Resguardo inicial

linux-7.2.3/drivers/iommu/intel/Kconfig

@@ -0,0 +1,103 @@
+# SPDX-License-Identifier: GPL-2.0-only
+# Intel IOMMU support
+config DMAR_TABLE
+	bool
+
+config DMAR_PERF
+	bool
+
+config DMAR_DEBUG
+	bool
+
+config INTEL_IOMMU
+	bool "Support for Intel IOMMU using DMA Remapping Devices"
+	depends on PCI_MSI && ACPI && X86
+	select IOMMU_API
+	select GENERIC_PT
+	select IOMMU_PT
+	select IOMMU_PT_X86_64
+	select IOMMU_PT_VTDSS
+	select IOMMU_IOVA
+	select IOMMU_IOPF
+	select IOMMUFD_DRIVER if IOMMUFD
+	select NEED_DMA_MAP_STATE
+	select DMAR_TABLE
+	select SWIOTLB
+	select PCI_ATS
+	select PCI_PRI
+	select PCI_PASID
+	help
+	  DMA remapping (DMAR) devices support enables independent address
+	  translations for Direct Memory Access (DMA) from devices.
+	  These DMA remapping devices are reported via ACPI tables
+	  and include PCI device scope covered by these DMA
+	  remapping devices.
+
+if INTEL_IOMMU
+
+config INTEL_IOMMU_DEBUGFS
+	bool "Export Intel IOMMU internals in Debugfs"
+	depends on IOMMU_DEBUGFS
+	select DMAR_PERF
+	select DMAR_DEBUG
+	help
+	  !!!WARNING!!!
+
+	  DO NOT ENABLE THIS OPTION UNLESS YOU REALLY KNOW WHAT YOU ARE DOING!!!
+
+	  Expose Intel IOMMU internals in Debugfs.
+
+	  This option is -NOT- intended for production environments, and should
+	  only be enabled for debugging Intel IOMMU.
+
+config INTEL_IOMMU_SVM
+	bool "Support for Shared Virtual Memory with Intel IOMMU"
+	depends on X86_64
+	select MMU_NOTIFIER
+	select IOMMU_SVA
+	help
+	  Shared Virtual Memory (SVM) provides a facility for devices
+	  to access DMA resources through process address space by
+	  means of a Process Address Space ID (PASID).
+
+config INTEL_IOMMU_DEFAULT_ON
+	bool "Enable Intel DMA Remapping Devices by default"
+	default y
+	help
+	  Selecting this option will enable a DMAR device at boot time if
+	  one is found. If this option is not selected, DMAR support can
+	  be enabled by passing intel_iommu=on to the kernel.
+
+config INTEL_IOMMU_FLOPPY_WA
+	def_bool y
+	depends on X86 && BLK_DEV_FD
+	help
+	  Floppy disk drivers are known to bypass DMA API calls
+	  thereby failing to work when IOMMU is enabled. This
+	  workaround will setup a 1:1 mapping for the first
+	  16MiB to make floppy (an ISA device) work.
+
+config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON
+	bool "Enable Intel IOMMU scalable mode by default"
+	default y
+	help
+	  Selecting this option will enable by default the scalable mode if
+	  hardware presents the capability. The scalable mode is defined in
+	  VT-d 3.0. The scalable mode capability could be checked by reading
+	  /sys/devices/virtual/iommu/dmar*/intel-iommu/ecap. If this option
+	  is not selected, scalable mode support could also be enabled by
+	  passing intel_iommu=sm_on to the kernel. If not sure, please use
+	  the default value.
+
+config INTEL_IOMMU_PERF_EVENTS
+	bool "Intel IOMMU performance events"
+	default y
+	depends on INTEL_IOMMU && PERF_EVENTS
+	help
+	  Selecting this option will enable the performance monitoring
+	  infrastructure in the Intel IOMMU. It collects information about
+	  key events occurring during operation of the remapping hardware,
+	  to aid performance tuning and debug. These are available on modern
+	  processors which support Intel VT-d 4.0 and later.
+
+endif # INTEL_IOMMU

linux-7.2.3/drivers/iommu/intel/Makefile

@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-y += iommu.o pasid.o nested.o cache.o prq.o
+obj-$(CONFIG_DMAR_TABLE) += dmar.o trace.o
+obj-$(CONFIG_DMAR_PERF) += perf.o
+obj-$(CONFIG_INTEL_IOMMU_DEBUGFS) += debugfs.o
+obj-$(CONFIG_INTEL_IOMMU_SVM) += svm.o
+obj-$(CONFIG_IRQ_REMAP) += irq_remapping.o
+obj-$(CONFIG_INTEL_IOMMU_PERF_EVENTS) += perfmon.o

linux-7.2.3/drivers/iommu/intel/cache.c

@@ -0,0 +1,521 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * cache.c - Intel VT-d cache invalidation
+ *
+ * Copyright (C) 2024 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+#define pr_fmt(fmt)	"DMAR: " fmt
+
+#include <linux/dmar.h>
+#include <linux/iommu.h>
+#include <linux/memory.h>
+#include <linux/pci.h>
+#include <linux/spinlock.h>
+
+#include "iommu.h"
+#include "pasid.h"
+#include "trace.h"
+
+/* Check if an existing cache tag can be reused for a new association. */
+static bool cache_tage_match(struct cache_tag *tag, u16 domain_id,
+			     struct intel_iommu *iommu, struct device *dev,
+			     ioasid_t pasid, enum cache_tag_type type)
+{
+	if (tag->type != type)
+		return false;
+
+	if (tag->domain_id != domain_id || tag->pasid != pasid)
+		return false;
+
+	if (type == CACHE_TAG_IOTLB || type == CACHE_TAG_NESTING_IOTLB)
+		return tag->iommu == iommu;
+
+	if (type == CACHE_TAG_DEVTLB || type == CACHE_TAG_NESTING_DEVTLB)
+		return tag->dev == dev;
+
+	return false;
+}
+
+/* Assign a cache tag with specified type to domain. */
+int cache_tag_assign(struct dmar_domain *domain, u16 did, struct device *dev,
+		     ioasid_t pasid, enum cache_tag_type type)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	struct cache_tag *tag, *temp;
+	struct list_head *prev;
+	unsigned long flags;
+
+	tag = kzalloc_obj(*tag);
+	if (!tag)
+		return -ENOMEM;
+
+	tag->type = type;
+	tag->iommu = iommu;
+	tag->domain_id = did;
+	tag->pasid = pasid;
+	tag->users = 1;
+
+	if (type == CACHE_TAG_DEVTLB || type == CACHE_TAG_NESTING_DEVTLB)
+		tag->dev = dev;
+	else
+		tag->dev = iommu->iommu.dev;
+
+	spin_lock_irqsave(&domain->cache_lock, flags);
+	prev = &domain->cache_tags;
+	list_for_each_entry(temp, &domain->cache_tags, node) {
+		if (cache_tage_match(temp, did, iommu, dev, pasid, type)) {
+			temp->users++;
+			spin_unlock_irqrestore(&domain->cache_lock, flags);
+			kfree(tag);
+			trace_cache_tag_assign(temp);
+			return 0;
+		}
+		if (temp->iommu == iommu)
+			prev = &temp->node;
+	}
+	/*
+	 * Link cache tags of same iommu unit together, so corresponding
+	 * flush ops can be batched for iommu unit.
+	 */
+	list_add(&tag->node, prev);
+
+	spin_unlock_irqrestore(&domain->cache_lock, flags);
+	trace_cache_tag_assign(tag);
+
+	return 0;
+}
+
+/* Unassign a cache tag with specified type from domain. */
+static void cache_tag_unassign(struct dmar_domain *domain, u16 did,
+			       struct device *dev, ioasid_t pasid,
+			       enum cache_tag_type type)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	struct cache_tag *tag;
+	unsigned long flags;
+
+	spin_lock_irqsave(&domain->cache_lock, flags);
+	list_for_each_entry(tag, &domain->cache_tags, node) {
+		if (cache_tage_match(tag, did, iommu, dev, pasid, type)) {
+			trace_cache_tag_unassign(tag);
+			if (--tag->users == 0) {
+				list_del(&tag->node);
+				kfree(tag);
+			}
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&domain->cache_lock, flags);
+}
+
+/* domain->qi_batch will be freed in iommu_free_domain() path. */
+static int domain_qi_batch_alloc(struct dmar_domain *domain)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&domain->cache_lock, flags);
+	if (domain->qi_batch)
+		goto out_unlock;
+
+	domain->qi_batch = kzalloc_obj(*domain->qi_batch, GFP_ATOMIC);
+	if (!domain->qi_batch)
+		ret = -ENOMEM;
+out_unlock:
+	spin_unlock_irqrestore(&domain->cache_lock, flags);
+
+	return ret;
+}
+
+static int __cache_tag_assign_domain(struct dmar_domain *domain, u16 did,
+				     struct device *dev, ioasid_t pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	int ret;
+
+	ret = domain_qi_batch_alloc(domain);
+	if (ret)
+		return ret;
+
+	ret = cache_tag_assign(domain, did, dev, pasid, CACHE_TAG_IOTLB);
+	if (ret || !info->ats_enabled)
+		return ret;
+
+	ret = cache_tag_assign(domain, did, dev, pasid, CACHE_TAG_DEVTLB);
+	if (ret)
+		cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_IOTLB);
+
+	return ret;
+}
+
+static void __cache_tag_unassign_domain(struct dmar_domain *domain, u16 did,
+					struct device *dev, ioasid_t pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+
+	cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_IOTLB);
+
+	if (info->ats_enabled)
+		cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_DEVTLB);
+}
+
+static int __cache_tag_assign_parent_domain(struct dmar_domain *domain, u16 did,
+					    struct device *dev, ioasid_t pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	int ret;
+
+	ret = domain_qi_batch_alloc(domain);
+	if (ret)
+		return ret;
+
+	ret = cache_tag_assign(domain, did, dev, pasid, CACHE_TAG_NESTING_IOTLB);
+	if (ret || !info->ats_enabled)
+		return ret;
+
+	ret = cache_tag_assign(domain, did, dev, pasid, CACHE_TAG_NESTING_DEVTLB);
+	if (ret)
+		cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_NESTING_IOTLB);
+
+	return ret;
+}
+
+static void __cache_tag_unassign_parent_domain(struct dmar_domain *domain, u16 did,
+					       struct device *dev, ioasid_t pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+
+	cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_NESTING_IOTLB);
+
+	if (info->ats_enabled)
+		cache_tag_unassign(domain, did, dev, pasid, CACHE_TAG_NESTING_DEVTLB);
+}
+
+static u16 domain_get_id_for_dev(struct dmar_domain *domain, struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+
+	/*
+	 * The driver assigns different domain IDs for all domains except
+	 * the SVA type.
+	 */
+	if (domain->domain.type == IOMMU_DOMAIN_SVA)
+		return FLPT_DEFAULT_DID;
+
+	return domain_id_iommu(domain, iommu);
+}
+
+/*
+ * Assign cache tags to a domain when it's associated with a device's
+ * PASID using a specific domain ID.
+ *
+ * On success (return value of 0), cache tags are created and added to the
+ * domain's cache tag list. On failure (negative return value), an error
+ * code is returned indicating the reason for the failure.
+ */
+int cache_tag_assign_domain(struct dmar_domain *domain,
+			    struct device *dev, ioasid_t pasid)
+{
+	u16 did = domain_get_id_for_dev(domain, dev);
+	int ret;
+
+	ret = __cache_tag_assign_domain(domain, did, dev, pasid);
+	if (ret || domain->domain.type != IOMMU_DOMAIN_NESTED)
+		return ret;
+
+	ret = __cache_tag_assign_parent_domain(domain->s2_domain, did, dev, pasid);
+	if (ret)
+		__cache_tag_unassign_domain(domain, did, dev, pasid);
+
+	return ret;
+}
+
+/*
+ * Remove the cache tags associated with a device's PASID when the domain is
+ * detached from the device.
+ *
+ * The cache tags must be previously assigned to the domain by calling the
+ * assign interface.
+ */
+void cache_tag_unassign_domain(struct dmar_domain *domain,
+			       struct device *dev, ioasid_t pasid)
+{
+	u16 did = domain_get_id_for_dev(domain, dev);
+
+	__cache_tag_unassign_domain(domain, did, dev, pasid);
+	if (domain->domain.type == IOMMU_DOMAIN_NESTED)
+		__cache_tag_unassign_parent_domain(domain->s2_domain, did, dev, pasid);
+}
+
+static unsigned long calculate_psi_aligned_address(unsigned long start,
+						   unsigned long last,
+						   unsigned long *size_order)
+{
+	unsigned int sz_lg2;
+
+	/* Compute a sz_lg2 that spans start and last */
+	start &= GENMASK(BITS_PER_LONG - 1, VTD_PAGE_SHIFT);
+	sz_lg2 = fls_long(start ^ last);
+	if (sz_lg2 <= 12) {
+		*size_order = 0;
+		return start;
+	}
+	if (unlikely(sz_lg2 >= BITS_PER_LONG)) {
+		/*
+		 * MAX_AGAW_PFN_WIDTH triggers full invalidation in all
+		 * downstream users.
+		 */
+		*size_order = MAX_AGAW_PFN_WIDTH;
+		return 0;
+	}
+
+	*size_order = sz_lg2 - VTD_PAGE_SHIFT;
+	return start & GENMASK(BITS_PER_LONG - 1, sz_lg2);
+}
+
+static void qi_batch_flush_descs(struct intel_iommu *iommu, struct qi_batch *batch)
+{
+	if (!iommu || !batch->index)
+		return;
+
+	qi_submit_sync(iommu, batch->descs, batch->index, 0);
+
+	/* Reset the index value and clean the whole batch buffer. */
+	memset(batch, 0, sizeof(*batch));
+}
+
+static void qi_batch_increment_index(struct intel_iommu *iommu, struct qi_batch *batch)
+{
+	if (++batch->index == QI_MAX_BATCHED_DESC_COUNT)
+		qi_batch_flush_descs(iommu, batch);
+}
+
+static void qi_batch_add_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
+			       unsigned int size_order, u64 type,
+			       struct qi_batch *batch)
+{
+	qi_desc_iotlb(iommu, did, addr, size_order, type, &batch->descs[batch->index]);
+	qi_batch_increment_index(iommu, batch);
+}
+
+static void qi_batch_add_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 pfsid,
+				   u16 qdep, u64 addr, unsigned int mask,
+				   struct qi_batch *batch)
+{
+	/*
+	 * According to VT-d spec, software is recommended to not submit any Device-TLB
+	 * invalidation requests while address remapping hardware is disabled.
+	 */
+	if (!(iommu->gcmd & DMA_GCMD_TE))
+		return;
+
+	qi_desc_dev_iotlb(sid, pfsid, qdep, addr, mask, &batch->descs[batch->index]);
+	qi_batch_increment_index(iommu, batch);
+}
+
+static void qi_batch_add_piotlb_all(struct intel_iommu *iommu, u16 did,
+				    u32 pasid, struct qi_batch *batch)
+{
+	qi_desc_piotlb_all(did, pasid, &batch->descs[batch->index]);
+	qi_batch_increment_index(iommu, batch);
+}
+
+static void qi_batch_add_piotlb(struct intel_iommu *iommu, u16 did, u32 pasid,
+				u64 addr, unsigned int size_order, bool ih,
+				struct qi_batch *batch)
+{
+	qi_desc_piotlb(did, pasid, addr, size_order, ih,
+		       &batch->descs[batch->index]);
+	qi_batch_increment_index(iommu, batch);
+}
+
+static void qi_batch_add_pasid_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 pfsid,
+					 u32 pasid,  u16 qdep, u64 addr,
+					 unsigned int size_order, struct qi_batch *batch)
+{
+	/*
+	 * According to VT-d spec, software is recommended to not submit any
+	 * Device-TLB invalidation requests while address remapping hardware
+	 * is disabled.
+	 */
+	if (!(iommu->gcmd & DMA_GCMD_TE))
+		return;
+
+	qi_desc_dev_iotlb_pasid(sid, pfsid, pasid, qdep, addr, size_order,
+				&batch->descs[batch->index]);
+	qi_batch_increment_index(iommu, batch);
+}
+
+static bool intel_domain_use_piotlb(struct dmar_domain *domain)
+{
+	return domain->domain.type == IOMMU_DOMAIN_SVA ||
+			domain->domain.type == IOMMU_DOMAIN_NESTED ||
+			intel_domain_is_fs_paging(domain);
+}
+
+static void cache_tag_flush_iotlb(struct dmar_domain *domain, struct cache_tag *tag,
+				  unsigned long addr, unsigned long mask, int ih)
+{
+	struct intel_iommu *iommu = tag->iommu;
+	u64 type = DMA_TLB_PSI_FLUSH;
+
+	if (intel_domain_use_piotlb(domain)) {
+		if (mask >= MAX_AGAW_PFN_WIDTH)
+			qi_batch_add_piotlb_all(iommu, tag->domain_id,
+						tag->pasid, domain->qi_batch);
+		else
+			qi_batch_add_piotlb(iommu, tag->domain_id, tag->pasid,
+					    addr, mask, ih, domain->qi_batch);
+		return;
+	}
+
+	/*
+	 * Fallback to domain selective flush if no PSI support or the size
+	 * is too big.
+	 */
+	if (!cap_pgsel_inv(iommu->cap) ||
+	    mask > cap_max_amask_val(iommu->cap)) {
+		addr = 0;
+		mask = 0;
+		ih = 0;
+		type = DMA_TLB_DSI_FLUSH;
+	}
+
+	if (ecap_qis(iommu->ecap))
+		qi_batch_add_iotlb(iommu, tag->domain_id, addr | ih, mask, type,
+				   domain->qi_batch);
+	else
+		__iommu_flush_iotlb(iommu, tag->domain_id, addr | ih, mask, type);
+}
+
+static void cache_tag_flush_devtlb_psi(struct dmar_domain *domain, struct cache_tag *tag,
+				       unsigned long addr, unsigned long mask)
+{
+	struct intel_iommu *iommu = tag->iommu;
+	struct device_domain_info *info;
+	u16 sid;
+
+	info = dev_iommu_priv_get(tag->dev);
+	sid = PCI_DEVID(info->bus, info->devfn);
+
+	if (tag->pasid == IOMMU_NO_PASID) {
+		qi_batch_add_dev_iotlb(iommu, sid, info->pfsid, info->ats_qdep,
+				       addr, mask, domain->qi_batch);
+		if (info->dtlb_extra_inval)
+			qi_batch_add_dev_iotlb(iommu, sid, info->pfsid, info->ats_qdep,
+					       addr, mask, domain->qi_batch);
+		return;
+	}
+
+	qi_batch_add_pasid_dev_iotlb(iommu, sid, info->pfsid, tag->pasid,
+				     info->ats_qdep, addr, mask, domain->qi_batch);
+	if (info->dtlb_extra_inval)
+		qi_batch_add_pasid_dev_iotlb(iommu, sid, info->pfsid, tag->pasid,
+					     info->ats_qdep, addr, mask,
+					     domain->qi_batch);
+}
+
+/*
+ * Invalidates a range of IOVA from @start (inclusive) to @end (inclusive)
+ * when the memory mappings in the target domain have been modified.
+ */
+void cache_tag_flush_range(struct dmar_domain *domain, unsigned long start,
+			   unsigned long end, int ih)
+{
+	struct intel_iommu *iommu = NULL;
+	unsigned long mask, addr;
+	struct cache_tag *tag;
+	unsigned long flags;
+
+	addr = calculate_psi_aligned_address(start, end, &mask);
+
+	spin_lock_irqsave(&domain->cache_lock, flags);
+	list_for_each_entry(tag, &domain->cache_tags, node) {
+		if (iommu && iommu != tag->iommu)
+			qi_batch_flush_descs(iommu, domain->qi_batch);
+		iommu = tag->iommu;
+
+		switch (tag->type) {
+		case CACHE_TAG_IOTLB:
+		case CACHE_TAG_NESTING_IOTLB:
+			cache_tag_flush_iotlb(domain, tag, addr, mask, ih);
+			break;
+		case CACHE_TAG_NESTING_DEVTLB:
+			/*
+			 * Address translation cache in device side caches the
+			 * result of nested translation. There is no easy way
+			 * to identify the exact set of nested translations
+			 * affected by a change in S2. So just flush the entire
+			 * device cache.
+			 */
+			addr = 0;
+			mask = MAX_AGAW_PFN_WIDTH;
+			fallthrough;
+		case CACHE_TAG_DEVTLB:
+			cache_tag_flush_devtlb_psi(domain, tag, addr, mask);
+			break;
+		}
+
+		trace_cache_tag_flush_range(tag, start, end, addr, mask);
+	}
+	qi_batch_flush_descs(iommu, domain->qi_batch);
+	spin_unlock_irqrestore(&domain->cache_lock, flags);
+}
+
+/*
+ * Invalidates all ranges of IOVA when the memory mappings in the target
+ * domain have been modified.
+ */
+void cache_tag_flush_all(struct dmar_domain *domain)
+{
+	cache_tag_flush_range(domain, 0, ULONG_MAX, 0);
+}
+
+/*
+ * Invalidate a range of IOVA when new mappings are created in the target
+ * domain.
+ *
+ * - VT-d spec, Section 6.1 Caching Mode: When the CM field is reported as
+ *   Set, any software updates to remapping structures other than first-
+ *   stage mapping requires explicit invalidation of the caches.
+ * - VT-d spec, Section 6.8 Write Buffer Flushing: For hardware that requires
+ *   write buffer flushing, software must explicitly perform write-buffer
+ *   flushing, if cache invalidation is not required.
+ */
+void cache_tag_flush_range_np(struct dmar_domain *domain, unsigned long start,
+			      unsigned long end)
+{
+	struct intel_iommu *iommu = NULL;
+	unsigned long mask, addr;
+	struct cache_tag *tag;
+	unsigned long flags;
+
+	addr = calculate_psi_aligned_address(start, end, &mask);
+
+	spin_lock_irqsave(&domain->cache_lock, flags);
+	list_for_each_entry(tag, &domain->cache_tags, node) {
+		if (iommu && iommu != tag->iommu)
+			qi_batch_flush_descs(iommu, domain->qi_batch);
+		iommu = tag->iommu;
+
+		if (!cap_caching_mode(iommu->cap) ||
+		    intel_domain_is_fs_paging(domain)) {
+			iommu_flush_write_buffer(iommu);
+			continue;
+		}
+
+		if (tag->type == CACHE_TAG_IOTLB ||
+		    tag->type == CACHE_TAG_NESTING_IOTLB)
+			cache_tag_flush_iotlb(domain, tag, addr, mask, 0);
+
+		trace_cache_tag_flush_range_np(tag, start, end, addr, mask);
+	}
+	qi_batch_flush_descs(iommu, domain->qi_batch);
+	spin_unlock_irqrestore(&domain->cache_lock, flags);
+}

