PRG_IDX_SHIFT		0
+#define PRIQ_1_PRG_IDX_MASK		0x1ffUL
+#define PRIQ_1_ADDR_SHIFT		12
+#define PRIQ_1_ADDR_MASK		0xfffffffffffffUL
+
+/* High-level queue structures */
+#define ARM_SMMU_POLL_TIMEOUT_US	100
+
+static bool disable_bypass;
+module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
+MODULE_PARM_DESC(disable_bypass,
+	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");
+
+enum pri_resp {
+	PRI_RESP_DENY,
+	PRI_RESP_FAIL,
+	PRI_RESP_SUCC,
+};
+
+struct arm_smmu_cmdq_ent {
+	/* Common fields */
+	u8				opcode;
+	bool				substream_valid;
+
+	/* Command-specific fields */
+	union {
+		#define CMDQ_OP_PREFETCH_CFG	0x1
+		struct {
+			u32			sid;
+			u8			size;
+			u64			addr;
+		} prefetch;
+
+		#define CMDQ_OP_CFGI_STE	0x3
+		#define CMDQ_OP_CFGI_ALL	0x4
+		struct {
+			u32			sid;
+			union {
+				bool		leaf;
+				u8		span;
+			};
+		} cfgi;
+
+		#define CMDQ_OP_TLBI_NH_ASID	0x11
+		#define CMDQ_OP_TLBI_NH_VA	0x12
+		#define CMDQ_OP_TLBI_EL2_ALL	0x20
+		#define CMDQ_OP_TLBI_S12_VMALL	0x28
+		#define CMDQ_OP_TLBI_S2_IPA	0x2a
+		#define CMDQ_OP_TLBI_NSNH_ALL	0x30
+		struct {
+			u16			asid;
+			u16			vmid;
+			bool			leaf;
+			u64			addr;
+		} tlbi;
+
+		#define CMDQ_OP_PRI_RESP	0x41
+		struct {
+			u32			sid;
+			u32			ssid;
+			u16			grpid;
+			enum pri_resp		resp;
+		} pri;
+
+		#define CMDQ_OP_CMD_SYNC	0x46
+	};
+};
+
+struct arm_smmu_queue {
+	int				irq; /* Wired interrupt */
+
+	__le64				*base;
+	dma_addr_t			base_dma;
+	u64				q_base;
+
+	size_t				ent_dwords;
+	u32				max_n_shift;
+	u32				prod;
+	u32				cons;
+
+	u32 __iomem			*prod_reg;
+	u32 __iomem			*cons_reg;
+};
+
+struct arm_smmu_cmdq {
+	struct arm_smmu_queue		q;
+	spinlock_t			lock;
+};
+
+struct arm_smmu_evtq {
+	struct arm_smmu_queue		q;
+	u32				max_stalls;
+};
+
+struct arm_smmu_priq {
+	struct arm_smmu_queue		q;
+};
+
+/* High-level stream table and context descriptor structures */
+struct arm_smmu_strtab_l1_desc {
+	u8				span;
+
+	__le64				*l2ptr;
+	dma_addr_t			l2ptr_dma;
+};
+
+struct arm_smmu_s1_cfg {
+	__le64				*cdptr;
+	dma_addr_t			cdptr_dma;
+
+	struct arm_smmu_ctx_desc {
+		u16	asid;
+		u64	ttbr;
+		u64	tcr;
+		u64	mair;
+	}				cd;
+};
+
+struct arm_smmu_s2_cfg {
+	u16				vmid;
+	u64				vttbr;
+	u64				vtcr;
+};
+
+struct arm_smmu_strtab_ent {
+	bool				valid;
+
+	bool				bypass;	/* Overrides s1/s2 config */
+	struct arm_smmu_s1_cfg		*s1_cfg;
+	struct arm_smmu_s2_cfg		*s2_cfg;
+};
+
+struct arm_smmu_strtab_cfg {
+	__le64				*strtab;
+	dma_addr_t			strtab_dma;
+	struct arm_smmu_strtab_l1_desc	*l1_desc;
+	unsigned int			num_l1_ents;
+
+	u64				strtab_base;
+	u32				strtab_base_cfg;
+};
+
+/* An SMMUv3 instance */
+struct arm_smmu_device {
+	struct device			*dev;
+	void __iomem			*base;
+
+#define ARM_SMMU_FEAT_2_LVL_STRTAB	(1 << 0)
+#define ARM_SMMU_FEAT_2_LVL_CDTAB	(1 << 1)
+#define ARM_SMMU_FEAT_TT_LE		(1 << 2)
+#define ARM_SMMU_FEAT_TT_BE		(1 << 3)
+#define ARM_SMMU_FEAT_PRI		(1 << 4)
+#define ARM_SMMU_FEAT_ATS		(1 << 5)
+#define ARM_SMMU_FEAT_SEV		(1 << 6)
+#define ARM_SMMU_FEAT_MSI		(1 << 7)
+#define ARM_SMMU_FEAT_COHERENCY		(1 << 8)
+#define ARM_SMMU_FEAT_TRANS_S1		(1 << 9)
+#define ARM_SMMU_FEAT_TRANS_S2		(1 << 10)
+#define ARM_SMMU_FEAT_STALLS		(1 << 11)
+#define ARM_SMMU_FEAT_HYP		(1 << 12)
+	u32				features;
+
+#define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
+	u32				options;
+
+	struct arm_smmu_cmdq		cmdq;
+	struct arm_smmu_evtq		evtq;
+	struct arm_smmu_priq		priq;
+
+	int				gerr_irq;
+
+	unsigned long			ias; /* IPA */
+	unsigned long			oas; /* PA */
+
+#define ARM_SMMU_MAX_ASIDS		(1 << 16)
+	unsigned int			asid_bits;
+	DECLARE_BITMAP(asid_map, ARM_SMMU_MAX_ASIDS);
+
+#define ARM_SMMU_MAX_VMIDS		(1 << 16)
+	unsigned int			vmid_bits;
+	DECLARE_BITMAP(vmid_map, ARM_SMMU_MAX_VMIDS);
+
+	unsigned int			ssid_bits;
+	unsigned int			sid_bits;
+
+	struct arm_smmu_strtab_cfg	strtab_cfg;
+	struct list_head		list;
+};
+
+/* SMMU private data for an IOMMU group */
+struct arm_smmu_group {
+	struct arm_smmu_device		*smmu;
+	struct arm_smmu_domain		*domain;
+	int				num_sids;
+	u32				*sids;
+	struct arm_smmu_strtab_ent	ste;
+};
+
+/* SMMU private data for an IOMMU domain */
+enum arm_smmu_domain_stage {
+	ARM_SMMU_DOMAIN_S1 = 0,
+	ARM_SMMU_DOMAIN_S2,
+	ARM_SMMU_DOMAIN_NESTED,
+};
+
+struct arm_smmu_domain {
+	struct arm_smmu_device		*smmu;
+	struct mutex			init_mutex; /* Protects smmu pointer */
+
+	struct io_pgtable_ops		*pgtbl_ops;
+	spinlock_t			pgtbl_lock;
+
+	enum arm_smmu_domain_stage	stage;
+	union {
+		struct arm_smmu_s1_cfg	s1_cfg;
+		struct arm_smmu_s2_cfg	s2_cfg;
+	};
+
+	struct iommu_domain		domain;
+};
+
+/* Our list of SMMU instances */
+static DEFINE_SPINLOCK(arm_smmu_devices_lock);
+static LIST_HEAD(arm_smmu_devices);
+
+struct arm_smmu_option_prop {
+	u32 opt;
+	const char *prop;
+};
+
+static struct arm_smmu_option_prop arm_smmu_options[] = {
+	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
+	{ 0, NULL},
+};
+
+static struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom)
+{
+	return container_of(dom, struct arm_smmu_domain, domain);
+}
+
+static void parse_driver_options(struct arm_smmu_device *smmu)
+{
+	int i = 0;
+
+	do {
+		if (of_property_read_bool(smmu->dev->of_node,
+						arm_smmu_options[i].prop)) {
+			smmu->options |= arm_smmu_options[i].opt;
+			dev_notice(smmu->dev, "option %s\n",
+				arm_smmu_options[i].prop);
+		}
+	} while (arm_smmu_options[++i].opt);
+}
+
+/* Low-level queue manipulation functions */
+static bool queue_full(struct arm_smmu_queue *q)
+{
+	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
+	       Q_WRP(q, q->prod) != Q_WRP(q, q->cons);
+}
+
+static bool queue_empty(struct arm_smmu_queue *q)
+{
+	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
+	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
+}
+
+static void queue_sync_cons(struct arm_smmu_queue *q)
+{
+	q->cons = readl_relaxed(q->cons_reg);
+}
+
+static void queue_inc_cons(struct arm_smmu_queue *q)
+{
+	u32 cons = (Q_WRP(q, q->cons) | Q_IDX(q, q->cons)) + 1;
+
+	q->cons = Q_OVF(q, q->cons) | Q_WRP(q, cons) | Q_IDX(q, cons);
+	writel(q->cons, q->cons_reg);
+}
+
+static int queue_sync_prod(struct arm_smmu_queue *q)
+{
+	int ret = 0;
+	u32 prod = readl_relaxed(q->prod_reg);
+
+	if (Q_OVF(q, prod) != Q_OVF(q, q->prod))
+		ret = -EOVERFLOW;
+
+	q->prod = prod;
+	return ret;
+}
+
+static void queue_inc_prod(struct arm_smmu_queue *q)
+{
+	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + 1;
+
+	q->prod = Q_OVF(q, q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
+	writel(q->prod, q->prod_reg);
+}
+
+static bool __queue_cons_before(struct arm_smmu_queue *q, u32 until)
+{
+	if (Q_WRP(q, q->cons) == Q_WRP(q, until))
+		return Q_IDX(q, q->cons) < Q_IDX(q, until);
+
+	return Q_IDX(q, q->cons) >= Q_IDX(q, until);
+}
+
+static int queue_poll_cons(struct arm_smmu_queue *q, u32 until, bool wfe)
+{
+	ktime_t timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
+
+	while (queue_sync_cons(q), __queue_cons_before(q, until)) {
+		if (ktime_compare(ktime_get(), timeout) > 0)
+			return -ETIMEDOUT;
+
+		if (wfe) {
+			wfe();
+		} else {
+			cpu_relax();
+			udelay(1);
+		}
+	}
+
+	return 0;
+}
+
+static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
+{
+	int i;
+
+	for (i = 0; i < n_dwords; ++i)
+		*dst++ = cpu_to_le64(*src++);
+}
+
+static int queue_insert_raw(struct arm_smmu_queue *q, u64 *ent)
+{
+	if (queue_full(q))
+		return -ENOSPC;
+
+	queue_write(Q_ENT(q, q->prod), ent, q->ent_dwords);
+	queue_inc_prod(q);
+	return 0;
+}
+
+static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
+{
+	int i;
+
+	for (i = 0; i < n_dwords; ++i)
+		*dst++ = le64_to_cpu(*src++);
+}
+
+static int queue_remove_raw(struct arm_smmu_queue *q, u64 *ent)
+{
+	if (queue_empty(q))
+		return -EAGAIN;
+
+	queue_read(ent, Q_ENT(q, q->cons), q->ent_dwords);
+	queue_inc_cons(q);
+	return 0;
+}
+
+/* High-level queue accessors */
+static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
+{
+	memset(cmd, 0, CMDQ_ENT_DWORDS << 3);
+	cmd[0] |= (ent->opcode & CMDQ_0_OP_MASK) << CMDQ_0_OP_SHIFT;
+
+	switch (ent->opcode) {
+	case CMDQ_OP_TLBI_EL2_ALL:
+	case CMDQ_OP_TLBI_NSNH_ALL:
+		break;
+	case CMDQ_OP_PREFETCH_CFG:
+		cmd[0] |= (u64)ent->prefetch.sid << CMDQ_PREFETCH_0_SID_SHIFT;
+		cmd[1] |= ent->prefetch.size << CMDQ_PREFETCH_1_SIZE_SHIFT;
+		cmd[1] |= ent->prefetch.addr & CMDQ_PREFETCH_1_ADDR_MASK;
+		break;
+	case CMDQ_OP_CFGI_STE:
+		cmd[0] |= (u64)ent->cfgi.sid << CMDQ_CFGI_0_SID_SHIFT;
+		cmd[1] |= ent->cfgi.leaf ? CMDQ_CFGI_1_LEAF : 0;
+		break;
+	case CMDQ_OP_CFGI_ALL:
+		/* Cover the entire SID range */
+		cmd[1] |= CMDQ_CFGI_1_RANGE_MASK << CMDQ_CFGI_1_RANGE_SHIFT;
+		break;
+	case CMDQ_OP_TLBI_NH_VA:
+		cmd[0] |= (u64)ent->tlbi.asid << CMDQ_TLBI_0_ASID_SHIFT;
+		/* Fallthrough */
+	case CMDQ_OP_TLBI_S2_IPA:
+		cmd[0] |= (u64)ent->tlbi.vmid << CMDQ_TLBI_0_VMID_SHIFT;
+		cmd[1] |= ent->tlbi.leaf ? CMDQ_TLBI_1_LEAF : 0;
+		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_ADDR_MASK;
+		break;
+	case CMDQ_OP_TLBI_NH_ASID:
+		cmd[0] |= (u64)ent->tlbi.asid << CMDQ_TLBI_0_ASID_SHIFT;
+		/* Fallthrough */
+	case CMDQ_OP_TLBI_S12_VMALL:
+		cmd[0] |= (u64)ent->tlbi.vmid << CMDQ_TLBI_0_VMID_SHIFT;
+		break;
+	case CMDQ_OP_PRI_RESP:
+		cmd[0] |= ent->substream_valid ? CMDQ_0_SSV : 0;
+		cmd[0] |= ent->pri.ssid << CMDQ_PRI_0_SSID_SHIFT;
+		cmd[0] |= (u64)ent->pri.sid << CMDQ_PRI_0_SID_SHIFT;
+		cmd[1] |= ent->pri.grpid << CMDQ_PRI_1_GRPID_SHIFT;
+		switch (ent->pri.resp) {
+		case PRI_RESP_DENY:
+			cmd[1] |= CMDQ_PRI_1_RESP_DENY;
+			break;
+		case PRI_RESP_FAIL:
+			cmd[1] |= CMDQ_PRI_1_RESP_FAIL;
+			break;
+		case PRI_RESP_SUCC:
+			cmd[1] |= CMDQ_PRI_1_RESP_SUCC;
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case CMDQ_OP_CMD_SYNC:
+		cmd[0] |= CMDQ_SYNC_0_CS_SEV;
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
+static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
+{
+	static const char *cerror_str[] = {
+		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
+		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
+		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
+	};
+
+	int i;
+	u64 cmd[CMDQ_ENT_DWORDS];
+	struct arm_smmu_queue *q = &smmu->cmdq.q;
+	u32 cons = readl_relaxed(q->cons_reg);
+	u32 idx = cons >> CMDQ_ERR_SHIFT & CMDQ_ERR_MASK;
+	struct arm_smmu_cmdq_ent cmd_sync = {
+		.opcode = CMDQ_OP_CMD_SYNC,
+	};
+
+	dev_err(smmu->dev, "CMDQ error (cons 0x%08x): %s\n", cons,
+		cerror_str[idx]);
+
+	switch (idx) {
+	case CMDQ_ERR_CERROR_ILL_IDX:
+		break;
+	case CMDQ_ERR_CERROR_ABT_IDX:
+		dev_err(smmu->dev, "retrying command fetch\n");
+	case CMDQ_ERR_CERROR_NONE_IDX:
+		return;
+	}
+
+	/*
+	 * We may have concurrent producers, so we need to be careful
+	 * not to touch any of the shadow cmdq state.
+	 */
+	queue_read(cmd, Q_ENT(q, idx), q->ent_dwords);
+	dev_err(smmu->dev, "skipping command in error state:\n");
+	for (i = 0; i < ARRAY_SIZE(cmd); ++i)
+		dev_err(smmu->dev, "\t0x%016llx\n", (unsigned long long)cmd[i]);
+
+	/* Convert the erroneous command into a CMD_SYNC */
+	if (arm_smmu_cmdq_build_cmd(cmd, &cmd_sync)) {
+		dev_err(smmu->dev, "failed to convert to CMD_SYNC\n");
+		return;
+	}
+
+	queue_write(cmd, Q_ENT(q, idx), q->ent_dwords);
+}
+
+static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
+				    struct arm_smmu_cmdq_ent *ent)
+{
+	u32 until;
+	u64 cmd[CMDQ_ENT_DWORDS];
+	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
+	struct arm_smmu_queue *q = &smmu->cmdq.q;
+
+	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
+		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
+			 ent->opcode);
+		return;
+	}
+
+	spin_lock(&smmu->cmdq.lock);
+	while (until = q->prod + 1, queue_insert_raw(q, cmd) == -ENOSPC) {
+		/*
+		 * Keep the queue locked, otherwise the producer could wrap
+		 * twice and we could see a future consumer pointer that looks
+		 * like it's behind us.
+		 */
+		if (queue_poll_cons(q, until, wfe))
+			dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
+	}
+
+	if (ent->opcode == CMDQ_OP_CMD_SYNC && queue_poll_cons(q, until, wfe))
+		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
+	spin_unlock(&smmu->cmdq.lock);
+}
+
+/* Context descriptor manipulation functions */
+static u64 arm_smmu_cpu_tcr_to_cd(u64 tcr)
+{
+	u64 val = 0;
+
+	/* Repack the TCR. Just care about TTBR0 for now */
+	val |= ARM_SMMU_TCR2CD(tcr, T0SZ);
+	val |= ARM_SMMU_TCR2CD(tcr, TG0);
+	val |= ARM_SMMU_TCR2CD(tcr, IRGN0);
+	val |= ARM_SMMU_TCR2CD(tcr, ORGN0);
+	val |= ARM_SMMU_TCR2CD(tcr, SH0);
+	val |= ARM_SMMU_TCR2CD(tcr, EPD0);
+	val |= ARM_SMMU_TCR2CD(tcr, EPD1);
+	val |= ARM_SMMU_TCR2CD(tcr, IPS);
+	val |= ARM_SMMU_TCR2CD(tcr, TBI0);
+
+	return val;
+}
+
+static void arm_smmu_write_ctx_desc(struct arm_smmu_device *smmu,
+				    struct arm_smmu_s1_cfg *cfg)
+{
+	u64 val;
+
+	/*
+	 * We don't need to issue any invalidation here, as we'll invalidate
+	 * the STE when installing the new entry anyway.
+	 */
+	val = arm_smmu_cpu_tcr_to_cd(cfg->cd.tcr) |
+#ifdef __BIG_ENDIAN
+	      CTXDESC_CD_0_ENDI |
+#endif
+	      CTXDESC_CD_0_R | CTXDESC_CD_0_A | CTXDESC_CD_0_ASET_PRIVATE |
+	      CTXDESC_CD_0_AA64 | (u64)cfg->cd.asid << CTXDESC_CD_0_ASID_SHIFT |
+	      CTXDESC_CD_0_V;
+	cfg->cdptr[0] = cpu_to_le64(val);
+
+	val = cfg->cd.ttbr & CTXDESC_CD_1_TTB0_MASK << CTXDESC_CD_1_TTB0_SHIFT;
+	cfg->cdptr[1] = cpu_to_le64(val);
+
+	cfg->cdptr[3] = cpu_to_le64(cfg->cd.mair << CTXDESC_CD_3_MAIR_SHIFT);
+}
+
+/* Stream table manipulation functions */
+static void
+arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
+{
+	u64 val = 0;
+
+	val |= (desc->span & STRTAB_L1_DESC_SPAN_MASK)
+		<< STRTAB_L1_DESC_SPAN_SHIFT;
+	val |= desc->l2ptr_dma &
+	       STRTAB_L1_DESC_L2PTR_MASK << STRTAB_L1_DESC_L2PTR_SHIFT;
+
+	*dst = cpu_to_le64(val);
+}
+
+static void arm_smmu_sync_ste_for_sid(struct arm_smmu_device *smmu, u32 sid)
+{
+	struct arm_smmu_cmdq_ent cmd = {
+		.opcode	= CMDQ_OP_CFGI_STE,
+		.cfgi	= {
+			.sid	= sid,
+			.leaf	= true,
+		},
+	};
+
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	cmd.opcode = CMDQ_OP_CMD_SYNC;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+}
+
+static void arm_smmu_write_strtab_ent(struct arm_smmu_device *smmu, u32 sid,
+				      __le64 *dst, struct arm_smmu_strtab_ent *ste)
+{
+	/*
+	 * This is hideously complicated, but we only really care about
+	 * three cases at the moment:
+	 *
+	 * 1. Invalid (all zero) -> bypass  (init)
+	 * 2. Bypass -> translation (attach)
+	 * 3. Translation -> bypass (detach)
+	 *
+	 * Given that we can't update the STE atomically and the SMMU
+	 * doesn't read the thing in a defined order, that leaves us
+	 * with the following maintenance requirements:
+	 *
+	 * 1. Update Config, return (init time STEs aren't live)
+	 * 2. Write everything apart from dword 0, sync, write dword 0, sync
+	 * 3. Update Config, sync
+	 */
+	u64 val = le64_to_cpu(dst[0]);
+	bool ste_live = false;
+	struct arm_smmu_cmdq_ent prefetch_cmd = {
+		.opcode		= CMDQ_OP_PREFETCH_CFG,
+		.prefetch	= {
+			.sid	= sid,
+		},
+	};
+
+	if (val & STRTAB_STE_0_V) {
+		u64 cfg;
+
+		cfg = val & STRTAB_STE_0_CFG_MASK << STRTAB_STE_0_CFG_SHIFT;
+		switch (cfg) {
+		case STRTAB_STE_0_CFG_BYPASS:
+			break;
+		case STRTAB_STE_0_CFG_S1_TRANS:
+		case STRTAB_STE_0_CFG_S2_TRANS:
+			ste_live = true;
+			break;
+		default:
+			BUG(); /* STE corruption */
+		}
+	}
+
+	/* Nuke the existing Config, as we're going to rewrite it */
+	val &= ~(STRTAB_STE_0_CFG_MASK << STRTAB_STE_0_CFG_SHIFT);
+
+	if (ste->valid)
+		val |= STRTAB_STE_0_V;
+	else
+		val &= ~STRTAB_STE_0_V;
+
+	if (ste->bypass) {
+		val |= disable_bypass ? STRTAB_STE_0_CFG_ABORT
+				      : STRTAB_STE_0_CFG_BYPASS;
+		dst[0] = cpu_to_le64(val);
+		dst[2] = 0; /* Nuke the VMID */
+		if (ste_live)
+			arm_smmu_sync_ste_for_sid(smmu, sid);
+		return;
+	}
+
+	if (ste->s1_cfg) {
+		BUG_ON(ste_live);
+		dst[1] = cpu_to_le64(
+			 STRTAB_STE_1_S1C_CACHE_WBRA
+			 << STRTAB_STE_1_S1CIR_SHIFT |
+			 STRTAB_STE_1_S1C_CACHE_WBRA
+			 << STRTAB_STE_1_S1COR_SHIFT |
+			 STRTAB_STE_1_S1C_SH_ISH << STRTAB_STE_1_S1CSH_SHIFT |
+			 STRTAB_STE_1_S1STALLD |
+#ifdef CONFIG_PCI_ATS
+			 STRTAB_STE_1_EATS_TRANS << STRTAB_STE_1_EATS_SHIFT |
+#endif
+			 STRTAB_STE_1_STRW_NSEL1 << STRTAB_STE_1_STRW_SHIFT);
+
+		val |= (ste->s1_cfg->cdptr_dma & STRTAB_STE_0_S1CTXPTR_MASK
+		        << STRTAB_STE_0_S1CTXPTR_SHIFT) |
+			STRTAB_STE_0_CFG_S1_TRANS;
+
+	}
+
+	if (ste->s2_cfg) {
+		BUG_ON(ste_live);
+		dst[2] = cpu_to_le64(
+			 ste->s2_cfg->vmid << STRTAB_STE_2_S2VMID_SHIFT |
+			 (ste->s2_cfg->vtcr & STRTAB_STE_2_VTCR_MASK)
+			  << STRTAB_STE_2_VTCR_SHIFT |
+#ifdef __BIG_ENDIAN
+			 STRTAB_STE_2_S2ENDI |
+#endif
+			 STRTAB_STE_2_S2PTW | STRTAB_STE_2_S2AA64 |
+			 STRTAB_STE_2_S2R);
+
+		dst[3] = cpu_to_le64(ste->s2_cfg->vttbr &
+			 STRTAB_STE_3_S2TTB_MASK << STRTAB_STE_3_S2TTB_SHIFT);
+
+		val |= STRTAB_STE_0_CFG_S2_TRANS;
+	}
+
+	arm_smmu_sync_ste_for_sid(smmu, sid);
+	dst[0] = cpu_to_le64(val);
+	arm_smmu_sync_ste_for_sid(smmu, sid);
+
+	/* It's likely that we'll want to use the new STE soon */
+	if (!(smmu->options & ARM_SMMU_OPT_SKIP_PREFETCH))
+		arm_smmu_cmdq_issue_cmd(smmu, &prefetch_cmd);
+}
+
+static void arm_smmu_init_bypass_stes(u64 *strtab, unsigned int nent)
+{
+	unsigned int i;
+	struct arm_smmu_strtab_ent ste = {
+		.valid	= true,
+		.bypass	= true,
+	};
+
+	for (i = 0; i < nent; ++i) {
+		arm_smmu_write_strtab_ent(NULL, -1, strtab, &ste);
+		strtab += STRTAB_STE_DWORDS;
+	}
+}
+
+static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
+{
+	size_t size;
+	void *strtab;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+	struct arm_smmu_strtab_l1_desc *desc = &cfg->l1_desc[sid >> STRTAB_SPLIT];
+
+	if (desc->l2ptr)
+		return 0;
+
+	size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3);
+	strtab = &cfg->strtab[(sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS];
+
+	desc->span = STRTAB_SPLIT + 1;
+	desc->l2ptr = dma_zalloc_coherent(smmu->dev, size, &desc->l2ptr_dma,
+					  GFP_KERNEL);
+	if (!desc->l2ptr) {
+		dev_err(smmu->dev,
+			"failed to allocate l2 stream table for SID %u\n",
+			sid);
+		return -ENOMEM;
+	}
+
+	arm_smmu_init_bypass_stes(desc->l2ptr, 1 << STRTAB_SPLIT);
+	arm_smmu_write_strtab_l1_desc(strtab, desc);
+	return 0;
+}
+
+/* IRQ and event handlers */
+static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
+{
+	int i;
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->evtq.q;
+	u64 evt[EVTQ_ENT_DWORDS];
+
+	while (!queue_remove_raw(q, evt)) {
+		u8 id = evt[0] >> EVTQ_0_ID_SHIFT & EVTQ_0_ID_MASK;
+
+		dev_info(smmu->dev, "event 0x%02x received:\n", id);
+		for (i = 0; i < ARRAY_SIZE(evt); ++i)
+			dev_info(smmu->dev, "\t0x%016llx\n",
+				 (unsigned long long)evt[i]);
+	}
+
+	/* Sync our overflow flag, as we believe we're up to speed */
+	q->cons = Q_OVF(q, q->prod) | Q_WRP(q, q->cons) | Q_IDX(q, q->cons);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t arm_smmu_evtq_handler(int irq, void *dev)
+{
+	irqreturn_t ret = IRQ_WAKE_THREAD;
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->evtq.q;
+
+	/*
+	 * Not much we can do on overflow, so scream and pretend we're
+	 * trying harder.
+	 */
+	if (queue_sync_prod(q) == -EOVERFLOW)
+		dev_err(smmu->dev, "EVTQ overflow detected -- events lost\n");
+	else if (queue_empty(q))
+		ret = IRQ_NONE;
+
+	return ret;
+}
+
+static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
+{
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->priq.q;
+	u64 evt[PRIQ_ENT_DWORDS];
+
+	while (!queue_remove_raw(q, evt)) {
+		u32 sid, ssid;
+		u16 grpid;
+		bool ssv, last;
+
+		sid = evt[0] >> PRIQ_0_SID_SHIFT & PRIQ_0_SID_MASK;
+		ssv = evt[0] & PRIQ_0_SSID_V;
+		ssid = ssv ? evt[0] >> PRIQ_0_SSID_SHIFT & PRIQ_0_SSID_MASK : 0;
+		last = evt[0] & PRIQ_0_PRG_LAST;
+		grpid = evt[1] >> PRIQ_1_PRG_IDX_SHIFT & PRIQ_1_PRG_IDX_MASK;
+
+		dev_info(smmu->dev, "unexpected PRI request received:\n");
+		dev_info(smmu->dev,
+			 "\tsid 0x%08x.0x%05x: [%u%s] %sprivileged %s%s%s access at iova 0x%016llx\n",
+			 sid, ssid, grpid, last ? "L" : "",
+			 evt[0] & PRIQ_0_PERM_PRIV ? "" : "un",
+			 evt[0] & PRIQ_0_PERM_READ ? "R" : "",
+			 evt[0] & PRIQ_0_PERM_WRITE ? "W" : "",
+			 evt[0] & PRIQ_0_PERM_EXEC ? "X" : "",
+			 evt[1] & PRIQ_1_ADDR_MASK << PRIQ_1_ADDR_SHIFT);
+
+		if (last) {
+			struct arm_smmu_cmdq_ent cmd = {
+				.opcode			= CMDQ_OP_PRI_RESP,
+				.substream_valid	= ssv,
+				.pri			= {
+					.sid	= sid,
+					.ssid	= ssid,
+					.grpid	= grpid,
+					.resp	= PRI_RESP_DENY,
+				},
+			};
+
+			arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+		}
+	}
+
+	/* Sync our overflow flag, as we believe we're up to speed */
+	q->cons = Q_OVF(q, q->prod) | Q_WRP(q, q->cons) | Q_IDX(q, q->cons);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t arm_smmu_priq_handler(int irq, void *dev)
+{
+	irqreturn_t ret = IRQ_WAKE_THREAD;
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->priq.q;
+
+	/* PRIQ overflow indicates a programming error */
+	if (queue_sync_prod(q) == -EOVERFLOW)
+		dev_err(smmu->dev, "PRIQ overflow detected -- requests lost\n");
+	else if (queue_empty(q))
+		ret = IRQ_NONE;
+
+	return ret;
+}
+
+static irqreturn_t arm_smmu_cmdq_sync_handler(int irq, void *dev)
+{
+	/* We don't actually use CMD_SYNC interrupts for anything */
+	return IRQ_HANDLED;
+}
+
+static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
+
+static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
+{
+	u32 gerror, gerrorn;
+	struct arm_smmu_device *smmu = dev;
+
+	gerror = readl_relaxed(smmu->base + ARM_SMMU_GERROR);
+	gerrorn = readl_relaxed(smmu->base + ARM_SMMU_GERRORN);
+
+	gerror ^= gerrorn;
+	if (!(gerror & GERROR_ERR_MASK))
+		return IRQ_NONE; /* No errors pending */
+
+	dev_warn(smmu->dev,
+		 "unexpected global error reported (0x%08x), this could be serious\n",
+		 gerror);
+
+	if (gerror & GERROR_SFM_ERR) {
+		dev_err(smmu->dev, "device has entered Service Failure Mode!\n");
+		arm_smmu_device_disable(smmu);
+	}
+
+	if (gerror & GERROR_MSI_GERROR_ABT_ERR)
+		dev_warn(smmu->dev, "GERROR MSI write aborted\n");
+
+	if (gerror & GERROR_MSI_PRIQ_ABT_ERR) {
+		dev_warn(smmu->dev, "PRIQ MSI write aborted\n");
+		arm_smmu_priq_handler(irq, smmu->dev);
+	}
+
+	if (gerror & GERROR_MSI_EVTQ_ABT_ERR) {
+		dev_warn(smmu->dev, "EVTQ MSI write aborted\n");
+		arm_smmu_evtq_handler(irq, smmu->dev);
+	}
+
+	if (gerror & GERROR_MSI_CMDQ_ABT_ERR) {
+		dev_warn(smmu->dev, "CMDQ MSI write aborted\n");
+		arm_smmu_cmdq_sync_handler(irq, smmu->dev);
+	}
+
+	if (gerror & GERROR_PRIQ_ABT_ERR)
+		dev_err(smmu->dev, "PRIQ write aborted -- events may have been lost\n");
+
+	if (gerror & GERROR_EVTQ_ABT_ERR)
+		dev_err(smmu->dev, "EVTQ write aborted -- events may have been lost\n");
+
+	if (gerror & GERROR_CMDQ_ERR)
+		arm_smmu_cmdq_skip_err(smmu);
+
+	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
+	return IRQ_HANDLED;
+}
+
+/* IO_PGTABLE API */
+static void __arm_smmu_tlb_sync(struct arm_smmu_device *smmu)
+{
+	struct arm_smmu_cmdq_ent cmd;
+
+	cmd.opcode = CMDQ_OP_CMD_SYNC;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+}
+
+static void arm_smmu_tlb_sync(void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	__arm_smmu_tlb_sync(smmu_domain->smmu);
+}
+
+static void arm_smmu_tlb_inv_context(void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_cmdq_ent cmd;
+
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		cmd.opcode	= CMDQ_OP_TLBI_NH_ASID;
+		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
+		cmd.tlbi.vmid	= 0;
+	} else {
+		cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
+		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
+	}
+
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	__arm_smmu_tlb_sync(smmu);
+}
+
+static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
+					  bool leaf, void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_cmdq_ent cmd = {
+		.tlbi = {
+			.leaf	= leaf,
+			.addr	= iova,
+		},
+	};
+
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
+		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
+	} else {
+		cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
+		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
+	}
+
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+}
+
+static void arm_smmu_flush_pgtable(void *addr, size_t size, void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	unsigned long offset = (unsigned long)addr & ~PAGE_MASK;
+
+	if (smmu->features & ARM_SMMU_FEAT_COHERENCY) {
+		dsb(ishst);
+	} else {
+		dma_addr_t dma_addr;
+		struct device *dev = smmu->dev;
+
+		dma_addr = dma_map_page(dev, virt_to_page(addr), offset, size,
+					DMA_TO_DEVICE);
+
+		if (dma_mapping_error(dev, dma_addr))
+			dev_err(dev, "failed to flush pgtable at %p\n", addr);
+		else
+			dma_unmap_page(dev, dma_addr, size, DMA_TO_DEVICE);
+	}
+}
+
+static struct iommu_gather_ops arm_smmu_gather_ops = {
+	.tlb_flush_all	= arm_smmu_tlb_inv_context,
+	.tlb_add_flush	= arm_smmu_tlb_inv_range_nosync,
+	.tlb_sync	= arm_smmu_tlb_sync,
+	.flush_pgtable	= arm_smmu_flush_pgtable,
+};
+
+/* IOMMU API */
+static bool arm_smmu_capable(enum iommu_cap cap)
+{
+	switch (cap) {
+	case IOMMU_CAP_CACHE_COHERENCY:
+		return true;
+	case IOMMU_CAP_INTR_REMAP:
+		return true; /* MSIs are just memory writes */
+	case IOMMU_CAP_NOEXEC:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
+{
+	struct arm_smmu_domain *smmu_domain;
+
+	if (type != IOMMU_DOMAIN_UNMANAGED)
+		return NULL;
+
+	/*
+	 * Allocate the domain and initialise some of its data structures.
+	 * We can't really do anything meaningful until we've added a
+	 * master.
+	 */
+	smmu_domain = kzalloc(sizeof(*smmu_domain), GFP_KERNEL);
+	if (!smmu_domain)
+		return NULL;
+
+	mutex_init(&smmu_domain->init_mutex);
+	spin_lock_init(&smmu_domain->pgtbl_lock);
+	return &smmu_domain->domain;
+}
+
+static int arm_smmu_bitmap_alloc(unsigned long *map, int span)
+{
+	int idx, size = 1 << span;
+
+	do {
+		idx = find_first_zero_bit(map, size);
+		if (idx == size)
+			return -ENOSPC;
+	} while (test_and_set_bit(idx, map));
+
+	return idx;
+}
+
+static void arm_smmu_bitmap_free(unsigned long *map, int idx)
+{
+	clear_bit(idx, map);
+}
+
+static void arm_smmu_domain_free(struct iommu_domain *domain)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
+
+	/* Free the CD and ASID, if we allocated them */
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+
+		if (cfg->cdptr) {
+			dma_free_coherent(smmu_domain->smmu->dev,
+					  CTXDESC_CD_DWORDS << 3,
+					  cfg->cdptr,
+					  cfg->cdptr_dma);
+
+			arm_smmu_bitmap_free(smmu->asid_map, cfg->cd.asid);
+		}
+	} else {
+		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
+		if (cfg->vmid)
+			arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
+	}
+
+	kfree(smmu_domain);
+}
+
+static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
+				       struct io_pgtable_cfg *pgtbl_cfg)
+{
+	int ret;
+	u16 asid;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+
+	asid = arm_smmu_bitmap_alloc(smmu->asid_map, smmu->asid_bits);
+	if (IS_ERR_VALUE(asid))
+		return asid;
+
+	cfg->cdptr = dma_zalloc_coherent(smmu->dev, CTXDESC_CD_DWORDS << 3,
+					 &cfg->cdptr_dma, GFP_KERNEL);
+	if (!cfg->cdptr) {
+		dev_warn(smmu->dev, "failed to allocate context descriptor\n");
+		goto out_free_asid;
+	}
+
+	cfg->cd.asid	= asid;
+	cfg->cd.ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr[0];
+	cfg->cd.tcr	= pgtbl_cfg->arm_lpae_s1_cfg.tcr;
+	cfg->cd.mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair[0];
+	return 0;
+
+out_free_asid:
+	arm_smmu_bitmap_free(smmu->asid_map, asid);
+	return ret;
+}
+
+static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
+				       struct io_pgtable_cfg *pgtbl_cfg)
+{
+	u16 vmid;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
+
+	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
+	if (IS_ERR_VALUE(vmid))
+		return vmid;
+
+	cfg->vmid	= vmid;
+	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
+	cfg->vtcr	= pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
+	return 0;
+}
+
+static struct iommu_ops arm_smmu_ops;
+
+static int arm_smmu_domain_finalise(struct iommu_domain *domain)
+{
+	int ret;
+	unsigned long ias, oas;
+	enum io_pgtable_fmt fmt;
+	struct io_pgtable_cfg pgtbl_cfg;
+	struct io_pgtable_ops *pgtbl_ops;
+	int (*finalise_stage_fn)(struct arm_smmu_domain *,
+				 struct io_pgtable_cfg *);
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	/* Restrict the stage to what we can actually support */
+	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
+		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
+	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
+		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
+
+	switch (smmu_domain->stage) {
+	case ARM_SMMU_DOMAIN_S1:
+		ias = VA_BITS;
+		oas = smmu->ias;
+		fmt = ARM_64_LPAE_S1;
+		finalise_stage_fn = arm_smmu_domain_finalise_s1;
+		break;
+	case ARM_SMMU_DOMAIN_NESTED:
+	case ARM_SMMU_DOMAIN_S2:
+		ias = smmu->ias;
+		oas = smmu->oas;
+		fmt = ARM_64_LPAE_S2;
+		finalise_stage_fn = arm_smmu_domain_finalise_s2;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	pgtbl_cfg = (struct io_pgtable_cfg) {
+		.pgsize_bitmap	= arm_smmu_ops.pgsize_bitmap,
+		.ias		= ias,
+		.oas		= oas,
+		.tlb		= &arm_smmu_gather_ops,
+	};
+
+	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
+	if (!pgtbl_ops)
+		return -ENOMEM;
+
+	arm_smmu_ops.pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
+	smmu_domain->pgtbl_ops = pgtbl_ops;
+
+	ret = finalise_stage_fn(smmu_domain, &pgtbl_cfg);
+	if (IS_ERR_VALUE(ret))
+		free_io_pgtable_ops(pgtbl_ops);
+
+	return ret;
+}
+
+static struct arm_smmu_group *arm_smmu_group_get(struct device *dev)
+{
+	struct iommu_group *group;
+	struct arm_smmu_group *smmu_group;
+
+	group = iommu_group_get(dev);
+	if (!group)
+		return NULL;
+
+	smmu_group = iommu_group_get_iommudata(group);
+	iommu_group_put(group);
+	return smmu_group;
+}
+
+static __le64 *arm_smmu_get_step_for_sid(struct arm_smmu_device *smmu, u32 sid)
+{
+	__le64 *step;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
+		struct arm_smmu_strtab_l1_desc *l1_desc;
+		int idx;
+
+		/* Two-level walk */
+		idx = (sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS;
+		l1_desc = &cfg->l1_desc[idx];
+		idx = (sid & ((1 << STRTAB_SPLIT) - 1)) * STRTAB_STE_DWORDS;
+		step = &l1_desc->l2ptr[idx];
+	} else {
+		/* Simple linear lookup */
+		step = &cfg->strtab[sid * STRTAB_STE_DWORDS];
+	}
+
+	return step;
+}
+
+static int arm_smmu_install_ste_for_group(struct arm_smmu_group *smmu_group)
+{
+	int i;
+	struct arm_smmu_domain *smmu_domain = smmu_group->domain;
+	struct arm_smmu_strtab_ent *ste = &smmu_group->ste;
+	struct arm_smmu_device *smmu = smmu_group->smmu;
+
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		ste->s1_cfg = &smmu_domain->s1_cfg;
+		ste->s2_cfg = NULL;
+		arm_smmu_write_ctx_desc(smmu, ste->s1_cfg);
+	} else {
+		ste->s1_cfg = NULL;
+		ste->s2_cfg = &smmu_domain->s2_cfg;
+	}
+
+	for (i = 0; i < smmu_group->num_sids; ++i) {
+		u32 sid = smmu_group->sids[i];
+		__le64 *step = arm_smmu_get_step_for_sid(smmu, sid);
+
+		arm_smmu_write_strtab_ent(smmu, sid, step, ste);
+	}
+
+	return 0;
+}
+
+static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
+{
+	int ret = 0;
+	struct arm_smmu_device *smmu;
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_group *smmu_group = arm_smmu_group_get(dev);
+
+	if (!smmu_group)
+		return -ENOENT;
+
+	/* Already attached to a different domain? */
+	if (smmu_group->domain && smmu_group->domain != smmu_domain)
+		return -EEXIST;
+
+	smmu = smmu_group->smmu;
+	mutex_lock(&smmu_domain->init_mutex);
+
+	if (!smmu_domain->smmu) {
+		smmu_domain->smmu = smmu;
+		ret = arm_smmu_domain_finalise(domain);
+		if (ret) {
+			smmu_domain->smmu = NULL;
+			goto out_unlock;
+		}
+	} else if (smmu_domain->smmu != smmu) {
+		dev_err(dev,
+			"cannot attach to SMMU %s (upstream of %s)\n",
+			dev_name(smmu_domain->smmu->dev),
+			dev_name(smmu->dev));
+		ret = -ENXIO;
+		goto out_unlock;
+	}
+
+	/* Group already attached to this domain? */
+	if (smmu_group->domain)
+		goto out_unlock;
+
+	smmu_group->domain	= smmu_domain;
+	smmu_group->ste.bypass	= false;
+
+	ret = arm_smmu_install_ste_for_group(smmu_group);
+	if (IS_ERR_VALUE(ret))
+		smmu_group->domain = NULL;
+
+out_unlock:
+	mutex_unlock(&smmu_domain->init_mutex);
+	return ret;
+}
+
+static void arm_smmu_detach_dev(struct iommu_domain *domain, struct device *dev)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_group *smmu_group = arm_smmu_group_get(dev);
+
+	BUG_ON(!smmu_domain);
+	BUG_ON(!smmu_group);
+
+	mutex_lock(&smmu_domain->init_mutex);
+	BUG_ON(smmu_group->domain != smmu_domain);
+
+	smmu_group->ste.bypass = true;
+	if (IS_ERR_VALUE(arm_smmu_install_ste_for_group(smmu_group)))
+		dev_warn(dev, "failed to install bypass STE\n");
+
+	smmu_group->domain = NULL;
+	mutex_unlock(&smmu_domain->init_mutex);
+}
+
+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;
+
+	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;
+}
+
+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;
+
+	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;
+}
+
+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;
+
+	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;
+}
+
+static int __arm_smmu_get_pci_sid(struct pci_dev *pdev, u16 alias, void *sidp)
+{
+	*(u32 *)sidp = alias;
+	return 0; /* Continue walking */
+}
+
+static void __arm_smmu_release_pci_iommudata(void *data)
+{
+	kfree(data);
+}
+
+static struct arm_smmu_device *arm_smmu_get_for_pci_dev(struct pci_dev *pdev)
+{
+	struct device_node *of_node;
+	struct arm_smmu_device *curr, *smmu = NULL;
+	struct pci_bus *bus = pdev->bus;
+
+	/* Walk up to the root bus */
+	while (!pci_is_root_bus(bus))
+		bus = bus->parent;
+
+	/* Follow the "iommus" phandle from the host controller */
+	of_node = of_parse_phandle(bus->bridge->parent->of_node, "iommus", 0);
+	if (!of_node)
+		return NULL;
+
+	/* See if we can find an SMMU corresponding to the phandle */
+	spin_lock(&arm_smmu_devices_lock);
+	list_for_each_entry(curr, &arm_smmu_devices, list) {
+		if (curr->dev->of_node == of_node) {
+			smmu = curr;
+			break;
+		}
+	}
+	spin_unlock(&arm_smmu_devices_lock);
+	of_node_put(of_node);
+	return smmu;
+}
+
+static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
+{
+	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
+		limit *= 1UL << STRTAB_SPLIT;
+
+	return sid < limit;
+}
+
+static int arm_smmu_add_device(struct device *dev)
+{
+	int i, ret;
+	u32 sid, *sids;
+	struct pci_dev *pdev;
+	struct iommu_group *group;
+	struct arm_smmu_group *smmu_group;
+	struct arm_smmu_device *smmu;
+
+	/* We only support PCI, for now */
+	if (!dev_is_pci(dev))
+		return -ENODEV;
+
+	pdev = to_pci_dev(dev);
+	group = iommu_group_get_for_dev(dev);
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	smmu_group = iommu_group_get_iommudata(group);
+	if (!smmu_group) {
+		smmu = arm_smmu_get_for_pci_dev(pdev);
+		if (!smmu) {
+			ret = -ENOENT;
+			goto out_put_group;
+		}
+
+		smmu_group = kzalloc(sizeof(*smmu_group), GFP_KERNEL);
+		if (!smmu_group) {
+			ret = -ENOMEM;
+			goto out_put_group;
+		}
+
+		smmu_group->ste.valid	= true;
+		smmu_group->smmu	= smmu;
+		iommu_group_set_iommudata(group, smmu_group,
+					  __arm_smmu_release_pci_iommudata);
+	} else {
+		smmu = smmu_group->smmu;
+	}
+
+	/* Assume SID == RID until firmware tells us otherwise */
+	pci_for_each_dma_alias(pdev, __arm_smmu_get_pci_sid, &sid);
+	for (i = 0; i < smmu_group->num_sids; ++i) {
+		/* If we already know about this SID, then we're done */
+		if (smmu_group->sids[i] == sid)
+			return 0;
+	}
+
+	/* Check the SID is in range of the SMMU and our stream table */
+	if (!arm_smmu_sid_in_range(smmu, sid)) {
+		ret = -ERANGE;
+		goto out_put_group;
+	}
+
+	/* Ensure l2 strtab is initialised */
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
+		ret = arm_smmu_init_l2_strtab(smmu, sid);
+		if (ret)
+			goto out_put_group;
+	}
+
+	/* Resize the SID array for the group */
+	smmu_group->num_sids++;
+	sids = krealloc(smmu_group->sids, smmu_group->num_sids * sizeof(*sids),
+			GFP_KERNEL);
+	if (!sids) {
+		smmu_group->num_sids--;
+		ret = -ENOMEM;
+		goto out_put_group;
+	}
+
+	/* Add the new SID */
+	sids[smmu_group->num_sids - 1] = sid;
+	smmu_group->sids = sids;
+	return 0;
+
+out_put_group:
+	iommu_group_put(group);
+	return ret;
+}
+
+static void arm_smmu_remove_device(struct device *dev)
+{
+	iommu_group_remove_device(dev);
+}
+
+static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
+				    enum iommu_attr attr, void *data)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	switch (attr) {
+	case DOMAIN_ATTR_NESTING:
+		*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
+		return 0;
+	default:
+		return -ENODEV;
+	}
+}
+
+static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
+				    enum iommu_attr attr, void *data)
+{
+	int ret = 0;
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	mutex_lock(&smmu_domain->init_mutex);
+
+	switch (attr) {
+	case DOMAIN_ATTR_NESTING:
+		if (smmu_domain->smmu) {
+			ret = -EPERM;
+			goto out_unlock;
+		}
+
+		if (*(int *)data)
+			smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
+		else
+			smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
+
+		break;
+	default:
+		ret = -ENODEV;
+	}
+
+out_unlock:
+	mutex_unlock(&smmu_domain->init_mutex);
+	return ret;
+}
+
+static struct iommu_ops arm_smmu_ops = {
+	.capable		= arm_smmu_capable,
+	.domain_alloc		= arm_smmu_domain_alloc,
+	.domain_free		= arm_smmu_domain_free,
+	.attach_dev		= arm_smmu_attach_dev,
+	.detach_dev		= arm_smmu_detach_dev,
+	.map			= arm_smmu_map,
+	.unmap			= arm_smmu_unmap,
+	.iova_to_phys		= arm_smmu_iova_to_phys,
+	.add_device		= arm_smmu_add_device,
+	.remove_device		= arm_smmu_remove_device,
+	.domain_get_attr	= arm_smmu_domain_get_attr,
+	.domain_set_attr	= arm_smmu_domain_set_attr,
+	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
+};
+
+/* Probing and initialisation functions */
+static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
+				   struct arm_smmu_queue *q,
+				   unsigned long prod_off,
+				   unsigned long cons_off,
+				   size_t dwords)
+{
+	size_t qsz = ((1 << q->max_n_shift) * dwords) << 3;
+
+	q->base = dma_alloc_coherent(smmu->dev, qsz, &q->base_dma, GFP_KERNEL);
+	if (!q->base) {
+		dev_err(smmu->dev, "failed to allocate queue (0x%zx bytes)\n",
+			qsz);
+		return -ENOMEM;
+	}
+
+	q->prod_reg	= smmu->base + prod_off;
+	q->cons_reg	= smmu->base + cons_off;
+	q->ent_dwords	= dwords;
+
+	q->q_base  = Q_BASE_RWA;
+	q->q_base |= q->base_dma & Q_BASE_ADDR_MASK << Q_BASE_ADDR_SHIFT;
+	q->q_base |= (q->max_n_shift & Q_BASE_LOG2SIZE_MASK)
+		     << Q_BASE_LOG2SIZE_SHIFT;
+
+	q->prod = q->cons = 0;
+	return 0;
+}
+
+static void arm_smmu_free_one_queue(struct arm_smmu_device *smmu,
+				    struct arm_smmu_queue *q)
+{
+	size_t qsz = ((1 << q->max_n_shift) * q->ent_dwords) << 3;
+
+	dma_free_coherent(smmu->dev, qsz, q->base, q->base_dma);
+}
+
+static void arm_smmu_free_queues(struct arm_smmu_device *smmu)
+{
+	arm_smmu_free_one_queue(smmu, &smmu->cmdq.q);
+	arm_smmu_free_one_queue(smmu, &smmu->evtq.q);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		arm_smmu_free_one_queue(smmu, &smmu->priq.q);
+}
+
+static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	/* cmdq */
+	spin_lock_init(&smmu->cmdq.lock);
+	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
+				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS);
+	if (ret)
+		goto out;
+
+	/* evtq */
+	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
+				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS);
+	if (ret)
+		goto out_free_cmdq;
+
+	/* priq */
+	if (!(smmu->features & ARM_SMMU_FEAT_PRI))
+		return 0;
+
+	ret = arm_smmu_init_one_queue(smmu, &smmu->priq.q, ARM_SMMU_PRIQ_PROD,
+				      ARM_SMMU_PRIQ_CONS, PRIQ_ENT_DWORDS);
+	if (ret)
+		goto out_free_evtq;
+
+	return 0;
+
+out_free_evtq:
+	arm_smmu_free_one_queue(smmu, &smmu->evtq.q);
+out_free_cmdq:
+	arm_smmu_free_one_queue(smmu, &smmu->cmdq.q);
+out:
+	return ret;
+}
+
+static void arm_smmu_free_l2_strtab(struct arm_smmu_device *smmu)
+{
+	int i;
+	size_t size;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3);
+	for (i = 0; i < cfg->num_l1_ents; ++i) {
+		struct arm_smmu_strtab_l1_desc *desc = &cfg->l1_desc[i];
+
+		if (!desc->l2ptr)
+			continue;
+
+		dma_free_coherent(smmu->dev, size, desc->l2ptr,
+				  desc->l2ptr_dma);
+	}
+}
+
+static int arm_smmu_init_l1_strtab(struct arm_smmu_device *smmu)
+{
+	unsigned int i;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+	size_t size = sizeof(*cfg->l1_desc) * cfg->num_l1_ents;
+	void *strtab = smmu->strtab_cfg.strtab;
+
+	cfg->l1_desc = devm_kzalloc(smmu->dev, size, GFP_KERNEL);
+	if (!cfg->l1_desc) {
+		dev_err(smmu->dev, "failed to allocate l1 stream table desc\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < cfg->num_l1_ents; ++i) {
+		arm_smmu_write_strtab_l1_desc(strtab, &cfg->l1_desc[i]);
+		strtab += STRTAB_L1_DESC_DWORDS << 3;
+	}
+
+	return 0;
+}
+
+static int arm_smmu_init_strtab_2lvl(struct arm_smmu_device *smmu)
+{
+	void *strtab;
+	u64 reg;
+	u32 size, l1size;
+	int ret;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	/* Calculate the L1 size, capped to the SIDSIZE */
+	size = STRTAB_L1_SZ_SHIFT - (ilog2(STRTAB_L1_DESC_DWORDS) + 3);
+	size = min(size, smmu->sid_bits - STRTAB_SPLIT);
+	cfg->num_l1_ents = 1 << size;
+
+	size += STRTAB_SPLIT;
+	if (size < smmu->sid_bits)
+		dev_warn(smmu->dev,
+			 "2-level strtab only covers %u/%u bits of SID\n",
+			 size, smmu->sid_bits);
+
+	l1size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3);
+	strtab = dma_zalloc_coherent(smmu->dev, l1size, &cfg->strtab_dma,
+				     GFP_KERNEL);
+	if (!strtab) {
+		dev_err(smmu->dev,
+			"failed to allocate l1 stream table (%u bytes)\n",
+			size);
+		return -ENOMEM;
+	}
+	cfg->strtab = strtab;
+
+	/* Configure strtab_base_cfg for 2 levels */
+	reg  = STRTAB_BASE_CFG_FMT_2LVL;
+	reg |= (size & STRTAB_BASE_CFG_LOG2SIZE_MASK)
+		<< STRTAB_BASE_CFG_LOG2SIZE_SHIFT;
+	reg |= (STRTAB_SPLIT & STRTAB_BASE_CFG_SPLIT_MASK)
+		<< STRTAB_BASE_CFG_SPLIT_SHIFT;
+	cfg->strtab_base_cfg = reg;
+
+	ret = arm_smmu_init_l1_strtab(smmu);
+	if (ret)
+		dma_free_coherent(smmu->dev,
+				  l1size,
+				  strtab,
+				  cfg->strtab_dma);
+	return ret;
+}
+
+static int arm_smmu_init_strtab_linear(struct arm_smmu_device *smmu)
+{
+	void *strtab;
+	u64 reg;
+	u32 size;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	size = (1 << smmu->sid_bits) * (STRTAB_STE_DWORDS << 3);
+	strtab = dma_zalloc_coherent(smmu->dev, size, &cfg->strtab_dma,
+				     GFP_KERNEL);
+	if (!strtab) {
+		dev_err(smmu->dev,
+			"failed to allocate linear stream table (%u bytes)\n",
+			size);
+		return -ENOMEM;
+	}
+	cfg->strtab = strtab;
+	cfg->num_l1_ents = 1 << smmu->sid_bits;
+
+	/* Configure strtab_base_cfg for a linear table covering all SIDs */
+	reg  = STRTAB_BASE_CFG_FMT_LINEAR;
+	reg |= (smmu->sid_bits & STRTAB_BASE_CFG_LOG2SIZE_MASK)
+		<< STRTAB_BASE_CFG_LOG2SIZE_SHIFT;
+	cfg->strtab_base_cfg = reg;
+
+	arm_smmu_init_bypass_stes(strtab, cfg->num_l1_ents);
+	return 0;
+}
+
+static int arm_smmu_init_strtab(struct arm_smmu_device *smmu)
+{
+	u64 reg;
+	int ret;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
+		ret = arm_smmu_init_strtab_2lvl(smmu);
+	else
+		ret = arm_smmu_init_strtab_linear(smmu);
+
+	if (ret)
+		return ret;
+
+	/* Set the strtab base address */
+	reg  = smmu->strtab_cfg.strtab_dma &
+	       STRTAB_BASE_ADDR_MASK << STRTAB_BASE_ADDR_SHIFT;
+	reg |= STRTAB_BASE_RA;
+	smmu->strtab_cfg.strtab_base = reg;
+
+	/* Allocate the first VMID for stage-2 bypass STEs */
+	set_bit(0, smmu->vmid_map);
+	return 0;
+}
+
+static void arm_smmu_free_strtab(struct arm_smmu_device *smmu)
+{
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+	u32 size = cfg->num_l1_ents;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
+		arm_smmu_free_l2_strtab(smmu);
+		size *= STRTAB_L1_DESC_DWORDS << 3;
+	} else {
+		size *= STRTAB_STE_DWORDS * 3;
+	}
+
+	dma_free_coherent(smmu->dev, size, cfg->strtab, cfg->strtab_dma);
+}
+
+static int arm_smmu_init_structures(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	ret = arm_smmu_init_queues(smmu);
+	if (ret)
+		return ret;
+
+	ret = arm_smmu_init_strtab(smmu);
+	if (ret)
+		goto out_free_queues;
+
+	return 0;
+
+out_free_queues:
+	arm_smmu_free_queues(smmu);
+	return ret;
+}
+
+static void arm_smmu_free_structures(struct arm_smmu_device *smmu)
+{
+	arm_smmu_free_strtab(smmu);
+	arm_smmu_free_queues(smmu);
+}
+
+static int arm_smmu_write_reg_sync(struct arm_smmu_device *smmu, u32 val,
+				   unsigned int reg_off, unsigned int ack_off)
+{
+	u32 reg;
+
+	writel_relaxed(val, smmu->base + reg_off);
+	return readl_relaxed_poll_timeout(smmu->base + ack_off, reg, reg == val,
+					  1, ARM_SMMU_POLL_TIMEOUT_US);
+}
+
+static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
+{
+	int ret, irq;
+
+	/* 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;
+	}
+
+	/* Clear the MSI address regs */
+	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
+	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
+
+	/* Request wired interrupt lines */
+	irq = smmu->evtq.q.irq;
+	if (irq) {
+		ret = devm_request_threaded_irq(smmu->dev, irq,
+						arm_smmu_evtq_handler,
+						arm_smmu_evtq_thread,
+						0, "arm-smmu-v3-evtq", smmu);
+		if (IS_ERR_VALUE(ret))
+			dev_warn(smmu->dev, "failed to enable evtq irq\n");
+	}
+
+	irq = smmu->cmdq.q.irq;
+	if (irq) {
+		ret = devm_request_irq(smmu->dev, irq,
+				       arm_smmu_cmdq_sync_handler, 0,
+				       "arm-smmu-v3-cmdq-sync", smmu);
+		if (IS_ERR_VALUE(ret))
+			dev_warn(smmu->dev, "failed to enable cmdq-sync irq\n");
+	}
+
+	irq = smmu->gerr_irq;
+	if (irq) {
+		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
+				       0, "arm-smmu-v3-gerror", smmu);
+		if (IS_ERR_VALUE(ret))
+			dev_warn(smmu->dev, "failed to enable gerror irq\n");
+	}
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI) {
+		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
+
+		irq = smmu->priq.q.irq;
+		if (irq) {
+			ret = devm_request_threaded_irq(smmu->dev, irq,
+							arm_smmu_priq_handler,
+							arm_smmu_priq_thread,
+							0, "arm-smmu-v3-priq",
+							smmu);
+			if (IS_ERR_VALUE(ret))
+				dev_warn(smmu->dev,
+					 "failed to enable priq irq\n");
+		}
+	}
+
+	/* Enable interrupt generation on the SMMU */
+	ret = arm_smmu_write_reg_sync(smmu,
+				      IRQ_CTRL_EVTQ_IRQEN |
+				      IRQ_CTRL_GERROR_IRQEN,
+				      ARM_SMMU_IRQ_CTRL, ARM_SMMU_IRQ_CTRLACK);
+	if (ret)
+		dev_warn(smmu->dev, "failed to enable irqs\n");
+
+	return 0;
+}
+
+static int arm_smmu_device_disable(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_CR0, ARM_SMMU_CR0ACK);
+	if (ret)
+		dev_err(smmu->dev, "failed to clear cr0\n");
+
+	return ret;
+}
+
+static int arm_smmu_device_reset(struct arm_smmu_device *smmu)
+{
+	int ret;
+	u32 reg, enables;
+	struct arm_smmu_cmdq_ent cmd;
+
+	/* Clear CR0 and sync (disables SMMU and queue processing) */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
+	if (reg & CR0_SMMUEN)
+		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
+
+	ret = arm_smmu_device_disable(smmu);
+	if (ret)
+		return ret;
+
+	/* CR1 (table and queue memory attributes) */
+	reg = (CR1_SH_ISH << CR1_TABLE_SH_SHIFT) |
+	      (CR1_CACHE_WB << CR1_TABLE_OC_SHIFT) |
+	      (CR1_CACHE_WB << CR1_TABLE_IC_SHIFT) |
+	      (CR1_SH_ISH << CR1_QUEUE_SH_SHIFT) |
+	      (CR1_CACHE_WB << CR1_QUEUE_OC_SHIFT) |
+	      (CR1_CACHE_WB << CR1_QUEUE_IC_SHIFT);
+	writel_relaxed(reg, smmu->base + ARM_SMMU_CR1);
+
+	/* CR2 (random crap) */
+	reg = CR2_PTM | CR2_RECINVSID | CR2_E2H;
+	writel_relaxed(reg, smmu->base + ARM_SMMU_CR2);
+
+	/* Stream table */
+	writeq_relaxed(smmu->strtab_cfg.strtab_base,
+		       smmu->base + ARM_SMMU_STRTAB_BASE);
+	writel_relaxed(smmu->strtab_cfg.strtab_base_cfg,
+		       smmu->base + ARM_SMMU_STRTAB_BASE_CFG);
+
+	/* Command queue */
+	writeq_relaxed(smmu->cmdq.q.q_base, smmu->base + ARM_SMMU_CMDQ_BASE);
+	writel_relaxed(smmu->cmdq.q.prod, smmu->base + ARM_SMMU_CMDQ_PROD);
+	writel_relaxed(smmu->cmdq.q.cons, smmu->base + ARM_SMMU_CMDQ_CONS);
+
+	enables = CR0_CMDQEN;
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable command queue\n");
+		return ret;
+	}
+
+	/* Invalidate any cached configuration */
+	cmd.opcode = CMDQ_OP_CFGI_ALL;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	cmd.opcode = CMDQ_OP_CMD_SYNC;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+
+	/* Invalidate any stale TLB entries */
+	if (smmu->features & ARM_SMMU_FEAT_HYP) {
+		cmd.opcode = CMDQ_OP_TLBI_EL2_ALL;
+		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	}
+
+	cmd.opcode = CMDQ_OP_TLBI_NSNH_ALL;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	cmd.opcode = CMDQ_OP_CMD_SYNC;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+
+	/* 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);
+
+	enables |= CR0_EVTQEN;
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable event queue\n");
+		return ret;
+	}
+
+	/* PRI queue */
+	if (smmu->features & ARM_SMMU_FEAT_PRI) {
+		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);
+		writel_relaxed(smmu->priq.q.cons,
+			       smmu->base + ARM_SMMU_PRIQ_CONS);
+
+		enables |= CR0_PRIQEN;
+		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+					      ARM_SMMU_CR0ACK);
+		if (ret) {
+			dev_err(smmu->dev, "failed to enable PRI queue\n");
+			return ret;
+		}
+	}
+
+	ret = arm_smmu_setup_irqs(smmu);
+	if (ret) {
+		dev_err(smmu->dev, "failed to setup irqs\n");
+		return ret;
+	}
+
+	/* Enable the SMMU interface */
+	enables |= CR0_SMMUEN;
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable SMMU interface\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int arm_smmu_device_probe(struct arm_smmu_device *smmu)
+{
+	u32 reg;
+	bool coherent;
+	unsigned long pgsize_bitmap = 0;
+
+	/* IDR0 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR0);
+
+	/* 2-level structures */
+	if ((reg & IDR0_ST_LVL_MASK << IDR0_ST_LVL_SHIFT) == IDR0_ST_LVL_2LVL)
+		smmu->features |= ARM_SMMU_FEAT_2_LVL_STRTAB;
+
+	if (reg & IDR0_CD2L)
+		smmu->features |= ARM_SMMU_FEAT_2_LVL_CDTAB;
+
+	/*
+	 * Translation table endianness.
+	 * We currently require the same endianness as the CPU, but this
+	 * could be changed later by adding a new IO_PGTABLE_QUIRK.
+	 */
+	switch (reg & IDR0_TTENDIAN_MASK << IDR0_TTENDIAN_SHIFT) {
+	case IDR0_TTENDIAN_MIXED:
+		smmu->features |= ARM_SMMU_FEAT_TT_LE | ARM_SMMU_FEAT_TT_BE;
+		break;
+#ifdef __BIG_ENDIAN
+	case IDR0_TTENDIAN_BE:
+		smmu->features |= ARM_SMMU_FEAT_TT_BE;
+		break;
+#else
+	case IDR0_TTENDIAN_LE:
+		smmu->features |= ARM_SMMU_FEAT_TT_LE;
+		break;
+#endif
+	default:
+		dev_err(smmu->dev, "unknown/unsupported TT endianness!\n");
+		return -ENXIO;
+	}
+
+	/* Boolean feature flags */
+	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
+		smmu->features |= ARM_SMMU_FEAT_PRI;
+
+	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
+		smmu->features |= ARM_SMMU_FEAT_ATS;
+
+	if (reg & IDR0_SEV)
+		smmu->features |= ARM_SMMU_FEAT_SEV;
+
+	if (reg & IDR0_MSI)
+		smmu->features |= ARM_SMMU_FEAT_MSI;
+
+	if (reg & IDR0_HYP)
+		smmu->features |= ARM_SMMU_FEAT_HYP;
+
+	/*
+	 * The dma-coherent property is used in preference to the ID
+	 * register, but warn on mismatch.
+	 */
+	coherent = of_dma_is_coherent(smmu->dev->of_node);
+	if (coherent)
+		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
+
+	if (!!(reg & IDR0_COHACC) != coherent)
+		dev_warn(smmu->dev, "IDR0.COHACC overridden by dma-coherent property (%s)\n",
+			 coherent ? "true" : "false");
+
+	if (reg & IDR0_STALL_MODEL)
+		smmu->features |= ARM_SMMU_FEAT_STALLS;
+
+	if (reg & IDR0_S1P)
+		smmu->features |= ARM_SMMU_FEAT_TRANS_S1;
+
+	if (reg & IDR0_S2P)
+		smmu->features |= ARM_SMMU_FEAT_TRANS_S2;
+
+	if (!(reg & (IDR0_S1P | IDR0_S2P))) {
+		dev_err(smmu->dev, "no translation support!\n");
+		return -ENXIO;
+	}
+
+	/* We only support the AArch64 table format at present */
+	if ((reg & IDR0_TTF_MASK << IDR0_TTF_SHIFT) < IDR0_TTF_AARCH64) {
+		dev_err(smmu->dev, "AArch64 table format not supported!\n");
+		return -ENXIO;
+	}
+
+	/* ASID/VMID sizes */
+	smmu->asid_bits = reg & IDR0_ASID16 ? 16 : 8;
+	smmu->vmid_bits = reg & IDR0_VMID16 ? 16 : 8;
+
+	/* IDR1 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR1);
+	if (reg & (IDR1_TABLES_PRESET | IDR1_QUEUES_PRESET | IDR1_REL)) {
+		dev_err(smmu->dev, "embedded implementation not supported\n");
+		return -ENXIO;
+	}
+
+	/* Queue sizes, capped at 4k */
+	smmu->cmdq.q.max_n_shift = min((u32)CMDQ_MAX_SZ_SHIFT,
+				       reg >> IDR1_CMDQ_SHIFT & IDR1_CMDQ_MASK);
+	if (!smmu->cmdq.q.max_n_shift) {
+		/* Odd alignment restrictions on the base, so ignore for now */
+		dev_err(smmu->dev, "unit-length command queue not supported\n");
+		return -ENXIO;
+	}
+
+	smmu->evtq.q.max_n_shift = min((u32)EVTQ_MAX_SZ_SHIFT,
+				       reg >> IDR1_EVTQ_SHIFT & IDR1_EVTQ_MASK);
+	smmu->priq.q.max_n_shift = min((u32)PRIQ_MAX_SZ_SHIFT,
+				       reg >> IDR1_PRIQ_SHIFT & IDR1_PRIQ_MASK);
+
+	/* SID/SSID sizes */
+	smmu->ssid_bits = reg >> IDR1_SSID_SHIFT & IDR1_SSID_MASK;
+	smmu->sid_bits = reg >> IDR1_SID_SHIFT & IDR1_SID_MASK;
+
+	/* IDR5 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR5);
+
+	/* Maximum number of outstanding stalls */
+	smmu->evtq.max_stalls = reg >> IDR5_STALL_MAX_SHIFT
+				& IDR5_STALL_MAX_MASK;
+
+	/* Page sizes */
+	if (reg & IDR5_GRAN64K)
+		pgsize_bitmap |= SZ_64K | SZ_512M;
+	if (reg & IDR5_GRAN16K)
+		pgsize_bitmap |= SZ_16K | SZ_32M;
+	if (reg & IDR5_GRAN4K)
+		pgsize_bitmap |= SZ_4K | SZ_2M | SZ_1G;
+
+	arm_smmu_ops.pgsize_bitmap &= pgsize_bitmap;
+
+	/* Output address size */
+	switch (reg & IDR5_OAS_MASK << IDR5_OAS_SHIFT) {
+	case IDR5_OAS_32_BIT:
+		smmu->oas = 32;
+		break;
+	case IDR5_OAS_36_BIT:
+		smmu->oas = 36;
+		break;
+	case IDR5_OAS_40_BIT:
+		smmu->oas = 40;
+		break;
+	case IDR5_OAS_42_BIT:
+		smmu->oas = 42;
+		break;
+	case IDR5_OAS_44_BIT:
+		smmu->oas = 44;
+		break;
+	case IDR5_OAS_48_BIT:
+		smmu->oas = 48;
+		break;
+	default:
+		dev_err(smmu->dev, "unknown output address size!\n");
+		return -ENXIO;
+	}
+
+	/* Set the DMA mask for our table walker */
+	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
+		dev_warn(smmu->dev,
+			 "failed to set DMA mask for table walker\n");
+
+	if (!smmu->ias)
+		smmu->ias = smmu->oas;
+
+	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
+		 smmu->ias, smmu->oas, smmu->features);
+	return 0;
+}
+
+static int arm_smmu_device_dt_probe(struct platform_device *pdev)
+{
+	int irq, ret;
+	struct resource *res;
+	struct arm_smmu_device *smmu;
+	struct device *dev = &pdev->dev;
+
+	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
+	if (!smmu) {
+		dev_err(dev, "failed to allocate arm_smmu_device\n");
+		return -ENOMEM;
+	}
+	smmu->dev = dev;
+
+	/* Base address */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (resource_size(res) + 1 < SZ_128K) {
+		dev_err(dev, "MMIO region too small (%pr)\n", res);
+		return -EINVAL;
+	}
+
+	smmu->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(smmu->base))
+		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");
+	if (irq > 0)
+		smmu->priq.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, "gerror");
+	if (irq > 0)
+		smmu->gerr_irq = irq;
+
+	parse_driver_options(smmu);
+
+	/* Probe the h/w */
+	ret = arm_smmu_device_probe(smmu);
+	if (ret)
+		return ret;
+
+	/* Initialise in-memory data structures */
+	ret = arm_smmu_init_structures(smmu);
+	if (ret)
+		return ret;
+
+	/* Reset the device */
+	ret = arm_smmu_device_reset(smmu);
+	if (ret)
+		goto out_free_structures;
+
+	/* Record our private device structure */
+	INIT_LIST_HEAD(&smmu->list);
+	spin_lock(&arm_smmu_devices_lock);
+	list_add(&smmu->list, &arm_smmu_devices);
+	spin_unlock(&arm_smmu_devices_lock);
+	return 0;
+
+out_free_structures:
+	arm_smmu_free_structures(smmu);
+	return ret;
+}
+
+static int arm_smmu_device_remove(struct platform_device *pdev)
+{
+	struct arm_smmu_device *curr, *smmu = NULL;
+	struct device *dev = &pdev->dev;
+
+	spin_lock(&arm_smmu_devices_lock);
+	list_for_each_entry(curr, &arm_smmu_devices, list) {
+		if (curr->dev == dev) {
+			smmu = curr;
+			list_del(&smmu->list);
+			break;
+		}
+	}
+	spin_unlock(&arm_smmu_devices_lock);
+
+	if (!smmu)
+		return -ENODEV;
+
+	arm_smmu_device_disable(smmu);
+	arm_smmu_free_structures(smmu);
+	return 0;
+}
+
+static struct of_device_id arm_smmu_of_match[] = {
+	{ .compatible = "arm,smmu-v3", },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
+
+static struct platform_driver arm_smmu_driver = {
+	.driver	= {
+		.name		= "arm-smmu-v3",
+		.of_match_table	= of_match_ptr(arm_smmu_of_match),
+	},
+	.probe	= arm_smmu_device_dt_probe,
+	.remove	= arm_smmu_device_remove,
+};
+
+static int __init arm_smmu_init(void)
+{
+	struct device_node *np;
+	int ret;
+
+	np = of_find_matching_node(NULL, arm_smmu_of_match);
+	if (!np)
+		return 0;
+
+	of_node_put(np);
+
+	ret = platform_driver_register(&arm_smmu_driver);
+	if (ret)
+		return ret;
+
+	return bus_set_iommu(&pci_bus_type, &arm_smmu_ops);
+}
+
+static void __exit arm_smmu_exit(void)
+{
+	return platform_driver_unregister(&arm_smmu_driver);
+}
+
+subsys_initcall(arm_smmu_init);
+module_exit(arm_smmu_exit);
+
+MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
+MODULE_AUTHOR("Will Deacon <will.deacon@arm.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 65075ef75..4cd0c29cb 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -202,8 +202,7 @@
 #define ARM_SMMU_CB_S1_TLBIVAL		0x620
 #define ARM_SMMU_CB_S2_TLBIIPAS2	0x630
 #define ARM_SMMU_CB_S2_TLBIIPAS2L	0x638
-#define ARM_SMMU_CB_ATS1PR_LO		0x800
-#define ARM_SMMU_CB_ATS1PR_HI		0x804
+#define ARM_SMMU_CB_ATS1PR		0x800
 #define ARM_SMMU_CB_ATSR		0x8f0
 