linux-7.2.3/drivers/iommu/intel/debugfs.c

@@ -0,0 +1,815 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright © 2018 Intel Corporation.
+ *
+ * Authors: Gayatri Kammela <gayatri.kammela@intel.com>
+ *	    Sohil Mehta <sohil.mehta@intel.com>
+ *	    Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *	    Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+#include <linux/debugfs.h>
+#include <linux/dmar.h>
+#include <linux/pci.h>
+
+#include <asm/irq_remapping.h>
+
+#include "iommu.h"
+#include "pasid.h"
+#include "perf.h"
+
+struct tbl_walk {
+	u16 bus;
+	u16 devfn;
+	u32 pasid;
+	struct root_entry *rt_entry;
+	struct context_entry *ctx_entry;
+	struct pasid_entry *pasid_tbl_entry;
+};
+
+struct iommu_regset {
+	int offset;
+	const char *regs;
+};
+
+#define DEBUG_BUFFER_SIZE	1024
+static char debug_buf[DEBUG_BUFFER_SIZE];
+
+#define IOMMU_REGSET_ENTRY(_reg_)					\
+	{ DMAR_##_reg_##_REG, __stringify(_reg_) }
+
+static const struct iommu_regset iommu_regs_32[] = {
+	IOMMU_REGSET_ENTRY(VER),
+	IOMMU_REGSET_ENTRY(GCMD),
+	IOMMU_REGSET_ENTRY(GSTS),
+	IOMMU_REGSET_ENTRY(FSTS),
+	IOMMU_REGSET_ENTRY(FECTL),
+	IOMMU_REGSET_ENTRY(FEDATA),
+	IOMMU_REGSET_ENTRY(FEADDR),
+	IOMMU_REGSET_ENTRY(FEUADDR),
+	IOMMU_REGSET_ENTRY(PMEN),
+	IOMMU_REGSET_ENTRY(PLMBASE),
+	IOMMU_REGSET_ENTRY(PLMLIMIT),
+	IOMMU_REGSET_ENTRY(ICS),
+	IOMMU_REGSET_ENTRY(PRS),
+	IOMMU_REGSET_ENTRY(PECTL),
+	IOMMU_REGSET_ENTRY(PEDATA),
+	IOMMU_REGSET_ENTRY(PEADDR),
+	IOMMU_REGSET_ENTRY(PEUADDR),
+};
+
+static const struct iommu_regset iommu_regs_64[] = {
+	IOMMU_REGSET_ENTRY(CAP),
+	IOMMU_REGSET_ENTRY(ECAP),
+	IOMMU_REGSET_ENTRY(RTADDR),
+	IOMMU_REGSET_ENTRY(PHMBASE),
+	IOMMU_REGSET_ENTRY(PHMLIMIT),
+	IOMMU_REGSET_ENTRY(IQH),
+	IOMMU_REGSET_ENTRY(IQT),
+	IOMMU_REGSET_ENTRY(IQA),
+	IOMMU_REGSET_ENTRY(IRTA),
+	IOMMU_REGSET_ENTRY(PQH),
+	IOMMU_REGSET_ENTRY(PQT),
+	IOMMU_REGSET_ENTRY(PQA),
+	IOMMU_REGSET_ENTRY(MTRRCAP),
+	IOMMU_REGSET_ENTRY(MTRRDEF),
+	IOMMU_REGSET_ENTRY(MTRR_FIX64K_00000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX16K_80000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX16K_A0000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_C0000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_C8000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_D0000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_D8000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_E0000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_E8000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_F0000),
+	IOMMU_REGSET_ENTRY(MTRR_FIX4K_F8000),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE0),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK0),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE1),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK1),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE2),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK2),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE3),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK3),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE4),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK4),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE5),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK5),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE6),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK6),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE7),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK7),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE8),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK8),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSBASE9),
+	IOMMU_REGSET_ENTRY(MTRR_PHYSMASK9),
+};
+
+static struct dentry *intel_iommu_debug;
+
+static int iommu_regset_show(struct seq_file *m, void *unused)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	unsigned long flag;
+	int i, ret = 0;
+	u64 value;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd) {
+		if (!drhd->reg_base_addr) {
+			seq_puts(m, "IOMMU: Invalid base address\n");
+			ret = -EINVAL;
+			goto out;
+		}
+
+		seq_printf(m, "IOMMU: %s Register Base Address: %llx\n",
+			   iommu->name, drhd->reg_base_addr);
+		seq_puts(m, "Name\t\t\tOffset\t\tContents\n");
+		/*
+		 * Publish the contents of the 64-bit hardware registers
+		 * by adding the offset to the pointer (virtual address).
+		 */
+		raw_spin_lock_irqsave(&iommu->register_lock, flag);
+		for (i = 0 ; i < ARRAY_SIZE(iommu_regs_32); i++) {
+			value = readl(iommu->reg + iommu_regs_32[i].offset);
+			seq_printf(m, "%-16s\t0x%02x\t\t0x%016llx\n",
+				   iommu_regs_32[i].regs, iommu_regs_32[i].offset,
+				   value);
+		}
+		for (i = 0 ; i < ARRAY_SIZE(iommu_regs_64); i++) {
+			value = readq(iommu->reg + iommu_regs_64[i].offset);
+			seq_printf(m, "%-16s\t0x%02x\t\t0x%016llx\n",
+				   iommu_regs_64[i].regs, iommu_regs_64[i].offset,
+				   value);
+		}
+		raw_spin_unlock_irqrestore(&iommu->register_lock, flag);
+		seq_putc(m, '\n');
+	}
+out:
+	rcu_read_unlock();
+
+	return ret;
+}
+DEFINE_SHOW_ATTRIBUTE(iommu_regset);
+
+static inline void print_tbl_walk(struct seq_file *m)
+{
+	struct tbl_walk *tbl_wlk = m->private;
+
+	seq_printf(m, "%02x:%02x.%x\t0x%016llx:0x%016llx\t0x%016llx:0x%016llx\t",
+		   tbl_wlk->bus, PCI_SLOT(tbl_wlk->devfn),
+		   PCI_FUNC(tbl_wlk->devfn), tbl_wlk->rt_entry->hi,
+		   tbl_wlk->rt_entry->lo, tbl_wlk->ctx_entry->hi,
+		   tbl_wlk->ctx_entry->lo);
+
+	/*
+	 * A legacy mode DMAR doesn't support PASID, hence default it to -1
+	 * indicating that it's invalid. Also, default all PASID related fields
+	 * to 0.
+	 */
+	if (!tbl_wlk->pasid_tbl_entry)
+		seq_printf(m, "%-6d\t0x%016llx:0x%016llx:0x%016llx\n", -1,
+			   (u64)0, (u64)0, (u64)0);
+	else
+		seq_printf(m, "%-6d\t0x%016llx:0x%016llx:0x%016llx\n",
+			   tbl_wlk->pasid, tbl_wlk->pasid_tbl_entry->val[2],
+			   tbl_wlk->pasid_tbl_entry->val[1],
+			   tbl_wlk->pasid_tbl_entry->val[0]);
+}
+
+static void pasid_tbl_walk(struct seq_file *m, struct pasid_entry *tbl_entry,
+			   u16 dir_idx)
+{
+	struct tbl_walk *tbl_wlk = m->private;
+	u8 tbl_idx;
+
+	for (tbl_idx = 0; tbl_idx < PASID_TBL_ENTRIES; tbl_idx++) {
+		if (pasid_pte_is_present(tbl_entry)) {
+			tbl_wlk->pasid_tbl_entry = tbl_entry;
+			tbl_wlk->pasid = (dir_idx << PASID_PDE_SHIFT) + tbl_idx;
+			print_tbl_walk(m);
+		}
+
+		tbl_entry++;
+	}
+}
+
+static void pasid_dir_walk(struct seq_file *m, u64 pasid_dir_ptr,
+			   u16 pasid_dir_size)
+{
+	struct pasid_dir_entry *dir_entry = phys_to_virt(pasid_dir_ptr);
+	struct pasid_entry *pasid_tbl;
+	u16 dir_idx;
+
+	for (dir_idx = 0; dir_idx < pasid_dir_size; dir_idx++) {
+		pasid_tbl = get_pasid_table_from_pde(dir_entry);
+		if (pasid_tbl)
+			pasid_tbl_walk(m, pasid_tbl, dir_idx);
+
+		dir_entry++;
+	}
+}
+
+static void ctx_tbl_walk(struct seq_file *m, struct intel_iommu *iommu, u16 bus)
+{
+	struct context_entry *context;
+	u16 devfn, pasid_dir_size;
+	u64 pasid_dir_ptr;
+
+	for (devfn = 0; devfn < 256; devfn++) {
+		struct tbl_walk tbl_wlk = {0};
+
+		/*
+		 * Scalable mode root entry points to upper scalable mode
+		 * context table and lower scalable mode context table. Each
+		 * scalable mode context table has 128 context entries where as
+		 * legacy mode context table has 256 context entries. So in
+		 * scalable mode, the context entries for former 128 devices are
+		 * in the lower scalable mode context table, while the latter
+		 * 128 devices are in the upper scalable mode context table.
+		 * In scalable mode, when devfn > 127, iommu_context_addr()
+		 * automatically refers to upper scalable mode context table and
+		 * hence the caller doesn't have to worry about differences
+		 * between scalable mode and non scalable mode.
+		 */
+		context = iommu_context_addr(iommu, bus, devfn, 0);
+		if (!context)
+			return;
+
+		if (!context_present(context))
+			continue;
+
+		tbl_wlk.bus = bus;
+		tbl_wlk.devfn = devfn;
+		tbl_wlk.rt_entry = &iommu->root_entry[bus];
+		tbl_wlk.ctx_entry = context;
+		m->private = &tbl_wlk;
+
+		if (readq(iommu->reg + DMAR_RTADDR_REG) & DMA_RTADDR_SMT) {
+			pasid_dir_ptr = context->lo & VTD_PAGE_MASK;
+			pasid_dir_size = get_pasid_dir_size(context);
+			pasid_dir_walk(m, pasid_dir_ptr, pasid_dir_size);
+			continue;
+		}
+
+		print_tbl_walk(m);
+	}
+}
+
+static void root_tbl_walk(struct seq_file *m, struct intel_iommu *iommu)
+{
+	u16 bus;
+
+	spin_lock(&iommu->lock);
+	seq_printf(m, "IOMMU %s: Root Table Address: 0x%llx\n", iommu->name,
+		   (u64)virt_to_phys(iommu->root_entry));
+	seq_puts(m, "B.D.F\tRoot_entry\t\t\t\tContext_entry\t\t\t\tPASID\tPASID_table_entry\n");
+
+	/*
+	 * No need to check if the root entry is present or not because
+	 * iommu_context_addr() performs the same check before returning
+	 * context entry.
+	 */
+	for (bus = 0; bus < 256; bus++)
+		ctx_tbl_walk(m, iommu, bus);
+	spin_unlock(&iommu->lock);
+}
+
+static int dmar_translation_struct_show(struct seq_file *m, void *unused)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	u32 sts;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd) {
+		sts = readl(iommu->reg + DMAR_GSTS_REG);
+		if (!(sts & DMA_GSTS_TES)) {
+			seq_printf(m, "DMA Remapping is not enabled on %s\n",
+				   iommu->name);
+			continue;
+		}
+		root_tbl_walk(m, iommu);
+		seq_putc(m, '\n');
+	}
+	rcu_read_unlock();
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(dmar_translation_struct);
+
+static inline unsigned long level_to_directory_size(int level)
+{
+	return BIT_ULL(VTD_PAGE_SHIFT + VTD_STRIDE_SHIFT * (level - 1));
+}
+
+static inline void
+dump_page_info(struct seq_file *m, unsigned long iova, u64 *path)
+{
+	seq_printf(m, "0x%013lx |\t0x%016llx\t0x%016llx\t0x%016llx",
+		   iova >> VTD_PAGE_SHIFT, path[5], path[4], path[3]);
+	if (path[2]) {
+		seq_printf(m, "\t0x%016llx", path[2]);
+		if (path[1])
+			seq_printf(m, "\t0x%016llx", path[1]);
+	}
+	seq_putc(m, '\n');
+}
+
+static void pgtable_walk_level(struct seq_file *m, struct dma_pte *pde,
+			       int level, unsigned long start,
+			       u64 *path)
+{
+	int i;
+
+	if (level > 5 || level < 1)
+		return;
+
+	for (i = 0; i < BIT_ULL(VTD_STRIDE_SHIFT);
+			i++, pde++, start += level_to_directory_size(level)) {
+		if (!dma_pte_present(pde))
+			continue;
+
+		path[level] = pde->val;
+		if (dma_pte_superpage(pde) || level == 1)
+			dump_page_info(m, start, path);
+		else
+			pgtable_walk_level(m, phys_to_virt(dma_pte_addr(pde)),
+					   level - 1, start, path);
+		path[level] = 0;
+	}
+}
+
+static int domain_translation_struct_show(struct seq_file *m,
+					  struct device_domain_info *info,
+					  ioasid_t pasid)
+{
+	bool scalable, found = false;
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	u16 devfn, bus, seg;
+
+	bus = info->bus;
+	devfn = info->devfn;
+	seg = info->segment;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd) {
+		struct context_entry *context;
+		u64 pgd, path[6] = { 0 };
+		u32 sts, agaw;
+
+		if (seg != iommu->segment)
+			continue;
+
+		sts = readl(iommu->reg + DMAR_GSTS_REG);
+		if (!(sts & DMA_GSTS_TES)) {
+			seq_printf(m, "DMA Remapping is not enabled on %s\n",
+				   iommu->name);
+			continue;
+		}
+		if (readq(iommu->reg + DMAR_RTADDR_REG) & DMA_RTADDR_SMT)
+			scalable = true;
+		else
+			scalable = false;
+
+		/*
+		 * The iommu->lock is held across the callback, which will
+		 * block calls to domain_attach/domain_detach. Hence,
+		 * the domain of the device will not change during traversal.
+		 *
+		 * Traversing page table possibly races with the iommu_unmap()
+		 * interface. This could be solved by RCU-freeing the page
+		 * table pages in the iommu_unmap() path.
+		 */
+		spin_lock(&iommu->lock);
+
+		context = iommu_context_addr(iommu, bus, devfn, 0);
+		if (!context || !context_present(context))
+			goto iommu_unlock;
+
+		if (scalable) {	/* scalable mode */
+			struct pasid_entry *pasid_tbl, *pasid_tbl_entry;
+			struct pasid_dir_entry *dir_tbl, *dir_entry;
+			u16 dir_idx, tbl_idx, pgtt;
+			u64 pasid_dir_ptr;
+
+			pasid_dir_ptr = context->lo & VTD_PAGE_MASK;
+
+			/* Dump specified device domain mappings with PASID. */
+			dir_idx = pasid >> PASID_PDE_SHIFT;
+			tbl_idx = pasid & PASID_PTE_MASK;
+
+			dir_tbl = phys_to_virt(pasid_dir_ptr);
+			dir_entry = &dir_tbl[dir_idx];
+
+			pasid_tbl = get_pasid_table_from_pde(dir_entry);
+			if (!pasid_tbl)
+				goto iommu_unlock;
+
+			pasid_tbl_entry = &pasid_tbl[tbl_idx];
+			if (!pasid_pte_is_present(pasid_tbl_entry))
+				goto iommu_unlock;
+
+			/*
+			 * According to PASID Granular Translation Type(PGTT),
+			 * get the page table pointer.
+			 */
+			pgtt = (u16)(pasid_tbl_entry->val[0] & GENMASK_ULL(8, 6)) >> 6;
+			agaw = (u8)(pasid_tbl_entry->val[0] & GENMASK_ULL(4, 2)) >> 2;
+
+			switch (pgtt) {
+			case PASID_ENTRY_PGTT_FL_ONLY:
+				pgd = pasid_tbl_entry->val[2];
+				break;
+			case PASID_ENTRY_PGTT_SL_ONLY:
+			case PASID_ENTRY_PGTT_NESTED:
+				pgd = pasid_tbl_entry->val[0];
+				break;
+			default:
+				goto iommu_unlock;
+			}
+			pgd &= VTD_PAGE_MASK;
+		} else { /* legacy mode */
+			u8 tt = (u8)(context->lo & GENMASK_ULL(3, 2)) >> 2;
+
+			/*
+			 * According to Translation Type(TT),
+			 * get the page table pointer(SSPTPTR).
+			 */
+			switch (tt) {
+			case CONTEXT_TT_MULTI_LEVEL:
+			case CONTEXT_TT_DEV_IOTLB:
+				pgd = context->lo & VTD_PAGE_MASK;
+				agaw = context->hi & 7;
+				break;
+			default:
+				goto iommu_unlock;
+			}
+		}
+
+		seq_printf(m, "Device %04x:%02x:%02x.%x ",
+			   iommu->segment, bus, PCI_SLOT(devfn), PCI_FUNC(devfn));
+
+		if (scalable)
+			seq_printf(m, "with pasid %x @0x%llx\n", pasid, pgd);
+		else
+			seq_printf(m, "@0x%llx\n", pgd);
+
+		seq_printf(m, "%-17s\t%-18s\t%-18s\t%-18s\t%-18s\t%-s\n",
+			   "IOVA_PFN", "PML5E", "PML4E", "PDPE", "PDE", "PTE");
+		pgtable_walk_level(m, phys_to_virt(pgd), agaw + 2, 0, path);
+
+		found = true;
+iommu_unlock:
+		spin_unlock(&iommu->lock);
+		if (found)
+			break;
+	}
+	rcu_read_unlock();
+
+	return 0;
+}
+
+static int dev_domain_translation_struct_show(struct seq_file *m, void *unused)
+{
+	struct device_domain_info *info = (struct device_domain_info *)m->private;
+
+	return domain_translation_struct_show(m, info, IOMMU_NO_PASID);
+}
+DEFINE_SHOW_ATTRIBUTE(dev_domain_translation_struct);
+
+static int pasid_domain_translation_struct_show(struct seq_file *m, void *unused)
+{
+	struct dev_pasid_info *dev_pasid = (struct dev_pasid_info *)m->private;
+	struct device_domain_info *info = dev_iommu_priv_get(dev_pasid->dev);
+
+	return domain_translation_struct_show(m, info, dev_pasid->pasid);
+}
+DEFINE_SHOW_ATTRIBUTE(pasid_domain_translation_struct);
+
+static void invalidation_queue_entry_show(struct seq_file *m,
+					  struct intel_iommu *iommu)
+{
+	int index, shift = qi_shift(iommu);
+	struct qi_desc *desc;
+	int offset;
+
+	if (ecap_smts(iommu->ecap))
+		seq_puts(m, "Index\t\tqw0\t\t\tqw1\t\t\tqw2\t\t\tqw3\t\t\tstatus\n");
+	else
+		seq_puts(m, "Index\t\tqw0\t\t\tqw1\t\t\tstatus\n");
+
+	for (index = 0; index < QI_LENGTH; index++) {
+		offset = index << shift;
+		desc = iommu->qi->desc + offset;
+		if (ecap_smts(iommu->ecap))
+			seq_printf(m, "%5d\t%016llx\t%016llx\t%016llx\t%016llx\t%016x\n",
+				   index, desc->qw0, desc->qw1,
+				   desc->qw2, desc->qw3,
+				   iommu->qi->desc_status[index]);
+		else
+			seq_printf(m, "%5d\t%016llx\t%016llx\t%016x\n",
+				   index, desc->qw0, desc->qw1,
+				   iommu->qi->desc_status[index]);
+	}
+}
+
+static int invalidation_queue_show(struct seq_file *m, void *unused)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	unsigned long flags;
+	struct q_inval *qi;
+	int shift;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd) {
+		qi = iommu->qi;
+		shift = qi_shift(iommu);
+
+		if (!qi || !ecap_qis(iommu->ecap))
+			continue;
+
+		seq_printf(m, "Invalidation queue on IOMMU: %s\n", iommu->name);
+
+		raw_spin_lock_irqsave(&qi->q_lock, flags);
+		seq_printf(m, " Base: 0x%llx\tHead: %lld\tTail: %lld\n",
+			   (u64)virt_to_phys(qi->desc),
+			   readq(iommu->reg + DMAR_IQH_REG) >> shift,
+			   readq(iommu->reg + DMAR_IQT_REG) >> shift);
+		invalidation_queue_entry_show(m, iommu);
+		raw_spin_unlock_irqrestore(&qi->q_lock, flags);
+		seq_putc(m, '\n');
+	}
+	rcu_read_unlock();
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(invalidation_queue);
+
+#ifdef CONFIG_IRQ_REMAP
+static void ir_tbl_remap_entry_show(struct seq_file *m,
+				    struct intel_iommu *iommu)
+{
+	struct irte *ri_entry;
+	unsigned long flags;
+	int idx;
+
+	seq_puts(m, " Entry SrcID   DstID    Vct IRTE_high\t\tIRTE_low\n");
+
+	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
+	for (idx = 0; idx < INTR_REMAP_TABLE_ENTRIES; idx++) {
+		ri_entry = &iommu->ir_table->base[idx];
+		if (!ri_entry->present || ri_entry->p_pst)
+			continue;
+
+		seq_printf(m, " %-5d %02x:%02x.%01x %08x %02x  %016llx\t%016llx\n",
+			   idx, PCI_BUS_NUM(ri_entry->sid),
+			   PCI_SLOT(ri_entry->sid), PCI_FUNC(ri_entry->sid),
+			   ri_entry->dest_id, ri_entry->vector,
+			   ri_entry->high, ri_entry->low);
+	}
+	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+}
+
+static void ir_tbl_posted_entry_show(struct seq_file *m,
+				     struct intel_iommu *iommu)
+{
+	struct irte *pi_entry;
+	unsigned long flags;
+	int idx;
+
+	seq_puts(m, " Entry SrcID   PDA_high PDA_low  Vct IRTE_high\t\tIRTE_low\n");
+
+	raw_spin_lock_irqsave(&irq_2_ir_lock, flags);
+	for (idx = 0; idx < INTR_REMAP_TABLE_ENTRIES; idx++) {
+		pi_entry = &iommu->ir_table->base[idx];
+		if (!pi_entry->present || !pi_entry->p_pst)
+			continue;
+
+		seq_printf(m, " %-5d %02x:%02x.%01x %08x %08x %02x  %016llx\t%016llx\n",
+			   idx, PCI_BUS_NUM(pi_entry->sid),
+			   PCI_SLOT(pi_entry->sid), PCI_FUNC(pi_entry->sid),
+			   pi_entry->pda_h, pi_entry->pda_l << 6,
+			   pi_entry->vector, pi_entry->high,
+			   pi_entry->low);
+	}
+	raw_spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+}
+
+/*
+ * For active IOMMUs go through the Interrupt remapping
+ * table and print valid entries in a table format for
+ * Remapped and Posted Interrupts.
+ */
+static int ir_translation_struct_show(struct seq_file *m, void *unused)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	u64 irta;
+	u32 sts;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd) {
+		if (!ecap_ir_support(iommu->ecap))
+			continue;
+
+		seq_printf(m, "Remapped Interrupt supported on IOMMU: %s\n",
+			   iommu->name);
+
+		sts = readl(iommu->reg + DMAR_GSTS_REG);
+		if (iommu->ir_table && (sts & DMA_GSTS_IRES)) {
+			irta = virt_to_phys(iommu->ir_table->base);
+			seq_printf(m, " IR table address:%llx\n", irta);
+			ir_tbl_remap_entry_show(m, iommu);
+		} else {
+			seq_puts(m, "Interrupt Remapping is not enabled\n");
+		}
+		seq_putc(m, '\n');
+	}
+
+	seq_puts(m, "****\n\n");
+
+	for_each_active_iommu(iommu, drhd) {
+		if (!cap_pi_support(iommu->cap))
+			continue;
+
+		seq_printf(m, "Posted Interrupt supported on IOMMU: %s\n",
+			   iommu->name);
+
+		if (iommu->ir_table) {
+			irta = virt_to_phys(iommu->ir_table->base);
+			seq_printf(m, " IR table address:%llx\n", irta);
+			ir_tbl_posted_entry_show(m, iommu);
+		} else {
+			seq_puts(m, "Interrupt Remapping is not enabled\n");
+		}
+		seq_putc(m, '\n');
+	}
+	rcu_read_unlock();
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(ir_translation_struct);
+#endif
+
+static void latency_show_one(struct seq_file *m, struct intel_iommu *iommu,
+			     struct dmar_drhd_unit *drhd)
+{
+	seq_printf(m, "IOMMU: %s Register Base Address: %llx\n",
+		   iommu->name, drhd->reg_base_addr);
+
+	dmar_latency_snapshot(iommu, debug_buf, DEBUG_BUFFER_SIZE);
+	seq_printf(m, "%s\n", debug_buf);
+}
+
+static int latency_show(struct seq_file *m, void *v)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+
+	rcu_read_lock();
+	for_each_active_iommu(iommu, drhd)
+		latency_show_one(m, iommu, drhd);
+	rcu_read_unlock();
+
+	return 0;
+}
+
+static int dmar_perf_latency_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, latency_show, NULL);
+}
+
+static ssize_t dmar_perf_latency_write(struct file *filp,
+				       const char __user *ubuf,
+				       size_t cnt, loff_t *ppos)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	int counting;
+	char buf[64];
+
+	if (cnt > 63)
+		cnt = 63;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	buf[cnt] = 0;
+
+	if (kstrtoint(buf, 0, &counting))
+		return -EINVAL;
+
+	switch (counting) {
+	case 0:
+		rcu_read_lock();
+		for_each_active_iommu(iommu, drhd) {
+			dmar_latency_disable(iommu, DMAR_LATENCY_INV_IOTLB);
+			dmar_latency_disable(iommu, DMAR_LATENCY_INV_DEVTLB);
+			dmar_latency_disable(iommu, DMAR_LATENCY_INV_IEC);
+		}
+		rcu_read_unlock();
+		break;
+	case 1:
+		rcu_read_lock();
+		for_each_active_iommu(iommu, drhd)
+			dmar_latency_enable(iommu, DMAR_LATENCY_INV_IOTLB);
+		rcu_read_unlock();
+		break;
+	case 2:
+		rcu_read_lock();
+		for_each_active_iommu(iommu, drhd)
+			dmar_latency_enable(iommu, DMAR_LATENCY_INV_DEVTLB);
+		rcu_read_unlock();
+		break;
+	case 3:
+		rcu_read_lock();
+		for_each_active_iommu(iommu, drhd)
+			dmar_latency_enable(iommu, DMAR_LATENCY_INV_IEC);
+		rcu_read_unlock();
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	*ppos += cnt;
+	return cnt;
+}
+
+static const struct file_operations dmar_perf_latency_fops = {
+	.open		= dmar_perf_latency_open,
+	.write		= dmar_perf_latency_write,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+void __init intel_iommu_debugfs_init(void)
+{
+	intel_iommu_debug = debugfs_create_dir("intel", iommu_debugfs_dir);
+
+	debugfs_create_file("iommu_regset", 0444, intel_iommu_debug, NULL,
+			    &iommu_regset_fops);
+	debugfs_create_file("dmar_translation_struct", 0444, intel_iommu_debug,
+			    NULL, &dmar_translation_struct_fops);
+	debugfs_create_file("invalidation_queue", 0444, intel_iommu_debug,
+			    NULL, &invalidation_queue_fops);
+#ifdef CONFIG_IRQ_REMAP
+	debugfs_create_file("ir_translation_struct", 0444, intel_iommu_debug,
+			    NULL, &ir_translation_struct_fops);
+#endif
+	debugfs_create_file("dmar_perf_latency", 0644, intel_iommu_debug,
+			    NULL, &dmar_perf_latency_fops);
+}
+
+/*
+ * Create a debugfs directory for each device, and then create a
+ * debugfs file in this directory for users to dump the page table
+ * of the default domain. e.g.
+ * /sys/kernel/debug/iommu/intel/0000:00:01.0/domain_translation_struct
+ */
+void intel_iommu_debugfs_create_dev(struct device_domain_info *info)
+{
+	info->debugfs_dentry = debugfs_create_dir(dev_name(info->dev), intel_iommu_debug);
+
+	debugfs_create_file("domain_translation_struct", 0444, info->debugfs_dentry,
+			    info, &dev_domain_translation_struct_fops);
+}
+
+/* Remove the device debugfs directory. */
+void intel_iommu_debugfs_remove_dev(struct device_domain_info *info)
+{
+	debugfs_remove_recursive(info->debugfs_dentry);
+}
+
+/*
+ * Create a debugfs directory per pair of {device, pasid}, then create the
+ * corresponding debugfs file in this directory for users to dump its page
+ * table. e.g.
+ * /sys/kernel/debug/iommu/intel/0000:00:01.0/1/domain_translation_struct
+ *
+ * The debugfs only dumps the page tables whose mappings are created and
+ * destroyed by the iommu_map/unmap() interfaces. Check the mapping type
+ * of the domain before creating debugfs directory.
+ */
+void intel_iommu_debugfs_create_dev_pasid(struct dev_pasid_info *dev_pasid)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev_pasid->dev);
+	char dir_name[10];
+
+	sprintf(dir_name, "%x", dev_pasid->pasid);
+	dev_pasid->debugfs_dentry = debugfs_create_dir(dir_name, info->debugfs_dentry);
+
+	debugfs_create_file("domain_translation_struct", 0444, dev_pasid->debugfs_dentry,
+			    dev_pasid, &pasid_domain_translation_struct_fops);
+}
+
+/* Remove the device pasid debugfs directory. */
+void intel_iommu_debugfs_remove_dev_pasid(struct dev_pasid_info *dev_pasid)
+{
+	debugfs_remove_recursive(dev_pasid->debugfs_dentry);
+}

linux-7.2.3/drivers/iommu/intel/nested.c

@@ -0,0 +1,241 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * nested.c - nested mode translation support
+ *
+ * Copyright (C) 2023 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ *         Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *         Yi Liu <yi.l.liu@intel.com>
+ */
+
+#define pr_fmt(fmt)	"DMAR: " fmt
+
+#include <linux/iommu.h>
+#include <linux/pci.h>
+#include <linux/pci-ats.h>
+
+#include "iommu.h"
+#include "pasid.h"
+
+static int intel_nested_attach_dev(struct iommu_domain *domain,
+				   struct device *dev, struct iommu_domain *old)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+	struct intel_iommu *iommu = info->iommu;
+	unsigned long flags;
+	int ret = 0;
+
+	device_block_translation(dev);
+
+	/*
+	 * Stage-1 domain cannot work alone, it is nested on a s2_domain.
+	 * The s2_domain will be used in nested translation, hence needs
+	 * to ensure the s2_domain is compatible with this IOMMU.
+	 */
+	ret = paging_domain_compatible(&dmar_domain->s2_domain->domain, dev);
+	if (ret) {
+		dev_err_ratelimited(dev, "s2 domain is not compatible\n");
+		return ret;
+	}
+
+	ret = domain_attach_iommu(dmar_domain, iommu);
+	if (ret) {
+		dev_err_ratelimited(dev, "Failed to attach domain to iommu\n");
+		return ret;
+	}
+
+	ret = cache_tag_assign_domain(dmar_domain, dev, IOMMU_NO_PASID);
+	if (ret)
+		goto detach_iommu;
+
+	ret = iopf_for_domain_set(domain, dev);
+	if (ret)
+		goto unassign_tag;
+
+	ret = intel_pasid_setup_nested(iommu, dev,
+				       IOMMU_NO_PASID, dmar_domain);
+	if (ret)
+		goto disable_iopf;
+
+	info->domain = dmar_domain;
+	info->domain_attached = true;
+	spin_lock_irqsave(&dmar_domain->lock, flags);
+	list_add(&info->link, &dmar_domain->devices);
+	spin_unlock_irqrestore(&dmar_domain->lock, flags);
+
+	return 0;
+disable_iopf:
+	iopf_for_domain_remove(domain, dev);
+unassign_tag:
+	cache_tag_unassign_domain(dmar_domain, dev, IOMMU_NO_PASID);
+detach_iommu:
+	domain_detach_iommu(dmar_domain, iommu);
+
+	return ret;
+}
+
+static void intel_nested_domain_free(struct iommu_domain *domain)
+{
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+	struct dmar_domain *s2_domain = dmar_domain->s2_domain;
+
+	spin_lock(&s2_domain->s1_lock);
+	list_del(&dmar_domain->s2_link);
+	spin_unlock(&s2_domain->s1_lock);
+	kfree(dmar_domain->qi_batch);
+	kfree(dmar_domain);
+}
+
+static int intel_nested_cache_invalidate_user(struct iommu_domain *domain,
+					      struct iommu_user_data_array *array)
+{
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+	struct iommu_hwpt_vtd_s1_invalidate inv_entry;
+	u32 index, processed = 0;
+	int ret = 0;
+
+	if (array->type != IOMMU_HWPT_INVALIDATE_DATA_VTD_S1) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	for (index = 0; index < array->entry_num; index++) {
+		ret = iommu_copy_struct_from_user_array(&inv_entry, array,
+							IOMMU_HWPT_INVALIDATE_DATA_VTD_S1,
+							index, __reserved);
+		if (ret)
+			break;
+
+		if ((inv_entry.flags & ~IOMMU_VTD_INV_FLAGS_LEAF) ||
+		    inv_entry.__reserved) {
+			ret = -EOPNOTSUPP;
+			break;
+		}
+
+		if (!IS_ALIGNED(inv_entry.addr, VTD_PAGE_SIZE) ||
+		    ((inv_entry.npages == U64_MAX) && inv_entry.addr)) {
+			ret = -EINVAL;
+			break;
+		}
+
+		cache_tag_flush_range(dmar_domain, inv_entry.addr,
+				      inv_entry.addr + nrpages_to_size(inv_entry.npages) - 1,
+				      inv_entry.flags & IOMMU_VTD_INV_FLAGS_LEAF);
+		processed++;
+	}
+
+out:
+	array->entry_num = processed;
+	return ret;
+}
+
+static int domain_setup_nested(struct intel_iommu *iommu,
+			       struct dmar_domain *domain,
+			       struct device *dev, ioasid_t pasid,
+			       struct iommu_domain *old)
+{
+	if (old)
+		intel_pasid_tear_down_entry(iommu, dev, pasid, false);
+
+	return intel_pasid_setup_nested(iommu, dev, pasid, domain);
+}
+
+static int intel_nested_set_dev_pasid(struct iommu_domain *domain,
+				      struct device *dev, ioasid_t pasid,
+				      struct iommu_domain *old)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+	struct intel_iommu *iommu = info->iommu;
+	struct dev_pasid_info *dev_pasid;
+	int ret;
+
+	if (!pasid_supported(iommu) || dev_is_real_dma_subdevice(dev))
+		return -EOPNOTSUPP;
+
+	if (context_copied(iommu, info->bus, info->devfn))
+		return -EBUSY;
+
+	ret = paging_domain_compatible(&dmar_domain->s2_domain->domain, dev);
+	if (ret)
+		return ret;
+
+	dev_pasid = domain_add_dev_pasid(domain, dev, pasid);
+	if (IS_ERR(dev_pasid))
+		return PTR_ERR(dev_pasid);
+
+	ret = iopf_for_domain_replace(domain, old, dev);
+	if (ret)
+		goto out_remove_dev_pasid;
+
+	ret = domain_setup_nested(iommu, dmar_domain, dev, pasid, old);
+	if (ret)
+		goto out_unwind_iopf;
+
+	domain_remove_dev_pasid(old, dev, pasid);
+
+	return 0;
+
+out_unwind_iopf:
+	iopf_for_domain_replace(old, domain, dev);
+out_remove_dev_pasid:
+	domain_remove_dev_pasid(domain, dev, pasid);
+	return ret;
+}
+
+static const struct iommu_domain_ops intel_nested_domain_ops = {
+	.attach_dev		= intel_nested_attach_dev,
+	.set_dev_pasid		= intel_nested_set_dev_pasid,
+	.free			= intel_nested_domain_free,
+	.cache_invalidate_user	= intel_nested_cache_invalidate_user,
+};
+
+struct iommu_domain *
+intel_iommu_domain_alloc_nested(struct device *dev, struct iommu_domain *parent,
+				u32 flags,
+				const struct iommu_user_data *user_data)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct dmar_domain *s2_domain = to_dmar_domain(parent);
+	struct intel_iommu *iommu = info->iommu;
+	struct iommu_hwpt_vtd_s1 vtd;
+	struct dmar_domain *domain;
+	int ret;
+
+	if (!nested_supported(iommu) || flags & ~IOMMU_HWPT_ALLOC_PASID)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	/* Must be nested domain */
+	if (user_data->type != IOMMU_HWPT_DATA_VTD_S1)
+		return ERR_PTR(-EOPNOTSUPP);
+	if (!intel_domain_is_ss_paging(s2_domain) || !s2_domain->nested_parent)
+		return ERR_PTR(-EINVAL);
+
+	ret = iommu_copy_struct_from_user(&vtd, user_data,
+					  IOMMU_HWPT_DATA_VTD_S1, __reserved);
+	if (ret)
+		return ERR_PTR(ret);
+
+	domain = kzalloc_obj(*domain, GFP_KERNEL_ACCOUNT);
+	if (!domain)
+		return ERR_PTR(-ENOMEM);
+
+	domain->s2_domain = s2_domain;
+	domain->s1_cfg = vtd;
+	domain->domain.ops = &intel_nested_domain_ops;
+	domain->domain.type = IOMMU_DOMAIN_NESTED;
+	INIT_LIST_HEAD(&domain->devices);
+	INIT_LIST_HEAD(&domain->dev_pasids);
+	INIT_LIST_HEAD(&domain->cache_tags);
+	spin_lock_init(&domain->lock);
+	spin_lock_init(&domain->cache_lock);
+	xa_init(&domain->iommu_array);
+
+	spin_lock(&s2_domain->s1_lock);
+	list_add(&domain->s2_link, &s2_domain->s1_domains);
+	spin_unlock(&s2_domain->s1_lock);
+
+	return &domain->domain;
+}