 #define SCTLR_S1_ASIDPNE		(1 << 12)
@@ -247,7 +246,7 @@
 #define FSYNR0_WNR			(1 << 4)
 
 static int force_stage;
-module_param_named(force_stage, force_stage, int, S_IRUGO | S_IWUSR);
+module_param_named(force_stage, force_stage, int, S_IRUGO);
 MODULE_PARM_DESC(force_stage,
 	"Force SMMU mappings to be installed at a particular stage of translation. A value of '1' or '2' forces the corresponding stage. All other values are ignored (i.e. no stage is forced). Note that selecting a specific stage will disable support for nested translation.");
 
@@ -1229,18 +1228,18 @@ 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;
 
 	cb_base = ARM_SMMU_CB_BASE(smmu) + ARM_SMMU_CB(smmu, cfg->cbndx);
 
-	if (smmu->version == 1) {
-		u32 reg = iova & ~0xfff;
-		writel_relaxed(reg, cb_base + ARM_SMMU_CB_ATS1PR_LO);
-	} else {
-		u32 reg = iova & ~0xfff;
-		writel_relaxed(reg, cb_base + ARM_SMMU_CB_ATS1PR_LO);
-		reg = ((u64)iova & ~0xfff) >> 32;
-		writel_relaxed(reg, cb_base + ARM_SMMU_CB_ATS1PR_HI);
-	}
+	/* ATS1 registers can only be written atomically */
+	va = iova & ~0xfffUL;
+#ifdef CONFIG_64BIT
+	if (smmu->version == ARM_SMMU_V2)
+		writeq_relaxed(va, cb_base + ARM_SMMU_CB_ATS1PR);
+	else
+#endif
+		writel_relaxed(va, cb_base + ARM_SMMU_CB_ATS1PR);
 
 	if (readl_poll_timeout_atomic(cb_base + ARM_SMMU_CB_ATSR, tmp,
 				      !(tmp & ATSR_ACTIVE), 5, 50)) {
diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c
index 984761308..c9db04d4e 100644
--- a/drivers/iommu/dmar.c
+++ b/drivers/iommu/dmar.c
@@ -26,7 +26,7 @@
  * These routines are used by both DMA-remapping and Interrupt-remapping
  */
 
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt /* has to precede printk.h */
+#define pr_fmt(fmt)     "DMAR: " fmt
 
 #include <linux/pci.h>
 #include <linux/dmar.h>
@@ -555,7 +555,7 @@ static int dmar_walk_remapping_entries(struct acpi_dmar_header *start,
 			break;
 		} else if (next > end) {
 			/* Avoid passing table end */
-			pr_warn(FW_BUG "record passes table end\n");
+			pr_warn(FW_BUG "Record passes table end\n");
 			ret = -EINVAL;
 			break;
 		}
@@ -802,7 +802,7 @@ int __init dmar_table_init(void)
 		ret = parse_dmar_table();
 		if (ret < 0) {
 			if (ret != -ENODEV)
-				pr_info("parse DMAR table failure.\n");
+				pr_info("Parse DMAR table failure.\n");
 		} else  if (list_empty(&dmar_drhd_units)) {
 			pr_info("No DMAR devices found\n");
 			ret = -ENODEV;
@@ -847,7 +847,7 @@ dmar_validate_one_drhd(struct acpi_dmar_header *entry, void *arg)
 	else
 		addr = early_ioremap(drhd->address, VTD_PAGE_SIZE);
 	if (!addr) {
-		pr_warn("IOMMU: can't validate: %llx\n", drhd->address);
+		pr_warn("Can't validate DRHD address: %llx\n", drhd->address);
 		return -EINVAL;
 	}
 
@@ -921,14 +921,14 @@ static int map_iommu(struct intel_iommu *iommu, u64 phys_addr)
 	iommu->reg_size = VTD_PAGE_SIZE;
 
 	if (!request_mem_region(iommu->reg_phys, iommu->reg_size, iommu->name)) {
-		pr_err("IOMMU: can't reserve memory\n");
+		pr_err("Can't reserve memory\n");
 		err = -EBUSY;
 		goto out;
 	}
 
 	iommu->reg = ioremap(iommu->reg_phys, iommu->reg_size);
 	if (!iommu->reg) {
-		pr_err("IOMMU: can't map the region\n");
+		pr_err("Can't map the region\n");
 		err = -ENOMEM;
 		goto release;
 	}
@@ -952,13 +952,13 @@ static int map_iommu(struct intel_iommu *iommu, u64 phys_addr)
 		iommu->reg_size = map_size;
 		if (!request_mem_region(iommu->reg_phys, iommu->reg_size,
 					iommu->name)) {
-			pr_err("IOMMU: can't reserve memory\n");
+			pr_err("Can't reserve memory\n");
 			err = -EBUSY;
 			goto out;
 		}
 		iommu->reg = ioremap(iommu->reg_phys, iommu->reg_size);
 		if (!iommu->reg) {
-			pr_err("IOMMU: can't map the region\n");
+			pr_err("Can't map the region\n");
 			err = -ENOMEM;
 			goto release;
 		}
@@ -1014,14 +1014,14 @@ static int alloc_iommu(struct dmar_drhd_unit *drhd)
 		return -ENOMEM;
 
 	if (dmar_alloc_seq_id(iommu) < 0) {
-		pr_err("IOMMU: failed to allocate seq_id\n");
+		pr_err("Failed to allocate seq_id\n");
 		err = -ENOSPC;
 		goto error;
 	}
 
 	err = map_iommu(iommu, drhd->reg_base_addr);
 	if (err) {
-		pr_err("IOMMU: failed to map %s\n", iommu->name);
+		pr_err("Failed to map %s\n", iommu->name);
 		goto error_free_seq_id;
 	}
 
@@ -1045,8 +1045,8 @@ static int alloc_iommu(struct dmar_drhd_unit *drhd)
 	iommu->node = -1;
 
 	ver = readl(iommu->reg + DMAR_VER_REG);
-	pr_info("IOMMU %d: reg_base_addr %llx ver %d:%d cap %llx ecap %llx\n",
-		iommu->seq_id,
+	pr_info("%s: reg_base_addr %llx ver %d:%d cap %llx ecap %llx\n",
+		iommu->name,
 		(unsigned long long)drhd->reg_base_addr,
 		DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver),
 		(unsigned long long)iommu->cap,
@@ -1087,8 +1087,8 @@ static void free_iommu(struct intel_iommu *iommu)
 
 	if (iommu->irq) {
 		free_irq(iommu->irq, iommu);
-		irq_set_handler_data(iommu->irq, NULL);
 		dmar_free_hwirq(iommu->irq);
+		iommu->irq = 0;
 	}
 
 	if (iommu->qi) {
@@ -1642,26 +1642,17 @@ int dmar_set_interrupt(struct intel_iommu *iommu)
 	if (iommu->irq)
 		return 0;
 
-	irq = dmar_alloc_hwirq();
-	if (irq <= 0) {
-		pr_err("IOMMU: no free vectors\n");
+	irq = dmar_alloc_hwirq(iommu->seq_id, iommu->node, iommu);
+	if (irq > 0) {
+		iommu->irq = irq;
+	} else {
+		pr_err("No free IRQ vectors\n");
 		return -EINVAL;
 	}
 
-	irq_set_handler_data(irq, iommu);
-	iommu->irq = irq;
-
-	ret = arch_setup_dmar_msi(irq);
-	if (ret) {
-		irq_set_handler_data(irq, NULL);
-		iommu->irq = 0;
-		dmar_free_hwirq(irq);
-		return ret;
-	}
-
 	ret = request_irq(irq, dmar_fault, IRQF_NO_THREAD, iommu->name, iommu);
 	if (ret)
-		pr_err("IOMMU: can't request irq\n");
+		pr_err("Can't request irq\n");
 	return ret;
 }
 
diff --git a/drivers/iommu/exynos-iommu.c b/drivers/iommu/exynos-iommu.c
index 3e898504a..97c41b8ab 100644
--- a/drivers/iommu/exynos-iommu.c
+++ b/drivers/iommu/exynos-iommu.c
@@ -12,21 +12,22 @@
 #define DEBUG
 #endif
 
-#include <linux/io.h>
-#include <linux/interrupt.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-#include <linux/pm_runtime.h>
 #include <linux/clk.h>
+#include <linux/dma-mapping.h>
 #include <linux/err.h>
-#include <linux/mm.h>
+#include <linux/io.h>
 #include <linux/iommu.h>
-#include <linux/errno.h>
+#include <linux/interrupt.h>
 #include <linux/list.h>
-#include <linux/memblock.h>
-#include <linux/export.h>
+#include <linux/of.h>
+#include <linux/of_iommu.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/slab.h>
 
 #include <asm/cacheflush.h>
+#include <asm/dma-iommu.h>
 #include <asm/pgtable.h>
 
 typedef u32 sysmmu_iova_t;
@@ -184,35 +185,50 @@ static char *sysmmu_fault_name[SYSMMU_FAULTS_NUM] = {
 	"UNKNOWN FAULT"
 };
 
-/* attached to dev.archdata.iommu of the master device */
+/*
+ * This structure is attached to dev.archdata.iommu of the master device
+ * on device add, contains a list of SYSMMU controllers defined by device tree,
+ * which are bound to given master device. It is usually referenced by 'owner'
+ * pointer.
+*/
 struct exynos_iommu_owner {
-	struct list_head client; /* entry of exynos_iommu_domain.clients */
-	struct device *dev;
-	struct device *sysmmu;
-	struct iommu_domain *domain;
-	void *vmm_data;         /* IO virtual memory manager's data */
-	spinlock_t lock;        /* Lock to preserve consistency of System MMU */
+	struct list_head controllers;	/* list of sysmmu_drvdata.owner_node */
 };
 
+/*
+ * This structure exynos specific generalization of struct iommu_domain.
+ * It contains list of SYSMMU controllers from all master devices, which has
+ * been attached to this domain and page tables of IO address space defined by
+ * it. It is usually referenced by 'domain' pointer.
+ */
 struct exynos_iommu_domain {
-	struct list_head clients; /* list of sysmmu_drvdata.node */
-	sysmmu_pte_t *pgtable; /* lv1 page table, 16KB */
-	short *lv2entcnt; /* free lv2 entry counter for each section */
-	spinlock_t lock; /* lock for this structure */
-	spinlock_t pgtablelock; /* lock for modifying page table @ pgtable */
+	struct list_head clients; /* list of sysmmu_drvdata.domain_node */
+	sysmmu_pte_t *pgtable;	/* lv1 page table, 16KB */
+	short *lv2entcnt;	/* free lv2 entry counter for each section */
+	spinlock_t lock;	/* lock for modyfying list of clients */
+	spinlock_t pgtablelock;	/* lock for modifying page table @ pgtable */
 	struct iommu_domain domain; /* generic domain data structure */
 };
 
+/*
+ * This structure hold all data of a single SYSMMU controller, this includes
+ * hw resources like registers and clocks, pointers and list nodes to connect
+ * it to all other structures, internal state and parameters read from device
+ * tree. It is usually referenced by 'data' pointer.
+ */
 struct sysmmu_drvdata {
-	struct device *sysmmu;	/* System MMU's device descriptor */
-	struct device *master;	/* Owner of system MMU */
-	void __iomem *sfrbase;
-	struct clk *clk;
-	struct clk *clk_master;
-	int activations;
-	spinlock_t lock;
-	struct iommu_domain *domain;
-	phys_addr_t pgtable;
+	struct device *sysmmu;		/* SYSMMU controller device */
+	struct device *master;		/* master device (owner) */
+	void __iomem *sfrbase;		/* our registers */
+	struct clk *clk;		/* SYSMMU's clock */
+	struct clk *clk_master;		/* master's device clock */
+	int activations;		/* number of calls to sysmmu_enable */
+	spinlock_t lock;		/* lock for modyfying state */
+	struct exynos_iommu_domain *domain; /* domain we belong to */
+	struct list_head domain_node;	/* node for domain clients list */
+	struct list_head owner_node;	/* node for owner controllers list */
+	phys_addr_t pgtable;		/* assigned page table structure */
+	unsigned int version;		/* our version */
 };
 
 static struct exynos_iommu_domain *to_exynos_domain(struct iommu_domain *dom)
@@ -244,11 +260,6 @@ static void sysmmu_unblock(void __iomem *sfrbase)
 	__raw_writel(CTRL_ENABLE, sfrbase + REG_MMU_CTRL);
 }
 
-static unsigned int __raw_sysmmu_version(struct sysmmu_drvdata *data)
-{
-	return MMU_RAW_VER(__raw_readl(data->sfrbase + REG_MMU_VERSION));
-}
-
 static bool sysmmu_block(void __iomem *sfrbase)
 {
 	int i = 120;
@@ -345,7 +356,7 @@ static irqreturn_t exynos_sysmmu_irq(int irq, void *dev_id)
 		show_fault_information(dev_name(data->sysmmu),
 					itype, base, addr);
 		if (data->domain)
-			ret = report_iommu_fault(data->domain,
+			ret = report_iommu_fault(&data->domain->domain,
 					data->master, addr, itype);
 	}
 
@@ -408,7 +419,7 @@ static void __sysmmu_init_config(struct sysmmu_drvdata *data)
 	unsigned int cfg = CFG_LRU | CFG_QOS(15);
 	unsigned int ver;
 
-	ver = __raw_sysmmu_version(data);
+	ver = MMU_RAW_VER(__raw_readl(data->sfrbase + REG_MMU_VERSION));
 	if (MMU_MAJ_VER(ver) == 3) {
 		if (MMU_MIN_VER(ver) >= 2) {
 			cfg |= CFG_FLPDCACHE;
@@ -422,6 +433,7 @@ static void __sysmmu_init_config(struct sysmmu_drvdata *data)
 	}
 
 	__raw_writel(cfg, data->sfrbase + REG_MMU_CFG);
+	data->version = ver;
 }
 
 static void __sysmmu_enable_nocount(struct sysmmu_drvdata *data)
@@ -442,8 +454,8 @@ static void __sysmmu_enable_nocount(struct sysmmu_drvdata *data)
 		clk_disable(data->clk_master);
 }
 