linux-7.2.3/drivers/iommu/intel/pasid.c

@@ -0,0 +1,986 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * intel-pasid.c - PASID idr, table and entry manipulation
+ *
+ * Copyright (C) 2018 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+#define pr_fmt(fmt)	"DMAR: " fmt
+
+#include <linux/bitops.h>
+#include <linux/cpufeature.h>
+#include <linux/dmar.h>
+#include <linux/iommu.h>
+#include <linux/memory.h>
+#include <linux/pci.h>
+#include <linux/pci-ats.h>
+#include <linux/spinlock.h>
+
+#include "iommu.h"
+#include "pasid.h"
+#include "../iommu-pages.h"
+
+/*
+ * Intel IOMMU system wide PASID name space:
+ */
+u32 intel_pasid_max_id = PASID_MAX;
+
+/*
+ * Per device pasid table management:
+ */
+
+/*
+ * Allocate a pasid table for @dev. It should be called in a
+ * single-thread context.
+ */
+int intel_pasid_alloc_table(struct device *dev)
+{
+	struct device_domain_info *info;
+	struct pasid_table *pasid_table;
+	struct pasid_dir_entry *dir;
+	u32 max_pasid = 0;
+	int order, size;
+
+	might_sleep();
+	info = dev_iommu_priv_get(dev);
+	if (WARN_ON(!info || !dev_is_pci(dev)))
+		return -ENODEV;
+	if (WARN_ON(info->pasid_table))
+		return -EEXIST;
+
+	pasid_table = kzalloc_obj(*pasid_table);
+	if (!pasid_table)
+		return -ENOMEM;
+
+	if (info->pasid_supported)
+		max_pasid = min_t(u32, pci_max_pasids(to_pci_dev(dev)),
+				  intel_pasid_max_id);
+
+	size = max_pasid >> (PASID_PDE_SHIFT - 3);
+	order = size ? get_order(size) : 0;
+	dir = iommu_alloc_pages_node_sz(info->iommu->node, GFP_KERNEL,
+					1 << (order + PAGE_SHIFT));
+	if (!dir) {
+		kfree(pasid_table);
+		return -ENOMEM;
+	}
+
+	pasid_table->table = dir;
+	pasid_table->max_pasid = 1 << (order + PAGE_SHIFT + 3);
+	info->pasid_table = pasid_table;
+
+	if (!ecap_coherent(info->iommu->ecap))
+		clflush_cache_range(pasid_table->table, (1 << order) * PAGE_SIZE);
+
+	return 0;
+}
+
+void intel_pasid_free_table(struct device *dev)
+{
+	struct device_domain_info *info;
+	struct pasid_table *pasid_table;
+	struct pasid_dir_entry *dir;
+	struct pasid_entry *table;
+	int i, max_pde;
+
+	info = dev_iommu_priv_get(dev);
+	if (!info || !dev_is_pci(dev) || !info->pasid_table)
+		return;
+
+	pasid_table = info->pasid_table;
+	info->pasid_table = NULL;
+
+	/* Free scalable mode PASID directory tables: */
+	dir = pasid_table->table;
+	max_pde = pasid_table->max_pasid >> PASID_PDE_SHIFT;
+	for (i = 0; i < max_pde; i++) {
+		table = get_pasid_table_from_pde(&dir[i]);
+		iommu_free_pages(table);
+	}
+
+	iommu_free_pages(pasid_table->table);
+	kfree(pasid_table);
+}
+
+struct pasid_table *intel_pasid_get_table(struct device *dev)
+{
+	struct device_domain_info *info;
+
+	info = dev_iommu_priv_get(dev);
+	if (!info)
+		return NULL;
+
+	return info->pasid_table;
+}
+
+static int intel_pasid_get_dev_max_id(struct device *dev)
+{
+	struct device_domain_info *info;
+
+	info = dev_iommu_priv_get(dev);
+	if (!info || !info->pasid_table)
+		return 0;
+
+	return info->pasid_table->max_pasid;
+}
+
+static struct pasid_entry *intel_pasid_get_entry(struct device *dev, u32 pasid)
+{
+	struct device_domain_info *info;
+	struct pasid_table *pasid_table;
+	struct pasid_dir_entry *dir;
+	struct pasid_entry *entries;
+	int dir_index, index;
+
+	pasid_table = intel_pasid_get_table(dev);
+	if (WARN_ON(!pasid_table || pasid >= intel_pasid_get_dev_max_id(dev)))
+		return NULL;
+
+	dir = pasid_table->table;
+	info = dev_iommu_priv_get(dev);
+	dir_index = pasid >> PASID_PDE_SHIFT;
+	index = pasid & PASID_PTE_MASK;
+
+retry:
+	entries = get_pasid_table_from_pde(&dir[dir_index]);
+	if (!entries) {
+		u64 tmp;
+
+		entries = iommu_alloc_pages_node_sz(info->iommu->node,
+						    GFP_ATOMIC, SZ_4K);
+		if (!entries)
+			return NULL;
+
+		if (!ecap_coherent(info->iommu->ecap))
+			clflush_cache_range(entries, VTD_PAGE_SIZE);
+
+		/*
+		 * The pasid directory table entry won't be freed after
+		 * allocation. No worry about the race with free and
+		 * clear. However, this entry might be populated by others
+		 * while we are preparing it. Use theirs with a retry.
+		 */
+		tmp = 0ULL;
+		if (!try_cmpxchg64(&dir[dir_index].val, &tmp,
+				   (u64)virt_to_phys(entries) | PASID_PTE_PRESENT)) {
+			iommu_free_pages(entries);
+			goto retry;
+		}
+		if (!ecap_coherent(info->iommu->ecap))
+			clflush_cache_range(&dir[dir_index].val, sizeof(*dir));
+	}
+
+	return &entries[index];
+}
+
+/*
+ * Interfaces for PASID table entry manipulation:
+ */
+static void
+intel_pasid_clear_entry(struct device *dev, u32 pasid, bool fault_ignore)
+{
+	struct pasid_entry *pe;
+
+	pe = intel_pasid_get_entry(dev, pasid);
+	if (WARN_ON(!pe))
+		return;
+
+	if (fault_ignore && pasid_pte_is_present(pe))
+		pasid_clear_entry_with_fpd(pe);
+	else
+		pasid_clear_entry(pe);
+}
+
+static void
+pasid_cache_invalidation_with_pasid(struct intel_iommu *iommu,
+				    u16 did, u32 pasid)
+{
+	struct qi_desc desc;
+
+	desc.qw0 = QI_PC_DID(did) | QI_PC_GRAN(QI_PC_PASID_SEL) |
+		QI_PC_PASID(pasid) | QI_PC_TYPE;
+	desc.qw1 = 0;
+	desc.qw2 = 0;
+	desc.qw3 = 0;
+
+	qi_submit_sync(iommu, &desc, 1, 0);
+}
+
+static void
+devtlb_invalidation_with_pasid(struct intel_iommu *iommu,
+			       struct device *dev, u32 pasid)
+{
+	struct device_domain_info *info;
+	u16 sid, qdep, pfsid;
+
+	info = dev_iommu_priv_get(dev);
+	if (!info || !info->ats_enabled)
+		return;
+
+	if (!pci_device_is_present(to_pci_dev(dev)))
+		return;
+
+	sid = PCI_DEVID(info->bus, info->devfn);
+	qdep = info->ats_qdep;
+	pfsid = info->pfsid;
+
+	/*
+	 * When PASID 0 is used, it indicates RID2PASID(DMA request w/o PASID),
+	 * devTLB flush w/o PASID should be used. For non-zero PASID under
+	 * SVA usage, device could do DMA with multiple PASIDs. It is more
+	 * efficient to flush devTLB specific to the PASID.
+	 */
+	if (pasid == IOMMU_NO_PASID)
+		qi_flush_dev_iotlb(iommu, sid, pfsid, qdep, 0, 64 - VTD_PAGE_SHIFT);
+	else
+		qi_flush_dev_iotlb_pasid(iommu, sid, pfsid, pasid, qdep, 0, 64 - VTD_PAGE_SHIFT);
+}
+
+void intel_pasid_tear_down_entry(struct intel_iommu *iommu, struct device *dev,
+				 u32 pasid, bool fault_ignore)
+{
+	struct pasid_entry *pte;
+	u16 did, pgtt;
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (WARN_ON(!pte)) {
+		spin_unlock(&iommu->lock);
+		return;
+	}
+
+	if (!pasid_pte_is_present(pte)) {
+		if (!pasid_pte_is_fault_disabled(pte)) {
+			WARN_ON(READ_ONCE(pte->val[0]) != 0);
+			spin_unlock(&iommu->lock);
+			return;
+		}
+
+		/*
+		 * When a PASID is used for SVA by a device, it's possible
+		 * that the pasid entry is non-present with the Fault
+		 * Processing Disabled bit set. Clear the pasid entry and
+		 * drain the PRQ for the PASID before return.
+		 */
+		pasid_clear_entry(pte);
+		spin_unlock(&iommu->lock);
+		intel_iommu_drain_pasid_prq(dev, pasid);
+
+		return;
+	}
+
+	did = pasid_get_domain_id(pte);
+	pgtt = pasid_pte_get_pgtt(pte);
+	pasid_clear_present(pte);
+	spin_unlock(&iommu->lock);
+
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(pte, sizeof(*pte));
+
+	pasid_cache_invalidation_with_pasid(iommu, did, pasid);
+
+	if (pgtt == PASID_ENTRY_PGTT_PT || pgtt == PASID_ENTRY_PGTT_FL_ONLY)
+		qi_flush_piotlb_all(iommu, did, pasid);
+	else
+		iommu->flush.flush_iotlb(iommu, did, 0, 0, DMA_TLB_DSI_FLUSH);
+
+	devtlb_invalidation_with_pasid(iommu, dev, pasid);
+	intel_pasid_clear_entry(dev, pasid, fault_ignore);
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(pte, sizeof(*pte));
+
+	if (!fault_ignore)
+		intel_iommu_drain_pasid_prq(dev, pasid);
+}
+
+/*
+ * This function flushes cache for a newly setup pasid table entry.
+ * Caller of it should not modify the in-use pasid table entries.
+ */
+static void pasid_flush_caches(struct intel_iommu *iommu,
+				struct pasid_entry *pte,
+			       u32 pasid, u16 did)
+{
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(pte, sizeof(*pte));
+
+	if (cap_caching_mode(iommu->cap)) {
+		pasid_cache_invalidation_with_pasid(iommu, did, pasid);
+		qi_flush_piotlb_all(iommu, did, pasid);
+	} else {
+		iommu_flush_write_buffer(iommu);
+	}
+}
+
+/*
+ * This function is supposed to be used after caller updates the fields
+ * except for the SSADE and P bit of a pasid table entry. It does the
+ * below:
+ * - Flush cacheline if needed
+ * - Flush the caches per Table 28 ”Guidance to Software for Invalidations“
+ *   of VT-d spec 5.0.
+ */
+static void intel_pasid_flush_present(struct intel_iommu *iommu,
+				      struct device *dev,
+				      u32 pasid, u16 did,
+				      struct pasid_entry *pte)
+{
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(pte, sizeof(*pte));
+
+	/*
+	 * VT-d spec 5.0 table28 states guides for cache invalidation:
+	 *
+	 * - PASID-selective-within-Domain PASID-cache invalidation
+	 * - PASID-selective PASID-based IOTLB invalidation
+	 * - If (pasid is RID_PASID)
+	 *    - Global Device-TLB invalidation to affected functions
+	 *   Else
+	 *    - PASID-based Device-TLB invalidation (with S=1 and
+	 *      Addr[63:12]=0x7FFFFFFF_FFFFF) to affected functions
+	 */
+	pasid_cache_invalidation_with_pasid(iommu, did, pasid);
+	qi_flush_piotlb_all(iommu, did, pasid);
+
+	devtlb_invalidation_with_pasid(iommu, dev, pasid);
+}
+
+/*
+ * Set up the scalable mode pasid table entry for first only
+ * translation type.
+ */
+static void pasid_pte_config_first_level(struct intel_iommu *iommu,
+					 struct pasid_entry *pte,
+					 phys_addr_t fsptptr, u16 did,
+					 int flags)
+{
+	lockdep_assert_held(&iommu->lock);
+
+	pasid_clear_entry(pte);
+
+	/* Setup the first level page table pointer: */
+	pasid_set_flptr(pte, fsptptr);
+
+	if (flags & PASID_FLAG_FL5LP)
+		pasid_set_flpm(pte, 1);
+
+	if (flags & PASID_FLAG_PAGE_SNOOP)
+		pasid_set_pgsnp(pte);
+
+	pasid_set_domain_id(pte, did);
+	pasid_set_address_width(pte, iommu->agaw);
+	pasid_set_page_snoop(pte, flags & PASID_FLAG_PWSNP);
+
+	/* Setup Present and PASID Granular Transfer Type: */
+	pasid_set_translation_type(pte, PASID_ENTRY_PGTT_FL_ONLY);
+	pasid_set_present(pte);
+}
+
+int intel_pasid_setup_first_level(struct intel_iommu *iommu, struct device *dev,
+				  phys_addr_t fsptptr, u32 pasid, u16 did,
+				  int flags)
+{
+	struct pasid_entry *pte;
+
+	if (!ecap_flts(iommu->ecap)) {
+		pr_err("No first level translation support on %s\n",
+		       iommu->name);
+		return -EINVAL;
+	}
+
+	if ((flags & PASID_FLAG_FL5LP) && !cap_fl5lp_support(iommu->cap)) {
+		pr_err("No 5-level paging support for first-level on %s\n",
+		       iommu->name);
+		return -EINVAL;
+	}
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (!pte) {
+		spin_unlock(&iommu->lock);
+		return -ENODEV;
+	}
+
+	if (pasid_pte_is_present(pte)) {
+		spin_unlock(&iommu->lock);
+		return -EBUSY;
+	}
+
+	pasid_pte_config_first_level(iommu, pte, fsptptr, did, flags);
+
+	spin_unlock(&iommu->lock);
+
+	pasid_flush_caches(iommu, pte, pasid, did);
+
+	return 0;
+}
+
+/*
+ * Set up the scalable mode pasid entry for second only translation type.
+ */
+static void pasid_pte_config_second_level(struct intel_iommu *iommu,
+					  struct pasid_entry *pte,
+					  struct dmar_domain *domain, u16 did)
+{
+	struct pt_iommu_vtdss_hw_info pt_info;
+
+	lockdep_assert_held(&iommu->lock);
+
+	pt_iommu_vtdss_hw_info(&domain->sspt, &pt_info);
+	pasid_clear_entry(pte);
+	pasid_set_domain_id(pte, did);
+	pasid_set_slptr(pte, pt_info.ssptptr);
+	pasid_set_address_width(pte, pt_info.aw);
+	pasid_set_translation_type(pte, PASID_ENTRY_PGTT_SL_ONLY);
+	pasid_set_fault_enable(pte);
+	pasid_set_page_snoop(pte, !(domain->sspt.vtdss_pt.common.features &
+				    BIT(PT_FEAT_DMA_INCOHERENT)));
+	if (domain->dirty_tracking)
+		pasid_set_ssade(pte);
+
+	pasid_set_present(pte);
+}
+
+int intel_pasid_setup_second_level(struct intel_iommu *iommu,
+				   struct dmar_domain *domain,
+				   struct device *dev, u32 pasid)
+{
+	struct pasid_entry *pte;
+	u16 did;
+
+
+	/*
+	 * If hardware advertises no support for second level
+	 * translation, return directly.
+	 */
+	if (!ecap_slts(iommu->ecap)) {
+		pr_err("No second level translation support on %s\n",
+		       iommu->name);
+		return -EINVAL;
+	}
+
+	did = domain_id_iommu(domain, iommu);
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (!pte) {
+		spin_unlock(&iommu->lock);
+		return -ENODEV;
+	}
+
+	if (pasid_pte_is_present(pte)) {
+		spin_unlock(&iommu->lock);
+		return -EBUSY;
+	}
+
+	pasid_pte_config_second_level(iommu, pte, domain, did);
+	spin_unlock(&iommu->lock);
+
+	pasid_flush_caches(iommu, pte, pasid, did);
+
+	return 0;
+}
+
+/*
+ * Set up dirty tracking on a second only or nested translation type.
+ */
+int intel_pasid_setup_dirty_tracking(struct intel_iommu *iommu,
+				     struct device *dev, u32 pasid,
+				     bool enabled)
+{
+	struct pasid_entry *pte;
+	u16 did, pgtt;
+
+	spin_lock(&iommu->lock);
+
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (!pte) {
+		spin_unlock(&iommu->lock);
+		dev_err_ratelimited(
+			dev, "Failed to get pasid entry of PASID %d\n", pasid);
+		return -ENODEV;
+	}
+
+	did = pasid_get_domain_id(pte);
+	pgtt = pasid_pte_get_pgtt(pte);
+	if (pgtt != PASID_ENTRY_PGTT_SL_ONLY &&
+	    pgtt != PASID_ENTRY_PGTT_NESTED) {
+		spin_unlock(&iommu->lock);
+		dev_err_ratelimited(
+			dev,
+			"Dirty tracking not supported on translation type %d\n",
+			pgtt);
+		return -EOPNOTSUPP;
+	}
+
+	if (pasid_get_ssade(pte) == enabled) {
+		spin_unlock(&iommu->lock);
+		return 0;
+	}
+
+	if (enabled)
+		pasid_set_ssade(pte);
+	else
+		pasid_clear_ssade(pte);
+	spin_unlock(&iommu->lock);
+
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(pte, sizeof(*pte));
+
+	/*
+	 * From VT-d spec table 25 "Guidance to Software for Invalidations":
+	 *
+	 * - PASID-selective-within-Domain PASID-cache invalidation
+	 *   If (PGTT=SS or Nested)
+	 *    - Domain-selective IOTLB invalidation
+	 *   Else
+	 *    - PASID-selective PASID-based IOTLB invalidation
+	 * - If (pasid is RID_PASID)
+	 *    - Global Device-TLB invalidation to affected functions
+	 *   Else
+	 *    - PASID-based Device-TLB invalidation (with S=1 and
+	 *      Addr[63:12]=0x7FFFFFFF_FFFFF) to affected functions
+	 */
+	pasid_cache_invalidation_with_pasid(iommu, did, pasid);
+
+	iommu->flush.flush_iotlb(iommu, did, 0, 0, DMA_TLB_DSI_FLUSH);
+
+	devtlb_invalidation_with_pasid(iommu, dev, pasid);
+
+	return 0;
+}
+
+/*
+ * Set up the scalable mode pasid entry for passthrough translation type.
+ */
+static void pasid_pte_config_pass_through(struct intel_iommu *iommu,
+					  struct pasid_entry *pte, u16 did)
+{
+	lockdep_assert_held(&iommu->lock);
+
+	pasid_clear_entry(pte);
+	pasid_set_domain_id(pte, did);
+	pasid_set_address_width(pte, iommu->agaw);
+	pasid_set_translation_type(pte, PASID_ENTRY_PGTT_PT);
+	pasid_set_fault_enable(pte);
+	pasid_set_page_snoop(pte, !!ecap_smpwc(iommu->ecap));
+	pasid_set_present(pte);
+}
+
+int intel_pasid_setup_pass_through(struct intel_iommu *iommu,
+				   struct device *dev, u32 pasid)
+{
+	u16 did = FLPT_DEFAULT_DID;
+	struct pasid_entry *pte;
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (!pte) {
+		spin_unlock(&iommu->lock);
+		return -ENODEV;
+	}
+
+	if (pasid_pte_is_present(pte)) {
+		spin_unlock(&iommu->lock);
+		return -EBUSY;
+	}
+
+	pasid_pte_config_pass_through(iommu, pte, did);
+	spin_unlock(&iommu->lock);
+
+	pasid_flush_caches(iommu, pte, pasid, did);
+
+	return 0;
+}
+
+/*
+ * Set the page snoop control for a pasid entry which has been set up.
+ */
+void intel_pasid_setup_page_snoop_control(struct intel_iommu *iommu,
+					  struct device *dev, u32 pasid)
+{
+	struct pasid_entry *pte;
+	u16 did;
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (WARN_ON(!pte || !pasid_pte_is_present(pte))) {
+		spin_unlock(&iommu->lock);
+		return;
+	}
+
+	pasid_set_pgsnp(pte);
+	did = pasid_get_domain_id(pte);
+	spin_unlock(&iommu->lock);
+
+	intel_pasid_flush_present(iommu, dev, pasid, did, pte);
+}
+
+static void pasid_pte_config_nested(struct intel_iommu *iommu,
+				    struct pasid_entry *pte,
+				    struct iommu_hwpt_vtd_s1 *s1_cfg,
+				    struct dmar_domain *s2_domain,
+				    u16 did)
+{
+	struct pt_iommu_vtdss_hw_info pt_info;
+
+	lockdep_assert_held(&iommu->lock);
+
+	pt_iommu_vtdss_hw_info(&s2_domain->sspt, &pt_info);
+
+	pasid_clear_entry(pte);
+
+	if (s1_cfg->addr_width == ADDR_WIDTH_5LEVEL)
+		pasid_set_flpm(pte, 1);
+
+	pasid_set_flptr(pte, s1_cfg->pgtbl_addr);
+
+	if (s1_cfg->flags & IOMMU_VTD_S1_SRE) {
+		pasid_set_sre(pte);
+		if (s1_cfg->flags & IOMMU_VTD_S1_WPE)
+			pasid_set_wpe(pte);
+	}
+
+	if (s1_cfg->flags & IOMMU_VTD_S1_EAFE)
+		pasid_set_eafe(pte);
+
+	if (s2_domain->force_snooping)
+		pasid_set_pgsnp(pte);
+
+	pasid_set_slptr(pte, pt_info.ssptptr);
+	pasid_set_fault_enable(pte);
+	pasid_set_domain_id(pte, did);
+	pasid_set_address_width(pte, pt_info.aw);
+	pasid_set_page_snoop(pte, !(s2_domain->sspt.vtdss_pt.common.features &
+				    BIT(PT_FEAT_DMA_INCOHERENT)));
+	if (s2_domain->dirty_tracking)
+		pasid_set_ssade(pte);
+	pasid_set_translation_type(pte, PASID_ENTRY_PGTT_NESTED);
+	pasid_set_present(pte);
+}
+
+/**
+ * intel_pasid_setup_nested() - Set up PASID entry for nested translation.
+ * @iommu:      IOMMU which the device belong to
+ * @dev:        Device to be set up for translation
+ * @pasid:      PASID to be programmed in the device PASID table
+ * @domain:     User stage-1 domain nested on a stage-2 domain
+ *
+ * This is used for nested translation. The input domain should be
+ * nested type and nested on a parent with 'is_nested_parent' flag
+ * set.
+ */
+int intel_pasid_setup_nested(struct intel_iommu *iommu, struct device *dev,
+			     u32 pasid, struct dmar_domain *domain)
+{
+	struct iommu_hwpt_vtd_s1 *s1_cfg = &domain->s1_cfg;
+	struct dmar_domain *s2_domain = domain->s2_domain;
+	u16 did = domain_id_iommu(domain, iommu);
+	struct pasid_entry *pte;
+
+	/* Address width should match the address width supported by hardware */
+	switch (s1_cfg->addr_width) {
+	case ADDR_WIDTH_4LEVEL:
+		break;
+	case ADDR_WIDTH_5LEVEL:
+		if (!cap_fl5lp_support(iommu->cap)) {
+			dev_err_ratelimited(dev,
+					    "5-level paging not supported\n");
+			return -EINVAL;
+		}
+		break;
+	default:
+		dev_err_ratelimited(dev, "Invalid stage-1 address width %d\n",
+				    s1_cfg->addr_width);
+		return -EINVAL;
+	}
+
+	if ((s1_cfg->flags & IOMMU_VTD_S1_SRE) && !ecap_srs(iommu->ecap)) {
+		pr_err_ratelimited("No supervisor request support on %s\n",
+				   iommu->name);
+		return -EINVAL;
+	}
+
+	if ((s1_cfg->flags & IOMMU_VTD_S1_EAFE) && !ecap_eafs(iommu->ecap)) {
+		pr_err_ratelimited("No extended access flag support on %s\n",
+				   iommu->name);
+		return -EINVAL;
+	}
+
+	spin_lock(&iommu->lock);
+	pte = intel_pasid_get_entry(dev, pasid);
+	if (!pte) {
+		spin_unlock(&iommu->lock);
+		return -ENODEV;
+	}
+	if (pasid_pte_is_present(pte)) {
+		spin_unlock(&iommu->lock);
+		return -EBUSY;
+	}
+
+	pasid_pte_config_nested(iommu, pte, s1_cfg, s2_domain, did);
+	spin_unlock(&iommu->lock);
+
+	pasid_flush_caches(iommu, pte, pasid, did);
+
+	return 0;
+}
+
+/*
+ * Interfaces to setup or teardown a pasid table to the scalable-mode
+ * context table entry:
+ */
+
+static void device_pasid_table_teardown(struct device *dev, u8 bus, u8 devfn)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	struct context_entry *context;
+	u16 did;
+
+	spin_lock(&iommu->lock);
+	context = iommu_context_addr(iommu, bus, devfn, false);
+	if (!context) {
+		spin_unlock(&iommu->lock);
+		return;
+	}
+
+	did = context_domain_id(context);
+	context_clear_present(context);
+	__iommu_flush_cache(iommu, context, sizeof(*context));
+	spin_unlock(&iommu->lock);
+	intel_context_flush_no_pasid(info, context, did);
+	context_clear_entry(context);
+	__iommu_flush_cache(iommu, context, sizeof(*context));
+}
+
+static int pci_pasid_table_teardown(struct pci_dev *pdev, u16 alias, void *data)
+{
+	struct device *dev = data;
+
+	if (dev == &pdev->dev)
+		device_pasid_table_teardown(dev, PCI_BUS_NUM(alias), alias & 0xff);
+
+	return 0;
+}
+
+void intel_pasid_teardown_sm_context(struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+
+	if (!dev_is_pci(dev)) {
+		device_pasid_table_teardown(dev, info->bus, info->devfn);
+		return;
+	}
+
+	pci_for_each_dma_alias(to_pci_dev(dev), pci_pasid_table_teardown, dev);
+}
+
+/*
+ * Get the PASID directory size for scalable mode context entry.
+ * Value of X in the PDTS field of a scalable mode context entry
+ * indicates PASID directory with 2^(X + 7) entries.
+ */
+static unsigned long context_get_sm_pds(struct pasid_table *table)
+{
+	unsigned long pds, max_pde;
+
+	max_pde = table->max_pasid >> PASID_PDE_SHIFT;
+	pds = find_first_bit(&max_pde, MAX_NR_PASID_BITS);
+	if (pds < 7)
+		return 0;
+
+	return pds - 7;
+}
+
+static int context_entry_set_pasid_table(struct context_entry *context,
+					 struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct pasid_table *table = info->pasid_table;
+	struct intel_iommu *iommu = info->iommu;
+	unsigned long pds;
+
+	context_clear_entry(context);
+
+	pds = context_get_sm_pds(table);
+	context->lo = (u64)virt_to_phys(table->table) | context_pdts(pds);
+	context_set_sm_rid2pasid(context, IOMMU_NO_PASID);
+
+	if (info->ats_supported)
+		context_set_sm_dte(context);
+	if (info->pasid_supported)
+		context_set_pasid(context);
+	if (info->pri_supported)
+		context_set_sm_pre(context);
+
+	context_set_fault_enable(context);
+	context_set_present(context);
+	__iommu_flush_cache(iommu, context, sizeof(*context));
+
+	return 0;
+}
+
+static int device_pasid_table_setup(struct device *dev, u8 bus, u8 devfn)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	struct context_entry *context;
+
+	spin_lock(&iommu->lock);
+	context = iommu_context_addr(iommu, bus, devfn, true);
+	if (!context) {
+		spin_unlock(&iommu->lock);
+		return -ENOMEM;
+	}
+
+	if (context_present(context) && !context_copied(iommu, bus, devfn)) {
+		spin_unlock(&iommu->lock);
+		return 0;
+	}
+
+	if (context_copied(iommu, bus, devfn)) {
+		context_clear_present(context);
+		__iommu_flush_cache(iommu, context, sizeof(*context));
+
+		/*
+		 * For kdump cases, old valid entries may be cached due to
+		 * the in-flight DMA and copied pgtable, but there is no
+		 * unmapping behaviour for them, thus we need explicit cache
+		 * flushes for all affected domain IDs and PASIDs used in
+		 * the copied PASID table. Given that we have no idea about
+		 * which domain IDs and PASIDs were used in the copied tables,
+		 * upgrade them to global PASID and IOTLB cache invalidation.
+		 */
+		iommu->flush.flush_context(iommu, 0,
+					   PCI_DEVID(bus, devfn),
+					   DMA_CCMD_MASK_NOBIT,
+					   DMA_CCMD_DEVICE_INVL);
+		qi_flush_pasid_cache(iommu, 0, QI_PC_GLOBAL, 0);
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH);
+		devtlb_invalidation_with_pasid(iommu, dev, IOMMU_NO_PASID);
+
+		context_clear_entry(context);
+		__iommu_flush_cache(iommu, context, sizeof(*context));
+
+		/*
+		 * At this point, the device is supposed to finish reset at
+		 * its driver probe stage, so no in-flight DMA will exist,
+		 * and we don't need to worry anymore hereafter.
+		 */
+		clear_context_copied(iommu, bus, devfn);
+	}
+
+	context_entry_set_pasid_table(context, dev);
+	spin_unlock(&iommu->lock);
+
+	/*
+	 * It's a non-present to present mapping. If hardware doesn't cache
+	 * non-present entry we don't need to flush the caches. If it does
+	 * cache non-present entries, then it does so in the special
+	 * domain #0, which we have to flush:
+	 */
+	if (cap_caching_mode(iommu->cap)) {
+		iommu->flush.flush_context(iommu, 0,
+					   PCI_DEVID(bus, devfn),
+					   DMA_CCMD_MASK_NOBIT,
+					   DMA_CCMD_DEVICE_INVL);
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH);
+	}
+
+	return 0;
+}
+
+static int pci_pasid_table_setup(struct pci_dev *pdev, u16 alias, void *data)
+{
+	struct device *dev = data;
+
+	if (dev != &pdev->dev)
+		return 0;
+
+	return device_pasid_table_setup(dev, PCI_BUS_NUM(alias), alias & 0xff);
+}
+
+/*
+ * Set the device's PASID table to its context table entry.
+ *
+ * The PASID table is set to the context entries of both device itself
+ * and its alias requester ID for DMA.
+ */
+int intel_pasid_setup_sm_context(struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+
+	if (!dev_is_pci(dev))
+		return device_pasid_table_setup(dev, info->bus, info->devfn);
+
+	return pci_for_each_dma_alias(to_pci_dev(dev), pci_pasid_table_setup, dev);
+}
+
+/*
+ * Global Device-TLB invalidation following changes in a context entry which
+ * was present.
+ */
+static void __context_flush_dev_iotlb(struct device_domain_info *info)
+{
+	if (!info->ats_enabled)
+		return;
+
+	/*
+	 * Skip dev-IOTLB flush for inaccessible PCIe devices to prevent the
+	 * Intel IOMMU from waiting indefinitely for an ATS invalidation that
+	 * cannot complete.
+	 */
+	if (!pci_device_is_present(to_pci_dev(info->dev)))
+		return;
+
+	qi_flush_dev_iotlb(info->iommu, PCI_DEVID(info->bus, info->devfn),
+			   info->pfsid, info->ats_qdep, 0, MAX_AGAW_PFN_WIDTH);
+
+	/*
+	 * There is no guarantee that the device DMA is stopped when it reaches
+	 * here. Therefore, always attempt the extra device TLB invalidation
+	 * quirk. The impact on performance is acceptable since this is not a
+	 * performance-critical path.
+	 */
+	quirk_extra_dev_tlb_flush(info, 0, MAX_AGAW_PFN_WIDTH, IOMMU_NO_PASID,
+				  info->ats_qdep);
+}
+
+/*
+ * Cache invalidations after change in a context table entry that was present
+ * according to the Spec 6.5.3.3 (Guidance to Software for Invalidations).
+ * This helper can only be used when IOMMU is working in the legacy mode or
+ * IOMMU is in scalable mode but all PASID table entries of the device are
+ * non-present.
+ */
+void intel_context_flush_no_pasid(struct device_domain_info *info,
+				  struct context_entry *context, u16 did)
+{
+	struct intel_iommu *iommu = info->iommu;
+
+	/*
+	 * Device-selective context-cache invalidation. The Domain-ID field
+	 * of the Context-cache Invalidate Descriptor is ignored by hardware
+	 * when operating in scalable mode. Therefore the @did value doesn't
+	 * matter in scalable mode.
+	 */
+	iommu->flush.flush_context(iommu, did, PCI_DEVID(info->bus, info->devfn),
+				   DMA_CCMD_MASK_NOBIT, DMA_CCMD_DEVICE_INVL);
+
+	/*
+	 * For legacy mode:
+	 * - Domain-selective IOTLB invalidation
+	 * - Global Device-TLB invalidation to all affected functions
+	 */
+	if (!sm_supported(iommu)) {
+		iommu->flush.flush_iotlb(iommu, did, 0, 0, DMA_TLB_DSI_FLUSH);
+		__context_flush_dev_iotlb(info);
+
+		return;
+	}
+
+	__context_flush_dev_iotlb(info);
+}

linux-7.2.3/drivers/iommu/intel/pasid.h

@@ -0,0 +1,326 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * pasid.h - PASID idr, table and entry header
+ *
+ * Copyright (C) 2018 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+#ifndef __INTEL_PASID_H
+#define __INTEL_PASID_H
+
+#define PASID_MAX			0x100000
+#define PASID_PTE_MASK			0x3F
+#define PASID_PTE_PRESENT		1
+#define PASID_PTE_FPD			2
+#define PDE_PFN_MASK			PAGE_MASK
+#define PASID_PDE_SHIFT			6
+#define MAX_NR_PASID_BITS		20
+#define PASID_TBL_ENTRIES		BIT(PASID_PDE_SHIFT)
+
+#define is_pasid_enabled(entry)		(((entry)->lo >> 3) & 0x1)
+#define get_pasid_dir_size(entry)	(1 << ((((entry)->lo >> 9) & 0x7) + 7))
+
+#define PASID_FLAG_NESTED		BIT(1)
+#define PASID_FLAG_PAGE_SNOOP		BIT(2)
+#define PASID_FLAG_PWSNP		BIT(3)
+
+/*
+ * The PASID_FLAG_FL5LP flag Indicates using 5-level paging for first-
+ * level translation, otherwise, 4-level paging will be used.
+ */
+#define PASID_FLAG_FL5LP		BIT(1)
+
+struct pasid_dir_entry {
+	u64 val;
+};
+
+struct pasid_entry {
+	u64 val[8];
+};
+
+#define PASID_ENTRY_PGTT_FL_ONLY	(1)
+#define PASID_ENTRY_PGTT_SL_ONLY	(2)
+#define PASID_ENTRY_PGTT_NESTED		(3)
+#define PASID_ENTRY_PGTT_PT		(4)
+
+/* The representative of a PASID table */
+struct pasid_table {
+	void			*table;		/* pasid table pointer */
+	u32			max_pasid;	/* max pasid */
+};
+
+/* Get PRESENT bit of a PASID directory entry. */
+static inline bool pasid_pde_is_present(struct pasid_dir_entry *pde)
+{
+	return READ_ONCE(pde->val) & PASID_PTE_PRESENT;
+}
+
+/* Get PASID table from a PASID directory entry. */
+static inline struct pasid_entry *
+get_pasid_table_from_pde(struct pasid_dir_entry *pde)
+{
+	if (!pasid_pde_is_present(pde))
+		return NULL;
+
+	return phys_to_virt(READ_ONCE(pde->val) & PDE_PFN_MASK);
+}
+
+/* Get PRESENT bit of a PASID table entry. */
+static inline bool pasid_pte_is_present(struct pasid_entry *pte)
+{
+	return READ_ONCE(pte->val[0]) & PASID_PTE_PRESENT;
+}
+
+/* Get FPD(Fault Processing Disable) bit of a PASID table entry */
+static inline bool pasid_pte_is_fault_disabled(struct pasid_entry *pte)
+{
+	return READ_ONCE(pte->val[0]) & PASID_PTE_FPD;
+}
+
+/* Get PGTT field of a PASID table entry */
+static inline u16 pasid_pte_get_pgtt(struct pasid_entry *pte)
+{
+	return (u16)((READ_ONCE(pte->val[0]) >> 6) & 0x7);
+}
+
+static inline void pasid_clear_entry(struct pasid_entry *pe)
+{
+	WRITE_ONCE(pe->val[0], 0);
+	WRITE_ONCE(pe->val[1], 0);
+	WRITE_ONCE(pe->val[2], 0);
+	WRITE_ONCE(pe->val[3], 0);
+	WRITE_ONCE(pe->val[4], 0);
+	WRITE_ONCE(pe->val[5], 0);
+	WRITE_ONCE(pe->val[6], 0);
+	WRITE_ONCE(pe->val[7], 0);
+}
+
+static inline void pasid_clear_entry_with_fpd(struct pasid_entry *pe)
+{
+	WRITE_ONCE(pe->val[0], PASID_PTE_FPD);
+	WRITE_ONCE(pe->val[1], 0);
+	WRITE_ONCE(pe->val[2], 0);
+	WRITE_ONCE(pe->val[3], 0);
+	WRITE_ONCE(pe->val[4], 0);
+	WRITE_ONCE(pe->val[5], 0);
+	WRITE_ONCE(pe->val[6], 0);
+	WRITE_ONCE(pe->val[7], 0);
+}
+
+static inline void pasid_set_bits(u64 *ptr, u64 mask, u64 bits)
+{
+	u64 old;
+
+	old = READ_ONCE(*ptr);
+	WRITE_ONCE(*ptr, (old & ~mask) | bits);
+}
+
+static inline u64 pasid_get_bits(u64 *ptr)
+{
+	return READ_ONCE(*ptr);
+}
+
+/*
+ * Setup the DID(Domain Identifier) field (Bit 64~79) of scalable mode
+ * PASID entry.
+ */
+static inline void
+pasid_set_domain_id(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[1], GENMASK_ULL(15, 0), value);
+}
+
+/*
+ * Get domain ID value of a scalable mode PASID entry.
+ */
+static inline u16
+pasid_get_domain_id(struct pasid_entry *pe)
+{
+	return (u16)(READ_ONCE(pe->val[1]) & GENMASK_ULL(15, 0));
+}
+
+/*
+ * Setup the SLPTPTR(Second Level Page Table Pointer) field (Bit 12~63)
+ * of a scalable mode PASID entry.
+ */
+static inline void
+pasid_set_slptr(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[0], VTD_PAGE_MASK, value);
+}
+
+/*
+ * Setup the AW(Address Width) field (Bit 2~4) of a scalable mode PASID
+ * entry.
+ */
+static inline void
+pasid_set_address_width(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[0], GENMASK_ULL(4, 2), value << 2);
+}
+
+/*
+ * Setup the PGTT(PASID Granular Translation Type) field (Bit 6~8)
+ * of a scalable mode PASID entry.
+ */
+static inline void
+pasid_set_translation_type(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[0], GENMASK_ULL(8, 6), value << 6);
+}
+
+/*
+ * Enable fault processing by clearing the FPD(Fault Processing
+ * Disable) field (Bit 1) of a scalable mode PASID entry.
+ */
+static inline void pasid_set_fault_enable(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[0], 1 << 1, 0);
+}
+
+/*
+ * Enable second level A/D bits by setting the SLADE (Second Level
+ * Access Dirty Enable) field (Bit 9) of a scalable mode PASID
+ * entry.
+ */
+static inline void pasid_set_ssade(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[0], 1 << 9, 1 << 9);
+}
+
+/*
+ * Disable second level A/D bits by clearing the SLADE (Second Level
+ * Access Dirty Enable) field (Bit 9) of a scalable mode PASID
+ * entry.
+ */
+static inline void pasid_clear_ssade(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[0], 1 << 9, 0);
+}
+
+/*
+ * Checks if second level A/D bits specifically the SLADE (Second Level
+ * Access Dirty Enable) field (Bit 9) of a scalable mode PASID
+ * entry is set.
+ */
+static inline bool pasid_get_ssade(struct pasid_entry *pe)
+{
+	return pasid_get_bits(&pe->val[0]) & (1 << 9);
+}
+
+/*
+ * Setup the SRE(Supervisor Request Enable) field (Bit 128) of a
+ * scalable mode PASID entry.
+ */
+static inline void pasid_set_sre(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[2], 1 << 0, 1);
+}
+
+/*
+ * Setup the WPE(Write Protect Enable) field (Bit 132) of a
+ * scalable mode PASID entry.
+ */
+static inline void pasid_set_wpe(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[2], 1 << 4, 1 << 4);
+}
+
+/*
+ * Setup the P(Present) field (Bit 0) of a scalable mode PASID
+ * entry.
+ */
+static inline void pasid_set_present(struct pasid_entry *pe)
+{
+	dma_wmb();
+	pasid_set_bits(&pe->val[0], 1 << 0, 1);
+}
+
+/*
+ * Clear the Present (P) bit (bit 0) of a scalable-mode PASID table entry.
+ * This initiates the transition of the entry's ownership from hardware
+ * to software. The caller is responsible for fulfilling the invalidation
+ * handshake recommended by the VT-d spec, Section 6.5.3.3 (Guidance to
+ * Software for Invalidations).
+ */
+static inline void pasid_clear_present(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[0], 1 << 0, 0);
+	dma_wmb();
+}
+
+/*
+ * Setup Page Walk Snoop bit (Bit 87) of a scalable mode PASID
+ * entry.
+ */
+static inline void pasid_set_page_snoop(struct pasid_entry *pe, bool value)
+{
+	pasid_set_bits(&pe->val[1], 1 << 23, value << 23);
+}
+
+/*
+ * Setup the Page Snoop (PGSNP) field (Bit 88) of a scalable mode
+ * PASID entry.
+ */
+static inline void
+pasid_set_pgsnp(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[1], 1ULL << 24, 1ULL << 24);
+}
+
+/*
+ * Setup the First Level Page table Pointer field (Bit 140~191)
+ * of a scalable mode PASID entry.
+ */
+static inline void
+pasid_set_flptr(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[2], VTD_PAGE_MASK, value);
+}
+
+/*
+ * Setup the First Level Paging Mode field (Bit 130~131) of a
+ * scalable mode PASID entry.
+ */
+static inline void
+pasid_set_flpm(struct pasid_entry *pe, u64 value)
+{
+	pasid_set_bits(&pe->val[2], GENMASK_ULL(3, 2), value << 2);
+}
+
+/*
+ * Setup the Extended Access Flag Enable (EAFE) field (Bit 135)
+ * of a scalable mode PASID entry.
+ */
+static inline void pasid_set_eafe(struct pasid_entry *pe)
+{
+	pasid_set_bits(&pe->val[2], 1 << 7, 1 << 7);
+}
+
+extern unsigned int intel_pasid_max_id;
+int intel_pasid_alloc_table(struct device *dev);
+void intel_pasid_free_table(struct device *dev);
+struct pasid_table *intel_pasid_get_table(struct device *dev);
+int intel_pasid_setup_first_level(struct intel_iommu *iommu, struct device *dev,
+				  phys_addr_t fsptptr, u32 pasid, u16 did,
+				  int flags);
+int intel_pasid_setup_second_level(struct intel_iommu *iommu,
+				   struct dmar_domain *domain,
+				   struct device *dev, u32 pasid);
+int intel_pasid_setup_dirty_tracking(struct intel_iommu *iommu,
+				     struct device *dev, u32 pasid,
+				     bool enabled);
+int intel_pasid_setup_pass_through(struct intel_iommu *iommu,
+				   struct device *dev, u32 pasid);
+int intel_pasid_setup_nested(struct intel_iommu *iommu, struct device *dev,
+			     u32 pasid, struct dmar_domain *domain);
+void intel_pasid_tear_down_entry(struct intel_iommu *iommu,
+				 struct device *dev, u32 pasid,
+				 bool fault_ignore);
+void intel_pasid_setup_page_snoop_control(struct intel_iommu *iommu,
+					  struct device *dev, u32 pasid);
+int intel_pasid_setup_sm_context(struct device *dev);
+void intel_pasid_teardown_sm_context(struct device *dev);
+#endif /* __INTEL_PASID_H */

linux-7.2.3/drivers/iommu/intel/perf.c

@@ -0,0 +1,164 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * perf.c - performance monitor
+ *
+ * Copyright (C) 2021 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ *         Fenghua Yu <fenghua.yu@intel.com>
+ */
+
+#include <linux/spinlock.h>
+
+#include "iommu.h"
+#include "perf.h"
+
+static DEFINE_SPINLOCK(latency_lock);
+
+bool dmar_latency_enabled(struct intel_iommu *iommu, enum latency_type type)
+{
+	struct latency_statistic *lstat = iommu->perf_statistic;
+
+	return lstat && lstat[type].enabled;
+}
+
+int dmar_latency_enable(struct intel_iommu *iommu, enum latency_type type)
+{
+	struct latency_statistic *lstat;
+	unsigned long flags;
+	int ret = -EBUSY;
+
+	if (dmar_latency_enabled(iommu, type))
+		return 0;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	if (!iommu->perf_statistic) {
+		iommu->perf_statistic = kzalloc_objs(*lstat, DMAR_LATENCY_NUM,
+						     GFP_ATOMIC);
+		if (!iommu->perf_statistic) {
+			ret = -ENOMEM;
+			goto unlock_out;
+		}
+	}
+
+	lstat = iommu->perf_statistic;
+
+	if (!lstat[type].enabled) {
+		lstat[type].enabled = true;
+		lstat[type].counter[COUNTS_MIN] = UINT_MAX;
+		ret = 0;
+	}
+unlock_out:
+	spin_unlock_irqrestore(&latency_lock, flags);
+
+	return ret;
+}
+
+void dmar_latency_disable(struct intel_iommu *iommu, enum latency_type type)
+{
+	struct latency_statistic *lstat = iommu->perf_statistic;
+	unsigned long flags;
+
+	if (!dmar_latency_enabled(iommu, type))
+		return;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	memset(&lstat[type], 0, sizeof(*lstat));
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+void dmar_latency_update(struct intel_iommu *iommu, enum latency_type type, u64 latency)
+{
+	struct latency_statistic *lstat = iommu->perf_statistic;
+	unsigned long flags;
+	u64 min, max;
+
+	if (!dmar_latency_enabled(iommu, type))
+		return;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	if (latency < 100)
+		lstat[type].counter[COUNTS_10e2]++;
+	else if (latency < 1000)
+		lstat[type].counter[COUNTS_10e3]++;
+	else if (latency < 10000)
+		lstat[type].counter[COUNTS_10e4]++;
+	else if (latency < 100000)
+		lstat[type].counter[COUNTS_10e5]++;
+	else if (latency < 1000000)
+		lstat[type].counter[COUNTS_10e6]++;
+	else if (latency < 10000000)
+		lstat[type].counter[COUNTS_10e7]++;
+	else
+		lstat[type].counter[COUNTS_10e8_plus]++;
+
+	min = lstat[type].counter[COUNTS_MIN];
+	max = lstat[type].counter[COUNTS_MAX];
+	lstat[type].counter[COUNTS_MIN] = min_t(u64, min, latency);
+	lstat[type].counter[COUNTS_MAX] = max_t(u64, max, latency);
+	lstat[type].counter[COUNTS_SUM] += latency;
+	lstat[type].samples++;
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static char *latency_counter_names[] = {
+	"                  <0.1us",
+	"   0.1us-1us", "    1us-10us", "  10us-100us",
+	"   100us-1ms", "    1ms-10ms", "      >=10ms",
+	"     min(us)", "     max(us)", " average(us)"
+};
+
+static char *latency_type_names[] = {
+	"   inv_iotlb", "  inv_devtlb", "     inv_iec",
+	"     svm_prq"
+};
+
+void dmar_latency_snapshot(struct intel_iommu *iommu, char *str, size_t size)
+{
+	struct latency_statistic *lstat = iommu->perf_statistic;
+	unsigned long flags;
+	int bytes = 0, i, j;
+
+	memset(str, 0, size);
+
+	for (i = 0; i < COUNTS_NUM; i++)
+		bytes += scnprintf(str + bytes, size - bytes,
+				  "%s", latency_counter_names[i]);
+
+	spin_lock_irqsave(&latency_lock, flags);
+	for (i = 0; i < DMAR_LATENCY_NUM; i++) {
+		if (!dmar_latency_enabled(iommu, i))
+			continue;
+
+		bytes += scnprintf(str + bytes, size - bytes,
+				  "\n%s", latency_type_names[i]);
+
+		for (j = 0; j < COUNTS_NUM; j++) {
+			u64 val = lstat[i].counter[j];
+
+			switch (j) {
+			case COUNTS_MIN:
+				if (val == UINT_MAX)
+					val = 0;
+				else
+					val = div_u64(val, 1000);
+				break;
+			case COUNTS_MAX:
+				val = div_u64(val, 1000);
+				break;
+			case COUNTS_SUM:
+				if (lstat[i].samples)
+					val = div_u64(val, (lstat[i].samples * 1000));
+				else
+					val = 0;
+				break;
+			default:
+				break;
+			}
+
+			bytes += scnprintf(str + bytes, size - bytes,
+					  "%12lld", val);
+		}
+	}
+	spin_unlock_irqrestore(&latency_lock, flags);
+}