-static int __sysmmu_enable(struct sysmmu_drvdata *data,
-			phys_addr_t pgtable, struct iommu_domain *domain)
+static int __sysmmu_enable(struct sysmmu_drvdata *data, phys_addr_t pgtable,
+			   struct exynos_iommu_domain *domain)
 {
 	int ret = 0;
 	unsigned long flags;
@@ -470,77 +482,17 @@ static int __sysmmu_enable(struct sysmmu_drvdata *data,
 	return ret;
 }
 
-/* __exynos_sysmmu_enable: Enables System MMU
- *
- * returns -error if an error occurred and System MMU is not enabled,
- * 0 if the System MMU has been just enabled and 1 if System MMU was already
- * enabled before.
- */
-static int __exynos_sysmmu_enable(struct device *dev, phys_addr_t pgtable,
-				  struct iommu_domain *domain)
-{
-	int ret = 0;
-	unsigned long flags;
-	struct exynos_iommu_owner *owner = dev->archdata.iommu;
-	struct sysmmu_drvdata *data;
-
-	BUG_ON(!has_sysmmu(dev));
-
-	spin_lock_irqsave(&owner->lock, flags);
-
-	data = dev_get_drvdata(owner->sysmmu);
-
-	ret = __sysmmu_enable(data, pgtable, domain);
-	if (ret >= 0)
-		data->master = dev;
-
-	spin_unlock_irqrestore(&owner->lock, flags);
-
-	return ret;
-}
-
-int exynos_sysmmu_enable(struct device *dev, phys_addr_t pgtable)
-{
-	BUG_ON(!memblock_is_memory(pgtable));
-
-	return __exynos_sysmmu_enable(dev, pgtable, NULL);
-}
-
-static bool exynos_sysmmu_disable(struct device *dev)
-{
-	unsigned long flags;
-	bool disabled = true;
-	struct exynos_iommu_owner *owner = dev->archdata.iommu;
-	struct sysmmu_drvdata *data;
-
-	BUG_ON(!has_sysmmu(dev));
-
-	spin_lock_irqsave(&owner->lock, flags);
-
-	data = dev_get_drvdata(owner->sysmmu);
-
-	disabled = __sysmmu_disable(data);
-	if (disabled)
-		data->master = NULL;
-
-	spin_unlock_irqrestore(&owner->lock, flags);
-
-	return disabled;
-}
-
 static void __sysmmu_tlb_invalidate_flpdcache(struct sysmmu_drvdata *data,
 					      sysmmu_iova_t iova)
 {
-	if (__raw_sysmmu_version(data) == MAKE_MMU_VER(3, 3))
+	if (data->version == MAKE_MMU_VER(3, 3))
 		__raw_writel(iova | 0x1, data->sfrbase + REG_MMU_FLUSH_ENTRY);
 }
 
-static void sysmmu_tlb_invalidate_flpdcache(struct device *dev,
+static void sysmmu_tlb_invalidate_flpdcache(struct sysmmu_drvdata *data,
 					    sysmmu_iova_t iova)
 {
 	unsigned long flags;
-	struct exynos_iommu_owner *owner = dev->archdata.iommu;
-	struct sysmmu_drvdata *data = dev_get_drvdata(owner->sysmmu);
 
 	if (!IS_ERR(data->clk_master))
 		clk_enable(data->clk_master);
@@ -554,14 +506,10 @@ static void sysmmu_tlb_invalidate_flpdcache(struct device *dev,
 		clk_disable(data->clk_master);
 }
 
-static void sysmmu_tlb_invalidate_entry(struct device *dev, sysmmu_iova_t iova,
-					size_t size)
+static void sysmmu_tlb_invalidate_entry(struct sysmmu_drvdata *data,
+					sysmmu_iova_t iova, size_t size)
 {
-	struct exynos_iommu_owner *owner = dev->archdata.iommu;
 	unsigned long flags;
-	struct sysmmu_drvdata *data;
-
-	data = dev_get_drvdata(owner->sysmmu);
 
 	spin_lock_irqsave(&data->lock, flags);
 	if (is_sysmmu_active(data)) {
@@ -580,7 +528,7 @@ static void sysmmu_tlb_invalidate_entry(struct device *dev, sysmmu_iova_t iova,
 		 * 1MB page can be cached in one of all sets.
 		 * 64KB page can be one of 16 consecutive sets.
 		 */
-		if (MMU_MAJ_VER(__raw_sysmmu_version(data)) == 2)
+		if (MMU_MAJ_VER(data->version) == 2)
 			num_inv = min_t(unsigned int, size / PAGE_SIZE, 64);
 
 		if (sysmmu_block(data->sfrbase)) {
@@ -591,32 +539,8 @@ static void sysmmu_tlb_invalidate_entry(struct device *dev, sysmmu_iova_t iova,
 		if (!IS_ERR(data->clk_master))
 			clk_disable(data->clk_master);
 	} else {
-		dev_dbg(dev, "disabled. Skipping TLB invalidation @ %#x\n",
-			iova);
-	}
-	spin_unlock_irqrestore(&data->lock, flags);
-}
-
-void exynos_sysmmu_tlb_invalidate(struct device *dev)
-{
-	struct exynos_iommu_owner *owner = dev->archdata.iommu;
-	unsigned long flags;
-	struct sysmmu_drvdata *data;
-
-	data = dev_get_drvdata(owner->sysmmu);
-
-	spin_lock_irqsave(&data->lock, flags);
-	if (is_sysmmu_active(data)) {
-		if (!IS_ERR(data->clk_master))
-			clk_enable(data->clk_master);
-		if (sysmmu_block(data->sfrbase)) {
-			__sysmmu_tlb_invalidate(data->sfrbase);
-			sysmmu_unblock(data->sfrbase);
-		}
-		if (!IS_ERR(data->clk_master))
-			clk_disable(data->clk_master);
-	} else {
-		dev_dbg(dev, "disabled. Skipping TLB invalidation\n");
+		dev_dbg(data->master,
+			"disabled. Skipping TLB invalidation @ %#x\n", iova);
 	}
 	spin_unlock_irqrestore(&data->lock, flags);
 }
@@ -682,6 +606,36 @@ static int __init exynos_sysmmu_probe(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int exynos_sysmmu_suspend(struct device *dev)
+{
+	struct sysmmu_drvdata *data = dev_get_drvdata(dev);
+
+	dev_dbg(dev, "suspend\n");
+	if (is_sysmmu_active(data)) {
+		__sysmmu_disable_nocount(data);
+		pm_runtime_put(dev);
+	}
+	return 0;
+}
+
+static int exynos_sysmmu_resume(struct device *dev)
+{
+	struct sysmmu_drvdata *data = dev_get_drvdata(dev);
+
+	dev_dbg(dev, "resume\n");
+	if (is_sysmmu_active(data)) {
+		pm_runtime_get_sync(dev);
+		__sysmmu_enable_nocount(data);
+	}
+	return 0;
+}
+#endif
+
+static const struct dev_pm_ops sysmmu_pm_ops = {
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(exynos_sysmmu_suspend, exynos_sysmmu_resume)
+};
+
 static const struct of_device_id sysmmu_of_match[] __initconst = {
 	{ .compatible	= "samsung,exynos-sysmmu", },
 	{ },
@@ -692,6 +646,7 @@ static struct platform_driver exynos_sysmmu_driver __refdata = {
 	.driver	= {
 		.name		= "exynos-sysmmu",
 		.of_match_table	= sysmmu_of_match,
+		.pm		= &sysmmu_pm_ops,
 	}
 };
 
@@ -704,104 +659,108 @@ static inline void pgtable_flush(void *vastart, void *vaend)
 
 static struct iommu_domain *exynos_iommu_domain_alloc(unsigned type)
 {
-	struct exynos_iommu_domain *exynos_domain;
+	struct exynos_iommu_domain *domain;
 	int i;
 
 	if (type != IOMMU_DOMAIN_UNMANAGED)
 		return NULL;
 
-	exynos_domain = kzalloc(sizeof(*exynos_domain), GFP_KERNEL);
-	if (!exynos_domain)
+	domain = kzalloc(sizeof(*domain), GFP_KERNEL);
+	if (!domain)
 		return NULL;
 
-	exynos_domain->pgtable = (sysmmu_pte_t *)__get_free_pages(GFP_KERNEL, 2);
-	if (!exynos_domain->pgtable)
+	domain->pgtable = (sysmmu_pte_t *)__get_free_pages(GFP_KERNEL, 2);
+	if (!domain->pgtable)
 		goto err_pgtable;
 
-	exynos_domain->lv2entcnt = (short *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, 1);
-	if (!exynos_domain->lv2entcnt)
+	domain->lv2entcnt = (short *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, 1);
+	if (!domain->lv2entcnt)
 		goto err_counter;
 
 	/* Workaround for System MMU v3.3 to prevent caching 1MiB mapping */
 	for (i = 0; i < NUM_LV1ENTRIES; i += 8) {
-		exynos_domain->pgtable[i + 0] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 1] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 2] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 3] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 4] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 5] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 6] = ZERO_LV2LINK;
-		exynos_domain->pgtable[i + 7] = ZERO_LV2LINK;
+		domain->pgtable[i + 0] = ZERO_LV2LINK;
+		domain->pgtable[i + 1] = ZERO_LV2LINK;
+		domain->pgtable[i + 2] = ZERO_LV2LINK;
+		domain->pgtable[i + 3] = ZERO_LV2LINK;
+		domain->pgtable[i + 4] = ZERO_LV2LINK;
+		domain->pgtable[i + 5] = ZERO_LV2LINK;
+		domain->pgtable[i + 6] = ZERO_LV2LINK;
+		domain->pgtable[i + 7] = ZERO_LV2LINK;
 	}
 
-	pgtable_flush(exynos_domain->pgtable, exynos_domain->pgtable + NUM_LV1ENTRIES);
+	pgtable_flush(domain->pgtable, domain->pgtable + NUM_LV1ENTRIES);
 
-	spin_lock_init(&exynos_domain->lock);
-	spin_lock_init(&exynos_domain->pgtablelock);
-	INIT_LIST_HEAD(&exynos_domain->clients);
+	spin_lock_init(&domain->lock);
+	spin_lock_init(&domain->pgtablelock);
+	INIT_LIST_HEAD(&domain->clients);
 
-	exynos_domain->domain.geometry.aperture_start = 0;
-	exynos_domain->domain.geometry.aperture_end   = ~0UL;
-	exynos_domain->domain.geometry.force_aperture = true;
+	domain->domain.geometry.aperture_start = 0;
+	domain->domain.geometry.aperture_end   = ~0UL;
+	domain->domain.geometry.force_aperture = true;
 
-	return &exynos_domain->domain;
+	return &domain->domain;
 
 err_counter:
-	free_pages((unsigned long)exynos_domain->pgtable, 2);
+	free_pages((unsigned long)domain->pgtable, 2);
 err_pgtable:
-	kfree(exynos_domain);
+	kfree(domain);
 	return NULL;
 }
 
-static void exynos_iommu_domain_free(struct iommu_domain *domain)
+static void exynos_iommu_domain_free(struct iommu_domain *iommu_domain)
 {
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
-	struct exynos_iommu_owner *owner;
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
+	struct sysmmu_drvdata *data, *next;
 	unsigned long flags;
 	int i;
 
-	WARN_ON(!list_empty(&priv->clients));
+	WARN_ON(!list_empty(&domain->clients));
 
-	spin_lock_irqsave(&priv->lock, flags);
+	spin_lock_irqsave(&domain->lock, flags);
 
-	list_for_each_entry(owner, &priv->clients, client) {
-		while (!exynos_sysmmu_disable(owner->dev))
-			; /* until System MMU is actually disabled */
+	list_for_each_entry_safe(data, next, &domain->clients, domain_node) {
+		if (__sysmmu_disable(data))
+			data->master = NULL;
+		list_del_init(&data->domain_node);
 	}
 
-	while (!list_empty(&priv->clients))
-		list_del_init(priv->clients.next);
-
-	spin_unlock_irqrestore(&priv->lock, flags);
+	spin_unlock_irqrestore(&domain->lock, flags);
 
 	for (i = 0; i < NUM_LV1ENTRIES; i++)
-		if (lv1ent_page(priv->pgtable + i))
+		if (lv1ent_page(domain->pgtable + i))
 			kmem_cache_free(lv2table_kmem_cache,
-				phys_to_virt(lv2table_base(priv->pgtable + i)));
+				phys_to_virt(lv2table_base(domain->pgtable + i)));
 
-	free_pages((unsigned long)priv->pgtable, 2);
-	free_pages((unsigned long)priv->lv2entcnt, 1);
-	kfree(priv);
+	free_pages((unsigned long)domain->pgtable, 2);
+	free_pages((unsigned long)domain->lv2entcnt, 1);
+	kfree(domain);
 }
 
-static int exynos_iommu_attach_device(struct iommu_domain *domain,
+static int exynos_iommu_attach_device(struct iommu_domain *iommu_domain,
 				   struct device *dev)
 {
 	struct exynos_iommu_owner *owner = dev->archdata.iommu;
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
-	phys_addr_t pagetable = virt_to_phys(priv->pgtable);
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
+	struct sysmmu_drvdata *data;
+	phys_addr_t pagetable = virt_to_phys(domain->pgtable);
 	unsigned long flags;
-	int ret;
+	int ret = -ENODEV;
 
-	spin_lock_irqsave(&priv->lock, flags);
+	if (!has_sysmmu(dev))
+		return -ENODEV;
 
-	ret = __exynos_sysmmu_enable(dev, pagetable, domain);
-	if (ret == 0) {
-		list_add_tail(&owner->client, &priv->clients);
-		owner->domain = domain;
-	}
+	list_for_each_entry(data, &owner->controllers, owner_node) {
+		pm_runtime_get_sync(data->sysmmu);
+		ret = __sysmmu_enable(data, pagetable, domain);
+		if (ret >= 0) {
+			data->master = dev;
 
-	spin_unlock_irqrestore(&priv->lock, flags);
+			spin_lock_irqsave(&domain->lock, flags);
+			list_add_tail(&data->domain_node, &domain->clients);
+			spin_unlock_irqrestore(&domain->lock, flags);
+		}
+	}
 
 	if (ret < 0) {
 		dev_err(dev, "%s: Failed to attach IOMMU with pgtable %pa\n",
@@ -815,36 +774,39 @@ static int exynos_iommu_attach_device(struct iommu_domain *domain,
 	return ret;
 }
 
-static void exynos_iommu_detach_device(struct iommu_domain *domain,
+static void exynos_iommu_detach_device(struct iommu_domain *iommu_domain,
 				    struct device *dev)
 {
-	struct exynos_iommu_owner *owner;
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
-	phys_addr_t pagetable = virt_to_phys(priv->pgtable);
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
+	phys_addr_t pagetable = virt_to_phys(domain->pgtable);
+	struct sysmmu_drvdata *data, *next;
 	unsigned long flags;
+	bool found = false;
 
-	spin_lock_irqsave(&priv->lock, flags);
+	if (!has_sysmmu(dev))
+		return;
 
-	list_for_each_entry(owner, &priv->clients, client) {
-		if (owner == dev->archdata.iommu) {
-			if (exynos_sysmmu_disable(dev)) {
-				list_del_init(&owner->client);
-				owner->domain = NULL;
+	spin_lock_irqsave(&domain->lock, flags);
+	list_for_each_entry_safe(data, next, &domain->clients, domain_node) {
+		if (data->master == dev) {
+			if (__sysmmu_disable(data)) {
+				data->master = NULL;
+				list_del_init(&data->domain_node);
 			}
-			break;
+			pm_runtime_put(data->sysmmu);
+			found = true;
 		}
 	}
+	spin_unlock_irqrestore(&domain->lock, flags);
 
-	spin_unlock_irqrestore(&priv->lock, flags);
-
-	if (owner == dev->archdata.iommu)
+	if (found)
 		dev_dbg(dev, "%s: Detached IOMMU with pgtable %pa\n",
 					__func__, &pagetable);
 	else
 		dev_err(dev, "%s: No IOMMU is attached\n", __func__);
 }
 
-static sysmmu_pte_t *alloc_lv2entry(struct exynos_iommu_domain *priv,
+static sysmmu_pte_t *alloc_lv2entry(struct exynos_iommu_domain *domain,
 		sysmmu_pte_t *sent, sysmmu_iova_t iova, short *pgcounter)
 {
 	if (lv1ent_section(sent)) {
@@ -862,6 +824,7 @@ static sysmmu_pte_t *alloc_lv2entry(struct exynos_iommu_domain *priv,
 			return ERR_PTR(-ENOMEM);
 
 		*sent = mk_lv1ent_page(virt_to_phys(pent));
+		kmemleak_ignore(pent);
 		*pgcounter = NUM_LV2ENTRIES;
 		pgtable_flush(pent, pent + NUM_LV2ENTRIES);
 		pgtable_flush(sent, sent + 1);
@@ -884,20 +847,19 @@ static sysmmu_pte_t *alloc_lv2entry(struct exynos_iommu_domain *priv,
 		 * not currently mapped.
 		 */
 		if (need_flush_flpd_cache) {
-			struct exynos_iommu_owner *owner;
+			struct sysmmu_drvdata *data;
 
-			spin_lock(&priv->lock);
-			list_for_each_entry(owner, &priv->clients, client)
-				sysmmu_tlb_invalidate_flpdcache(
-							owner->dev, iova);
-			spin_unlock(&priv->lock);
+			spin_lock(&domain->lock);
+			list_for_each_entry(data, &domain->clients, domain_node)
+				sysmmu_tlb_invalidate_flpdcache(data, iova);
+			spin_unlock(&domain->lock);
 		}
 	}
 
 	return page_entry(sent, iova);
 }
 
-static int lv1set_section(struct exynos_iommu_domain *priv,
+static int lv1set_section(struct exynos_iommu_domain *domain,
 			  sysmmu_pte_t *sent, sysmmu_iova_t iova,
 			  phys_addr_t paddr, short *pgcnt)
 {
@@ -922,17 +884,17 @@ static int lv1set_section(struct exynos_iommu_domain *priv,
 
 	pgtable_flush(sent, sent + 1);
 
-	spin_lock(&priv->lock);
+	spin_lock(&domain->lock);
 	if (lv1ent_page_zero(sent)) {
-		struct exynos_iommu_owner *owner;
+		struct sysmmu_drvdata *data;
 		/*
 		 * Flushing FLPD cache in System MMU v3.3 that may cache a FLPD
 		 * entry by speculative prefetch of SLPD which has no mapping.
 		 */
-		list_for_each_entry(owner, &priv->clients, client)
-			sysmmu_tlb_invalidate_flpdcache(owner->dev, iova);
+		list_for_each_entry(data, &domain->clients, domain_node)
+			sysmmu_tlb_invalidate_flpdcache(data, iova);
 	}
-	spin_unlock(&priv->lock);
+	spin_unlock(&domain->lock);
 
 	return 0;
 }
@@ -992,74 +954,75 @@ static int lv2set_page(sysmmu_pte_t *pent, phys_addr_t paddr, size_t size,
  *   than or equal to 128KiB.
  * - Start address of an I/O virtual region must be aligned by 128KiB.
  */
-static int exynos_iommu_map(struct iommu_domain *domain, unsigned long l_iova,
-			 phys_addr_t paddr, size_t size, int prot)
+static int exynos_iommu_map(struct iommu_domain *iommu_domain,
+			    unsigned long l_iova, phys_addr_t paddr, size_t size,
+			    int prot)
 {
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
 	sysmmu_pte_t *entry;
 	sysmmu_iova_t iova = (sysmmu_iova_t)l_iova;
 	unsigned long flags;
 	int ret = -ENOMEM;
 
-	BUG_ON(priv->pgtable == NULL);
+	BUG_ON(domain->pgtable == NULL);
 
-	spin_lock_irqsave(&priv->pgtablelock, flags);
+	spin_lock_irqsave(&domain->pgtablelock, flags);
 
-	entry = section_entry(priv->pgtable, iova);
+	entry = section_entry(domain->pgtable, iova);
 
 	if (size == SECT_SIZE) {
-		ret = lv1set_section(priv, entry, iova, paddr,
-					&priv->lv2entcnt[lv1ent_offset(iova)]);
+		ret = lv1set_section(domain, entry, iova, paddr,
+				     &domain->lv2entcnt[lv1ent_offset(iova)]);
 	} else {
 		sysmmu_pte_t *pent;
 
-		pent = alloc_lv2entry(priv, entry, iova,
-					&priv->lv2entcnt[lv1ent_offset(iova)]);
+		pent = alloc_lv2entry(domain, entry, iova,
+				      &domain->lv2entcnt[lv1ent_offset(iova)]);
 
 		if (IS_ERR(pent))
 			ret = PTR_ERR(pent);
 		else
 			ret = lv2set_page(pent, paddr, size,
-					&priv->lv2entcnt[lv1ent_offset(iova)]);
+				       &domain->lv2entcnt[lv1ent_offset(iova)]);
 	}
 
 	if (ret)
 		pr_err("%s: Failed(%d) to map %#zx bytes @ %#x\n",
 			__func__, ret, size, iova);
 
-	spin_unlock_irqrestore(&priv->pgtablelock, flags);
+	spin_unlock_irqrestore(&domain->pgtablelock, flags);
 
 	return ret;
 }
 
-static void exynos_iommu_tlb_invalidate_entry(struct exynos_iommu_domain *priv,
-						sysmmu_iova_t iova, size_t size)
+static void exynos_iommu_tlb_invalidate_entry(struct exynos_iommu_domain *domain,
+					      sysmmu_iova_t iova, size_t size)
 {
-	struct exynos_iommu_owner *owner;
+	struct sysmmu_drvdata *data;
 	unsigned long flags;
 
-	spin_lock_irqsave(&priv->lock, flags);
+	spin_lock_irqsave(&domain->lock, flags);
 
-	list_for_each_entry(owner, &priv->clients, client)
-		sysmmu_tlb_invalidate_entry(owner->dev, iova, size);
+	list_for_each_entry(data, &domain->clients, domain_node)
+		sysmmu_tlb_invalidate_entry(data, iova, size);
 
-	spin_unlock_irqrestore(&priv->lock, flags);
+	spin_unlock_irqrestore(&domain->lock, flags);
 }
 
-static size_t exynos_iommu_unmap(struct iommu_domain *domain,
-					unsigned long l_iova, size_t size)
+static size_t exynos_iommu_unmap(struct iommu_domain *iommu_domain,
+				 unsigned long l_iova, size_t size)
 {
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
 	sysmmu_iova_t iova = (sysmmu_iova_t)l_iova;
 	sysmmu_pte_t *ent;
 	size_t err_pgsize;
 	unsigned long flags;
 
-	BUG_ON(priv->pgtable == NULL);
+	BUG_ON(domain->pgtable == NULL);
 
-	spin_lock_irqsave(&priv->pgtablelock, flags);
+	spin_lock_irqsave(&domain->pgtablelock, flags);
 
-	ent = section_entry(priv->pgtable, iova);
+	ent = section_entry(domain->pgtable, iova);
 
 	if (lv1ent_section(ent)) {
 		if (WARN_ON(size < SECT_SIZE)) {
@@ -1093,7 +1056,7 @@ static size_t exynos_iommu_unmap(struct iommu_domain *domain,
 		*ent = 0;
 		size = SPAGE_SIZE;
 		pgtable_flush(ent, ent + 1);
-		priv->lv2entcnt[lv1ent_offset(iova)] += 1;
+		domain->lv2entcnt[lv1ent_offset(iova)] += 1;
 		goto done;
 	}
 
@@ -1107,15 +1070,15 @@ static size_t exynos_iommu_unmap(struct iommu_domain *domain,
 	pgtable_flush(ent, ent + SPAGES_PER_LPAGE);
 
 	size = LPAGE_SIZE;
-	priv->lv2entcnt[lv1ent_offset(iova)] += SPAGES_PER_LPAGE;
+	domain->lv2entcnt[lv1ent_offset(iova)] += SPAGES_PER_LPAGE;
 done:
-	spin_unlock_irqrestore(&priv->pgtablelock, flags);
+	spin_unlock_irqrestore(&domain->pgtablelock, flags);
 
-	exynos_iommu_tlb_invalidate_entry(priv, iova, size);
+	exynos_iommu_tlb_invalidate_entry(domain, iova, size);
 
 	return size;
 err:
-	spin_unlock_irqrestore(&priv->pgtablelock, flags);
+	spin_unlock_irqrestore(&domain->pgtablelock, flags);
 
 	pr_err("%s: Failed: size(%#zx) @ %#x is smaller than page size %#zx\n",
 		__func__, size, iova, err_pgsize);
@@ -1123,17 +1086,17 @@ err:
 	return 0;
 }
 
-static phys_addr_t exynos_iommu_iova_to_phys(struct iommu_domain *domain,
+static phys_addr_t exynos_iommu_iova_to_phys(struct iommu_domain *iommu_domain,
 					  dma_addr_t iova)
 {
-	struct exynos_iommu_domain *priv = to_exynos_domain(domain);
+	struct exynos_iommu_domain *domain = to_exynos_domain(iommu_domain);
 	sysmmu_pte_t *entry;
 	unsigned long flags;
 	phys_addr_t phys = 0;
 
-	spin_lock_irqsave(&priv->pgtablelock, flags);
+	spin_lock_irqsave(&domain->pgtablelock, flags);
 
-	entry = section_entry(priv->pgtable, iova);
+	entry = section_entry(domain->pgtable, iova);
 
 	if (lv1ent_section(entry)) {
 		phys = section_phys(entry) + section_offs(iova);
@@ -1146,7 +1109,7 @@ static phys_addr_t exynos_iommu_iova_to_phys(struct iommu_domain *domain,
 			phys = spage_phys(entry) + spage_offs(iova);
 	}
 
-	spin_unlock_irqrestore(&priv->pgtablelock, flags);
+	spin_unlock_irqrestore(&domain->pgtablelock, flags);
 
 	return phys;
 }
@@ -1156,6 +1119,9 @@ static int exynos_iommu_add_device(struct device *dev)
 	struct iommu_group *group;
 	int ret;
 
+	if (!has_sysmmu(dev))
+		return -ENODEV;
+
 	group = iommu_group_get(dev);
 
 	if (!group) {
@@ -1174,10 +1140,40 @@ static int exynos_iommu_add_device(struct device *dev)
 
 static void exynos_iommu_remove_device(struct device *dev)
 {
+	if (!has_sysmmu(dev))
+		return;
+
 	iommu_group_remove_device(dev);
 }
 
-static const struct iommu_ops exynos_iommu_ops = {
+static int exynos_iommu_of_xlate(struct device *dev,
+				 struct of_phandle_args *spec)
+{
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
+	struct platform_device *sysmmu = of_find_device_by_node(spec->np);
+	struct sysmmu_drvdata *data;
+
+	if (!sysmmu)
+		return -ENODEV;
+
+	data = platform_get_drvdata(sysmmu);
+	if (!data)
+		return -ENODEV;
+
+	if (!owner) {
+		owner = kzalloc(sizeof(*owner), GFP_KERNEL);
+		if (!owner)
+			return -ENOMEM;
+
+		INIT_LIST_HEAD(&owner->controllers);
+		dev->archdata.iommu = owner;
+	}
+
+	list_add_tail(&data->owner_node, &owner->controllers);
+	return 0;
+}
+
+static struct iommu_ops exynos_iommu_ops = {
 	.domain_alloc = exynos_iommu_domain_alloc,
 	.domain_free = exynos_iommu_domain_free,
 	.attach_dev = exynos_iommu_attach_device,
@@ -1189,19 +1185,15 @@ static const struct iommu_ops exynos_iommu_ops = {
 	.add_device = exynos_iommu_add_device,
 	.remove_device = exynos_iommu_remove_device,
 	.pgsize_bitmap = SECT_SIZE | LPAGE_SIZE | SPAGE_SIZE,
+	.of_xlate = exynos_iommu_of_xlate,
 };
 
+static bool init_done;
+
 static int __init exynos_iommu_init(void)
 {
-	struct device_node *np;
 	int ret;
 
-	np = of_find_matching_node(NULL, sysmmu_of_match);
-	if (!np)
-		return 0;
-
-	of_node_put(np);
-
 	lv2table_kmem_cache = kmem_cache_create("exynos-iommu-lv2table",
 				LV2TABLE_SIZE, LV2TABLE_SIZE, 0, NULL);
 	if (!lv2table_kmem_cache) {
@@ -1230,6 +1222,8 @@ static int __init exynos_iommu_init(void)
 		goto err_set_iommu;
 	}
 
+	init_done = true;
+
 	return 0;
 err_set_iommu:
 	kmem_cache_free(lv2table_kmem_cache, zero_lv2_table);
@@ -1239,4 +1233,21 @@ err_reg_driver:
 	kmem_cache_destroy(lv2table_kmem_cache);
 	return ret;
 }
-subsys_initcall(exynos_iommu_init);
+
+static int __init exynos_iommu_of_setup(struct device_node *np)
+{
+	struct platform_device *pdev;
+
+	if (!init_done)
+		exynos_iommu_init();
+
+	pdev = of_platform_device_create(np, NULL, platform_bus_type.dev_root);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	of_iommu_set_ops(np, &exynos_iommu_ops);
+	return 0;
+}
+
+IOMMU_OF_DECLARE(exynos_iommu_of, "samsung,exynos-sysmmu",
+		 exynos_iommu_of_setup);
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index c87c4b1bf..0649b94f5 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -15,8 +15,11 @@
  *          Shaohua Li <shaohua.li@intel.com>,
  *          Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>,
  *          Fenghua Yu <fenghua.yu@intel.com>
+ *          Joerg Roedel <jroedel@suse.de>
  */
 
+#define pr_fmt(fmt)     "DMAR: " fmt
+
 #include <linux/init.h>
 #include <linux/bitmap.h>
 #include <linux/debugfs.h>
@@ -40,6 +43,7 @@
 #include <linux/pci-ats.h>
 #include <linux/memblock.h>
 #include <linux/dma-contiguous.h>
+#include <linux/crash_dump.h>
 #include <asm/irq_remapping.h>
 #include <asm/cacheflush.h>
 #include <asm/iommu.h>
@@ -190,7 +194,29 @@ struct root_entry {
 };
 #define ROOT_ENTRY_NR (VTD_PAGE_SIZE/sizeof(struct root_entry))
 