linux-7.2.3/drivers/iommu/intel/perf.h

@@ -0,0 +1,71 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * perf.h - performance monitor header
+ *
+ * Copyright (C) 2021 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+enum latency_type {
+	DMAR_LATENCY_INV_IOTLB = 0,
+	DMAR_LATENCY_INV_DEVTLB,
+	DMAR_LATENCY_INV_IEC,
+	DMAR_LATENCY_NUM
+};
+
+enum latency_count {
+	COUNTS_10e2 = 0,	/* < 0.1us	*/
+	COUNTS_10e3,		/* 0.1us ~ 1us	*/
+	COUNTS_10e4,		/* 1us ~ 10us	*/
+	COUNTS_10e5,		/* 10us ~ 100us	*/
+	COUNTS_10e6,		/* 100us ~ 1ms	*/
+	COUNTS_10e7,		/* 1ms ~ 10ms	*/
+	COUNTS_10e8_plus,	/* 10ms and plus*/
+	COUNTS_MIN,
+	COUNTS_MAX,
+	COUNTS_SUM,
+	COUNTS_NUM
+};
+
+struct latency_statistic {
+	bool enabled;
+	u64 counter[COUNTS_NUM];
+	u64 samples;
+};
+
+#ifdef CONFIG_DMAR_PERF
+int dmar_latency_enable(struct intel_iommu *iommu, enum latency_type type);
+void dmar_latency_disable(struct intel_iommu *iommu, enum latency_type type);
+bool dmar_latency_enabled(struct intel_iommu *iommu, enum latency_type type);
+void dmar_latency_update(struct intel_iommu *iommu, enum latency_type type,
+			 u64 latency);
+void dmar_latency_snapshot(struct intel_iommu *iommu, char *str, size_t size);
+#else
+static inline int
+dmar_latency_enable(struct intel_iommu *iommu, enum latency_type type)
+{
+	return -EINVAL;
+}
+
+static inline void
+dmar_latency_disable(struct intel_iommu *iommu, enum latency_type type)
+{
+}
+
+static inline bool
+dmar_latency_enabled(struct intel_iommu *iommu, enum latency_type type)
+{
+	return false;
+}
+
+static inline void
+dmar_latency_update(struct intel_iommu *iommu, enum latency_type type, u64 latency)
+{
+}
+
+static inline void
+dmar_latency_snapshot(struct intel_iommu *iommu, char *str, size_t size)
+{
+}
+#endif /* CONFIG_DMAR_PERF */

linux-7.2.3/drivers/iommu/intel/perfmon.c

@@ -0,0 +1,790 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Support Intel IOMMU PerfMon
+ * Copyright(c) 2023 Intel Corporation.
+ */
+#define pr_fmt(fmt)	"DMAR: " fmt
+#define dev_fmt(fmt)	pr_fmt(fmt)
+
+#include <linux/dmar.h>
+#include "iommu.h"
+#include "perfmon.h"
+
+PMU_FORMAT_ATTR(event,		"config:0-27");		/* ES: Events Select */
+PMU_FORMAT_ATTR(event_group,	"config:28-31");	/* EGI: Event Group Index */
+
+static struct attribute *iommu_pmu_format_attrs[] = {
+	&format_attr_event_group.attr,
+	&format_attr_event.attr,
+	NULL
+};
+
+static struct attribute_group iommu_pmu_format_attr_group = {
+	.name = "format",
+	.attrs = iommu_pmu_format_attrs,
+};
+
+/* The available events are added in attr_update later */
+static struct attribute *attrs_empty[] = {
+	NULL
+};
+
+static struct attribute_group iommu_pmu_events_attr_group = {
+	.name = "events",
+	.attrs = attrs_empty,
+};
+
+static const struct attribute_group *iommu_pmu_attr_groups[] = {
+	&iommu_pmu_format_attr_group,
+	&iommu_pmu_events_attr_group,
+	NULL
+};
+
+static inline struct iommu_pmu *dev_to_iommu_pmu(struct device *dev)
+{
+	/*
+	 * The perf_event creates its own dev for each PMU.
+	 * See pmu_dev_alloc()
+	 */
+	return container_of(dev_get_drvdata(dev), struct iommu_pmu, pmu);
+}
+
+#define IOMMU_PMU_ATTR(_name, _format, _filter)				\
+	PMU_FORMAT_ATTR(_name, _format);				\
+									\
+static struct attribute *_name##_attr[] = {				\
+	&format_attr_##_name.attr,					\
+	NULL								\
+};									\
+									\
+static umode_t								\
+_name##_is_visible(struct kobject *kobj, struct attribute *attr, int i)	\
+{									\
+	struct device *dev = kobj_to_dev(kobj);				\
+	struct iommu_pmu *iommu_pmu = dev_to_iommu_pmu(dev);		\
+									\
+	if (!iommu_pmu)							\
+		return 0;						\
+	return (iommu_pmu->filter & _filter) ? attr->mode : 0;		\
+}									\
+									\
+static struct attribute_group _name = {					\
+	.name		= "format",					\
+	.attrs		= _name##_attr,					\
+	.is_visible	= _name##_is_visible,				\
+};
+
+IOMMU_PMU_ATTR(filter_requester_id_en,	"config1:0",		IOMMU_PMU_FILTER_REQUESTER_ID);
+IOMMU_PMU_ATTR(filter_domain_en,	"config1:1",		IOMMU_PMU_FILTER_DOMAIN);
+IOMMU_PMU_ATTR(filter_pasid_en,		"config1:2",		IOMMU_PMU_FILTER_PASID);
+IOMMU_PMU_ATTR(filter_ats_en,		"config1:3",		IOMMU_PMU_FILTER_ATS);
+IOMMU_PMU_ATTR(filter_page_table_en,	"config1:4",		IOMMU_PMU_FILTER_PAGE_TABLE);
+IOMMU_PMU_ATTR(filter_requester_id,	"config1:16-31",	IOMMU_PMU_FILTER_REQUESTER_ID);
+IOMMU_PMU_ATTR(filter_domain,		"config1:32-47",	IOMMU_PMU_FILTER_DOMAIN);
+IOMMU_PMU_ATTR(filter_pasid,		"config2:0-21",		IOMMU_PMU_FILTER_PASID);
+IOMMU_PMU_ATTR(filter_ats,		"config2:24-28",	IOMMU_PMU_FILTER_ATS);
+IOMMU_PMU_ATTR(filter_page_table,	"config2:32-36",	IOMMU_PMU_FILTER_PAGE_TABLE);
+
+#define iommu_pmu_en_requester_id(e)		((e) & 0x1)
+#define iommu_pmu_en_domain(e)			(((e) >> 1) & 0x1)
+#define iommu_pmu_en_pasid(e)			(((e) >> 2) & 0x1)
+#define iommu_pmu_en_ats(e)			(((e) >> 3) & 0x1)
+#define iommu_pmu_en_page_table(e)		(((e) >> 4) & 0x1)
+#define iommu_pmu_get_requester_id(filter)	(((filter) >> 16) & 0xffff)
+#define iommu_pmu_get_domain(filter)		(((filter) >> 32) & 0xffff)
+#define iommu_pmu_get_pasid(filter)		((filter) & 0x3fffff)
+#define iommu_pmu_get_ats(filter)		(((filter) >> 24) & 0x1f)
+#define iommu_pmu_get_page_table(filter)	(((filter) >> 32) & 0x1f)
+
+#define iommu_pmu_set_filter(_name, _config, _filter, _idx, _econfig)		\
+{										\
+	if ((iommu_pmu->filter & _filter) && iommu_pmu_en_##_name(_econfig)) {	\
+		writel(iommu_pmu_get_##_name(_config) | IOMMU_PMU_FILTER_EN,	\
+		       iommu_pmu->cfg_reg + _idx * IOMMU_PMU_CFG_OFFSET +	\
+		       IOMMU_PMU_CFG_SIZE +					\
+		       (ffs(_filter) - 1) * IOMMU_PMU_CFG_FILTERS_OFFSET);	\
+	}									\
+}
+
+#define iommu_pmu_clear_filter(_filter, _idx)					\
+{										\
+	if (iommu_pmu->filter & _filter) {					\
+		writel(0,							\
+		       iommu_pmu->cfg_reg + _idx * IOMMU_PMU_CFG_OFFSET +	\
+		       IOMMU_PMU_CFG_SIZE +					\
+		       (ffs(_filter) - 1) * IOMMU_PMU_CFG_FILTERS_OFFSET);	\
+	}									\
+}
+
+/*
+ * Define the event attr related functions
+ * Input: _name: event attr name
+ *        _string: string of the event in sysfs
+ *        _g_idx: event group encoding
+ *        _event: event encoding
+ */
+#define IOMMU_PMU_EVENT_ATTR(_name, _string, _g_idx, _event)			\
+	PMU_EVENT_ATTR_STRING(_name, event_attr_##_name, _string)		\
+										\
+static struct attribute *_name##_attr[] = {					\
+	&event_attr_##_name.attr.attr,						\
+	NULL									\
+};										\
+										\
+static umode_t									\
+_name##_is_visible(struct kobject *kobj, struct attribute *attr, int i)		\
+{										\
+	struct device *dev = kobj_to_dev(kobj);					\
+	struct iommu_pmu *iommu_pmu = dev_to_iommu_pmu(dev);			\
+										\
+	if (!iommu_pmu)								\
+		return 0;							\
+	return (iommu_pmu->evcap[_g_idx] & _event) ? attr->mode : 0;		\
+}										\
+										\
+static struct attribute_group _name = {						\
+	.name		= "events",						\
+	.attrs		= _name##_attr,						\
+	.is_visible	= _name##_is_visible,					\
+};
+
+IOMMU_PMU_EVENT_ATTR(iommu_clocks,		"event_group=0x0,event=0x001", 0x0, 0x001)
+IOMMU_PMU_EVENT_ATTR(iommu_requests,		"event_group=0x0,event=0x002", 0x0, 0x002)
+IOMMU_PMU_EVENT_ATTR(pw_occupancy,		"event_group=0x0,event=0x004", 0x0, 0x004)
+IOMMU_PMU_EVENT_ATTR(ats_blocked,		"event_group=0x0,event=0x008", 0x0, 0x008)
+IOMMU_PMU_EVENT_ATTR(iommu_mrds,		"event_group=0x1,event=0x001", 0x1, 0x001)
+IOMMU_PMU_EVENT_ATTR(iommu_mem_blocked,		"event_group=0x1,event=0x020", 0x1, 0x020)
+IOMMU_PMU_EVENT_ATTR(pg_req_posted,		"event_group=0x1,event=0x040", 0x1, 0x040)
+IOMMU_PMU_EVENT_ATTR(ctxt_cache_lookup,		"event_group=0x2,event=0x001", 0x2, 0x001)
+IOMMU_PMU_EVENT_ATTR(ctxt_cache_hit,		"event_group=0x2,event=0x002", 0x2, 0x002)
+IOMMU_PMU_EVENT_ATTR(pasid_cache_lookup,	"event_group=0x2,event=0x004", 0x2, 0x004)
+IOMMU_PMU_EVENT_ATTR(pasid_cache_hit,		"event_group=0x2,event=0x008", 0x2, 0x008)
+IOMMU_PMU_EVENT_ATTR(ss_nonleaf_lookup,		"event_group=0x2,event=0x010", 0x2, 0x010)
+IOMMU_PMU_EVENT_ATTR(ss_nonleaf_hit,		"event_group=0x2,event=0x020", 0x2, 0x020)
+IOMMU_PMU_EVENT_ATTR(fs_nonleaf_lookup,		"event_group=0x2,event=0x040", 0x2, 0x040)
+IOMMU_PMU_EVENT_ATTR(fs_nonleaf_hit,		"event_group=0x2,event=0x080", 0x2, 0x080)
+IOMMU_PMU_EVENT_ATTR(hpt_nonleaf_lookup,	"event_group=0x2,event=0x100", 0x2, 0x100)
+IOMMU_PMU_EVENT_ATTR(hpt_nonleaf_hit,		"event_group=0x2,event=0x200", 0x2, 0x200)
+IOMMU_PMU_EVENT_ATTR(iotlb_lookup,		"event_group=0x3,event=0x001", 0x3, 0x001)
+IOMMU_PMU_EVENT_ATTR(iotlb_hit,			"event_group=0x3,event=0x002", 0x3, 0x002)
+IOMMU_PMU_EVENT_ATTR(hpt_leaf_lookup,		"event_group=0x3,event=0x004", 0x3, 0x004)
+IOMMU_PMU_EVENT_ATTR(hpt_leaf_hit,		"event_group=0x3,event=0x008", 0x3, 0x008)
+IOMMU_PMU_EVENT_ATTR(int_cache_lookup,		"event_group=0x4,event=0x001", 0x4, 0x001)
+IOMMU_PMU_EVENT_ATTR(int_cache_hit_nonposted,	"event_group=0x4,event=0x002", 0x4, 0x002)
+IOMMU_PMU_EVENT_ATTR(int_cache_hit_posted,	"event_group=0x4,event=0x004", 0x4, 0x004)
+
+static const struct attribute_group *iommu_pmu_attr_update[] = {
+	&filter_requester_id_en,
+	&filter_domain_en,
+	&filter_pasid_en,
+	&filter_ats_en,
+	&filter_page_table_en,
+	&filter_requester_id,
+	&filter_domain,
+	&filter_pasid,
+	&filter_ats,
+	&filter_page_table,
+	&iommu_clocks,
+	&iommu_requests,
+	&pw_occupancy,
+	&ats_blocked,
+	&iommu_mrds,
+	&iommu_mem_blocked,
+	&pg_req_posted,
+	&ctxt_cache_lookup,
+	&ctxt_cache_hit,
+	&pasid_cache_lookup,
+	&pasid_cache_hit,
+	&ss_nonleaf_lookup,
+	&ss_nonleaf_hit,
+	&fs_nonleaf_lookup,
+	&fs_nonleaf_hit,
+	&hpt_nonleaf_lookup,
+	&hpt_nonleaf_hit,
+	&iotlb_lookup,
+	&iotlb_hit,
+	&hpt_leaf_lookup,
+	&hpt_leaf_hit,
+	&int_cache_lookup,
+	&int_cache_hit_nonposted,
+	&int_cache_hit_posted,
+	NULL
+};
+
+static inline void __iomem *
+iommu_event_base(struct iommu_pmu *iommu_pmu, int idx)
+{
+	return iommu_pmu->cntr_reg + idx * iommu_pmu->cntr_stride;
+}
+
+static inline void __iomem *
+iommu_config_base(struct iommu_pmu *iommu_pmu, int idx)
+{
+	return iommu_pmu->cfg_reg + idx * IOMMU_PMU_CFG_OFFSET;
+}
+
+static inline struct iommu_pmu *iommu_event_to_pmu(struct perf_event *event)
+{
+	return container_of(event->pmu, struct iommu_pmu, pmu);
+}
+
+static inline u64 iommu_event_config(struct perf_event *event)
+{
+	u64 config = event->attr.config;
+
+	return (iommu_event_select(config) << IOMMU_EVENT_CFG_ES_SHIFT) |
+	       (iommu_event_group(config) << IOMMU_EVENT_CFG_EGI_SHIFT) |
+	       IOMMU_EVENT_CFG_INT;
+}
+
+static inline bool is_iommu_pmu_event(struct iommu_pmu *iommu_pmu,
+				      struct perf_event *event)
+{
+	return event->pmu == &iommu_pmu->pmu;
+}
+
+static int iommu_pmu_validate_event(struct perf_event *event)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	u32 event_group = iommu_event_group(event->attr.config);
+
+	if (event_group >= iommu_pmu->num_eg)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int iommu_pmu_validate_group(struct perf_event *event)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	struct perf_event *sibling;
+	int nr = 0;
+
+	/*
+	 * All events in a group must be scheduled simultaneously.
+	 * Check whether there is enough counters for all the events.
+	 */
+	for_each_sibling_event(sibling, event->group_leader) {
+		if (!is_iommu_pmu_event(iommu_pmu, sibling) ||
+		    sibling->state <= PERF_EVENT_STATE_OFF)
+			continue;
+
+		if (++nr > iommu_pmu->num_cntr)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int iommu_pmu_event_init(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/* sampling not supported */
+	if (event->attr.sample_period)
+		return -EINVAL;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	if (iommu_pmu_validate_event(event))
+		return -EINVAL;
+
+	hwc->config = iommu_event_config(event);
+
+	return iommu_pmu_validate_group(event);
+}
+
+static void iommu_pmu_event_update(struct perf_event *event)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	struct hw_perf_event *hwc = &event->hw;
+	u64 prev_count, new_count, delta;
+	int shift = 64 - iommu_pmu->cntr_width;
+
+again:
+	prev_count = local64_read(&hwc->prev_count);
+	new_count = readq(iommu_event_base(iommu_pmu, hwc->idx));
+	if (local64_xchg(&hwc->prev_count, new_count) != prev_count)
+		goto again;
+
+	/*
+	 * The counter width is enumerated. Always shift the counter
+	 * before using it.
+	 */
+	delta = (new_count << shift) - (prev_count << shift);
+	delta >>= shift;
+
+	local64_add(delta, &event->count);
+}
+
+static void iommu_pmu_start(struct perf_event *event, int flags)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	struct intel_iommu *iommu = iommu_pmu->iommu;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 count;
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	if (WARN_ON_ONCE(hwc->idx < 0 || hwc->idx >= IOMMU_PMU_IDX_MAX))
+		return;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
+
+	/* Always reprogram the period */
+	count = readq(iommu_event_base(iommu_pmu, hwc->idx));
+	local64_set((&hwc->prev_count), count);
+
+	/*
+	 * The error of ecmd will be ignored.
+	 * - The existing perf_event subsystem doesn't handle the error.
+	 *   Only IOMMU PMU returns runtime HW error. We don't want to
+	 *   change the existing generic interfaces for the specific case.
+	 * - It's a corner case caused by HW, which is very unlikely to
+	 *   happen. There is nothing SW can do.
+	 * - The worst case is that the user will get <not count> with
+	 *   perf command, which can give the user some hints.
+	 */
+	ecmd_submit_sync(iommu, DMA_ECMD_ENABLE, hwc->idx, 0);
+
+	perf_event_update_userpage(event);
+}
+
+static void iommu_pmu_stop(struct perf_event *event, int flags)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	struct intel_iommu *iommu = iommu_pmu->iommu;
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		ecmd_submit_sync(iommu, DMA_ECMD_DISABLE, hwc->idx, 0);
+
+		iommu_pmu_event_update(event);
+
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
+}
+
+static inline int
+iommu_pmu_validate_per_cntr_event(struct iommu_pmu *iommu_pmu,
+				  int idx, struct perf_event *event)
+{
+	u32 event_group = iommu_event_group(event->attr.config);
+	u32 select = iommu_event_select(event->attr.config);
+
+	if (!(iommu_pmu->cntr_evcap[idx][event_group] & select))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int iommu_pmu_assign_event(struct iommu_pmu *iommu_pmu,
+				  struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx;
+
+	/*
+	 * The counters which support limited events are usually at the end.
+	 * Schedule them first to accommodate more events.
+	 */
+	for (idx = iommu_pmu->num_cntr - 1; idx >= 0; idx--) {
+		if (test_and_set_bit(idx, iommu_pmu->used_mask))
+			continue;
+		/* Check per-counter event capabilities */
+		if (!iommu_pmu_validate_per_cntr_event(iommu_pmu, idx, event))
+			break;
+		clear_bit(idx, iommu_pmu->used_mask);
+	}
+	if (idx < 0)
+		return -EINVAL;
+
+	iommu_pmu->event_list[idx] = event;
+	hwc->idx = idx;
+
+	/* config events */
+	writeq(hwc->config, iommu_config_base(iommu_pmu, idx));
+
+	iommu_pmu_set_filter(requester_id, event->attr.config1,
+			     IOMMU_PMU_FILTER_REQUESTER_ID, idx,
+			     event->attr.config1);
+	iommu_pmu_set_filter(domain, event->attr.config1,
+			     IOMMU_PMU_FILTER_DOMAIN, idx,
+			     event->attr.config1);
+	iommu_pmu_set_filter(pasid, event->attr.config2,
+			     IOMMU_PMU_FILTER_PASID, idx,
+			     event->attr.config1);
+	iommu_pmu_set_filter(ats, event->attr.config2,
+			     IOMMU_PMU_FILTER_ATS, idx,
+			     event->attr.config1);
+	iommu_pmu_set_filter(page_table, event->attr.config2,
+			     IOMMU_PMU_FILTER_PAGE_TABLE, idx,
+			     event->attr.config1);
+
+	return 0;
+}
+
+static int iommu_pmu_add(struct perf_event *event, int flags)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	struct hw_perf_event *hwc = &event->hw;
+	int ret;
+
+	ret = iommu_pmu_assign_event(iommu_pmu, event);
+	if (ret < 0)
+		return ret;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	if (flags & PERF_EF_START)
+		iommu_pmu_start(event, 0);
+
+	return 0;
+}
+
+static void iommu_pmu_del(struct perf_event *event, int flags)
+{
+	struct iommu_pmu *iommu_pmu = iommu_event_to_pmu(event);
+	int idx = event->hw.idx;
+
+	iommu_pmu_stop(event, PERF_EF_UPDATE);
+
+	iommu_pmu_clear_filter(IOMMU_PMU_FILTER_REQUESTER_ID, idx);
+	iommu_pmu_clear_filter(IOMMU_PMU_FILTER_DOMAIN, idx);
+	iommu_pmu_clear_filter(IOMMU_PMU_FILTER_PASID, idx);
+	iommu_pmu_clear_filter(IOMMU_PMU_FILTER_ATS, idx);
+	iommu_pmu_clear_filter(IOMMU_PMU_FILTER_PAGE_TABLE, idx);
+
+	iommu_pmu->event_list[idx] = NULL;
+	event->hw.idx = -1;
+	clear_bit(idx, iommu_pmu->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
+static void iommu_pmu_enable(struct pmu *pmu)
+{
+	struct iommu_pmu *iommu_pmu = container_of(pmu, struct iommu_pmu, pmu);
+	struct intel_iommu *iommu = iommu_pmu->iommu;
+
+	ecmd_submit_sync(iommu, DMA_ECMD_UNFREEZE, 0, 0);
+}
+
+static void iommu_pmu_disable(struct pmu *pmu)
+{
+	struct iommu_pmu *iommu_pmu = container_of(pmu, struct iommu_pmu, pmu);
+	struct intel_iommu *iommu = iommu_pmu->iommu;
+
+	ecmd_submit_sync(iommu, DMA_ECMD_FREEZE, 0, 0);
+}
+
+static void iommu_pmu_counter_overflow(struct iommu_pmu *iommu_pmu)
+{
+	struct perf_event *event;
+	u64 status;
+	int i;
+
+	/*
+	 * Two counters may be overflowed very close. Always check
+	 * whether there are more to handle.
+	 */
+	while ((status = readq(iommu_pmu->overflow))) {
+		for_each_set_bit(i, (unsigned long *)&status, iommu_pmu->num_cntr) {
+			/*
+			 * Find the assigned event of the counter.
+			 * Accumulate the value into the event->count.
+			 */
+			event = iommu_pmu->event_list[i];
+			if (!event) {
+				pr_warn_once("Cannot find the assigned event for counter %d\n", i);
+				continue;
+			}
+			iommu_pmu_event_update(event);
+		}
+
+		writeq(status, iommu_pmu->overflow);
+	}
+}
+
+static irqreturn_t iommu_pmu_irq_handler(int irq, void *dev_id)
+{
+	struct intel_iommu *iommu = dev_id;
+
+	if (!readl(iommu->reg + DMAR_PERFINTRSTS_REG))
+		return IRQ_NONE;
+
+	iommu_pmu_counter_overflow(iommu->pmu);
+
+	/* Clear the status bit */
+	writel(DMA_PERFINTRSTS_PIS, iommu->reg + DMAR_PERFINTRSTS_REG);
+
+	return IRQ_HANDLED;
+}
+
+static int __iommu_pmu_register(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu = iommu->pmu;
+
+	iommu_pmu->pmu.name		= iommu->name;
+	iommu_pmu->pmu.task_ctx_nr	= perf_invalid_context;
+	iommu_pmu->pmu.event_init	= iommu_pmu_event_init;
+	iommu_pmu->pmu.pmu_enable	= iommu_pmu_enable;
+	iommu_pmu->pmu.pmu_disable	= iommu_pmu_disable;
+	iommu_pmu->pmu.add		= iommu_pmu_add;
+	iommu_pmu->pmu.del		= iommu_pmu_del;
+	iommu_pmu->pmu.start		= iommu_pmu_start;
+	iommu_pmu->pmu.stop		= iommu_pmu_stop;
+	iommu_pmu->pmu.read		= iommu_pmu_event_update;
+	iommu_pmu->pmu.attr_groups	= iommu_pmu_attr_groups;
+	iommu_pmu->pmu.attr_update	= iommu_pmu_attr_update;
+	iommu_pmu->pmu.capabilities	= PERF_PMU_CAP_NO_EXCLUDE;
+	iommu_pmu->pmu.scope		= PERF_PMU_SCOPE_SYS_WIDE;
+	iommu_pmu->pmu.module		= THIS_MODULE;
+
+	return perf_pmu_register(&iommu_pmu->pmu, iommu_pmu->pmu.name, -1);
+}
+
+static inline void __iomem *
+get_perf_reg_address(struct intel_iommu *iommu, u32 offset)
+{
+	u32 off = readl(iommu->reg + offset);
+
+	return iommu->reg + off;
+}
+
+int alloc_iommu_pmu(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu;
+	int i, j, ret;
+	u64 perfcap;
+	u32 cap;
+
+	if (!ecap_pms(iommu->ecap))
+		return 0;
+
+	/* The IOMMU PMU requires the ECMD support as well */
+	if (!cap_ecmds(iommu->cap))
+		return -ENODEV;
+
+	perfcap = readq(iommu->reg + DMAR_PERFCAP_REG);
+	/* The performance monitoring is not supported. */
+	if (!perfcap)
+		return -ENODEV;
+
+	/* Sanity check for the number of the counters and event groups */
+	if (!pcap_num_cntr(perfcap) || !pcap_num_event_group(perfcap))
+		return -ENODEV;
+
+	/* The interrupt on overflow is required */
+	if (!pcap_interrupt(perfcap))
+		return -ENODEV;
+
+	/* Check required Enhanced Command Capability */
+	if (!ecmd_has_pmu_essential(iommu))
+		return -ENODEV;
+
+	iommu_pmu = kzalloc_obj(*iommu_pmu);
+	if (!iommu_pmu)
+		return -ENOMEM;
+
+	iommu_pmu->num_cntr = pcap_num_cntr(perfcap);
+	if (iommu_pmu->num_cntr > IOMMU_PMU_IDX_MAX) {
+		pr_warn_once("The number of IOMMU counters %d > max(%d), clipping!",
+			     iommu_pmu->num_cntr, IOMMU_PMU_IDX_MAX);
+		iommu_pmu->num_cntr = IOMMU_PMU_IDX_MAX;
+	}
+
+	iommu_pmu->cntr_width = pcap_cntr_width(perfcap);
+	iommu_pmu->filter = pcap_filters_mask(perfcap);
+	iommu_pmu->cntr_stride = pcap_cntr_stride(perfcap);
+	iommu_pmu->num_eg = pcap_num_event_group(perfcap);
+
+	iommu_pmu->evcap = kcalloc(iommu_pmu->num_eg, sizeof(u64), GFP_KERNEL);
+	if (!iommu_pmu->evcap) {
+		ret = -ENOMEM;
+		goto free_pmu;
+	}
+
+	/* Parse event group capabilities */
+	for (i = 0; i < iommu_pmu->num_eg; i++) {
+		u64 pcap;
+
+		pcap = readq(iommu->reg + DMAR_PERFEVNTCAP_REG +
+			     i * IOMMU_PMU_CAP_REGS_STEP);
+		iommu_pmu->evcap[i] = pecap_es(pcap);
+	}
+
+	iommu_pmu->cntr_evcap = kcalloc(iommu_pmu->num_cntr, sizeof(u32 *), GFP_KERNEL);
+	if (!iommu_pmu->cntr_evcap) {
+		ret = -ENOMEM;
+		goto free_pmu_evcap;
+	}
+	for (i = 0; i < iommu_pmu->num_cntr; i++) {
+		iommu_pmu->cntr_evcap[i] = kcalloc(iommu_pmu->num_eg, sizeof(u32), GFP_KERNEL);
+		if (!iommu_pmu->cntr_evcap[i]) {
+			ret = -ENOMEM;
+			goto free_pmu_cntr_evcap;
+		}
+		/*
+		 * Set to the global capabilities, will adjust according
+		 * to per-counter capabilities later.
+		 */
+		for (j = 0; j < iommu_pmu->num_eg; j++)
+			iommu_pmu->cntr_evcap[i][j] = (u32)iommu_pmu->evcap[j];
+	}
+
+	iommu_pmu->cfg_reg = get_perf_reg_address(iommu, DMAR_PERFCFGOFF_REG);
+	iommu_pmu->cntr_reg = get_perf_reg_address(iommu, DMAR_PERFCNTROFF_REG);
+	iommu_pmu->overflow = get_perf_reg_address(iommu, DMAR_PERFOVFOFF_REG);
+
+	/*
+	 * Check per-counter capabilities. All counters should have the
+	 * same capabilities on Interrupt on Overflow Support and Counter
+	 * Width.
+	 */
+	for (i = 0; i < iommu_pmu->num_cntr; i++) {
+		cap = readl(iommu_pmu->cfg_reg +
+			    i * IOMMU_PMU_CFG_OFFSET +
+			    IOMMU_PMU_CFG_CNTRCAP_OFFSET);
+		if (!iommu_cntrcap_pcc(cap))
+			continue;
+
+		/*
+		 * It's possible that some counters have a different
+		 * capability because of e.g., HW bug. Check the corner
+		 * case here and simply drop those counters.
+		 */
+		if ((iommu_cntrcap_cw(cap) != iommu_pmu->cntr_width) ||
+		    !iommu_cntrcap_ios(cap)) {
+			iommu_pmu->num_cntr = i;
+			pr_warn("PMU counter capability inconsistent, counter number reduced to %d\n",
+				iommu_pmu->num_cntr);
+		}
+
+		/* Clear the pre-defined events group */
+		for (j = 0; j < iommu_pmu->num_eg; j++)
+			iommu_pmu->cntr_evcap[i][j] = 0;
+
+		/* Override with per-counter event capabilities */
+		for (j = 0; j < iommu_cntrcap_egcnt(cap); j++) {
+			cap = readl(iommu_pmu->cfg_reg + i * IOMMU_PMU_CFG_OFFSET +
+				    IOMMU_PMU_CFG_CNTREVCAP_OFFSET +
+				    (j * IOMMU_PMU_OFF_REGS_STEP));
+			iommu_pmu->cntr_evcap[i][iommu_event_group(cap)] = iommu_event_select(cap);
+			/*
+			 * Some events may only be supported by a specific counter.
+			 * Track them in the evcap as well.
+			 */
+			iommu_pmu->evcap[iommu_event_group(cap)] |= iommu_event_select(cap);
+		}
+	}
+
+	iommu_pmu->iommu = iommu;
+	iommu->pmu = iommu_pmu;
+
+	return 0;
+
+free_pmu_cntr_evcap:
+	for (i = 0; i < iommu_pmu->num_cntr; i++)
+		kfree(iommu_pmu->cntr_evcap[i]);
+	kfree(iommu_pmu->cntr_evcap);
+free_pmu_evcap:
+	kfree(iommu_pmu->evcap);
+free_pmu:
+	kfree(iommu_pmu);
+
+	return ret;
+}
+
+void free_iommu_pmu(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu = iommu->pmu;
+
+	if (!iommu_pmu)
+		return;
+
+	if (iommu_pmu->evcap) {
+		int i;
+
+		for (i = 0; i < iommu_pmu->num_cntr; i++)
+			kfree(iommu_pmu->cntr_evcap[i]);
+		kfree(iommu_pmu->cntr_evcap);
+	}
+	kfree(iommu_pmu->evcap);
+	kfree(iommu_pmu);
+	iommu->pmu = NULL;
+}
+
+static int iommu_pmu_set_interrupt(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu = iommu->pmu;
+	int irq, ret;
+
+	irq = dmar_alloc_hwirq(IOMMU_IRQ_ID_OFFSET_PERF + iommu->seq_id, iommu->node, iommu);
+	if (irq <= 0)
+		return -EINVAL;
+
+	snprintf(iommu_pmu->irq_name, sizeof(iommu_pmu->irq_name), "dmar%d-perf", iommu->seq_id);
+
+	iommu->perf_irq = irq;
+	ret = request_threaded_irq(irq, NULL, iommu_pmu_irq_handler,
+				   IRQF_ONESHOT, iommu_pmu->irq_name, iommu);
+	if (ret) {
+		dmar_free_hwirq(irq);
+		iommu->perf_irq = 0;
+		return ret;
+	}
+	return 0;
+}
+
+static void iommu_pmu_unset_interrupt(struct intel_iommu *iommu)
+{
+	if (!iommu->perf_irq)
+		return;
+
+	free_irq(iommu->perf_irq, iommu);
+	dmar_free_hwirq(iommu->perf_irq);
+	iommu->perf_irq = 0;
+}
+
+void iommu_pmu_register(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu = iommu->pmu;
+
+	if (!iommu_pmu)
+		return;
+
+	if (__iommu_pmu_register(iommu))
+		goto err;
+
+	/* Set interrupt for overflow */
+	if (iommu_pmu_set_interrupt(iommu))
+		goto unregister;
+
+	return;
+
+unregister:
+	perf_pmu_unregister(&iommu_pmu->pmu);
+err:
+	pr_err("Failed to register PMU for iommu (seq_id = %d)\n", iommu->seq_id);
+	free_iommu_pmu(iommu);
+}
+
+void iommu_pmu_unregister(struct intel_iommu *iommu)
+{
+	struct iommu_pmu *iommu_pmu = iommu->pmu;
+
+	if (!iommu_pmu)
+		return;
+
+	iommu_pmu_unset_interrupt(iommu);
+	perf_pmu_unregister(&iommu_pmu->pmu);
+}