+/*
+ * Take a root_entry and return the Lower Context Table Pointer (LCTP)
+ * if marked present.
+ */
+static phys_addr_t root_entry_lctp(struct root_entry *re)
+{
+	if (!(re->lo & 1))
+		return 0;
+
+	return re->lo & VTD_PAGE_MASK;
+}
+
+/*
+ * Take a root_entry and return the Upper Context Table Pointer (UCTP)
+ * if marked present.
+ */
+static phys_addr_t root_entry_uctp(struct root_entry *re)
+{
+	if (!(re->hi & 1))
+		return 0;
 
+	return re->hi & VTD_PAGE_MASK;
+}
 /*
  * low 64 bits:
  * 0: present
@@ -207,10 +233,38 @@ struct context_entry {
 	u64 hi;
 };
 
-static inline bool context_present(struct context_entry *context)
+static inline void context_clear_pasid_enable(struct context_entry *context)
+{
+	context->lo &= ~(1ULL << 11);
+}
+
+static inline bool context_pasid_enabled(struct context_entry *context)
+{
+	return !!(context->lo & (1ULL << 11));
+}
+
+static inline void context_set_copied(struct context_entry *context)
+{
+	context->hi |= (1ull << 3);
+}
+
+static inline bool context_copied(struct context_entry *context)
+{
+	return !!(context->hi & (1ULL << 3));
+}
+
+static inline bool __context_present(struct context_entry *context)
 {
 	return (context->lo & 1);
 }
+
+static inline bool context_present(struct context_entry *context)
+{
+	return context_pasid_enabled(context) ?
+	     __context_present(context) :
+	     __context_present(context) && !context_copied(context);
+}
+
 static inline void context_set_present(struct context_entry *context)
 {
 	context->lo |= 1;
@@ -247,6 +301,11 @@ static inline void context_set_domain_id(struct context_entry *context,
 	context->hi |= (value & ((1 << 16) - 1)) << 8;
 }
 
+static inline int context_domain_id(struct context_entry *c)
+{
+	return((c->hi >> 8) & 0xffff);
+}
+
 static inline void context_clear_entry(struct context_entry *context)
 {
 	context->lo = 0;
@@ -440,6 +499,25 @@ static LIST_HEAD(device_domain_list);
 
 static const struct iommu_ops intel_iommu_ops;
 
+static bool translation_pre_enabled(struct intel_iommu *iommu)
+{
+	return (iommu->flags & VTD_FLAG_TRANS_PRE_ENABLED);
+}
+
+static void clear_translation_pre_enabled(struct intel_iommu *iommu)
+{
+	iommu->flags &= ~VTD_FLAG_TRANS_PRE_ENABLED;
+}
+
+static void init_translation_status(struct intel_iommu *iommu)
+{
+	u32 gsts;
+
+	gsts = readl(iommu->reg + DMAR_GSTS_REG);
+	if (gsts & DMA_GSTS_TES)
+		iommu->flags |= VTD_FLAG_TRANS_PRE_ENABLED;
+}
+
 /* Convert generic 'struct iommu_domain to private struct dmar_domain */
 static struct dmar_domain *to_dmar_domain(struct iommu_domain *dom)
 {
@@ -453,25 +531,21 @@ static int __init intel_iommu_setup(char *str)
 	while (*str) {
 		if (!strncmp(str, "on", 2)) {
 			dmar_disabled = 0;
-			printk(KERN_INFO "Intel-IOMMU: enabled\n");
+			pr_info("IOMMU enabled\n");
 		} else if (!strncmp(str, "off", 3)) {
 			dmar_disabled = 1;
-			printk(KERN_INFO "Intel-IOMMU: disabled\n");
+			pr_info("IOMMU disabled\n");
 		} else if (!strncmp(str, "igfx_off", 8)) {
 			dmar_map_gfx = 0;
-			printk(KERN_INFO
-				"Intel-IOMMU: disable GFX device mapping\n");
+			pr_info("Disable GFX device mapping\n");
 		} else if (!strncmp(str, "forcedac", 8)) {
-			printk(KERN_INFO
-				"Intel-IOMMU: Forcing DAC for PCI devices\n");
+			pr_info("Forcing DAC for PCI devices\n");
 			dmar_forcedac = 1;
 		} else if (!strncmp(str, "strict", 6)) {
-			printk(KERN_INFO
-				"Intel-IOMMU: disable batched IOTLB flush\n");
+			pr_info("Disable batched IOTLB flush\n");
 			intel_iommu_strict = 1;
 		} else if (!strncmp(str, "sp_off", 6)) {
-			printk(KERN_INFO
-				"Intel-IOMMU: disable supported super page\n");
+			pr_info("Disable supported super page\n");
 			intel_iommu_superpage = 0;
 		} else if (!strncmp(str, "ecs_off", 7)) {
 			printk(KERN_INFO
@@ -1132,7 +1206,7 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 
 	root = (struct root_entry *)alloc_pgtable_page(iommu->node);
 	if (!root) {
-		pr_err("IOMMU: allocating root entry for %s failed\n",
+		pr_err("Allocating root entry for %s failed\n",
 			iommu->name);
 		return -ENOMEM;
 	}
@@ -1270,9 +1344,9 @@ static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
 
 	/* check IOTLB invalidation granularity */
 	if (DMA_TLB_IAIG(val) == 0)
-		printk(KERN_ERR"IOMMU: flush IOTLB failed\n");
+		pr_err("Flush IOTLB failed\n");
 	if (DMA_TLB_IAIG(val) != DMA_TLB_IIRG(type))
-		pr_debug("IOMMU: tlb flush request %Lx, actual %Lx\n",
+		pr_debug("TLB flush request %Lx, actual %Lx\n",
 			(unsigned long long)DMA_TLB_IIRG(type),
 			(unsigned long long)DMA_TLB_IAIG(val));
 }
@@ -1443,8 +1517,8 @@ static int iommu_init_domains(struct intel_iommu *iommu)
 	unsigned long nlongs;
 
 	ndomains = cap_ndoms(iommu->cap);
-	pr_debug("IOMMU%d: Number of Domains supported <%ld>\n",
-		 iommu->seq_id, ndomains);
+	pr_debug("%s: Number of Domains supported <%ld>\n",
+		 iommu->name, ndomains);
 	nlongs = BITS_TO_LONGS(ndomains);
 
 	spin_lock_init(&iommu->lock);
@@ -1454,15 +1528,15 @@ static int iommu_init_domains(struct intel_iommu *iommu)
 	 */
 	iommu->domain_ids = kcalloc(nlongs, sizeof(unsigned long), GFP_KERNEL);
 	if (!iommu->domain_ids) {
-		pr_err("IOMMU%d: allocating domain id array failed\n",
-		       iommu->seq_id);
+		pr_err("%s: Allocating domain id array failed\n",
+		       iommu->name);
 		return -ENOMEM;
 	}
 	iommu->domains = kcalloc(ndomains, sizeof(struct dmar_domain *),
 			GFP_KERNEL);
 	if (!iommu->domains) {
-		pr_err("IOMMU%d: allocating domain array failed\n",
-		       iommu->seq_id);
+		pr_err("%s: Allocating domain array failed\n",
+		       iommu->name);
 		kfree(iommu->domain_ids);
 		iommu->domain_ids = NULL;
 		return -ENOMEM;
@@ -1567,7 +1641,7 @@ static int iommu_attach_domain(struct dmar_domain *domain,
 	num = __iommu_attach_domain(domain, iommu);
 	spin_unlock_irqrestore(&iommu->lock, flags);
 	if (num < 0)
-		pr_err("IOMMU: no free domain ids\n");
+		pr_err("%s: No free domain ids\n", iommu->name);
 
 	return num;
 }
@@ -1659,7 +1733,7 @@ static int dmar_init_reserved_ranges(void)
 	iova = reserve_iova(&reserved_iova_list, IOVA_PFN(IOAPIC_RANGE_START),
 		IOVA_PFN(IOAPIC_RANGE_END));
 	if (!iova) {
-		printk(KERN_ERR "Reserve IOAPIC range failed\n");
+		pr_err("Reserve IOAPIC range failed\n");
 		return -ENODEV;
 	}
 
@@ -1675,7 +1749,7 @@ static int dmar_init_reserved_ranges(void)
 					    IOVA_PFN(r->start),
 					    IOVA_PFN(r->end));
 			if (!iova) {
-				printk(KERN_ERR "Reserve iova failed\n");
+				pr_err("Reserve iova failed\n");
 				return -ENODEV;
 			}
 		}
@@ -1722,7 +1796,7 @@ static int domain_init(struct dmar_domain *domain, int guest_width)
 	sagaw = cap_sagaw(iommu->cap);
 	if (!test_bit(agaw, &sagaw)) {
 		/* hardware doesn't support it, choose a bigger one */
-		pr_debug("IOMMU: hardware doesn't support agaw %d\n", agaw);
+		pr_debug("Hardware doesn't support agaw %d\n", agaw);
 		agaw = find_next_bit(&sagaw, 5, agaw);
 		if (agaw >= 5)
 			return -ENODEV;
@@ -1818,6 +1892,8 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 		return 0;
 	}
 
+	context_clear_entry(context);
+
 	id = domain->id;
 	pgd = domain->pgd;
 
@@ -1826,7 +1902,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 			id = iommu_attach_vm_domain(domain, iommu);
 			if (id < 0) {
 				spin_unlock_irqrestore(&iommu->lock, flags);
-				pr_err("IOMMU: no free domain ids\n");
+				pr_err("%s: No free domain ids\n", iommu->name);
 				return -EFAULT;
 			}
 		}
@@ -2053,8 +2129,8 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 		tmp = cmpxchg64_local(&pte->val, 0ULL, pteval);
 		if (tmp) {
 			static int dumps = 5;
-			printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx not %llx)\n",
-			       iov_pfn, tmp, (unsigned long long)pteval);
+			pr_crit("ERROR: DMA PTE for vPFN 0x%lx already set (to %llx not %llx)\n",
+				iov_pfn, tmp, (unsigned long long)pteval);
 			if (dumps) {
 				dumps--;
 				debug_dma_dump_mappings(NULL);
@@ -2326,7 +2402,7 @@ static int iommu_domain_identity_map(struct dmar_domain *domain,
 
 	if (!reserve_iova(&domain->iovad, dma_to_mm_pfn(first_vpfn),
 			  dma_to_mm_pfn(last_vpfn))) {
-		printk(KERN_ERR "IOMMU: reserve iova failed\n");
+		pr_err("Reserving iova failed\n");
 		return -ENOMEM;
 	}
 
@@ -2359,15 +2435,14 @@ static int iommu_prepare_identity_map(struct device *dev,
 	   range which is reserved in E820, so which didn't get set
 	   up to start with in si_domain */
 	if (domain == si_domain && hw_pass_through) {
-		printk("Ignoring identity map for HW passthrough device %s [0x%Lx - 0x%Lx]\n",
-		       dev_name(dev), start, end);
+		pr_warn("Ignoring identity map for HW passthrough device %s [0x%Lx - 0x%Lx]\n",
+			dev_name(dev), start, end);
 		return 0;
 	}
 
-	printk(KERN_INFO
-	       "IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
-	       dev_name(dev), start, end);
-	
+	pr_info("Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
+		dev_name(dev), start, end);
+
 	if (end < start) {
 		WARN(1, "Your BIOS is broken; RMRR ends before it starts!\n"
 			"BIOS vendor: %s; Ver: %s; Product Version: %s\n",
@@ -2424,12 +2499,11 @@ static inline void iommu_prepare_isa(void)
 	if (!pdev)
 		return;
 
-	printk(KERN_INFO "IOMMU: Prepare 0-16MiB unity mapping for LPC\n");
+	pr_info("Prepare 0-16MiB unity mapping for LPC\n");
 	ret = iommu_prepare_identity_map(&pdev->dev, 0, 16*1024*1024 - 1);
 
 	if (ret)
-		printk(KERN_ERR "IOMMU: Failed to create 0-16MiB identity map; "
-		       "floppy might not work\n");
+		pr_err("Failed to create 0-16MiB identity map - floppy might not work\n");
 
 	pci_dev_put(pdev);
 }
@@ -2473,7 +2547,7 @@ static int __init si_domain_init(int hw)
 		return -EFAULT;
 	}
 
-	pr_debug("IOMMU: identity mapping domain is domain %d\n",
+	pr_debug("Identity mapping domain is domain %d\n",
 		 si_domain->id);
 
 	if (hw)
@@ -2673,8 +2747,8 @@ static int __init dev_prepare_static_identity_mapping(struct device *dev, int hw
 				  hw ? CONTEXT_TT_PASS_THROUGH :
 				       CONTEXT_TT_MULTI_LEVEL);
 	if (!ret)
-		pr_info("IOMMU: %s identity mapping for device %s\n",
-			hw ? "hardware" : "software", dev_name(dev));
+		pr_info("%s identity mapping for device %s\n",
+			hw ? "Hardware" : "Software", dev_name(dev));
 	else if (ret == -ENODEV)
 		/* device not associated with an iommu */
 		ret = 0;
@@ -2692,10 +2766,6 @@ static int __init iommu_prepare_static_identity_mapping(int hw)
 	int i;
 	int ret = 0;
 
-	ret = si_domain_init(hw);
-	if (ret)
-		return -EFAULT;
-
 	for_each_pci_dev(pdev) {
 		ret = dev_prepare_static_identity_mapping(&pdev->dev, hw);
 		if (ret)
@@ -2709,7 +2779,7 @@ static int __init iommu_prepare_static_identity_mapping(int hw)
 
 			if (dev->bus != &acpi_bus_type)
 				continue;
-				
+
 			adev= to_acpi_device(dev);
 			mutex_lock(&adev->physical_node_lock);
 			list_for_each_entry(pn, &adev->physical_node_list, node) {
@@ -2751,19 +2821,200 @@ static void intel_iommu_init_qi(struct intel_iommu *iommu)
 		 */
 		iommu->flush.flush_context = __iommu_flush_context;
 		iommu->flush.flush_iotlb = __iommu_flush_iotlb;
-		pr_info("IOMMU: %s using Register based invalidation\n",
+		pr_info("%s: Using Register based invalidation\n",
 			iommu->name);
 	} else {
 		iommu->flush.flush_context = qi_flush_context;
 		iommu->flush.flush_iotlb = qi_flush_iotlb;
-		pr_info("IOMMU: %s using Queued invalidation\n", iommu->name);
+		pr_info("%s: Using Queued invalidation\n", iommu->name);
+	}
+}
+
+static int copy_context_table(struct intel_iommu *iommu,
+			      struct root_entry *old_re,
+			      struct context_entry **tbl,
+			      int bus, bool ext)
+{
+	struct context_entry *old_ce = NULL, *new_ce = NULL, ce;
+	int tbl_idx, pos = 0, idx, devfn, ret = 0, did;
+	phys_addr_t old_ce_phys;
+
+	tbl_idx = ext ? bus * 2 : bus;
+
+	for (devfn = 0; devfn < 256; devfn++) {
+		/* First calculate the correct index */
+		idx = (ext ? devfn * 2 : devfn) % 256;
+
+		if (idx == 0) {
+			/* First save what we may have and clean up */
+			if (new_ce) {
+				tbl[tbl_idx] = new_ce;
+				__iommu_flush_cache(iommu, new_ce,
+						    VTD_PAGE_SIZE);
+				pos = 1;
+			}
+
+			if (old_ce)
+				iounmap(old_ce);
+
+			ret = 0;
+			if (devfn < 0x80)
+				old_ce_phys = root_entry_lctp(old_re);
+			else
+				old_ce_phys = root_entry_uctp(old_re);
+
+			if (!old_ce_phys) {
+				if (ext && devfn == 0) {
+					/* No LCTP, try UCTP */
+					devfn = 0x7f;
+					continue;
+				} else {
+					goto out;
+				}
+			}
+
+			ret = -ENOMEM;
+			old_ce = ioremap_cache(old_ce_phys, PAGE_SIZE);
+			if (!old_ce)
+				goto out;
+
+			new_ce = alloc_pgtable_page(iommu->node);
+			if (!new_ce)
+				goto out_unmap;
+
+			ret = 0;
+		}
+
+		/* Now copy the context entry */
+		ce = old_ce[idx];
+
+		if (!__context_present(&ce))
+			continue;
+
+		did = context_domain_id(&ce);
+		if (did >= 0 && did < cap_ndoms(iommu->cap))
+			set_bit(did, iommu->domain_ids);
+
+		/*
+		 * We need a marker for copied context entries. This
+		 * marker needs to work for the old format as well as
+		 * for extended context entries.
+		 *
+		 * Bit 67 of the context entry is used. In the old
+		 * format this bit is available to software, in the
+		 * extended format it is the PGE bit, but PGE is ignored
+		 * by HW if PASIDs are disabled (and thus still
+		 * available).
+		 *
+		 * So disable PASIDs first and then mark the entry
+		 * copied. This means that we don't copy PASID
+		 * translations from the old kernel, but this is fine as
+		 * faults there are not fatal.
+		 */
+		context_clear_pasid_enable(&ce);
+		context_set_copied(&ce);
+
+		new_ce[idx] = ce;
+	}
+
+	tbl[tbl_idx + pos] = new_ce;
+
+	__iommu_flush_cache(iommu, new_ce, VTD_PAGE_SIZE);
+
+out_unmap:
+	iounmap(old_ce);
+
+out:
+	return ret;
+}
+
+static int copy_translation_tables(struct intel_iommu *iommu)
+{
+	struct context_entry **ctxt_tbls;
+	struct root_entry *old_rt;
+	phys_addr_t old_rt_phys;
+	int ctxt_table_entries;
+	unsigned long flags;
+	u64 rtaddr_reg;
+	int bus, ret;
+	bool new_ext, ext;
+
+	rtaddr_reg = dmar_readq(iommu->reg + DMAR_RTADDR_REG);
+	ext        = !!(rtaddr_reg & DMA_RTADDR_RTT);
+	new_ext    = !!ecap_ecs(iommu->ecap);
+
+	/*
+	 * The RTT bit can only be changed when translation is disabled,
+	 * but disabling translation means to open a window for data
+	 * corruption. So bail out and don't copy anything if we would
+	 * have to change the bit.
+	 */
+	if (new_ext != ext)
+		return -EINVAL;
+
+	old_rt_phys = rtaddr_reg & VTD_PAGE_MASK;
+	if (!old_rt_phys)
+		return -EINVAL;
+
+	old_rt = ioremap_cache(old_rt_phys, PAGE_SIZE);
+	if (!old_rt)
+		return -ENOMEM;
+
+	/* This is too big for the stack - allocate it from slab */
+	ctxt_table_entries = ext ? 512 : 256;
+	ret = -ENOMEM;
+	ctxt_tbls = kzalloc(ctxt_table_entries * sizeof(void *), GFP_KERNEL);
+	if (!ctxt_tbls)
+		goto out_unmap;
+
+	for (bus = 0; bus < 256; bus++) {
+		ret = copy_context_table(iommu, &old_rt[bus],
+					 ctxt_tbls, bus, ext);
+		if (ret) {
+			pr_err("%s: Failed to copy context table for bus %d\n",
+				iommu->name, bus);
+			continue;
+		}
 	}
+
+	spin_lock_irqsave(&iommu->lock, flags);
+
+	/* Context tables are copied, now write them to the root_entry table */
+	for (bus = 0; bus < 256; bus++) {
+		int idx = ext ? bus * 2 : bus;
+		u64 val;
+
+		if (ctxt_tbls[idx]) {
+			val = virt_to_phys(ctxt_tbls[idx]) | 1;
+			iommu->root_entry[bus].lo = val;
+		}
+
+		if (!ext || !ctxt_tbls[idx + 1])
+			continue;
+
+		val = virt_to_phys(ctxt_tbls[idx + 1]) | 1;
+		iommu->root_entry[bus].hi = val;
+	}
+
+	spin_unlock_irqrestore(&iommu->lock, flags);
+
+	kfree(ctxt_tbls);
+
+	__iommu_flush_cache(iommu, iommu->root_entry, PAGE_SIZE);
+
+	ret = 0;
+
+out_unmap:
+	iounmap(old_rt);
+
+	return ret;
 }
 
 static int __init init_dmars(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct dmar_rmrr_unit *rmrr;
+	bool copied_tables = false;
 	struct device *dev;
 	struct intel_iommu *iommu;
 	int i, ret;
@@ -2784,8 +3035,7 @@ static int __init init_dmars(void)
 			g_num_of_iommus++;
 			continue;
 		}
-		printk_once(KERN_ERR "intel-iommu: exceeded %d IOMMUs\n",
-			  DMAR_UNITS_SUPPORTED);
+		pr_err_once("Exceeded %d IOMMUs\n", DMAR_UNITS_SUPPORTED);
 	}
 
 	/* Preallocate enough resources for IOMMU hot-addition */
@@ -2795,7 +3045,7 @@ static int __init init_dmars(void)
 	g_iommus = kcalloc(g_num_of_iommus, sizeof(struct intel_iommu *),
 			GFP_KERNEL);
 	if (!g_iommus) {
-		printk(KERN_ERR "Allocating global iommu array failed\n");
+		pr_err("Allocating global iommu array failed\n");
 		ret = -ENOMEM;
 		goto error;
 	}
@@ -2810,10 +3060,21 @@ static int __init init_dmars(void)
 	for_each_active_iommu(iommu, drhd) {
 		g_iommus[iommu->seq_id] = iommu;
 
+		intel_iommu_init_qi(iommu);
+
 		ret = iommu_init_domains(iommu);
 		if (ret)
 			goto free_iommu;
 
+		init_translation_status(iommu);
+
+		if (translation_pre_enabled(iommu) && !is_kdump_kernel()) {
+			iommu_disable_translation(iommu);
+			clear_translation_pre_enabled(iommu);
+			pr_warn("Translation was enabled for %s but we are not in kdump mode\n",
+				iommu->name);
+		}
+
 		/*
 		 * TBD:
 		 * we could share the same root & context tables
@@ -2822,13 +3083,41 @@ static int __init init_dmars(void)
 		ret = iommu_alloc_root_entry(iommu);
 		if (ret)
 			goto free_iommu;
+
+		if (translation_pre_enabled(iommu)) {
+			pr_info("Translation already enabled - trying to copy translation structures\n");
+
+			ret = copy_translation_tables(iommu);
+			if (ret) {
+				/*
+				 * We found the IOMMU with translation
+				 * enabled - but failed to copy over the
+				 * old root-entry table. Try to proceed
+				 * by disabling translation now and
+				 * allocating a clean root-entry table.
+				 * This might cause DMAR faults, but
+				 * probably the dump will still succeed.
+				 */
+				pr_err("Failed to copy translation tables from previous kernel for %s\n",
+				       iommu->name);
+				iommu_disable_translation(iommu);
+				clear_translation_pre_enabled(iommu);
+			} else {
+				pr_info("Copied translation tables from previous kernel for %s\n",
+					iommu->name);
+				copied_tables = true;
+			}
+		}
+
+		iommu_flush_write_buffer(iommu);
+		iommu_set_root_entry(iommu);
+		iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL);
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH);
+
 		if (!ecap_pass_through(iommu->ecap))
 			hw_pass_through = 0;
 	}
 
-	for_each_active_iommu(iommu, drhd)
-		intel_iommu_init_qi(iommu);
-
 	if (iommu_pass_through)
 		iommu_identity_mapping |= IDENTMAP_ALL;
 
@@ -2836,9 +3125,24 @@ static int __init init_dmars(void)
 	iommu_identity_mapping |= IDENTMAP_GFX;
 #endif
 
+	if (iommu_identity_mapping) {
+		ret = si_domain_init(hw_pass_through);
+		if (ret)
+			goto free_iommu;
+	}
+
 	check_tylersburg_isoch();
 
 	/*
+	 * If we copied translations from a previous kernel in the kdump
+	 * case, we can not assign the devices to domains now, as that
+	 * would eliminate the old mappings. So skip this part and defer
+	 * the assignment to device driver initialization time.
+	 */
+	if (copied_tables)
+		goto domains_done;
+
+	/*
 	 * If pass through is not set or not enabled, setup context entries for
 	 * identity mappings for rmrr, gfx, and isa and may fall back to static
 	 * identity mapping if iommu_identity_mapping is set.
@@ -2846,7 +3150,7 @@ static int __init init_dmars(void)
 	if (iommu_identity_mapping) {
 		ret = iommu_prepare_static_identity_mapping(hw_pass_through);
 		if (ret) {
-			printk(KERN_CRIT "Failed to setup IOMMU pass-through\n");
+			pr_crit("Failed to setup IOMMU pass-through\n");
 			goto free_iommu;
 		}
 	}
@@ -2864,20 +3168,21 @@ static int __init init_dmars(void)
 	 *    endfor
 	 * endfor
 	 */
-	printk(KERN_INFO "IOMMU: Setting RMRR:\n");
+	pr_info("Setting RMRR:\n");
 	for_each_rmrr_units(rmrr) {
 		/* some BIOS lists non-exist devices in DMAR table. */
 		for_each_active_dev_scope(rmrr->devices, rmrr->devices_cnt,
 					  i, dev) {
 			ret = iommu_prepare_rmrr_dev(rmrr, dev);
 			if (ret)
-				printk(KERN_ERR
-				       "IOMMU: mapping reserved region failed\n");
+				pr_err("Mapping reserved region failed\n");
 		}
 	}
 
 	iommu_prepare_isa();
 
+domains_done:
+
 	/*
 	 * for each drhd
 	 *   enable fault log
@@ -2902,11 +3207,9 @@ static int __init init_dmars(void)
 		if (ret)
 			goto free_iommu;
 
-		iommu_set_root_entry(iommu);
+		if (!translation_pre_enabled(iommu))
+			iommu_enable_translation(iommu);
 
-		iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL);
-		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH);
-		iommu_enable_translation(iommu);
 		iommu_disable_protect_mem_regions(iommu);
 	}
 
@@ -2947,7 +3250,7 @@ static struct iova *intel_alloc_iova(struct device *dev,
 	}
 	iova = alloc_iova(&domain->iovad, nrpages, IOVA_PFN(dma_mask), 1);
 	if (unlikely(!iova)) {
-		printk(KERN_ERR "Allocating %ld-page iova for %s failed",
+		pr_err("Allocating %ld-page iova for %s failed",
 		       nrpages, dev_name(dev));
 		return NULL;
 	}
@@ -2962,7 +3265,7 @@ static struct dmar_domain *__get_valid_domain_for_dev(struct device *dev)
 
 	domain = get_domain_for_dev(dev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
 	if (!domain) {
-		printk(KERN_ERR "Allocating domain for %s failed",
+		pr_err("Allocating domain for %s failed\n",
 		       dev_name(dev));
 		return NULL;
 	}
@@ -2971,7 +3274,7 @@ static struct dmar_domain *__get_valid_domain_for_dev(struct device *dev)
 	if (unlikely(!domain_context_mapped(dev))) {
 		ret = domain_context_mapping(domain, dev, CONTEXT_TT_MULTI_LEVEL);
 		if (ret) {
-			printk(KERN_ERR "Domain context map for %s failed",
+			pr_err("Domain context map for %s failed\n",
 			       dev_name(dev));
 			return NULL;
 		}
@@ -3013,8 +3316,8 @@ static int iommu_no_mapping(struct device *dev)
 			 * to non-identity mapping.
 			 */
 			domain_remove_one_dev_info(si_domain, dev);
-			printk(KERN_INFO "32bit %s uses non-identity mapping\n",
-			       dev_name(dev));
+			pr_info("32bit %s uses non-identity mapping\n",
+				dev_name(dev));
 			return 0;
 		}
 	} else {
@@ -3029,8 +3332,8 @@ static int iommu_no_mapping(struct device *dev)
 						  CONTEXT_TT_PASS_THROUGH :
 						  CONTEXT_TT_MULTI_LEVEL);
 			if (!ret) {
-				printk(KERN_INFO "64bit %s uses identity mapping\n",
-				       dev_name(dev));
+				pr_info("64bit %s uses identity mapping\n",
+					dev_name(dev));
 				return 1;
 			}
 		}
@@ -3099,7 +3402,7 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr,
 error:
 	if (iova)
 		__free_iova(&domain->iovad, iova);
-	printk(KERN_ERR"Device %s request: %zx@%llx dir %d --- failed\n",
+	pr_err("Device %s request: %zx@%llx dir %d --- failed\n",
 		dev_name(dev), size, (unsigned long long)paddr, dir);
 	return 0;
 }
@@ -3414,7 +3717,7 @@ static inline int iommu_domain_cache_init(void)
 
 					 NULL);
 	if (!iommu_domain_cache) {
-		printk(KERN_ERR "Couldn't create iommu_domain cache\n");
+		pr_err("Couldn't create iommu_domain cache\n");
 		ret = -ENOMEM;
 	}
 
@@ -3431,7 +3734,7 @@ static inline int iommu_devinfo_cache_init(void)
 					 SLAB_HWCACHE_ALIGN,
 					 NULL);
 	if (!iommu_devinfo_cache) {
-		printk(KERN_ERR "Couldn't create devinfo cache\n");
+		pr_err("Couldn't create devinfo cache\n");
 		ret = -ENOMEM;
 	}
 