linux-7.2.3/drivers/iommu/intel/perfmon.h

@@ -0,0 +1,64 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/*
+ * PERFCFGOFF_REG, PERFFRZOFF_REG
+ * PERFOVFOFF_REG, PERFCNTROFF_REG
+ */
+#define IOMMU_PMU_NUM_OFF_REGS			4
+#define IOMMU_PMU_OFF_REGS_STEP			4
+
+#define IOMMU_PMU_FILTER_REQUESTER_ID		0x01
+#define IOMMU_PMU_FILTER_DOMAIN			0x02
+#define IOMMU_PMU_FILTER_PASID			0x04
+#define IOMMU_PMU_FILTER_ATS			0x08
+#define IOMMU_PMU_FILTER_PAGE_TABLE		0x10
+
+#define IOMMU_PMU_FILTER_EN			BIT(31)
+
+#define IOMMU_PMU_CFG_OFFSET			0x100
+#define IOMMU_PMU_CFG_CNTRCAP_OFFSET		0x80
+#define IOMMU_PMU_CFG_CNTREVCAP_OFFSET		0x84
+#define IOMMU_PMU_CFG_SIZE			0x8
+#define IOMMU_PMU_CFG_FILTERS_OFFSET		0x4
+
+#define IOMMU_PMU_CAP_REGS_STEP			8
+
+#define iommu_cntrcap_pcc(p)			((p) & 0x1)
+#define iommu_cntrcap_cw(p)			(((p) >> 8) & 0xff)
+#define iommu_cntrcap_ios(p)			(((p) >> 16) & 0x1)
+#define iommu_cntrcap_egcnt(p)			(((p) >> 28) & 0xf)
+
+#define IOMMU_EVENT_CFG_EGI_SHIFT		8
+#define IOMMU_EVENT_CFG_ES_SHIFT		32
+#define IOMMU_EVENT_CFG_INT			BIT_ULL(1)
+
+#define iommu_event_select(p)			((p) & 0xfffffff)
+#define iommu_event_group(p)			(((p) >> 28) & 0xf)
+
+#ifdef CONFIG_INTEL_IOMMU_PERF_EVENTS
+int alloc_iommu_pmu(struct intel_iommu *iommu);
+void free_iommu_pmu(struct intel_iommu *iommu);
+void iommu_pmu_register(struct intel_iommu *iommu);
+void iommu_pmu_unregister(struct intel_iommu *iommu);
+#else
+static inline int
+alloc_iommu_pmu(struct intel_iommu *iommu)
+{
+	return 0;
+}
+
+static inline void
+free_iommu_pmu(struct intel_iommu *iommu)
+{
+}
+
+static inline void
+iommu_pmu_register(struct intel_iommu *iommu)
+{
+}
+
+static inline void
+iommu_pmu_unregister(struct intel_iommu *iommu)
+{
+}
+#endif /* CONFIG_INTEL_IOMMU_PERF_EVENTS */

linux-7.2.3/drivers/iommu/intel/prq.c

@@ -0,0 +1,396 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015 Intel Corporation
+ *
+ * Originally split from drivers/iommu/intel/svm.c
+ */
+
+#include <linux/pci.h>
+#include <linux/pci-ats.h>
+
+#include "iommu.h"
+#include "pasid.h"
+#include "../iommu-pages.h"
+#include "trace.h"
+
+/* Page request queue descriptor */
+struct page_req_dsc {
+	union {
+		struct {
+			u64 type:8;
+			u64 pasid_present:1;
+			u64 rsvd:7;
+			u64 rid:16;
+			u64 pasid:20;
+			u64 exe_req:1;
+			u64 pm_req:1;
+			u64 rsvd2:10;
+		};
+		u64 qw_0;
+	};
+	union {
+		struct {
+			u64 rd_req:1;
+			u64 wr_req:1;
+			u64 lpig:1;
+			u64 prg_index:9;
+			u64 addr:52;
+		};
+		u64 qw_1;
+	};
+	u64 qw_2;
+	u64 qw_3;
+};
+
+/**
+ * intel_iommu_drain_pasid_prq - Drain page requests and responses for a pasid
+ * @dev: target device
+ * @pasid: pasid for draining
+ *
+ * Drain all pending page requests and responses related to @pasid in both
+ * software and hardware. This is supposed to be called after the device
+ * driver has stopped DMA, the pasid entry has been cleared, and both IOTLB
+ * and DevTLB have been invalidated.
+ *
+ * It waits until all pending page requests for @pasid in the page fault
+ * queue are completed by the prq handling thread. Then follow the steps
+ * described in VT-d spec CH7.10 to drain all page requests and page
+ * responses pending in the hardware.
+ */
+void intel_iommu_drain_pasid_prq(struct device *dev, u32 pasid)
+{
+	struct device_domain_info *info;
+	struct dmar_domain *domain;
+	struct intel_iommu *iommu;
+	struct qi_desc desc[3];
+	int head, tail;
+	u16 sid, did;
+
+	info = dev_iommu_priv_get(dev);
+	if (!info->iopf_refcount)
+		return;
+
+	iommu = info->iommu;
+	domain = info->domain;
+	sid = PCI_DEVID(info->bus, info->devfn);
+	did = domain ? domain_id_iommu(domain, iommu) : FLPT_DEFAULT_DID;
+
+	/*
+	 * Check and wait until all pending page requests in the queue are
+	 * handled by the prq handling thread.
+	 */
+prq_retry:
+	reinit_completion(&iommu->prq_complete);
+	tail = readq(iommu->reg + DMAR_PQT_REG) & PRQ_RING_MASK;
+	head = readq(iommu->reg + DMAR_PQH_REG) & PRQ_RING_MASK;
+	while (head != tail) {
+		struct page_req_dsc *req;
+
+		req = &iommu->prq[head / sizeof(*req)];
+		if (req->rid != sid ||
+		    (req->pasid_present && pasid != req->pasid) ||
+		    (!req->pasid_present && pasid != IOMMU_NO_PASID)) {
+			head = (head + sizeof(*req)) & PRQ_RING_MASK;
+			continue;
+		}
+
+		wait_for_completion(&iommu->prq_complete);
+		goto prq_retry;
+	}
+
+	iopf_queue_flush_dev(dev);
+
+	/*
+	 * Perform steps described in VT-d spec CH7.10 to drain page
+	 * requests and responses in hardware.
+	 */
+	memset(desc, 0, sizeof(desc));
+	desc[0].qw0 = QI_IWD_STATUS_DATA(QI_DONE) |
+			QI_IWD_FENCE |
+			QI_IWD_TYPE;
+	if (pasid == IOMMU_NO_PASID) {
+		qi_desc_iotlb(iommu, did, 0, 0, DMA_TLB_DSI_FLUSH, &desc[1]);
+		qi_desc_dev_iotlb(sid, info->pfsid, info->ats_qdep, 0,
+				  MAX_AGAW_PFN_WIDTH, &desc[2]);
+	} else {
+		qi_desc_piotlb_all(did, pasid, &desc[1]);
+		qi_desc_dev_iotlb_pasid(sid, info->pfsid, pasid, info->ats_qdep,
+					0, MAX_AGAW_PFN_WIDTH, &desc[2]);
+	}
+qi_retry:
+	reinit_completion(&iommu->prq_complete);
+	qi_submit_sync(iommu, desc, 3, QI_OPT_WAIT_DRAIN);
+	if (readl(iommu->reg + DMAR_PRS_REG) & DMA_PRS_PRO) {
+		wait_for_completion(&iommu->prq_complete);
+		goto qi_retry;
+	}
+}
+
+static bool is_canonical_address(u64 addr)
+{
+	int shift = 64 - (__VIRTUAL_MASK_SHIFT + 1);
+	long saddr = (long)addr;
+
+	return (((saddr << shift) >> shift) == saddr);
+}
+
+static void handle_bad_prq_event(struct intel_iommu *iommu,
+				 struct page_req_dsc *req, int result)
+{
+	struct qi_desc desc = { };
+
+	pr_err("%s: Invalid page request: %08llx %08llx\n",
+	       iommu->name, ((unsigned long long *)req)[0],
+	       ((unsigned long long *)req)[1]);
+
+	if (!req->lpig)
+		return;
+
+	desc.qw0 = QI_PGRP_PASID(req->pasid) |
+			QI_PGRP_DID(req->rid) |
+			QI_PGRP_PASID_P(req->pasid_present) |
+			QI_PGRP_RESP_CODE(result) |
+			QI_PGRP_RESP_TYPE;
+	desc.qw1 = QI_PGRP_IDX(req->prg_index);
+
+	qi_submit_sync(iommu, &desc, 1, 0);
+}
+
+static int prq_to_iommu_prot(struct page_req_dsc *req)
+{
+	int prot = 0;
+
+	if (req->rd_req)
+		prot |= IOMMU_FAULT_PERM_READ;
+	if (req->wr_req)
+		prot |= IOMMU_FAULT_PERM_WRITE;
+	if (req->exe_req)
+		prot |= IOMMU_FAULT_PERM_EXEC;
+	if (req->pm_req)
+		prot |= IOMMU_FAULT_PERM_PRIV;
+
+	return prot;
+}
+
+static void intel_prq_report(struct intel_iommu *iommu, struct device *dev,
+			     struct page_req_dsc *desc)
+{
+	struct iopf_fault event = { };
+
+	/* Fill in event data for device specific processing */
+	event.fault.type = IOMMU_FAULT_PAGE_REQ;
+	event.fault.prm.addr = (u64)desc->addr << VTD_PAGE_SHIFT;
+	event.fault.prm.pasid = desc->pasid;
+	event.fault.prm.grpid = desc->prg_index;
+	event.fault.prm.perm = prq_to_iommu_prot(desc);
+
+	if (desc->lpig)
+		event.fault.prm.flags |= IOMMU_FAULT_PAGE_REQUEST_LAST_PAGE;
+	if (desc->pasid_present) {
+		event.fault.prm.flags |= IOMMU_FAULT_PAGE_REQUEST_PASID_VALID;
+		event.fault.prm.flags |= IOMMU_FAULT_PAGE_RESPONSE_NEEDS_PASID;
+	}
+
+	iommu_report_device_fault(dev, &event);
+}
+
+static irqreturn_t prq_event_thread(int irq, void *d)
+{
+	struct intel_iommu *iommu = d;
+	struct page_req_dsc *req;
+	int head, tail, handled;
+	struct device *dev;
+	u64 address;
+
+	/*
+	 * Clear PPR bit before reading head/tail registers, to ensure that
+	 * we get a new interrupt if needed.
+	 */
+	writel(DMA_PRS_PPR, iommu->reg + DMAR_PRS_REG);
+
+	tail = readq(iommu->reg + DMAR_PQT_REG) & PRQ_RING_MASK;
+	head = readq(iommu->reg + DMAR_PQH_REG) & PRQ_RING_MASK;
+	handled = (head != tail);
+	while (head != tail) {
+		req = &iommu->prq[head / sizeof(*req)];
+		address = (u64)req->addr << VTD_PAGE_SHIFT;
+
+		if (unlikely(!is_canonical_address(address))) {
+			pr_err("IOMMU: %s: Address is not canonical\n",
+			       iommu->name);
+bad_req:
+			handle_bad_prq_event(iommu, req, QI_RESP_INVALID);
+			goto prq_advance;
+		}
+
+		if (unlikely(req->pm_req && (req->rd_req | req->wr_req))) {
+			pr_err("IOMMU: %s: Page request in Privilege Mode\n",
+			       iommu->name);
+			goto bad_req;
+		}
+
+		if (unlikely(req->exe_req && req->rd_req)) {
+			pr_err("IOMMU: %s: Execution request not supported\n",
+			       iommu->name);
+			goto bad_req;
+		}
+
+		/* Drop Stop Marker message. No need for a response. */
+		if (unlikely(req->lpig && !req->rd_req && !req->wr_req))
+			goto prq_advance;
+
+		/*
+		 * If prq is to be handled outside iommu driver via receiver of
+		 * the fault notifiers, we skip the page response here.
+		 */
+		mutex_lock(&iommu->iopf_lock);
+		dev = device_rbtree_find(iommu, req->rid);
+		if (!dev) {
+			mutex_unlock(&iommu->iopf_lock);
+			goto bad_req;
+		}
+
+		intel_prq_report(iommu, dev, req);
+		trace_prq_report(iommu, dev, req->qw_0, req->qw_1,
+				 req->qw_2, req->qw_3,
+				 iommu->prq_seq_number++);
+		mutex_unlock(&iommu->iopf_lock);
+prq_advance:
+		head = (head + sizeof(*req)) & PRQ_RING_MASK;
+	}
+
+	writeq(tail, iommu->reg + DMAR_PQH_REG);
+
+	/*
+	 * Clear the page request overflow bit and wake up all threads that
+	 * are waiting for the completion of this handling.
+	 */
+	if (readl(iommu->reg + DMAR_PRS_REG) & DMA_PRS_PRO) {
+		pr_info_ratelimited("IOMMU: %s: PRQ overflow detected\n",
+				    iommu->name);
+		head = readq(iommu->reg + DMAR_PQH_REG) & PRQ_RING_MASK;
+		tail = readq(iommu->reg + DMAR_PQT_REG) & PRQ_RING_MASK;
+		if (head == tail) {
+			iopf_queue_discard_partial(iommu->iopf_queue);
+			writel(DMA_PRS_PRO, iommu->reg + DMAR_PRS_REG);
+			pr_info_ratelimited("IOMMU: %s: PRQ overflow cleared",
+					    iommu->name);
+		}
+	}
+
+	if (!completion_done(&iommu->prq_complete))
+		complete(&iommu->prq_complete);
+
+	return IRQ_RETVAL(handled);
+}
+
+int intel_iommu_enable_prq(struct intel_iommu *iommu)
+{
+	struct iopf_queue *iopfq;
+	int irq, ret;
+
+	iommu->prq =
+		iommu_alloc_pages_node_sz(iommu->node, GFP_KERNEL, PRQ_SIZE);
+	if (!iommu->prq) {
+		pr_warn("IOMMU: %s: Failed to allocate page request queue\n",
+			iommu->name);
+		return -ENOMEM;
+	}
+
+	irq = dmar_alloc_hwirq(IOMMU_IRQ_ID_OFFSET_PRQ + iommu->seq_id, iommu->node, iommu);
+	if (irq <= 0) {
+		pr_err("IOMMU: %s: Failed to create IRQ vector for page request queue\n",
+		       iommu->name);
+		ret = -EINVAL;
+		goto free_prq;
+	}
+	iommu->pr_irq = irq;
+
+	snprintf(iommu->iopfq_name, sizeof(iommu->iopfq_name),
+		 "dmar%d-iopfq", iommu->seq_id);
+	iopfq = iopf_queue_alloc(iommu->iopfq_name);
+	if (!iopfq) {
+		pr_err("IOMMU: %s: Failed to allocate iopf queue\n", iommu->name);
+		ret = -ENOMEM;
+		goto free_hwirq;
+	}
+	iommu->iopf_queue = iopfq;
+
+	snprintf(iommu->prq_name, sizeof(iommu->prq_name), "dmar%d-prq", iommu->seq_id);
+
+	ret = request_threaded_irq(irq, NULL, prq_event_thread, IRQF_ONESHOT,
+				   iommu->prq_name, iommu);
+	if (ret) {
+		pr_err("IOMMU: %s: Failed to request IRQ for page request queue\n",
+		       iommu->name);
+		goto free_iopfq;
+	}
+	writeq(0ULL, iommu->reg + DMAR_PQH_REG);
+	writeq(0ULL, iommu->reg + DMAR_PQT_REG);
+	writeq(virt_to_phys(iommu->prq) | PRQ_ORDER, iommu->reg + DMAR_PQA_REG);
+
+	init_completion(&iommu->prq_complete);
+
+	return 0;
+
+free_iopfq:
+	iopf_queue_free(iommu->iopf_queue);
+	iommu->iopf_queue = NULL;
+free_hwirq:
+	dmar_free_hwirq(irq);
+	iommu->pr_irq = 0;
+free_prq:
+	iommu_free_pages(iommu->prq);
+	iommu->prq = NULL;
+
+	return ret;
+}
+
+int intel_iommu_finish_prq(struct intel_iommu *iommu)
+{
+	writeq(0ULL, iommu->reg + DMAR_PQH_REG);
+	writeq(0ULL, iommu->reg + DMAR_PQT_REG);
+	writeq(0ULL, iommu->reg + DMAR_PQA_REG);
+
+	if (iommu->pr_irq) {
+		free_irq(iommu->pr_irq, iommu);
+		dmar_free_hwirq(iommu->pr_irq);
+		iommu->pr_irq = 0;
+	}
+
+	if (iommu->iopf_queue) {
+		iopf_queue_free(iommu->iopf_queue);
+		iommu->iopf_queue = NULL;
+	}
+
+	iommu_free_pages(iommu->prq);
+	iommu->prq = NULL;
+
+	return 0;
+}
+
+void intel_iommu_page_response(struct device *dev, struct iopf_fault *evt,
+			       struct iommu_page_response *msg)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	u8 bus = info->bus, devfn = info->devfn;
+	struct iommu_fault_page_request *prm;
+	struct qi_desc desc;
+	bool pasid_present;
+	u16 sid;
+
+	prm = &evt->fault.prm;
+	sid = PCI_DEVID(bus, devfn);
+	pasid_present = prm->flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID;
+
+	desc.qw0 = QI_PGRP_PASID(prm->pasid) | QI_PGRP_DID(sid) |
+			QI_PGRP_PASID_P(pasid_present) |
+			QI_PGRP_RESP_CODE(msg->code) |
+			QI_PGRP_RESP_TYPE;
+	desc.qw1 = QI_PGRP_IDX(prm->grpid);
+	desc.qw2 = 0;
+	desc.qw3 = 0;
+
+	qi_submit_sync(iommu, &desc, 1, 0);
+}