@@ -3808,19 +4111,19 @@ static int intel_iommu_add(struct dmar_drhd_unit *dmaru)
 		return 0;
 
 	if (hw_pass_through && !ecap_pass_through(iommu->ecap)) {
-		pr_warn("IOMMU: %s doesn't support hardware pass through.\n",
+		pr_warn("%s: Doesn't support hardware pass through.\n",
 			iommu->name);
 		return -ENXIO;
 	}
 	if (!ecap_sc_support(iommu->ecap) &&
 	    domain_update_iommu_snooping(iommu)) {
-		pr_warn("IOMMU: %s doesn't support snooping.\n",
+		pr_warn("%s: Doesn't support snooping.\n",
 			iommu->name);
 		return -ENXIO;
 	}
 	sp = domain_update_iommu_superpage(iommu) - 1;
 	if (sp >= 0 && !(cap_super_page_val(iommu->cap) & (1 << sp))) {
-		pr_warn("IOMMU: %s doesn't support large page.\n",
+		pr_warn("%s: Doesn't support large page.\n",
 			iommu->name);
 		return -ENXIO;
 	}
@@ -4051,7 +4354,7 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb,
 		start = mhp->start_pfn << PAGE_SHIFT;
 		end = ((mhp->start_pfn + mhp->nr_pages) << PAGE_SHIFT) - 1;
 		if (iommu_domain_identity_map(si_domain, start, end)) {
-			pr_warn("dmar: failed to build identity map for [%llx-%llx]\n",
+			pr_warn("Failed to build identity map for [%llx-%llx]\n",
 				start, end);
 			return NOTIFY_BAD;
 		}
@@ -4069,7 +4372,7 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb,
 
 			iova = find_iova(&si_domain->iovad, start_vpfn);
 			if (iova == NULL) {
-				pr_debug("dmar: failed get IOVA for PFN %lx\n",
+				pr_debug("Failed get IOVA for PFN %lx\n",
 					 start_vpfn);
 				break;
 			}
@@ -4077,7 +4380,7 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb,
 			iova = split_and_remove_iova(&si_domain->iovad, iova,
 						     start_vpfn, last_vpfn);
 			if (iova == NULL) {
-				pr_warn("dmar: failed to split IOVA PFN [%lx-%lx]\n",
+				pr_warn("Failed to split IOVA PFN [%lx-%lx]\n",
 					start_vpfn, last_vpfn);
 				return NOTIFY_BAD;
 			}
@@ -4186,13 +4489,6 @@ int __init intel_iommu_init(void)
 		goto out_free_dmar;
 	}
 
-	/*
-	 * Disable translation if already enabled prior to OS handover.
-	 */
-	for_each_active_iommu(iommu, drhd)
-		if (iommu->gcmd & DMA_GCMD_TE)
-			iommu_disable_translation(iommu);
-
 	if (dmar_dev_scope_init() < 0) {
 		if (force_on)
 			panic("tboot: Failed to initialize DMAR device scope\n");
@@ -4203,10 +4499,10 @@ int __init intel_iommu_init(void)
 		goto out_free_dmar;
 
 	if (list_empty(&dmar_rmrr_units))
-		printk(KERN_INFO "DMAR: No RMRR found\n");
+		pr_info("No RMRR found\n");
 
 	if (list_empty(&dmar_atsr_units))
-		printk(KERN_INFO "DMAR: No ATSR found\n");
+		pr_info("No ATSR found\n");
 
 	if (dmar_init_reserved_ranges()) {
 		if (force_on)
@@ -4220,12 +4516,11 @@ int __init intel_iommu_init(void)
 	if (ret) {
 		if (force_on)
 			panic("tboot: Failed to initialize DMARs\n");
-		printk(KERN_ERR "IOMMU: dmar init failed\n");
+		pr_err("Initialization failed\n");
 		goto out_free_reserved_range;
 	}
 	up_write(&dmar_global_lock);
-	printk(KERN_INFO
-	"PCI-DMA: Intel(R) Virtualization Technology for Directed I/O\n");
+	pr_info("Intel(R) Virtualization Technology for Directed I/O\n");
 
 	init_timer(&unmap_timer);
 #ifdef CONFIG_SWIOTLB
@@ -4367,13 +4662,11 @@ static struct iommu_domain *intel_iommu_domain_alloc(unsigned type)
 
 	dmar_domain = alloc_domain(DOMAIN_FLAG_VIRTUAL_MACHINE);
 	if (!dmar_domain) {
-		printk(KERN_ERR
-			"intel_iommu_domain_init: dmar_domain == NULL\n");
+		pr_err("Can't allocate dmar_domain\n");
 		return NULL;
 	}
 	if (md_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
-		printk(KERN_ERR
-			"intel_iommu_domain_init() failed\n");
+		pr_err("Domain initialization failed\n");
 		domain_exit(dmar_domain);
 		return NULL;
 	}
@@ -4432,7 +4725,7 @@ static int intel_iommu_attach_device(struct iommu_domain *domain,
 		addr_width = cap_mgaw(iommu->cap);
 
 	if (dmar_domain->max_addr > (1LL << addr_width)) {
-		printk(KERN_ERR "%s: iommu width (%d) is not "
+		pr_err("%s: iommu width (%d) is not "
 		       "sufficient for the mapped address (%llx)\n",
 		       __func__, addr_width, dmar_domain->max_addr);
 		return -EFAULT;
@@ -4486,7 +4779,7 @@ static int intel_iommu_map(struct iommu_domain *domain,
 		/* check if minimum agaw is sufficient for mapped address */
 		end = __DOMAIN_MAX_ADDR(dmar_domain->gaw) + 1;
 		if (end < max_addr) {
-			printk(KERN_ERR "%s: iommu width (%d) is not "
+			pr_err("%s: iommu width (%d) is not "
 			       "sufficient for the mapped address (%llx)\n",
 			       __func__, dmar_domain->gaw, max_addr);
 			return -EFAULT;
@@ -4627,7 +4920,7 @@ static const struct iommu_ops intel_iommu_ops = {
 static void quirk_iommu_g4x_gfx(struct pci_dev *dev)
 {
 	/* G4x/GM45 integrated gfx dmar support is totally busted. */
-	printk(KERN_INFO "DMAR: Disabling IOMMU for graphics on this chipset\n");
+	pr_info("Disabling IOMMU for graphics on this chipset\n");
 	dmar_map_gfx = 0;
 }
 
@@ -4645,7 +4938,7 @@ static void quirk_iommu_rwbf(struct pci_dev *dev)
 	 * Mobile 4 Series Chipset neglects to set RWBF capability,
 	 * but needs it. Same seems to hold for the desktop versions.
 	 */
-	printk(KERN_INFO "DMAR: Forcing write-buffer flush capability\n");
+	pr_info("Forcing write-buffer flush capability\n");
 	rwbf_quirk = 1;
 }
 
@@ -4675,11 +4968,11 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev)
 		return;
 
 	if (!(ggc & GGC_MEMORY_VT_ENABLED)) {
-		printk(KERN_INFO "DMAR: BIOS has allocated no shadow GTT; disabling IOMMU for graphics\n");
+		pr_info("BIOS has allocated no shadow GTT; disabling IOMMU for graphics\n");
 		dmar_map_gfx = 0;
 	} else if (dmar_map_gfx) {
 		/* we have to ensure the gfx device is idle before we flush */
-		printk(KERN_INFO "DMAR: Disabling batched IOTLB flush on Ironlake\n");
+		pr_info("Disabling batched IOTLB flush on Ironlake\n");
 		intel_iommu_strict = 1;
        }
 }
@@ -4741,7 +5034,7 @@ static void __init check_tylersburg_isoch(void)
 		iommu_identity_mapping |= IDENTMAP_AZALIA;
 		return;
 	}
-	
-	printk(KERN_WARNING "DMAR: Recommended TLB entries for ISOCH unit is 16; your BIOS set %d\n",
+
+	pr_warn("Recommended TLB entries for ISOCH unit is 16; your BIOS set %d\n",
 	       vtisochctrl);
 }
diff --git a/drivers/iommu/intel_irq_remapping.c b/drivers/iommu/intel_irq_remapping.c
index 5709ae9c3..f15692a41 100644
--- a/drivers/iommu/intel_irq_remapping.c
+++ b/drivers/iommu/intel_irq_remapping.c
@@ -1,3 +1,6 @@
+
+#define pr_fmt(fmt)     "DMAR-IR: " fmt
+
 #include <linux/interrupt.h>
 #include <linux/dmar.h>
 #include <linux/spinlock.h>
@@ -8,6 +11,8 @@
 #include <linux/irq.h>
 #include <linux/intel-iommu.h>
 #include <linux/acpi.h>
+#include <linux/irqdomain.h>
+#include <linux/crash_dump.h>
 #include <asm/io_apic.h>
 #include <asm/smp.h>
 #include <asm/cpu.h>
@@ -17,6 +22,11 @@
 
 #include "irq_remapping.h"
 
+enum irq_mode {
+	IRQ_REMAPPING,
+	IRQ_POSTING,
+};
+
 struct ioapic_scope {
 	struct intel_iommu *iommu;
 	unsigned int id;
@@ -31,6 +41,22 @@ struct hpet_scope {
 	unsigned int devfn;
 };
 
+struct irq_2_iommu {
+	struct intel_iommu *iommu;
+	u16 irte_index;
+	u16 sub_handle;
+	u8  irte_mask;
+	enum irq_mode mode;
+};
+
+struct intel_ir_data {
+	struct irq_2_iommu			irq_2_iommu;
+	struct irte				irte_entry;
+	union {
+		struct msi_msg			msi_entry;
+	};
+};
+
 #define IR_X2APIC_MODE(mode) (mode ? (1 << 11) : 0)
 #define IRTE_DEST(dest) ((eim_mode) ? dest : dest << 8)
 
@@ -50,43 +76,34 @@ static struct hpet_scope ir_hpet[MAX_HPET_TBS];
  * the dmar_global_lock.
  */
 static DEFINE_RAW_SPINLOCK(irq_2_ir_lock);
+static struct irq_domain_ops intel_ir_domain_ops;
 
+static void iommu_disable_irq_remapping(struct intel_iommu *iommu);
 static int __init parse_ioapics_under_ir(void);
 
-static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
+static bool ir_pre_enabled(struct intel_iommu *iommu)
 {
-	struct irq_cfg *cfg = irq_cfg(irq);
-	return cfg ? &cfg->irq_2_iommu : NULL;
+	return (iommu->flags & VTD_FLAG_IRQ_REMAP_PRE_ENABLED);
 }
 
-static int get_irte(int irq, struct irte *entry)
+static void clear_ir_pre_enabled(struct intel_iommu *iommu)
 {
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-	unsigned long flags;
-	int index;
-
-	if (!entry || !irq_iommu)
-		return -1;
-
-	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
-
-	if (unlikely(!irq_iommu->iommu)) {
-		raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
+	iommu->flags &= ~VTD_FLAG_IRQ_REMAP_PRE_ENABLED;
+}
 
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
-	*entry = *(irq_iommu->iommu->ir_table->base + index);
+static void init_ir_status(struct intel_iommu *iommu)
+{
+	u32 gsts;
 
-	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-	return 0;
+	gsts = readl(iommu->reg + DMAR_GSTS_REG);
+	if (gsts & DMA_GSTS_IRES)
+		iommu->flags |= VTD_FLAG_IRQ_REMAP_PRE_ENABLED;
 }
 
-static int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
+static int alloc_irte(struct intel_iommu *iommu, int irq,
+		      struct irq_2_iommu *irq_iommu, u16 count)
 {
 	struct ir_table *table = iommu->ir_table;
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-	struct irq_cfg *cfg = irq_cfg(irq);
 	unsigned int mask = 0;
 	unsigned long flags;
 	int index;
@@ -100,8 +117,7 @@ static int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 	}
 
 	if (mask > ecap_max_handle_mask(iommu->ecap)) {
-		printk(KERN_ERR
-		       "Requested mask %x exceeds the max invalidation handle"
+		pr_err("Requested mask %x exceeds the max invalidation handle"
 		       " mask value %Lx\n", mask,
 		       ecap_max_handle_mask(iommu->ecap));
 		return -1;
@@ -113,11 +129,11 @@ static int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 	if (index < 0) {
 		pr_warn("IR%d: can't allocate an IRTE\n", iommu->seq_id);
 	} else {
-		cfg->remapped = 1;
 		irq_iommu->iommu = iommu;
 		irq_iommu->irte_index =  index;
 		irq_iommu->sub_handle = 0;
 		irq_iommu->irte_mask = mask;
+		irq_iommu->mode = IRQ_REMAPPING;
 	}
 	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
@@ -135,47 +151,9 @@ static int qi_flush_iec(struct intel_iommu *iommu, int index, int mask)
 	return qi_submit_sync(&desc, iommu);
 }
 
-static int map_irq_to_irte_handle(int irq, u16 *sub_handle)
-{
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-	unsigned long flags;
-	int index;
-
-	if (!irq_iommu)
-		return -1;
-
-	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
-	*sub_handle = irq_iommu->sub_handle;
-	index = irq_iommu->irte_index;
-	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-	return index;
-}
-
-static int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
-{
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-	struct irq_cfg *cfg = irq_cfg(irq);
-	unsigned long flags;
-
-	if (!irq_iommu)
-		return -1;
-
-	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
-
-	cfg->remapped = 1;
-	irq_iommu->iommu = iommu;
-	irq_iommu->irte_index = index;
-	irq_iommu->sub_handle = subhandle;
-	irq_iommu->irte_mask = 0;
-
-	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return 0;
-}
-
-static int modify_irte(int irq, struct irte *irte_modified)
+static int modify_irte(struct irq_2_iommu *irq_iommu,
+		       struct irte *irte_modified)
 {
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	struct intel_iommu *iommu;
 	unsigned long flags;
 	struct irte *irte;
@@ -196,6 +174,9 @@ static int modify_irte(int irq, struct irte *irte_modified)
 	__iommu_flush_cache(iommu, irte, sizeof(*irte));
 
 	rc = qi_flush_iec(iommu, index, 0);
+
+	/* Update iommu mode according to the IRTE mode */
+	irq_iommu->mode = irte->pst ? IRQ_POSTING : IRQ_REMAPPING;
 	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return rc;
@@ -242,7 +223,7 @@ static int clear_entries(struct irq_2_iommu *irq_iommu)
 		return 0;
 
 	iommu = irq_iommu->iommu;
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
+	index = irq_iommu->irte_index;
 
 	start = iommu->ir_table->base + index;
 	end = start + (1 << irq_iommu->irte_mask);
@@ -257,29 +238,6 @@ static int clear_entries(struct irq_2_iommu *irq_iommu)
 	return qi_flush_iec(iommu, index, irq_iommu->irte_mask);
 }
 
-static int free_irte(int irq)
-{
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-	unsigned long flags;
-	int rc;
-
-	if (!irq_iommu)
-		return -1;
-
-	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
-
-	rc = clear_entries(irq_iommu);
-
-	irq_iommu->iommu = NULL;
-	irq_iommu->irte_index = 0;
-	irq_iommu->sub_handle = 0;
-	irq_iommu->irte_mask = 0;
-
-	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return rc;
-}
-
 /*
  * source validation type
  */
@@ -333,7 +291,7 @@ static int set_ioapic_sid(struct irte *irte, int apic)
 	up_read(&dmar_global_lock);
 
 	if (sid == 0) {
-		pr_warning("Failed to set source-id of IOAPIC (%d)\n", apic);
+		pr_warn("Failed to set source-id of IOAPIC (%d)\n", apic);
 		return -1;
 	}
 
@@ -360,7 +318,7 @@ static int set_hpet_sid(struct irte *irte, u8 id)
 	up_read(&dmar_global_lock);
 
 	if (sid == 0) {
-		pr_warning("Failed to set source-id of HPET block (%d)\n", id);
+		pr_warn("Failed to set source-id of HPET block (%d)\n", id);
 		return -1;
 	}
 
@@ -424,11 +382,59 @@ static int set_msi_sid(struct irte *irte, struct pci_dev *dev)
 	return 0;
 }
 
+static int iommu_load_old_irte(struct intel_iommu *iommu)
+{
+	struct irte *old_ir_table;
+	phys_addr_t irt_phys;
+	unsigned int i;
+	size_t size;
+	u64 irta;
+
+	if (!is_kdump_kernel()) {
+		pr_warn("IRQ remapping was enabled on %s but we are not in kdump mode\n",
+			iommu->name);
+		clear_ir_pre_enabled(iommu);
+		iommu_disable_irq_remapping(iommu);
+		return -EINVAL;
+	}
+
+	/* Check whether the old ir-table has the same size as ours */
+	irta = dmar_readq(iommu->reg + DMAR_IRTA_REG);
+	if ((irta & INTR_REMAP_TABLE_REG_SIZE_MASK)
+	     != INTR_REMAP_TABLE_REG_SIZE)
+		return -EINVAL;
+
+	irt_phys = irta & VTD_PAGE_MASK;
+	size     = INTR_REMAP_TABLE_ENTRIES*sizeof(struct irte);
+
+	/* Map the old IR table */
+	old_ir_table = ioremap_cache(irt_phys, size);
+	if (!old_ir_table)
+		return -ENOMEM;
+
+	/* Copy data over */
+	memcpy(iommu->ir_table->base, old_ir_table, size);
+
+	__iommu_flush_cache(iommu, iommu->ir_table->base, size);
+
+	/*
+	 * Now check the table for used entries and mark those as
+	 * allocated in the bitmap
+	 */
+	for (i = 0; i < INTR_REMAP_TABLE_ENTRIES; i++) {
+		if (iommu->ir_table->base[i].present)
+			bitmap_set(iommu->ir_table->bitmap, i, 1);
+	}
+
+	return 0;
+}
+
+
 static void iommu_set_irq_remapping(struct intel_iommu *iommu, int mode)
 {
+	unsigned long flags;
 	u64 addr;
 	u32 sts;
-	unsigned long flags;
 
 	addr = virt_to_phys((void *)iommu->ir_table->base);
 
@@ -445,10 +451,16 @@ static void iommu_set_irq_remapping(struct intel_iommu *iommu, int mode)
 	raw_spin_unlock_irqrestore(&iommu->register_lock, flags);
 
 	/*
-	 * global invalidation of interrupt entry cache before enabling
-	 * interrupt-remapping.
+	 * Global invalidation of interrupt entry cache to make sure the
+	 * hardware uses the new irq remapping table.
 	 */
 	qi_global_iec(iommu);
+}
+
+static void iommu_enable_irq_remapping(struct intel_iommu *iommu)
+{
+	unsigned long flags;
+	u32 sts;
 
 	raw_spin_lock_irqsave(&iommu->register_lock, flags);
 
@@ -488,7 +500,6 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 
 	pages = alloc_pages_node(iommu->node, GFP_KERNEL | __GFP_ZERO,
 				 INTR_REMAP_PAGE_ORDER);
-
 	if (!pages) {
 		pr_err("IR%d: failed to allocate pages of order %d\n",
 		       iommu->seq_id, INTR_REMAP_PAGE_ORDER);
@@ -502,21 +513,75 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 		goto out_free_pages;
 	}
 
+	iommu->ir_domain = irq_domain_add_hierarchy(arch_get_ir_parent_domain(),
+						    0, INTR_REMAP_TABLE_ENTRIES,
+						    NULL, &intel_ir_domain_ops,
+						    iommu);
+	if (!iommu->ir_domain) {
+		pr_err("IR%d: failed to allocate irqdomain\n", iommu->seq_id);
+		goto out_free_bitmap;
+	}
+	iommu->ir_msi_domain = arch_create_msi_irq_domain(iommu->ir_domain);
+
 	ir_table->base = page_address(pages);
 	ir_table->bitmap = bitmap;
 	iommu->ir_table = ir_table;
+
+	/*
+	 * If the queued invalidation is already initialized,
+	 * shouldn't disable it.
+	 */
+	if (!iommu->qi) {
+		/*
+		 * Clear previous faults.
+		 */
+		dmar_fault(-1, iommu);
+		dmar_disable_qi(iommu);
+
+		if (dmar_enable_qi(iommu)) {
+			pr_err("Failed to enable queued invalidation\n");
+			goto out_free_bitmap;
+		}
+	}
+
+	init_ir_status(iommu);
+
+	if (ir_pre_enabled(iommu)) {
+		if (iommu_load_old_irte(iommu))
+			pr_err("Failed to copy IR table for %s from previous kernel\n",
+			       iommu->name);
+		else
+			pr_info("Copied IR table for %s from previous kernel\n",
+				iommu->name);
+	}
+
+	iommu_set_irq_remapping(iommu, eim_mode);
+
 	return 0;
 
+out_free_bitmap:
+	kfree(bitmap);
 out_free_pages:
 	__free_pages(pages, INTR_REMAP_PAGE_ORDER);
 out_free_table:
 	kfree(ir_table);
+
+	iommu->ir_table  = NULL;
+
 	return -ENOMEM;
 }
 
 static void intel_teardown_irq_remapping(struct intel_iommu *iommu)
 {
 	if (iommu && iommu->ir_table) {
+		if (iommu->ir_msi_domain) {
+			irq_domain_remove(iommu->ir_msi_domain);
+			iommu->ir_msi_domain = NULL;
+		}
+		if (iommu->ir_domain) {
+			irq_domain_remove(iommu->ir_domain);
+			iommu->ir_domain = NULL;
+		}
 		free_pages((unsigned long)iommu->ir_table->base,
 			   INTR_REMAP_PAGE_ORDER);
 		kfree(iommu->ir_table->bitmap);
@@ -580,17 +645,17 @@ static void __init intel_cleanup_irq_remapping(void)
 	}
 
 	if (x2apic_supported())
-		pr_warn("Failed to enable irq remapping.  You are vulnerable to irq-injection attacks.\n");
+		pr_warn("Failed to enable irq remapping. You are vulnerable to irq-injection attacks.\n");
 }
 
 static int __init intel_prepare_irq_remapping(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct intel_iommu *iommu;
+	int eim = 0;
 
 	if (irq_remap_broken) {
-		printk(KERN_WARNING
-			"This system BIOS has enabled interrupt remapping\n"
+		pr_warn("This system BIOS has enabled interrupt remapping\n"
 			"on a chipset that contains an erratum making that\n"
 			"feature unstable.  To maintain system stability\n"
 			"interrupt remapping is being disabled.  Please\n"
@@ -606,7 +671,7 @@ static int __init intel_prepare_irq_remapping(void)
 		return -ENODEV;
 
 	if (parse_ioapics_under_ir() != 1) {
-		printk(KERN_INFO "Not enabling interrupt remapping\n");
+		pr_info("Not enabling interrupt remapping\n");
 		goto error;
 	}
 
@@ -615,85 +680,74 @@ static int __init intel_prepare_irq_remapping(void)
 		if (!ecap_ir_support(iommu->ecap))
 			goto error;
 
-	/* Do the allocations early */
-	for_each_iommu(iommu, drhd)
-		if (intel_setup_irq_remapping(iommu))
-			goto error;
-
-	return 0;
-
-error:
-	intel_cleanup_irq_remapping();
-	return -ENODEV;
-}
-
-static int __init intel_enable_irq_remapping(void)
-{
-	struct dmar_drhd_unit *drhd;
-	struct intel_iommu *iommu;
-	bool setup = false;
-	int eim = 0;
-
+	/* Detect remapping mode: lapic or x2apic */
 	if (x2apic_supported()) {
 		eim = !dmar_x2apic_optout();
-		if (!eim)
-			pr_info("x2apic is disabled because BIOS sets x2apic opt out bit. You can use 'intremap=no_x2apic_optout' to override the BIOS setting.\n");
+		if (!eim) {
+			pr_info("x2apic is disabled because BIOS sets x2apic opt out bit.");
+			pr_info("Use 'intremap=no_x2apic_optout' to override the BIOS setting.\n");
+		}
 	}
 
 	for_each_iommu(iommu, drhd) {
-		/*
-		 * If the queued invalidation is already initialized,
-		 * shouldn't disable it.
-		 */
-		if (iommu->qi)
-			continue;
-
-		/*
-		 * Clear previous faults.
-		 */
-		dmar_fault(-1, iommu);
-
-		/*
-		 * Disable intr remapping and queued invalidation, if already
-		 * enabled prior to OS handover.
-		 */
-		iommu_disable_irq_remapping(iommu);
-
-		dmar_disable_qi(iommu);
-	}
-
-	/*
-	 * check for the Interrupt-remapping support
-	 */
-	for_each_iommu(iommu, drhd)
 		if (eim && !ecap_eim_support(iommu->ecap)) {
-			printk(KERN_INFO "DRHD %Lx: EIM not supported by DRHD, "
-			       " ecap %Lx\n", drhd->reg_base_addr, iommu->ecap);
+			pr_info("%s does not support EIM\n", iommu->name);
 			eim = 0;
 		}
+	}
+
 	eim_mode = eim;
 	if (eim)
 		pr_info("Queued invalidation will be enabled to support x2apic and Intr-remapping.\n");
 
-	/*
-	 * Enable queued invalidation for all the DRHD's.
-	 */
+	/* Do the initializations early */
 	for_each_iommu(iommu, drhd) {
-		int ret = dmar_enable_qi(iommu);
-
-		if (ret) {
-			printk(KERN_ERR "DRHD %Lx: failed to enable queued, "
-			       " invalidation, ecap %Lx, ret %d\n",
-			       drhd->reg_base_addr, iommu->ecap, ret);
+		if (intel_setup_irq_remapping(iommu)) {
+			pr_err("Failed to setup irq remapping for %s\n",
+			       iommu->name);
 			goto error;
 		}
 	}
 
+	return 0;
+
+error:
+	intel_cleanup_irq_remapping();
+	return -ENODEV;
+}
+
+/*
+ * Set Posted-Interrupts capability.
+ */
+static inline void set_irq_posting_cap(void)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+
+	if (!disable_irq_post) {
+		intel_irq_remap_ops.capability |= 1 << IRQ_POSTING_CAP;
+
+		for_each_iommu(iommu, drhd)
+			if (!cap_pi_support(iommu->cap)) {
+				intel_irq_remap_ops.capability &=
+						~(1 << IRQ_POSTING_CAP);
+				break;
+			}
+	}
+}
+
+static int __init intel_enable_irq_remapping(void)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	bool setup = false;
+
 	/*
 	 * Setup Interrupt-remapping for all the DRHD's now.
 	 */
 	for_each_iommu(iommu, drhd) {
-		iommu_set_irq_remapping(iommu, eim);
+		if (!ir_pre_enabled(iommu))
+			iommu_enable_irq_remapping(iommu);
 		setup = true;
 	}
 
@@ -702,16 +756,11 @@ static int __init intel_enable_irq_remapping(void)
 
 	irq_remapping_enabled = 1;
 
-	/*
-	 * VT-d has a different layout for IO-APIC entries when
-	 * interrupt remapping is enabled. So it needs a special routine
-	 * to print IO-APIC entries for debugging purposes too.
-	 */
-	x86_io_apic_ops.print_entries = intel_ir_io_apic_print_entries;
+	set_irq_posting_cap();
 
-	pr_info("Enabled IRQ remapping in %s mode\n", eim ? "x2apic" : "xapic");
+	pr_info("Enabled IRQ remapping in %s mode\n", eim_mode ? "x2apic" : "xapic");
 
-	return eim ? IRQ_REMAP_X2APIC_MODE : IRQ_REMAP_XAPIC_MODE;
+	return eim_mode ? IRQ_REMAP_X2APIC_MODE : IRQ_REMAP_XAPIC_MODE;
 
 error:
 	intel_cleanup_irq_remapping();
@@ -909,6 +958,12 @@ static void disable_irq_remapping(void)
 
 		iommu_disable_irq_remapping(iommu);
 	}
+
+	/*
+	 * Clear Posted-Interrupts capability.
+	 */
+	if (!disable_irq_post)
+		intel_irq_remap_ops.capability &= ~(1 << IRQ_POSTING_CAP);
 }
 
 static int reenable_irq_remapping(int eim)
@@ -930,12 +985,15 @@ static int reenable_irq_remapping(int eim)
 
 		/* Set up interrupt remapping for iommu.*/
 		iommu_set_irq_remapping(iommu, eim);
+		iommu_enable_irq_remapping(iommu);
 		setup = true;
 	}
 
 	if (!setup)
 		goto error;
 
+	set_irq_posting_cap();
+
 	return 0;
 
 error:
@@ -945,8 +1003,7 @@ error:
 	return -1;
 }
 
-static void prepare_irte(struct irte *irte, int vector,
-			 unsigned int dest)
+static void prepare_irte(struct irte *irte, int vector, unsigned int dest)
 {
 	memset(irte, 0, sizeof(*irte));
 
@@ -966,76 +1023,63 @@ static void prepare_irte(struct irte *irte, int vector,
 	irte->redir_hint = 1;
 }
 
-static int intel_setup_ioapic_entry(int irq,
-				    struct IO_APIC_route_entry *route_entry,
-				    unsigned int destination, int vector,
-				    struct io_apic_irq_attr *attr)
+static struct irq_domain *intel_get_ir_irq_domain(struct irq_alloc_info *info)
 {
-	int ioapic_id = mpc_ioapic_id(attr->ioapic);
-	struct intel_iommu *iommu;
-	struct IR_IO_APIC_route_entry *entry;
-	struct irte irte;
-	int index;
-
-	down_read(&dmar_global_lock);
-	iommu = map_ioapic_to_ir(ioapic_id);
-	if (!iommu) {
-		pr_warn("No mapping iommu for ioapic %d\n", ioapic_id);
-		index = -ENODEV;
-	} else {
-		index = alloc_irte(iommu, irq, 1);
-		if (index < 0) {
-			pr_warn("Failed to allocate IRTE for ioapic %d\n",
-				ioapic_id);
-			index = -ENOMEM;
-		}
-	}
-	up_read(&dmar_global_lock);
-	if (index < 0)
-		return index;
-
-	prepare_irte(&irte, vector, destination);
+	struct intel_iommu *iommu = NULL;
 
-	/* Set source-id of interrupt request */
-	set_ioapic_sid(&irte, ioapic_id);
+	if (!info)
+		return NULL;
 
-	modify_irte(irq, &irte);
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_IOAPIC:
+		iommu = map_ioapic_to_ir(info->ioapic_id);
+		break;
+	case X86_IRQ_ALLOC_TYPE_HPET:
+		iommu = map_hpet_to_ir(info->hpet_id);
+		break;
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		iommu = map_dev_to_ir(info->msi_dev);
+		break;
+	default:
+		BUG_ON(1);
+		break;
+	}
 
-	apic_printk(APIC_VERBOSE, KERN_DEBUG "IOAPIC[%d]: "
-		"Set IRTE entry (P:%d FPD:%d Dst_Mode:%d "
-		"Redir_hint:%d Trig_Mode:%d Dlvry_Mode:%X "
-		"Avail:%X Vector:%02X Dest:%08X "
-		"SID:%04X SQ:%X SVT:%X)\n",
-		attr->ioapic, irte.present, irte.fpd, irte.dst_mode,
-		irte.redir_hint, irte.trigger_mode, irte.dlvry_mode,
-		irte.avail, irte.vector, irte.dest_id,
-		irte.sid, irte.sq, irte.svt);
+	return iommu ? iommu->ir_domain : NULL;
+}
 
-	entry = (struct IR_IO_APIC_route_entry *)route_entry;
-	memset(entry, 0, sizeof(*entry));
+static struct irq_domain *intel_get_irq_domain(struct irq_alloc_info *info)
+{
+	struct intel_iommu *iommu;
 
-	entry->index2	= (index >> 15) & 0x1;
-	entry->zero	= 0;
-	entry->format	= 1;
-	entry->index	= (index & 0x7fff);
-	/*
-	 * IO-APIC RTE will be configured with virtual vector.
-	 * irq handler will do the explicit EOI to the io-apic.
-	 */
-	entry->vector	= attr->ioapic_pin;
-	entry->mask	= 0;			/* enable IRQ */
-	entry->trigger	= attr->trigger;
-	entry->polarity	= attr->polarity;
+	if (!info)
+		return NULL;
 
-	/* Mask level triggered irqs.
-	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
-	 */
-	if (attr->trigger)
-		entry->mask = 1;
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		iommu = map_dev_to_ir(info->msi_dev);
+		if (iommu)
+			return iommu->ir_msi_domain;
+		break;
+	default:
+		break;
+	}
 
-	return 0;
+	return NULL;
 }
 
+struct irq_remap_ops intel_irq_remap_ops = {
+	.prepare		= intel_prepare_irq_remapping,
+	.enable			= intel_enable_irq_remapping,
+	.disable		= disable_irq_remapping,
+	.reenable		= reenable_irq_remapping,
+	.enable_faulting	= enable_drhd_fault_handling,
+	.get_ir_irq_domain	= intel_get_ir_irq_domain,
+	.get_irq_domain		= intel_get_irq_domain,
+};
+
 /*
  * Migrate the IO-APIC irq in the presence of intr-remapping.
  *
@@ -1051,170 +1095,281 @@ static int intel_setup_ioapic_entry(int irq,
  * is used to migrate MSI irq's in the presence of interrupt-remapping.
  */
 static int
-intel_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
-			  bool force)
+intel_ir_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		      bool force)
 {
+	struct intel_ir_data *ir_data = data->chip_data;
+	struct irte *irte = &ir_data->irte_entry;
 	struct irq_cfg *cfg = irqd_cfg(data);
-	unsigned int dest, irq = data->irq;
-	struct irte irte;
-	int err;
-
-	if (!config_enabled(CONFIG_SMP))
-		return -EINVAL;
-
-	if (!cpumask_intersects(mask, cpu_online_mask))
-		return -EINVAL;
-
-	if (get_irte(irq, &irte))
-		return -EBUSY;
-
-	err = assign_irq_vector(irq, cfg, mask);
-	if (err)
-		return err;
-
-	err = apic->cpu_mask_to_apicid_and(cfg->domain, mask, &dest);
-	if (err) {
-		if (assign_irq_vector(irq, cfg, data->affinity))
-			pr_err("Failed to recover vector for irq %d\n", irq);
-		return err;
-	}
+	struct irq_data *parent = data->parent_data;
+	int ret;
 
-	irte.vector = cfg->vector;
-	irte.dest_id = IRTE_DEST(dest);
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret < 0 || ret == IRQ_SET_MASK_OK_DONE)
+		return ret;
 
 	/*
 	 * Atomically updates the IRTE with the new destination, vector
 	 * and flushes the interrupt entry cache.
 	 */
-	modify_irte(irq, &irte);
+	irte->vector = cfg->vector;
+	irte->dest_id = IRTE_DEST(cfg->dest_apicid);
+
+	/* Update the hardware only if the interrupt is in remapped mode. */
+	if (ir_data->irq_2_iommu.mode == IRQ_REMAPPING)
+		modify_irte(&ir_data->irq_2_iommu, irte);
 
 	/*
 	 * After this point, all the interrupts will start arriving
 	 * at the new destination. So, time to cleanup the previous
 	 * vector allocation.
 	 */
-	if (cfg->move_in_progress)
-		send_cleanup_vector(cfg);
+	send_cleanup_vector(cfg);
 
-	cpumask_copy(data->affinity, mask);
-	return 0;
+	return IRQ_SET_MASK_OK_DONE;
 }
 
-static void intel_compose_msi_msg(struct pci_dev *pdev,
-				  unsigned int irq, unsigned int dest,
-				  struct msi_msg *msg, u8 hpet_id)
+static void intel_ir_compose_msi_msg(struct irq_data *irq_data,
+				     struct msi_msg *msg)
 {
-	struct irq_cfg *cfg;
-	struct irte irte;
-	u16 sub_handle = 0;
-	int ir_index;
-
-	cfg = irq_cfg(irq);
+	struct intel_ir_data *ir_data = irq_data->chip_data;
 
-	ir_index = map_irq_to_irte_handle(irq, &sub_handle);
-	BUG_ON(ir_index == -1);
+	*msg = ir_data->msi_entry;
+}
 
-	prepare_irte(&irte, cfg->vector, dest);
+static int intel_ir_set_vcpu_affinity(struct irq_data *data, void *info)
+{
+	struct intel_ir_data *ir_data = data->chip_data;
+	struct vcpu_data *vcpu_pi_info = info;
 
-	/* Set source-id of interrupt request */
-	if (pdev)
-		set_msi_sid(&irte, pdev);
-	else
-		set_hpet_sid(&irte, hpet_id);
+	/* stop posting interrupts, back to remapping mode */
+	if (!vcpu_pi_info) {
+		modify_irte(&ir_data->irq_2_iommu, &ir_data->irte_entry);
+	} else {
+		struct irte irte_pi;
 
-	modify_irte(irq, &irte);
+		/*
+		 * We are not caching the posted interrupt entry. We
+		 * copy the data from the remapped entry and modify
+		 * the fields which are relevant for posted mode. The
+		 * cached remapped entry is used for switching back to
+		 * remapped mode.
+		 */
+		memset(&irte_pi, 0, sizeof(irte_pi));
+		dmar_copy_shared_irte(&irte_pi, &ir_data->irte_entry);
+
+		/* Update the posted mode fields */
+		irte_pi.p_pst = 1;
+		irte_pi.p_urgent = 0;
+		irte_pi.p_vector = vcpu_pi_info->vector;
+		irte_pi.pda_l = (vcpu_pi_info->pi_desc_addr >>
+				(32 - PDA_LOW_BIT)) & ~(-1UL << PDA_LOW_BIT);
+		irte_pi.pda_h = (vcpu_pi_info->pi_desc_addr >> 32) &
+				~(-1UL << PDA_HIGH_BIT);
+
+		modify_irte(&ir_data->irq_2_iommu, &irte_pi);
+	}
 
-	msg->address_hi = MSI_ADDR_BASE_HI;
-	msg->data = sub_handle;
-	msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT |
-			  MSI_ADDR_IR_SHV |
-			  MSI_ADDR_IR_INDEX1(ir_index) |
-			  MSI_ADDR_IR_INDEX2(ir_index);
+	return 0;
 }
 
-/*
- * Map the PCI dev to the corresponding remapping hardware unit
- * and allocate 'nvec' consecutive interrupt-remapping table entries
- * in it.
- */
-static int intel_msi_alloc_irq(struct pci_dev *dev, int irq, int nvec)
+static struct irq_chip intel_ir_chip = {
+	.irq_ack = ir_ack_apic_edge,
+	.irq_set_affinity = intel_ir_set_affinity,
+	.irq_compose_msi_msg = intel_ir_compose_msi_msg,
+	.irq_set_vcpu_affinity = intel_ir_set_vcpu_affinity,
+};
+
+static void intel_irq_remapping_prepare_irte(struct intel_ir_data *data,
+					     struct irq_cfg *irq_cfg,
+					     struct irq_alloc_info *info,
+					     int index, int sub_handle)
 {
-	struct intel_iommu *iommu;
-	int index;
+	struct IR_IO_APIC_route_entry *entry;
+	struct irte *irte = &data->irte_entry;
+	struct msi_msg *msg = &data->msi_entry;
+
+	prepare_irte(irte, irq_cfg->vector, irq_cfg->dest_apicid);
+	switch (info->type) {
+	case X86_IRQ_ALLOC_TYPE_IOAPIC:
+		/* Set source-id of interrupt request */
+		set_ioapic_sid(irte, info->ioapic_id);
+		apic_printk(APIC_VERBOSE, KERN_DEBUG "IOAPIC[%d]: Set IRTE entry (P:%d FPD:%d Dst_Mode:%d Redir_hint:%d Trig_Mode:%d Dlvry_Mode:%X Avail:%X Vector:%02X Dest:%08X SID:%04X SQ:%X SVT:%X)\n",
+			info->ioapic_id, irte->present, irte->fpd,
+			irte->dst_mode, irte->redir_hint,
+			irte->trigger_mode, irte->dlvry_mode,
+			irte->avail, irte->vector, irte->dest_id,
+			irte->sid, irte->sq, irte->svt);
+
+		entry = (struct IR_IO_APIC_route_entry *)info->ioapic_entry;
+		info->ioapic_entry = NULL;
+		memset(entry, 0, sizeof(*entry));
+		entry->index2	= (index >> 15) & 0x1;
+		entry->zero	= 0;
+		entry->format	= 1;
+		entry->index	= (index & 0x7fff);
+		/*
+		 * IO-APIC RTE will be configured with virtual vector.
+		 * irq handler will do the explicit EOI to the io-apic.
+		 */
+		entry->vector	= info->ioapic_pin;
+		entry->mask	= 0;			/* enable IRQ */
+		entry->trigger	= info->ioapic_trigger;
+		entry->polarity	= info->ioapic_polarity;
+		if (info->ioapic_trigger)
+			entry->mask = 1; /* Mask level triggered irqs. */
+		break;
+
+	case X86_IRQ_ALLOC_TYPE_HPET:
+	case X86_IRQ_ALLOC_TYPE_MSI:
+	case X86_IRQ_ALLOC_TYPE_MSIX:
+		if (info->type == X86_IRQ_ALLOC_TYPE_HPET)
+			set_hpet_sid(irte, info->hpet_id);
+		else
+			set_msi_sid(irte, info->msi_dev);
+
+		msg->address_hi = MSI_ADDR_BASE_HI;
+		msg->data = sub_handle;
+		msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT |
+				  MSI_ADDR_IR_SHV |
+				  MSI_ADDR_IR_INDEX1(index) |
+				  MSI_ADDR_IR_INDEX2(index);
+		break;
+
+	default:
+		BUG_ON(1);
+		break;
+	}
+}
 
-	down_read(&dmar_global_lock);
-	iommu = map_dev_to_ir(dev);
-	if (!iommu) {
-		printk(KERN_ERR
-		       "Unable to map PCI %s to iommu\n", pci_name(dev));
-		index = -ENOENT;
-	} else {
-		index = alloc_irte(iommu, irq, nvec);
-		if (index < 0) {
-			printk(KERN_ERR
-			       "Unable to allocate %d IRTE for PCI %s\n",
-			       nvec, pci_name(dev));
-			index = -ENOSPC;
+static void intel_free_irq_resources(struct irq_domain *domain,
+				     unsigned int virq, unsigned int nr_irqs)
+{
+	struct irq_data *irq_data;
+	struct intel_ir_data *data;
+	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
+	int i;
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(domain, virq  + i);
+		if (irq_data && irq_data->chip_data) {
+			data = irq_data->chip_data;
+			irq_iommu = &data->irq_2_iommu;
+			raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
+			clear_entries(irq_iommu);
+			raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+			irq_domain_reset_irq_data(irq_data);
+			kfree(data);
 		}
 	}
-	up_read(&dmar_global_lock);
-
-	return index;
 }
 
-static int intel_msi_setup_irq(struct pci_dev *pdev, unsigned int irq,
-			       int index, int sub_handle)
+static int intel_irq_remapping_alloc(struct irq_domain *domain,
+				     unsigned int virq, unsigned int nr_irqs,
+				     void *arg)
 {
-	struct intel_iommu *iommu;
-	int ret = -ENOENT;
+	struct intel_iommu *iommu = domain->host_data;
+	struct irq_alloc_info *info = arg;
+	struct intel_ir_data *data, *ird;
+	struct irq_data *irq_data;
+	struct irq_cfg *irq_cfg;
+	int i, ret, index;
+
+	if (!info || !iommu)
+		return -EINVAL;
+	if (nr_irqs > 1 && info->type != X86_IRQ_ALLOC_TYPE_MSI &&
+	    info->type != X86_IRQ_ALLOC_TYPE_MSIX)
+		return -EINVAL;
+
+	/*
+	 * With IRQ remapping enabled, don't need contiguous CPU vectors
+	 * to support multiple MSI interrupts.
+	 */
+	if (info->type == X86_IRQ_ALLOC_TYPE_MSI)
+		info->flags &= ~X86_IRQ_ALLOC_CONTIGUOUS_VECTORS;
+
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
+	if (ret < 0)
+		return ret;
+
+	ret = -ENOMEM;
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		goto out_free_parent;
 
 	down_read(&dmar_global_lock);
-	iommu = map_dev_to_ir(pdev);
-	if (iommu) {
-		/*
-		 * setup the mapping between the irq and the IRTE
-		 * base index, the sub_handle pointing to the
-		 * appropriate interrupt remap table entry.
-		 */
-		set_irte_irq(irq, iommu, index, sub_handle);
-		ret = 0;
-	}
+	index = alloc_irte(iommu, virq, &data->irq_2_iommu, nr_irqs);
 	up_read(&dmar_global_lock);
+	if (index < 0) {
+		pr_warn("Failed to allocate IRTE\n");
+		kfree(data);
+		goto out_free_parent;
+	}
+
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(domain, virq + i);
+		irq_cfg = irqd_cfg(irq_data);
+		if (!irq_data || !irq_cfg) {
+			ret = -EINVAL;
+			goto out_free_data;
+		}
+
+		if (i > 0) {
+			ird = kzalloc(sizeof(*ird), GFP_KERNEL);
+			if (!ird)
+				goto out_free_data;
+			/* Initialize the common data */
+			ird->irq_2_iommu = data->irq_2_iommu;
+			ird->irq_2_iommu.sub_handle = i;
+		} else {
+			ird = data;
+		}
 
+		irq_data->hwirq = (index << 16) + i;
+		irq_data->chip_data = ird;
+		irq_data->chip = &intel_ir_chip;
+		intel_irq_remapping_prepare_irte(ird, irq_cfg, info, index, i);
+		irq_set_status_flags(virq + i, IRQ_MOVE_PCNTXT);
+	}
+	return 0;
+
+out_free_data:
+	intel_free_irq_resources(domain, virq, i);
+out_free_parent:
+	irq_domain_free_irqs_common(domain, virq, nr_irqs);
 	return ret;
 }
 
-static int intel_alloc_hpet_msi(unsigned int irq, unsigned int id)
+static void intel_irq_remapping_free(struct irq_domain *domain,
+				     unsigned int virq, unsigned int nr_irqs)
 {
-	int ret = -1;
-	struct intel_iommu *iommu;
-	int index;
+	intel_free_irq_resources(domain, virq, nr_irqs);
+	irq_domain_free_irqs_common(domain, virq, nr_irqs);
+}
 
-	down_read(&dmar_global_lock);
-	iommu = map_hpet_to_ir(id);
-	if (iommu) {
-		index = alloc_irte(iommu, irq, 1);
-		if (index >= 0)
-			ret = 0;
-	}
-	up_read(&dmar_global_lock);
+static void intel_irq_remapping_activate(struct irq_domain *domain,
+					 struct irq_data *irq_data)
+{
+	struct intel_ir_data *data = irq_data->chip_data;
 
-	return ret;
+	modify_irte(&data->irq_2_iommu, &data->irte_entry);
 }
 
-struct irq_remap_ops intel_irq_remap_ops = {
-	.prepare		= intel_prepare_irq_remapping,
-	.enable			= intel_enable_irq_remapping,
-	.disable		= disable_irq_remapping,
-	.reenable		= reenable_irq_remapping,
-	.enable_faulting	= enable_drhd_fault_handling,
-	.setup_ioapic_entry	= intel_setup_ioapic_entry,
-	.set_affinity		= intel_ioapic_set_affinity,
-	.free_irq		= free_irte,
-	.compose_msi_msg	= intel_compose_msi_msg,
-	.msi_alloc_irq		= intel_msi_alloc_irq,
-	.msi_setup_irq		= intel_msi_setup_irq,
-	.alloc_hpet_msi		= intel_alloc_hpet_msi,
+static void intel_irq_remapping_deactivate(struct irq_domain *domain,
+					   struct irq_data *irq_data)
+{
+	struct intel_ir_data *data = irq_data->chip_data;
+	struct irte entry;
+
+	memset(&entry, 0, sizeof(entry));
+	modify_irte(&data->irq_2_iommu, &entry);
+}
+
+static struct irq_domain_ops intel_ir_domain_ops = {
+	.alloc = intel_irq_remapping_alloc,
+	.free = intel_irq_remapping_free,
+	.activate = intel_irq_remapping_activate,
+	.deactivate = intel_irq_remapping_deactivate,
 };
 
 /*
@@ -1242,28 +1397,12 @@ static int dmar_ir_add(struct dmar_drhd_unit *dmaru, struct intel_iommu *iommu)
 	/* Setup Interrupt-remapping now. */
 	ret = intel_setup_irq_remapping(iommu);
 	if (ret) {
-		pr_err("DRHD %Lx: failed to allocate resource\n",
-		       iommu->reg_phys);
-		ir_remove_ioapic_hpet_scope(iommu);
-		return ret;
-	}
-
-	if (!iommu->qi) {
-		/* Clear previous faults. */
-		dmar_fault(-1, iommu);
-		iommu_disable_irq_remapping(iommu);
-		dmar_disable_qi(iommu);
-	}
-
-	/* Enable queued invalidation */
-	ret = dmar_enable_qi(iommu);
-	if (!ret) {
-		iommu_set_irq_remapping(iommu, eim);
-	} else {
-		pr_err("DRHD %Lx: failed to enable queued invalidation, ecap %Lx, ret %d\n",
-		       iommu->reg_phys, iommu->ecap, ret);
+		pr_err("Failed to setup irq remapping for %s\n",
+		       iommu->name);
 		intel_teardown_irq_remapping(iommu);
 		ir_remove_ioapic_hpet_scope(iommu);
+	} else {
+		iommu_enable_irq_remapping(iommu);
 	}
 
 	return ret;
@@ -1280,6 +1419,9 @@ int dmar_ir_hotplug(struct dmar_drhd_unit *dmaru, bool insert)
 		return -EINVAL;
 	if (!ecap_ir_support(iommu->ecap))
 		return 0;
+	if (irq_remapping_cap(IRQ_POSTING_CAP) &&
+	    !cap_pi_support(iommu->cap))
+		return -EBUSY;
 
 	if (insert) {
 		if (!iommu->ir_table)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index d4f527e56..f28609093 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -16,7 +16,7 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  */
 
-#define pr_fmt(fmt)    "%s: " fmt, __func__
+#define pr_fmt(fmt)    "iommu: " fmt
 
 #include <linux/device.h>
 #include <linux/kernel.h>
@@ -51,6 +51,8 @@ struct iommu_group {
 	void (*iommu_data_release)(void *iommu_data);
 	char *name;
 	int id;
+	struct iommu_domain *default_domain;
+	struct iommu_domain *domain;
 };
 
 struct iommu_device {
@@ -75,6 +77,15 @@ struct iommu_group_attribute iommu_group_attr_##_name =		\
 #define to_iommu_group(_kobj)		\
 	container_of(_kobj, struct iommu_group, kobj)
 
+static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
+						 unsigned type);
+static int __iommu_attach_device(struct iommu_domain *domain,
+				 struct device *dev);
+static int __iommu_attach_group(struct iommu_domain *domain,
+				struct iommu_group *group);
+static void __iommu_detach_group(struct iommu_domain *domain,
+				 struct iommu_group *group);
+
 static ssize_t iommu_group_attr_show(struct kobject *kobj,
 				     struct attribute *__attr, char *buf)
 {
@@ -128,6 +139,8 @@ static void iommu_group_release(struct kobject *kobj)
 {
 	struct iommu_group *group = to_iommu_group(kobj);
 
+	pr_debug("Releasing group %d\n", group->id);
+
 	if (group->iommu_data_release)
 		group->iommu_data_release(group->iommu_data);
 
@@ -135,6 +148,9 @@ static void iommu_group_release(struct kobject *kobj)
 	ida_remove(&iommu_group_ida, group->id);
 	mutex_unlock(&iommu_group_mutex);
 
+	if (group->default_domain)
+		iommu_domain_free(group->default_domain);
+
 	kfree(group->name);
 	kfree(group);
 }
@@ -207,6 +223,8 @@ again:
 	 */
 	kobject_put(&group->kobj);
 
+	pr_debug("Allocated group %d\n", group->id);
+
 	return group;
 }
 EXPORT_SYMBOL_GPL(iommu_group_alloc);
@@ -307,6 +325,52 @@ int iommu_group_set_name(struct iommu_group *group, const char *name)
 }
 EXPORT_SYMBOL_GPL(iommu_group_set_name);
 
+static int iommu_group_create_direct_mappings(struct iommu_group *group,
+					      struct device *dev)
+{
+	struct iommu_domain *domain = group->default_domain;
+	struct iommu_dm_region *entry;
+	struct list_head mappings;
+	unsigned long pg_size;
+	int ret = 0;
+
+	if (!domain || domain->type != IOMMU_DOMAIN_DMA)
+		return 0;
+
+	BUG_ON(!domain->ops->pgsize_bitmap);
+
+	pg_size = 1UL << __ffs(domain->ops->pgsize_bitmap);
+	INIT_LIST_HEAD(&mappings);
+
+	iommu_get_dm_regions(dev, &mappings);
+
+	/* We need to consider overlapping regions for different devices */
+	list_for_each_entry(entry, &mappings, list) {
+		dma_addr_t start, end, addr;
+
+		start = ALIGN(entry->start, pg_size);
+		end   = ALIGN(entry->start + entry->length, pg_size);
+
+		for (addr = start; addr < end; addr += pg_size) {
+			phys_addr_t phys_addr;
+
+			phys_addr = iommu_iova_to_phys(domain, addr);
+			if (phys_addr)
+				continue;
+
+			ret = iommu_map(domain, addr, addr, pg_size, entry->prot);
+			if (ret)
+				goto out;
+		}
+
+	}
+
+out:
+	iommu_put_dm_regions(dev, &mappings);
+
+	return ret;
+}
+
 /**
  * iommu_group_add_device - add a device to an iommu group
  * @group: the group into which to add the device (reference should be held)
@@ -363,8 +427,12 @@ rename:
 
 	dev->iommu_group = group;
 
+	iommu_group_create_direct_mappings(group, dev);
+
 	mutex_lock(&group->mutex);
 	list_add_tail(&device->list, &group->devices);
+	if (group->domain)
+		__iommu_attach_device(group->domain, dev);
 	mutex_unlock(&group->mutex);
 
 	/* Notify any listeners about change to group. */
@@ -372,6 +440,9 @@ rename:
 				     IOMMU_GROUP_NOTIFY_ADD_DEVICE, dev);
 
 	trace_add_device_to_group(group->id, dev);
+
+	pr_info("Adding device %s to group %d\n", dev_name(dev), group->id);
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(iommu_group_add_device);
@@ -388,6 +459,8 @@ void iommu_group_remove_device(struct device *dev)
 	struct iommu_group *group = dev->iommu_group;
 	struct iommu_device *tmp_device, *device = NULL;
 
+	pr_info("Removing device %s from group %d\n", dev_name(dev), group->id);
+
 	/* Pre-notify listeners that a device is being removed. */
 	blocking_notifier_call_chain(&group->notifier,
 				     IOMMU_GROUP_NOTIFY_DEL_DEVICE, dev);
@@ -417,6 +490,17 @@ void iommu_group_remove_device(struct device *dev)
 }
 EXPORT_SYMBOL_GPL(iommu_group_remove_device);
 
+static int iommu_group_device_count(struct iommu_group *group)
+{
+	struct iommu_device *entry;
+	int ret = 0;
+
+	list_for_each_entry(entry, &group->devices, list)
+		ret++;
+
+	return ret;
+}
+
 /**
  * iommu_group_for_each_dev - iterate over each device in the group
  * @group: the group
@@ -428,19 +512,30 @@ EXPORT_SYMBOL_GPL(iommu_group_remove_device);
  * The group->mutex is held across callbacks, which will block calls to
  * iommu_group_add/remove_device.
  */
-int iommu_group_for_each_dev(struct iommu_group *group, void *data,
-			     int (*fn)(struct device *, void *))
+static int __iommu_group_for_each_dev(struct iommu_group *group, void *data,
+				      int (*fn)(struct device *, void *))
 {
 	struct iommu_device *device;
 	int ret = 0;
 
-	mutex_lock(&group->mutex);
 	list_for_each_entry(device, &group->devices, list) {
 		ret = fn(device->dev, data);
 		if (ret)
 			break;
 	}
+	return ret;
+}
+
+
+int iommu_group_for_each_dev(struct iommu_group *group, void *data,
+			     int (*fn)(struct device *, void *))
+{
+	int ret;
+
+	mutex_lock(&group->mutex);
+	ret = __iommu_group_for_each_dev(group, data, fn);
 	mutex_unlock(&group->mutex);
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(iommu_group_for_each_dev);
@@ -692,7 +787,19 @@ static struct iommu_group *iommu_group_get_for_pci_dev(struct pci_dev *pdev)
 		return group;
 
 	/* No shared group found, allocate new */
-	return iommu_group_alloc();
+	group = iommu_group_alloc();
+	if (IS_ERR(group))
+		return NULL;
+
+	/*
+	 * Try to allocate a default domain - needs support from the
+	 * IOMMU driver.
+	 */
+	group->default_domain = __iommu_domain_alloc(pdev->dev.bus,
+						     IOMMU_DOMAIN_DMA);
+	group->domain = group->default_domain;
+
+	return group;
 }
 
 /**
@@ -731,17 +838,42 @@ struct iommu_group *iommu_group_get_for_dev(struct device *dev)
 	return group;
 }
 
+struct iommu_domain *iommu_group_default_domain(struct iommu_group *group)
+{
+	return group->default_domain;
+}
+
 static int add_iommu_group(struct device *dev, void *data)
 {
 	struct iommu_callback_data *cb = data;
 	const struct iommu_ops *ops = cb->ops;
+	int ret;
 
 	if (!ops->add_device)
 		return 0;
 
 	WARN_ON(dev->iommu_group);
 
-	ops->add_device(dev);
+	ret = ops->add_device(dev);
+
+	/*
+	 * We ignore -ENODEV errors for now, as they just mean that the
+	 * device is not translated by an IOMMU. We still care about
+	 * other errors and fail to initialize when they happen.
+	 */
+	if (ret == -ENODEV)
+		ret = 0;
+
+	return ret;
+}
+
+static int remove_iommu_group(struct device *dev, void *data)
+{
+	struct iommu_callback_data *cb = data;
+	const struct iommu_ops *ops = cb->ops;
+
+	if (ops->remove_device && dev->iommu_group)
+		ops->remove_device(dev);
 
 	return 0;
 }
@@ -761,7 +893,7 @@ static int iommu_bus_notifier(struct notifier_block *nb,
 	if (action == BUS_NOTIFY_ADD_DEVICE) {
 		if (ops->add_device)
 			return ops->add_device(dev);
-	} else if (action == BUS_NOTIFY_DEL_DEVICE) {
+	} else if (action == BUS_NOTIFY_REMOVED_DEVICE) {
 		if (ops->remove_device && dev->iommu_group) {
 			ops->remove_device(dev);
 			return 0;
@@ -814,19 +946,25 @@ static int iommu_bus_init(struct bus_type *bus, const struct iommu_ops *ops)
 	nb->notifier_call = iommu_bus_notifier;
 
 	err = bus_register_notifier(bus, nb);
-	if (err) {
-		kfree(nb);
-		return err;
-	}
+	if (err)
+		goto out_free;
 
 	err = bus_for_each_dev(bus, NULL, &cb, add_iommu_group);
-	if (err) {
-		bus_unregister_notifier(bus, nb);
-		kfree(nb);
-		return err;
-	}
+	if (err)
+		goto out_err;
+
 
 	return 0;
+
+out_err:
+	/* Clean up */
+	bus_for_each_dev(bus, NULL, &cb, remove_iommu_group);
+	bus_unregister_notifier(bus, nb);
+
+out_free:
+	kfree(nb);
+
+	return err;
 }
 
 /**
@@ -898,22 +1036,28 @@ void iommu_set_fault_handler(struct iommu_domain *domain,
 }
 EXPORT_SYMBOL_GPL(iommu_set_fault_handler);
 
-struct iommu_domain *iommu_domain_alloc(struct bus_type *bus)
+static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
+						 unsigned type)
 {
 	struct iommu_domain *domain;
 
 	if (bus == NULL || bus->iommu_ops == NULL)
 		return NULL;
 
-	domain = bus->iommu_ops->domain_alloc(IOMMU_DOMAIN_UNMANAGED);
+	domain = bus->iommu_ops->domain_alloc(type);
 	if (!domain)
 		return NULL;
 
 	domain->ops  = bus->iommu_ops;
-	domain->type = IOMMU_DOMAIN_UNMANAGED;
+	domain->type = type;
 
 	return domain;
 }
+
+struct iommu_domain *iommu_domain_alloc(struct bus_type *bus)
+{
+	return __iommu_domain_alloc(bus, IOMMU_DOMAIN_UNMANAGED);
+}
 EXPORT_SYMBOL_GPL(iommu_domain_alloc);
 
 void iommu_domain_free(struct iommu_domain *domain)
@@ -922,7 +1066,8 @@ void iommu_domain_free(struct iommu_domain *domain)
 }
 EXPORT_SYMBOL_GPL(iommu_domain_free);
 