linux-7.2.3/drivers/iommu/intel/svm.c

@@ -0,0 +1,231 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright © 2015 Intel Corporation.
+ *
+ * Authors: David Woodhouse <dwmw2@infradead.org>
+ */
+
+#include <linux/mmu_notifier.h>
+#include <linux/sched.h>
+#include <linux/sched/mm.h>
+#include <linux/slab.h>
+#include <linux/rculist.h>
+#include <linux/pci.h>
+#include <linux/pci-ats.h>
+#include <linux/dmar.h>
+#include <linux/interrupt.h>
+#include <linux/mm_types.h>
+#include <linux/xarray.h>
+#include <asm/page.h>
+#include <asm/fpu/api.h>
+
+#include "iommu.h"
+#include "pasid.h"
+#include "perf.h"
+#include "../iommu-pages.h"
+#include "trace.h"
+
+void intel_svm_check(struct intel_iommu *iommu)
+{
+	if (!pasid_supported(iommu) || !ecap_smpwc(iommu->ecap))
+		return;
+
+	if (cpu_feature_enabled(X86_FEATURE_GBPAGES) &&
+	    !cap_fl1gp_support(iommu->cap)) {
+		pr_err("%s SVM disabled, incompatible 1GB page capability\n",
+		       iommu->name);
+		return;
+	}
+
+	if (cpu_feature_enabled(X86_FEATURE_LA57) &&
+	    !cap_fl5lp_support(iommu->cap)) {
+		pr_err("%s SVM disabled, incompatible paging mode\n",
+		       iommu->name);
+		return;
+	}
+
+	iommu->flags |= VTD_FLAG_SVM_CAPABLE;
+}
+
+/* Pages have been freed at this point */
+static void intel_arch_invalidate_secondary_tlbs(struct mmu_notifier *mn,
+					struct mm_struct *mm,
+					unsigned long start, unsigned long end)
+{
+	struct dmar_domain *domain = container_of(mn, struct dmar_domain, notifier);
+
+	if (start == 0 && end == ULONG_MAX) {
+		cache_tag_flush_all(domain);
+		return;
+	}
+
+	/*
+	 * The mm_types defines vm_end as the first byte after the end address,
+	 * different from IOMMU subsystem using the last address of an address
+	 * range.
+	 */
+	cache_tag_flush_range(domain, start, end - 1, 0);
+}
+
+static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
+{
+	struct dmar_domain *domain = container_of(mn, struct dmar_domain, notifier);
+	struct dev_pasid_info *dev_pasid;
+	struct device_domain_info *info;
+	unsigned long flags;
+
+	/* This might end up being called from exit_mmap(), *before* the page
+	 * tables are cleared. And __mmu_notifier_release() will delete us from
+	 * the list of notifiers so that our invalidate_range() callback doesn't
+	 * get called when the page tables are cleared. So we need to protect
+	 * against hardware accessing those page tables.
+	 *
+	 * We do it by clearing the entry in the PASID table and then flushing
+	 * the IOTLB and the PASID table caches. This might upset hardware;
+	 * perhaps we'll want to point the PASID to a dummy PGD (like the zero
+	 * page) so that we end up taking a fault that the hardware really
+	 * *has* to handle gracefully without affecting other processes.
+	 */
+	spin_lock_irqsave(&domain->lock, flags);
+	list_for_each_entry(dev_pasid, &domain->dev_pasids, link_domain) {
+		info = dev_iommu_priv_get(dev_pasid->dev);
+		intel_pasid_tear_down_entry(info->iommu, dev_pasid->dev,
+					    dev_pasid->pasid, true);
+	}
+	spin_unlock_irqrestore(&domain->lock, flags);
+
+}
+
+static void intel_mm_free_notifier(struct mmu_notifier *mn)
+{
+	struct dmar_domain *domain = container_of(mn, struct dmar_domain, notifier);
+
+	kfree(domain->qi_batch);
+	kfree(domain);
+}
+
+static const struct mmu_notifier_ops intel_mmuops = {
+	.release = intel_mm_release,
+	.arch_invalidate_secondary_tlbs = intel_arch_invalidate_secondary_tlbs,
+	.free_notifier = intel_mm_free_notifier,
+};
+
+static int intel_iommu_sva_supported(struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu;
+
+	if (!info || dmar_disabled)
+		return -EINVAL;
+
+	iommu = info->iommu;
+	if (!iommu)
+		return -EINVAL;
+
+	if (!(iommu->flags & VTD_FLAG_SVM_CAPABLE))
+		return -ENODEV;
+
+	if (!info->pasid_enabled || !info->ats_enabled)
+		return -EINVAL;
+
+	/*
+	 * Devices having device-specific I/O fault handling should not
+	 * support PCI/PRI. The IOMMU side has no means to check the
+	 * capability of device-specific IOPF.  Therefore, IOMMU can only
+	 * default that if the device driver enables SVA on a non-PRI
+	 * device, it will handle IOPF in its own way.
+	 */
+	if (!info->pri_supported)
+		return 0;
+
+	/* Devices supporting PRI should have it enabled. */
+	if (!info->pri_enabled)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int intel_svm_set_dev_pasid(struct iommu_domain *domain,
+				   struct device *dev, ioasid_t pasid,
+				   struct iommu_domain *old)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+	struct mm_struct *mm = domain->mm;
+	struct dev_pasid_info *dev_pasid;
+	unsigned long sflags;
+	int ret = 0;
+
+	ret = intel_iommu_sva_supported(dev);
+	if (ret)
+		return ret;
+
+	dev_pasid = domain_add_dev_pasid(domain, dev, pasid);
+	if (IS_ERR(dev_pasid))
+		return PTR_ERR(dev_pasid);
+
+	ret = iopf_for_domain_replace(domain, old, dev);
+	if (ret)
+		goto out_remove_dev_pasid;
+
+	/* Setup the pasid table: */
+	sflags = cpu_feature_enabled(X86_FEATURE_LA57) ? PASID_FLAG_FL5LP : 0;
+	sflags |= PASID_FLAG_PWSNP;
+	ret = __domain_setup_first_level(iommu, dev, pasid,
+					 FLPT_DEFAULT_DID, __pa(mm->pgd),
+					 sflags, old);
+	if (ret)
+		goto out_unwind_iopf;
+
+	domain_remove_dev_pasid(old, dev, pasid);
+
+	return 0;
+out_unwind_iopf:
+	iopf_for_domain_replace(old, domain, dev);
+out_remove_dev_pasid:
+	domain_remove_dev_pasid(domain, dev, pasid);
+	return ret;
+}
+
+static void intel_svm_domain_free(struct iommu_domain *domain)
+{
+	struct dmar_domain *dmar_domain = to_dmar_domain(domain);
+
+	/* dmar_domain free is deferred to the mmu free_notifier callback. */
+	mmu_notifier_put(&dmar_domain->notifier);
+}
+
+static const struct iommu_domain_ops intel_svm_domain_ops = {
+	.set_dev_pasid		= intel_svm_set_dev_pasid,
+	.free			= intel_svm_domain_free
+};
+
+struct iommu_domain *intel_svm_domain_alloc(struct device *dev,
+					    struct mm_struct *mm)
+{
+	struct dmar_domain *domain;
+	int ret;
+
+	ret = intel_iommu_sva_supported(dev);
+	if (ret)
+		return ERR_PTR(ret);
+
+	domain = kzalloc_obj(*domain);
+	if (!domain)
+		return ERR_PTR(-ENOMEM);
+
+	domain->domain.ops = &intel_svm_domain_ops;
+	INIT_LIST_HEAD(&domain->dev_pasids);
+	INIT_LIST_HEAD(&domain->cache_tags);
+	spin_lock_init(&domain->cache_lock);
+	spin_lock_init(&domain->lock);
+
+	domain->notifier.ops = &intel_mmuops;
+	ret = mmu_notifier_register(&domain->notifier, mm);
+	if (ret) {
+		kfree(domain);
+		return ERR_PTR(ret);
+	}
+
+	return &domain->domain;
+}

linux-7.2.3/drivers/iommu/intel/trace.c

@@ -0,0 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Intel IOMMU trace support
+ *
+ * Copyright (C) 2019 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+
+#define CREATE_TRACE_POINTS
+#include "trace.h"

linux-7.2.3/drivers/iommu/intel/trace.h

@@ -0,0 +1,189 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Intel IOMMU trace support
+ *
+ * Copyright (C) 2019 Intel Corporation
+ *
+ * Author: Lu Baolu <baolu.lu@linux.intel.com>
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM intel_iommu
+
+#if !defined(_TRACE_INTEL_IOMMU_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_INTEL_IOMMU_H
+
+#include <linux/tracepoint.h>
+
+#include "iommu.h"
+
+#define MSG_MAX		256
+
+TRACE_EVENT(qi_submit,
+	TP_PROTO(struct intel_iommu *iommu, u64 qw0, u64 qw1, u64 qw2, u64 qw3),
+
+	TP_ARGS(iommu, qw0, qw1, qw2, qw3),
+
+	TP_STRUCT__entry(
+		__field(u64, qw0)
+		__field(u64, qw1)
+		__field(u64, qw2)
+		__field(u64, qw3)
+		__string(iommu, iommu->name)
+	),
+
+	TP_fast_assign(
+		__assign_str(iommu);
+		__entry->qw0 = qw0;
+		__entry->qw1 = qw1;
+		__entry->qw2 = qw2;
+		__entry->qw3 = qw3;
+	),
+
+	TP_printk("%s %s: 0x%llx 0x%llx 0x%llx 0x%llx",
+		  __print_symbolic(__entry->qw0 & 0xf,
+				   { QI_CC_TYPE,	"cc_inv" },
+				   { QI_IOTLB_TYPE,	"iotlb_inv" },
+				   { QI_DIOTLB_TYPE,	"dev_tlb_inv" },
+				   { QI_IEC_TYPE,	"iec_inv" },
+				   { QI_IWD_TYPE,	"inv_wait" },
+				   { QI_EIOTLB_TYPE,	"p_iotlb_inv" },
+				   { QI_PC_TYPE,	"pc_inv" },
+				   { QI_DEIOTLB_TYPE,	"p_dev_tlb_inv" },
+				   { QI_PGRP_RESP_TYPE,	"page_grp_resp" }),
+		__get_str(iommu),
+		__entry->qw0, __entry->qw1, __entry->qw2, __entry->qw3
+	)
+);
+
+TRACE_EVENT(prq_report,
+	TP_PROTO(struct intel_iommu *iommu, struct device *dev,
+		 u64 dw0, u64 dw1, u64 dw2, u64 dw3,
+		 unsigned long seq),
+
+	TP_ARGS(iommu, dev, dw0, dw1, dw2, dw3, seq),
+
+	TP_STRUCT__entry(
+		__field(u64, dw0)
+		__field(u64, dw1)
+		__field(u64, dw2)
+		__field(u64, dw3)
+		__field(unsigned long, seq)
+		__string(iommu, iommu->name)
+		__string(dev, dev_name(dev))
+		__dynamic_array(char, buff, MSG_MAX)
+	),
+
+	TP_fast_assign(
+		__entry->dw0 = dw0;
+		__entry->dw1 = dw1;
+		__entry->dw2 = dw2;
+		__entry->dw3 = dw3;
+		__entry->seq = seq;
+		__assign_str(iommu);
+		__assign_str(dev);
+	),
+
+	TP_printk("%s/%s seq# %ld: %s",
+		__get_str(iommu), __get_str(dev), __entry->seq,
+		decode_prq_descriptor(__get_str(buff), MSG_MAX, __entry->dw0,
+				      __entry->dw1, __entry->dw2, __entry->dw3)
+	)
+);
+
+DECLARE_EVENT_CLASS(cache_tag_log,
+	TP_PROTO(struct cache_tag *tag),
+	TP_ARGS(tag),
+	TP_STRUCT__entry(
+		__string(iommu, tag->iommu->name)
+		__string(dev, dev_name(tag->dev))
+		__field(u16, type)
+		__field(u16, domain_id)
+		__field(u32, pasid)
+		__field(u32, users)
+	),
+	TP_fast_assign(
+		__assign_str(iommu);
+		__assign_str(dev);
+		__entry->type = tag->type;
+		__entry->domain_id = tag->domain_id;
+		__entry->pasid = tag->pasid;
+		__entry->users = tag->users;
+	),
+	TP_printk("%s/%s type %s did %d pasid %d ref %d",
+		  __get_str(iommu), __get_str(dev),
+		  __print_symbolic(__entry->type,
+			{ CACHE_TAG_IOTLB,		"iotlb" },
+			{ CACHE_TAG_DEVTLB,		"devtlb" },
+			{ CACHE_TAG_NESTING_IOTLB,	"nesting_iotlb" },
+			{ CACHE_TAG_NESTING_DEVTLB,	"nesting_devtlb" }),
+		__entry->domain_id, __entry->pasid, __entry->users
+	)
+);
+
+DEFINE_EVENT(cache_tag_log, cache_tag_assign,
+	TP_PROTO(struct cache_tag *tag),
+	TP_ARGS(tag)
+);
+
+DEFINE_EVENT(cache_tag_log, cache_tag_unassign,
+	TP_PROTO(struct cache_tag *tag),
+	TP_ARGS(tag)
+);
+
+DECLARE_EVENT_CLASS(cache_tag_flush,
+	TP_PROTO(struct cache_tag *tag, unsigned long start, unsigned long end,
+		 unsigned long addr, unsigned long mask),
+	TP_ARGS(tag, start, end, addr, mask),
+	TP_STRUCT__entry(
+		__string(iommu, tag->iommu->name)
+		__string(dev, dev_name(tag->dev))
+		__field(u16, type)
+		__field(u16, domain_id)
+		__field(u32, pasid)
+		__field(unsigned long, start)
+		__field(unsigned long, end)
+		__field(unsigned long, addr)
+		__field(unsigned long, mask)
+	),
+	TP_fast_assign(
+		__assign_str(iommu);
+		__assign_str(dev);
+		__entry->type = tag->type;
+		__entry->domain_id = tag->domain_id;
+		__entry->pasid = tag->pasid;
+		__entry->start = start;
+		__entry->end = end;
+		__entry->addr = addr;
+		__entry->mask = mask;
+	),
+	TP_printk("%s %s[%d] type %s did %d [0x%lx-0x%lx] addr 0x%lx mask 0x%lx",
+		  __get_str(iommu), __get_str(dev), __entry->pasid,
+		  __print_symbolic(__entry->type,
+			{ CACHE_TAG_IOTLB,		"iotlb" },
+			{ CACHE_TAG_DEVTLB,		"devtlb" },
+			{ CACHE_TAG_NESTING_IOTLB,	"nesting_iotlb" },
+			{ CACHE_TAG_NESTING_DEVTLB,	"nesting_devtlb" }),
+		__entry->domain_id, __entry->start, __entry->end,
+		__entry->addr, __entry->mask
+	)
+);
+
+DEFINE_EVENT(cache_tag_flush, cache_tag_flush_range,
+	TP_PROTO(struct cache_tag *tag, unsigned long start, unsigned long end,
+		 unsigned long addr, unsigned long mask),
+	TP_ARGS(tag, start, end, addr, mask)
+);
+
+DEFINE_EVENT(cache_tag_flush, cache_tag_flush_range_np,
+	TP_PROTO(struct cache_tag *tag, unsigned long start, unsigned long end,
+		 unsigned long addr, unsigned long mask),
+	TP_ARGS(tag, start, end, addr, mask)
+);
+#endif /* _TRACE_INTEL_IOMMU_H */
+
+/* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH ../../drivers/iommu/intel/
+#define TRACE_INCLUDE_FILE trace
+#include <trace/define_trace.h>