-int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
+static int __iommu_attach_device(struct iommu_domain *domain,
+				 struct device *dev)
 {
 	int ret;
 	if (unlikely(domain->ops->attach_dev == NULL))
@@ -933,9 +1078,38 @@ int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
 		trace_attach_device_to_domain(dev);
 	return ret;
 }
+
+int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
+{
+	struct iommu_group *group;
+	int ret;
+
+	group = iommu_group_get(dev);
+	/* FIXME: Remove this when groups a mandatory for iommu drivers */
+	if (group == NULL)
+		return __iommu_attach_device(domain, dev);
+
+	/*
+	 * We have a group - lock it to make sure the device-count doesn't
+	 * change while we are attaching
+	 */
+	mutex_lock(&group->mutex);
+	ret = -EINVAL;
+	if (iommu_group_device_count(group) != 1)
+		goto out_unlock;
+
+	ret = __iommu_attach_group(domain, group);
+
+out_unlock:
+	mutex_unlock(&group->mutex);
+	iommu_group_put(group);
+
+	return ret;
+}
 EXPORT_SYMBOL_GPL(iommu_attach_device);
 
-void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
+static void __iommu_detach_device(struct iommu_domain *domain,
+				  struct device *dev)
 {
 	if (unlikely(domain->ops->detach_dev == NULL))
 		return;
@@ -943,8 +1117,48 @@ void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
 	domain->ops->detach_dev(domain, dev);
 	trace_detach_device_from_domain(dev);
 }
+
+void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
+{
+	struct iommu_group *group;
+
+	group = iommu_group_get(dev);
+	/* FIXME: Remove this when groups a mandatory for iommu drivers */
+	if (group == NULL)
+		return __iommu_detach_device(domain, dev);
+
+	mutex_lock(&group->mutex);
+	if (iommu_group_device_count(group) != 1) {
+		WARN_ON(1);
+		goto out_unlock;
+	}
+
+	__iommu_detach_group(domain, group);
+
+out_unlock:
+	mutex_unlock(&group->mutex);
+	iommu_group_put(group);
+}
 EXPORT_SYMBOL_GPL(iommu_detach_device);
 
+struct iommu_domain *iommu_get_domain_for_dev(struct device *dev)
+{
+	struct iommu_domain *domain;
+	struct iommu_group *group;
+
+	group = iommu_group_get(dev);
+	/* FIXME: Remove this when groups a mandatory for iommu drivers */
+	if (group == NULL)
+		return NULL;
+
+	domain = group->domain;
+
+	iommu_group_put(group);
+
+	return domain;
+}
+EXPORT_SYMBOL_GPL(iommu_get_domain_for_dev);
+
 /*
  * IOMMU groups are really the natrual working unit of the IOMMU, but
  * the IOMMU API works on domains and devices.  Bridge that gap by
@@ -959,13 +1173,34 @@ static int iommu_group_do_attach_device(struct device *dev, void *data)
 {
 	struct iommu_domain *domain = data;
 
-	return iommu_attach_device(domain, dev);
+	return __iommu_attach_device(domain, dev);
+}
+
+static int __iommu_attach_group(struct iommu_domain *domain,
+				struct iommu_group *group)
+{
+	int ret;
+
+	if (group->default_domain && group->domain != group->default_domain)
+		return -EBUSY;
+
+	ret = __iommu_group_for_each_dev(group, domain,
+					 iommu_group_do_attach_device);
+	if (ret == 0)
+		group->domain = domain;
+
+	return ret;
 }
 
 int iommu_attach_group(struct iommu_domain *domain, struct iommu_group *group)
 {
-	return iommu_group_for_each_dev(group, domain,
-					iommu_group_do_attach_device);
+	int ret;
+
+	mutex_lock(&group->mutex);
+	ret = __iommu_attach_group(domain, group);
+	mutex_unlock(&group->mutex);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(iommu_attach_group);
 
@@ -973,14 +1208,40 @@ static int iommu_group_do_detach_device(struct device *dev, void *data)
 {
 	struct iommu_domain *domain = data;
 
-	iommu_detach_device(domain, dev);
+	__iommu_detach_device(domain, dev);
 
 	return 0;
 }
 
+static void __iommu_detach_group(struct iommu_domain *domain,
+				 struct iommu_group *group)
+{
+	int ret;
+
+	if (!group->default_domain) {
+		__iommu_group_for_each_dev(group, domain,
+					   iommu_group_do_detach_device);
+		group->domain = NULL;
+		return;
+	}
+
+	if (group->domain == group->default_domain)
+		return;
+
+	/* Detach by re-attaching to the default domain */
+	ret = __iommu_group_for_each_dev(group, group->default_domain,
+					 iommu_group_do_attach_device);
+	if (ret != 0)
+		WARN_ON(1);
+	else
+		group->domain = group->default_domain;
+}
+
 void iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group)
 {
-	iommu_group_for_each_dev(group, domain, iommu_group_do_detach_device);
+	mutex_lock(&group->mutex);
+	__iommu_detach_group(domain, group);
+	mutex_unlock(&group->mutex);
 }
 EXPORT_SYMBOL_GPL(iommu_detach_group);
 
@@ -1207,7 +1468,7 @@ static int __init iommu_init(void)
 
 	return 0;
 }
-arch_initcall(iommu_init);
+core_initcall(iommu_init);
 
 int iommu_domain_get_attr(struct iommu_domain *domain,
 			  enum iommu_attr attr, void *data)
@@ -1273,3 +1534,72 @@ int iommu_domain_set_attr(struct iommu_domain *domain,
 	return ret;
 }
 EXPORT_SYMBOL_GPL(iommu_domain_set_attr);
+
+void iommu_get_dm_regions(struct device *dev, struct list_head *list)
+{
+	const struct iommu_ops *ops = dev->bus->iommu_ops;
+
+	if (ops && ops->get_dm_regions)
+		ops->get_dm_regions(dev, list);
+}
+
+void iommu_put_dm_regions(struct device *dev, struct list_head *list)
+{
+	const struct iommu_ops *ops = dev->bus->iommu_ops;
+
+	if (ops && ops->put_dm_regions)
+		ops->put_dm_regions(dev, list);
+}
+
+/* Request that a device is direct mapped by the IOMMU */
+int iommu_request_dm_for_dev(struct device *dev)
+{
+	struct iommu_domain *dm_domain;
+	struct iommu_group *group;
+	int ret;
+
+	/* Device must already be in a group before calling this function */
+	group = iommu_group_get_for_dev(dev);
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	mutex_lock(&group->mutex);
+
+	/* Check if the default domain is already direct mapped */
+	ret = 0;
+	if (group->default_domain &&
+	    group->default_domain->type == IOMMU_DOMAIN_IDENTITY)
+		goto out;
+
+	/* Don't change mappings of existing devices */
+	ret = -EBUSY;
+	if (iommu_group_device_count(group) != 1)
+		goto out;
+
+	/* Allocate a direct mapped domain */
+	ret = -ENOMEM;
+	dm_domain = __iommu_domain_alloc(dev->bus, IOMMU_DOMAIN_IDENTITY);
+	if (!dm_domain)
+		goto out;
+
+	/* Attach the device to the domain */
+	ret = __iommu_attach_group(dm_domain, group);
+	if (ret) {
+		iommu_domain_free(dm_domain);
+		goto out;
+	}
+
+	/* Make the direct mapped domain the default for this group */
+	if (group->default_domain)
+		iommu_domain_free(group->default_domain);
+	group->default_domain = dm_domain;
+
+	pr_info("Using direct mapping for device %s\n", dev_name(dev));
+
+	ret = 0;
+out:
+	mutex_unlock(&group->mutex);
+	iommu_group_put(group);
+
+	return ret;
+}
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 9dd820831..b7c3d923f 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -227,6 +227,7 @@ iova_insert_rbtree(struct rb_root *root, struct iova *iova)
 	/* Figure out where to put new node */
 	while (*new) {
 		struct iova *this = container_of(*new, struct iova, node);
+
 		parent = *new;
 
 		if (iova->pfn_lo < this->pfn_lo)
@@ -350,6 +351,7 @@ void
 free_iova(struct iova_domain *iovad, unsigned long pfn)
 {
 	struct iova *iova = find_iova(iovad, pfn);
+
 	if (iova)
 		__free_iova(iovad, iova);
 
@@ -369,6 +371,7 @@ void put_iova_domain(struct iova_domain *iovad)
 	node = rb_first(&iovad->rbroot);
 	while (node) {
 		struct iova *iova = container_of(node, struct iova, node);
+
 		rb_erase(node, &iovad->rbroot);
 		free_iova_mem(iova);
 		node = rb_first(&iovad->rbroot);
@@ -482,6 +485,7 @@ copy_reserved_iova(struct iova_domain *from, struct iova_domain *to)
 	for (node = rb_first(&from->rbroot); node; node = rb_next(node)) {
 		struct iova *iova = container_of(node, struct iova, node);
 		struct iova *new_iova;
+
 		new_iova = reserve_iova(to, iova->pfn_lo, iova->pfn_hi);
 		if (!new_iova)
 			printk(KERN_ERR "Reserve iova range %lx@%lx failed\n",
diff --git a/drivers/iommu/irq_remapping.c b/drivers/iommu/irq_remapping.c
index 390079ee1..2d9993062 100644
--- a/drivers/iommu/irq_remapping.c
+++ b/drivers/iommu/irq_remapping.c
@@ -6,6 +6,7 @@
 #include <linux/msi.h>
 #include <linux/irq.h>
 #include <linux/pci.h>
+#include <linux/irqdomain.h>
 
 #include <asm/hw_irq.h>
 #include <asm/irq_remapping.h>
@@ -21,21 +22,11 @@ int irq_remap_broken;
 int disable_sourceid_checking;
 int no_x2apic_optout;
 
+int disable_irq_post = 1;
+
 static int disable_irq_remap;
 static struct irq_remap_ops *remap_ops;
 
-static int msi_alloc_remapped_irq(struct pci_dev *pdev, int irq, int nvec);
-static int msi_setup_remapped_irq(struct pci_dev *pdev, unsigned int irq,
-				  int index, int sub_handle);
-static int set_remapped_irq_affinity(struct irq_data *data,
-				     const struct cpumask *mask,
-				     bool force);
-
-static bool irq_remapped(struct irq_cfg *cfg)
-{
-	return (cfg->remapped == 1);
-}
-
 static void irq_remapping_disable_io_apic(void)
 {
 	/*
@@ -49,117 +40,9 @@ static void irq_remapping_disable_io_apic(void)
 		disconnect_bsp_APIC(0);
 }
 
-static int do_setup_msi_irqs(struct pci_dev *dev, int nvec)
-{
-	int ret, sub_handle, nvec_pow2, index = 0;
-	unsigned int irq;
-	struct msi_desc *msidesc;
-
-	msidesc = list_entry(dev->msi_list.next, struct msi_desc, list);
-
-	irq = irq_alloc_hwirqs(nvec, dev_to_node(&dev->dev));
-	if (irq == 0)
-		return -ENOSPC;
-
-	nvec_pow2 = __roundup_pow_of_two(nvec);
-	for (sub_handle = 0; sub_handle < nvec; sub_handle++) {
-		if (!sub_handle) {
-			index = msi_alloc_remapped_irq(dev, irq, nvec_pow2);
-			if (index < 0) {
-				ret = index;
-				goto error;
-			}
-		} else {
-			ret = msi_setup_remapped_irq(dev, irq + sub_handle,
-						     index, sub_handle);
-			if (ret < 0)
-				goto error;
-		}
-		ret = setup_msi_irq(dev, msidesc, irq, sub_handle);
-		if (ret < 0)
-			goto error;
-	}
-	return 0;
-
-error:
-	irq_free_hwirqs(irq, nvec);
-
-	/*
-	 * Restore altered MSI descriptor fields and prevent just destroyed
-	 * IRQs from tearing down again in default_teardown_msi_irqs()
-	 */
-	msidesc->irq = 0;
-
-	return ret;
-}
-
-static int do_setup_msix_irqs(struct pci_dev *dev, int nvec)
-{
-	int node, ret, sub_handle, index = 0;
-	struct msi_desc *msidesc;
-	unsigned int irq;
-
-	node		= dev_to_node(&dev->dev);
-	sub_handle	= 0;
-
-	list_for_each_entry(msidesc, &dev->msi_list, list) {
-
-		irq = irq_alloc_hwirq(node);
-		if (irq == 0)
-			return -1;
-
-		if (sub_handle == 0)
-			ret = index = msi_alloc_remapped_irq(dev, irq, nvec);
-		else
-			ret = msi_setup_remapped_irq(dev, irq, index, sub_handle);
-
-		if (ret < 0)
-			goto error;
-
-		ret = setup_msi_irq(dev, msidesc, irq, 0);
-		if (ret < 0)
-			goto error;
-
-		sub_handle += 1;
-		irq        += 1;
-	}
-
-	return 0;
-
-error:
-	irq_free_hwirq(irq);
-	return ret;
-}
-
-static int irq_remapping_setup_msi_irqs(struct pci_dev *dev,
-					int nvec, int type)
-{
-	if (type == PCI_CAP_ID_MSI)
-		return do_setup_msi_irqs(dev, nvec);
-	else
-		return do_setup_msix_irqs(dev, nvec);
-}
-
-static void eoi_ioapic_pin_remapped(int apic, int pin, int vector)
-{
-	/*
-	 * Intr-remapping uses pin number as the virtual vector
-	 * in the RTE. Actual vector is programmed in
-	 * intr-remapping table entry. Hence for the io-apic
-	 * EOI we use the pin number.
-	 */
-	io_apic_eoi(apic, pin);
-}
-
 static void __init irq_remapping_modify_x86_ops(void)
 {
 	x86_io_apic_ops.disable		= irq_remapping_disable_io_apic;
-	x86_io_apic_ops.set_affinity	= set_remapped_irq_affinity;
-	x86_io_apic_ops.setup_entry	= setup_ioapic_remapped_entry;
-	x86_io_apic_ops.eoi_ioapic_pin	= eoi_ioapic_pin_remapped;
-	x86_msi.setup_msi_irqs		= irq_remapping_setup_msi_irqs;
-	x86_msi.setup_hpet_msi		= setup_hpet_msi_remapped;
-	x86_msi.compose_msi_msg		= compose_remapped_msi_msg;
 }
 
 static __init int setup_nointremap(char *str)
@@ -198,6 +81,15 @@ void set_irq_remapping_broken(void)
 	irq_remap_broken = 1;
 }
 
+bool irq_remapping_cap(enum irq_remap_cap cap)
+{
+	if (!remap_ops || disable_irq_post)
+		return 0;
+
+	return (remap_ops->capability & (1 << cap));
+}
+EXPORT_SYMBOL_GPL(irq_remapping_cap);
+
 int __init irq_remapping_prepare(void)
 {
 	if (disable_irq_remap)
@@ -254,113 +146,48 @@ int __init irq_remap_enable_fault_handling(void)
 	return remap_ops->enable_faulting();
 }
 
-int setup_ioapic_remapped_entry(int irq,
-				struct IO_APIC_route_entry *entry,
-				unsigned int destination, int vector,
-				struct io_apic_irq_attr *attr)
-{
-	if (!remap_ops->setup_ioapic_entry)
-		return -ENODEV;
-
-	return remap_ops->setup_ioapic_entry(irq, entry, destination,
-					     vector, attr);
-}
-
-static int set_remapped_irq_affinity(struct irq_data *data,
-				     const struct cpumask *mask, bool force)
-{
-	if (!config_enabled(CONFIG_SMP) || !remap_ops->set_affinity)
-		return 0;
-
-	return remap_ops->set_affinity(data, mask, force);
-}
-
-void free_remapped_irq(int irq)
-{
-	struct irq_cfg *cfg = irq_cfg(irq);
-
-	if (irq_remapped(cfg) && remap_ops->free_irq)
-		remap_ops->free_irq(irq);
-}
-
-void compose_remapped_msi_msg(struct pci_dev *pdev,
-			      unsigned int irq, unsigned int dest,
-			      struct msi_msg *msg, u8 hpet_id)
-{
-	struct irq_cfg *cfg = irq_cfg(irq);
-
-	if (!irq_remapped(cfg))
-		native_compose_msi_msg(pdev, irq, dest, msg, hpet_id);
-	else if (remap_ops->compose_msi_msg)
-		remap_ops->compose_msi_msg(pdev, irq, dest, msg, hpet_id);
-}
-
-static int msi_alloc_remapped_irq(struct pci_dev *pdev, int irq, int nvec)
-{
-	if (!remap_ops->msi_alloc_irq)
-		return -ENODEV;
-
-	return remap_ops->msi_alloc_irq(pdev, irq, nvec);
-}
-
-static int msi_setup_remapped_irq(struct pci_dev *pdev, unsigned int irq,
-				  int index, int sub_handle)
-{
-	if (!remap_ops->msi_setup_irq)
-		return -ENODEV;
-
-	return remap_ops->msi_setup_irq(pdev, irq, index, sub_handle);
-}
-
-int setup_hpet_msi_remapped(unsigned int irq, unsigned int id)
-{
-	int ret;
-
-	if (!remap_ops->alloc_hpet_msi)
-		return -ENODEV;
-
-	ret = remap_ops->alloc_hpet_msi(irq, id);
-	if (ret)
-		return -EINVAL;
-
-	return default_setup_hpet_msi(irq, id);
-}
-
 void panic_if_irq_remap(const char *msg)
 {
 	if (irq_remapping_enabled)
 		panic(msg);
 }
 
-static void ir_ack_apic_edge(struct irq_data *data)
+void ir_ack_apic_edge(struct irq_data *data)
 {
 	ack_APIC_irq();
 }
 
-static void ir_ack_apic_level(struct irq_data *data)
+/**
+ * irq_remapping_get_ir_irq_domain - Get the irqdomain associated with the IOMMU
+ *				     device serving request @info
+ * @info: interrupt allocation information, used to identify the IOMMU device
+ *
+ * It's used to get parent irqdomain for HPET and IOAPIC irqdomains.
+ * Returns pointer to IRQ domain, or NULL on failure.
+ */
+struct irq_domain *
+irq_remapping_get_ir_irq_domain(struct irq_alloc_info *info)
 {
-	ack_APIC_irq();
-	eoi_ioapic_irq(data->irq, irqd_cfg(data));
-}
+	if (!remap_ops || !remap_ops->get_ir_irq_domain)
+		return NULL;
 
-static void ir_print_prefix(struct irq_data *data, struct seq_file *p)
-{
-	seq_printf(p, " IR-%s", data->chip->name);
+	return remap_ops->get_ir_irq_domain(info);
 }
 
-void irq_remap_modify_chip_defaults(struct irq_chip *chip)
+/**
+ * irq_remapping_get_irq_domain - Get the irqdomain serving the request @info
+ * @info: interrupt allocation information, used to identify the IOMMU device
+ *
+ * There will be one PCI MSI/MSIX irqdomain associated with each interrupt
+ * remapping device, so this interface is used to retrieve the PCI MSI/MSIX
+ * irqdomain serving request @info.
+ * Returns pointer to IRQ domain, or NULL on failure.
+ */
+struct irq_domain *
+irq_remapping_get_irq_domain(struct irq_alloc_info *info)
 {
-	chip->irq_print_chip = ir_print_prefix;
-	chip->irq_ack = ir_ack_apic_edge;
-	chip->irq_eoi = ir_ack_apic_level;
-	chip->irq_set_affinity = x86_io_apic_ops.set_affinity;
-}
+	if (!remap_ops || !remap_ops->get_irq_domain)
+		return NULL;
 
-bool setup_remapped_irq(int irq, struct irq_cfg *cfg, struct irq_chip *chip)
-{
-	if (!irq_remapped(cfg))
-		return false;
-	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-	irq_remap_modify_chip_defaults(chip);
-	return true;
+	return remap_ops->get_irq_domain(info);
 }
diff --git a/drivers/iommu/irq_remapping.h b/drivers/iommu/irq_remapping.h
index 7c70cc29f..039c7af7b 100644
--- a/drivers/iommu/irq_remapping.h
+++ b/drivers/iommu/irq_remapping.h
@@ -24,19 +24,22 @@
 
 #ifdef CONFIG_IRQ_REMAP
 
-struct IO_APIC_route_entry;
-struct io_apic_irq_attr;
 struct irq_data;
-struct cpumask;
-struct pci_dev;
 struct msi_msg;
+struct irq_domain;
+struct irq_alloc_info;
 
 extern int irq_remap_broken;
 extern int disable_sourceid_checking;
 extern int no_x2apic_optout;
 extern int irq_remapping_enabled;
 
+extern int disable_irq_post;
+
 struct irq_remap_ops {
+	/* The supported capabilities */
+	int capability;
+
 	/* Initializes hardware and makes it ready for remapping interrupts */
 	int  (*prepare)(void);
 
@@ -52,40 +55,23 @@ struct irq_remap_ops {
 	/* Enable fault handling */
 	int  (*enable_faulting)(void);
 
-	/* IO-APIC setup routine */
-	int (*setup_ioapic_entry)(int irq, struct IO_APIC_route_entry *,
-				  unsigned int, int,
-				  struct io_apic_irq_attr *);
-
-	/* Set the CPU affinity of a remapped interrupt */
-	int (*set_affinity)(struct irq_data *data, const struct cpumask *mask,
-			    bool force);
-
-	/* Free an IRQ */
-	int (*free_irq)(int);
+	/* Get the irqdomain associated the IOMMU device */
+	struct irq_domain *(*get_ir_irq_domain)(struct irq_alloc_info *);
 
-	/* Create MSI msg to use for interrupt remapping */
-	void (*compose_msi_msg)(struct pci_dev *,
-				unsigned int, unsigned int,
-				struct msi_msg *, u8);
-
-	/* Allocate remapping resources for MSI */
-	int (*msi_alloc_irq)(struct pci_dev *, int, int);
-
-	/* Setup the remapped MSI irq */
-	int (*msi_setup_irq)(struct pci_dev *, unsigned int, int, int);
-
-	/* Setup interrupt remapping for an HPET MSI */
-	int (*alloc_hpet_msi)(unsigned int, unsigned int);
+	/* Get the MSI irqdomain associated with the IOMMU device */
+	struct irq_domain *(*get_irq_domain)(struct irq_alloc_info *);
 };
 
 extern struct irq_remap_ops intel_irq_remap_ops;
 extern struct irq_remap_ops amd_iommu_irq_ops;
 
+extern void ir_ack_apic_edge(struct irq_data *data);
+
 #else  /* CONFIG_IRQ_REMAP */
 
 #define irq_remapping_enabled 0
 #define irq_remap_broken      0
+#define disable_irq_post      1
 
 #endif /* CONFIG_IRQ_REMAP */
 
diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c
index cab214544..ebf0adb8e 100644
--- a/drivers/iommu/rockchip-iommu.c
+++ b/drivers/iommu/rockchip-iommu.c
@@ -551,6 +551,15 @@ static void rk_iommu_zap_iova(struct rk_iommu_domain *rk_domain,
 	spin_unlock_irqrestore(&rk_domain->iommus_lock, flags);
 }
 
+static void rk_iommu_zap_iova_first_last(struct rk_iommu_domain *rk_domain,
+					 dma_addr_t iova, size_t size)
+{
+	rk_iommu_zap_iova(rk_domain, iova, SPAGE_SIZE);
+	if (size > SPAGE_SIZE)
+		rk_iommu_zap_iova(rk_domain, iova + size - SPAGE_SIZE,
+					SPAGE_SIZE);
+}
+
 static u32 *rk_dte_get_page_table(struct rk_iommu_domain *rk_domain,
 				  dma_addr_t iova)
 {
@@ -575,12 +584,6 @@ static u32 *rk_dte_get_page_table(struct rk_iommu_domain *rk_domain,
 	rk_table_flush(page_table, NUM_PT_ENTRIES);
 	rk_table_flush(dte_addr, 1);
 
-	/*
-	 * Zap the first iova of newly allocated page table so iommu evicts
-	 * old cached value of new dte from the iotlb.
-	 */
-	rk_iommu_zap_iova(rk_domain, iova, SPAGE_SIZE);
-
 done:
 	pt_phys = rk_dte_pt_address(dte);
 	return (u32 *)phys_to_virt(pt_phys);
@@ -630,6 +633,14 @@ static int rk_iommu_map_iova(struct rk_iommu_domain *rk_domain, u32 *pte_addr,
 
 	rk_table_flush(pte_addr, pte_count);
 
+	/*
+	 * Zap the first and last iova to evict from iotlb any previously
+	 * mapped cachelines holding stale values for its dte and pte.
+	 * We only zap the first and last iova, since only they could have
+	 * dte or pte shared with an existing mapping.
+	 */
+	rk_iommu_zap_iova_first_last(rk_domain, iova, size);
+
 	return 0;
 unwind:
 	/* Unmap the range of iovas that we just mapped */
@@ -774,7 +785,7 @@ static int rk_iommu_attach_device(struct iommu_domain *domain,
 	list_add_tail(&iommu->node, &rk_domain->iommus);
 	spin_unlock_irqrestore(&rk_domain->iommus_lock, flags);
 
-	dev_info(dev, "Attached to iommu domain\n");
+	dev_dbg(dev, "Attached to iommu domain\n");
 
 	rk_iommu_disable_stall(iommu);
 
@@ -808,7 +819,7 @@ static void rk_iommu_detach_device(struct iommu_domain *domain,
 
 	iommu->domain = NULL;
 
-	dev_info(dev, "Detached from iommu domain\n");
+	dev_dbg(dev, "Detached from iommu domain\n");
 }
 
 static struct iommu_domain *rk_iommu_domain_alloc(unsigned type)
diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
index c845d99ec..c1f2e521d 100644
--- a/drivers/iommu/tegra-smmu.c
+++ b/drivers/iommu/tegra-smmu.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/bitops.h>
+#include <linux/debugfs.h>
 #include <linux/err.h>
 #include <linux/iommu.h>
 #include <linux/kernel.h>
@@ -31,6 +32,8 @@ struct tegra_smmu {
 	struct mutex lock;
 
 	struct list_head list;
+
+	struct dentry *debugfs;
 };
 
 struct tegra_smmu_as {
@@ -673,6 +676,103 @@ static void tegra_smmu_ahb_enable(void)
 	}
 }
 
+static int tegra_smmu_swgroups_show(struct seq_file *s, void *data)
+{
+	struct tegra_smmu *smmu = s->private;
+	unsigned int i;
+	u32 value;
+
+	seq_printf(s, "swgroup    enabled  ASID\n");
+	seq_printf(s, "------------------------\n");
+
+	for (i = 0; i < smmu->soc->num_swgroups; i++) {
+		const struct tegra_smmu_swgroup *group = &smmu->soc->swgroups[i];
+		const char *status;
+		unsigned int asid;
+
+		value = smmu_readl(smmu, group->reg);
+
+		if (value & SMMU_ASID_ENABLE)
+			status = "yes";
+		else
+			status = "no";
+
+		asid = value & SMMU_ASID_MASK;
+
+		seq_printf(s, "%-9s  %-7s  %#04x\n", group->name, status,
+			   asid);
+	}
+
+	return 0;
+}
+
+static int tegra_smmu_swgroups_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, tegra_smmu_swgroups_show, inode->i_private);
+}
+
+static const struct file_operations tegra_smmu_swgroups_fops = {
+	.open = tegra_smmu_swgroups_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static int tegra_smmu_clients_show(struct seq_file *s, void *data)
+{
+	struct tegra_smmu *smmu = s->private;
+	unsigned int i;
+	u32 value;
+
+	seq_printf(s, "client       enabled\n");
+	seq_printf(s, "--------------------\n");
+
+	for (i = 0; i < smmu->soc->num_clients; i++) {
+		const struct tegra_mc_client *client = &smmu->soc->clients[i];
+		const char *status;
+
+		value = smmu_readl(smmu, client->smmu.reg);
+
+		if (value & BIT(client->smmu.bit))
+			status = "yes";
+		else
+			status = "no";
+
+		seq_printf(s, "%-12s %s\n", client->name, status);
+	}
+
+	return 0;
+}
+
+static int tegra_smmu_clients_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, tegra_smmu_clients_show, inode->i_private);
+}
+
+static const struct file_operations tegra_smmu_clients_fops = {
+	.open = tegra_smmu_clients_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static void tegra_smmu_debugfs_init(struct tegra_smmu *smmu)
+{
+	smmu->debugfs = debugfs_create_dir("smmu", NULL);
+	if (!smmu->debugfs)
+		return;
+
+	debugfs_create_file("swgroups", S_IRUGO, smmu->debugfs, smmu,
+			    &tegra_smmu_swgroups_fops);
+	debugfs_create_file("clients", S_IRUGO, smmu->debugfs, smmu,
+			    &tegra_smmu_clients_fops);
+}
+
+static void tegra_smmu_debugfs_exit(struct tegra_smmu *smmu)
+{
+	debugfs_remove_recursive(smmu->debugfs);
+}
+
 struct tegra_smmu *tegra_smmu_probe(struct device *dev,
 				    const struct tegra_smmu_soc *soc,
 				    struct tegra_mc *mc)
@@ -743,5 +843,14 @@ struct tegra_smmu *tegra_smmu_probe(struct device *dev,
 	if (err < 0)
 		return ERR_PTR(err);
 
+	if (IS_ENABLED(CONFIG_DEBUG_FS))
+		tegra_smmu_debugfs_init(smmu);
+
 	return smmu;
 }
+
+void tegra_smmu_remove(struct tegra_smmu *smmu)
+{
+	if (IS_ENABLED(CONFIG_DEBUG_FS))
+		tegra_smmu_debugfs_exit(smmu);
+}